RESILIENCE ENHANCEMENT-ORIENTED ENERGY STORAGE CONTROL METHOD AND SYSTEM FOR DISTRIBUTED GENERATOR (DG) IN DISTRIBUTION NETWORK

Abstract

A resilience enhancement-oriented energy storage control method and system for a distributed generator (DG) in a distribution network. The method includes: determining an objective function including a first sub-objective function and a second sub-objective function; determining an outer-level constraint based on a quantity of configured energy storage systems (ESSs), a quantity of hardened lines, and a rated power and capacity configuration of an ESS; determining a middle-level constraint based on a line failure; determining an inner-level constraint based on node power of a distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS; determining parameters based on the outer-level constraint, the first sub-objective function, the middle-level constraint, the inner-level constraint, and the second sub-objective function; and controlling energy storage of a DG in the distribution network.

Description

TECHNICAL FIELD

The present disclosure relates to the technical field of responding to extreme events of an alternating current (AC)/direct current (DC) distribution network, and more specifically, to a resilience enhancement-oriented energy storage control method and system for a distributed generator (DG) in a distribution network.

BACKGROUND

As an important infrastructure in modem society, a power system supports economic, commercial and social development and plays a vital role in normal operation of society and economy. Any large-scale power outage and paralysis in the power system cause a huge impact to society. With the rapid progress of society and the rapid development of a power grid, a higher requirement is imposed on reliability of electric power supply of the power system. Traditional reliability indexes mainly focus on a possible high-probability failure of the power grid. However, many global accidents in recent years show that it is far from enough to focus only on predictable and small-scale accidents under normal operating conditions. An extreme event often damages the power grid. Many components fail due to a disaster, and a large-scale power outage occurs in the power grid, which seriously affects post-disaster recovery and reconstruction of a facility affected by the disaster.

Two measures are usually taken to enhance resilience of a distribution network: a preventive measure before the disaster, and a recovery measure during and after the disaster. The preventive measure is to determine severity of an impact of a large disturbance in the system on the distribution network in advance based on relevant forecast information, quickly switch an operation mode of the distribution network, make the distribution network in an optimal operation state, narrow an outage range, and support power consumption of key loads. The recovery measure is a regulation mode in which the distribution network takes an active measure to prevent outages of the key loads and quickly recovers to an expected state of the system under normal conditions during or after the large disturbance in the system.

SUMMARY

The present disclosure provides a resilience enhancement-oriented energy storage control method and system for a DG in a distribution network, so as to control energy storage of a DG in a distribution network.

To resolve the above problem, an aspect of the present disclosure provides a resilience enhancement-oriented energy storage control method for a DG in a distribution network. The method includes:

• • determining an objective function, where the objective function includes a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize a load shedding (LS) cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level;
  • determining an outer-level constraint based on a quantity of configured energy storage systems (ESSs), a quantity of hardened lines, and a rated power and capacity configuration of an ESS;
  • determining a middle-level constraint based on a line failure;
  • determining an inner-level constraint based on node power of a distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS;
  • determining, based on the outer-level constraint and the first sub-objective function, a unit hardening investment cost c_ij^l of a line, a line length s_ij, a binary decision variable y_ij for line hardening, a first cost coefficient c^e, a second cost coefficient c^p, and a capacity E_i^R and rated power P_i^R of an ESS configured on node i; and determining a power output c_g^G of a DG, a unit cost coefficient c_i^ch of charging of the ESS, a unit cost coefficient c_i^dis of discharging of the ESS, a unit LS cost c_i^LS after a load weight is considered, an LS ratio ρ_i,t of node i, and load p_i,t^D of node i based on the middle-level constraint, the inner-level constraint, and the second sub-objective function; and
  • controlling energy storage of the DG in the distribution network based on the obtained parameters.

Preferably, the outer-level constraint includes: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS;

• • the sub-constraint on the quantity of configured ESSs includes:

$\sum _{i\in {\Omega }_B}{\sigma }_i\le N_{\mathrm{ES}}^{{ }_{}\mathrm{INV}}$

• • the sub-constraint on the quantity of hardened lines includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}{y}_{\mathrm{ij}}\le N_L^{{ }_{}\mathrm{INV}};$

and

• • the sub-constraint on the rated power and capacity configuration of the ESS includes:

$0\le P_i^{{ }_{}R}\le {\sigma }_i{P}_i^{R,\max },\forall i\in {\Omega }_B$ $0\le E_i^{{ }_{}R}\le {\sigma }_i{E}_i^{R,\max },\forall i\in {\Omega }_B$

• • where σ_i represents a binary decision variable for ESS configuration, where when a value of the variable is 1, node i is configured with the ESS, otherwise, the node is not configured with the ESS; N_ES^INV represents a maximum allowable quantity of configured ESSs; N_L^INV represents a maximum quantity of hardened lines; P_i^R,max and E_i^R,max respectively represent upper limits of configurable rated power and a configurable capacity of the ESS; when σ_i=0, node i is not configured with the ESS, which is equivalent to setting both the rated power and the capacity of the ESS to 0; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; Ω_B represents a node set of an AC/DC distribution network; Ω_L represents a line set of the AC/DC distribution network; and y_ij represents the binary decision variable for line hardening.

Preferably, the middle-level constraint includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}\left(-{\log }_2{p}_{\mathrm{ij}}^{\mathrm{dam}}\right){𝓏}_{\mathrm{ij},0}\le W_{\max }$

• • where p_ij^dam represents a line failure rate; _ij,0 represents a binary variable, indicating whether a failure occurs on the line, and a value of the binary variable is 1 when a failure occurs and 0 when no failure occurs; W_max represents an upper limit of an uncertain entropy value of the line failure in the distribution network; and Ω_L represents a line set of an AC/DC distribution network.

Preferably, the inner-level constraint includes: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint;

• • the state sub-constraint of the ESS includes:

$S_i\left(t\right)=S_i\left(t-1\right)+p_i^{{ }_{}\mathrm{ch}}\left(t\right){\eta }_{\mathrm{ch}}\Delta t-\frac{p_i^{\mathrm{dis}}\left(t\right)\Delta t}{{\eta }_{\mathrm{dis}}},\forall i\in {\Omega }_B,\forall t\in T$ ${\mathrm{SOC}}_{i,\min }\le \frac{S_i\left(t\right)}{E_i^{{ }_{}R}}\le {\mathrm{SOC}}_{i,\max },\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{{ }_{}\mathrm{ch}}\left(t\right)\le P_i^{{ }_{}R},\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{{ }_{}\mathrm{dis}}\left(t\right)\le P_i^{{ }_{}R},\forall i\in {\Omega }_B,\forall t\in T$ $p_i^{{ }_{}\mathrm{ch}}\left(t\right)·p_i^{{ }_{}\mathrm{dis}}\left(t\right)=0,\forall i\in {\Omega }_B,\forall t\in T;$

and

• • the LS sub-constraint includes:

$0\le {\rho }_{i,t}\le 1,\forall i\in {\Omega }_B,\forall t\in T$

• • where S_i(t) represents a remaining electricity quantity at time t; η_ch, η_dis represent charging efficiency and discharging efficiency respectively; Δt represents a time interval; p_i^ch(t) represents charging power; p_i^dis(t) represents discharging power; and P_i^R represents an upper charging/discharging limit.

Preferably, the first sub-objective function includes:

$\underset{y\in Y}{\min }{C}^{{ }_{}\mathrm{INV}}$ $C^{{ }_{}\mathrm{INV}}=C_L^{{ }_{}\mathrm{INV}}+C_{\mathrm{ES}}^{{ }_{}\mathrm{INV}}$ $C_L^{{ }_{}\mathrm{INV}}={\beta }_l\sum _{\left(i,j\right)\in {\Omega }_L}{c}_{\mathrm{ij}}^{{ }_{}l}{s}_{\mathrm{ij}}{y}_{\mathrm{ij}}$ $C_{\mathrm{ES}}^{{ }_{}\mathrm{INV}}={\beta }_{\mathrm{es}}{ }_{}\left(\sum _{i\in {\Omega }_B}{c}^{{ }_{}e}{E}_i^{{ }_{}R}+c^{{ }_{}p}{P}_i^{{ }_{}R}\right);$

and

• • the second sub-objective function includes:

${\rho }_{\mathrm{pre}}\tau \underset{u\in U}{\max }\underset{x\in F\left(y,u\right)}{\min }\left(C^{\mathrm{OPE}}+C^{\mathrm{LS}}\right)$ $C^{\mathrm{OPE}}=\sum _{t\in T}\left[\sum _{g\in {\Omega }_G}{c}_g^G{p}_{g,t}^G+\sum _{i\in {\Omega }_B}\left(c_i^{\mathrm{dis}}{p}_{i,t}^{\mathrm{dis}}+c_i^{\mathrm{ch}}{p}_{i,t}^{\mathrm{ch}}\right)\right]$ $C^{\mathrm{LS}}=\sum _{i\in {\Omega }_B,t\in T}{c}_i^{\mathrm{LS}}{\rho }_{i,t}{p}_{i,t}^D$

• • where Y, U, F represent variable sets of tri-level decision-making respectively; C^INV represents an investment cost; C^OPE and C^LS represent the post-disaster operation cost and the LS cost respectively; ρ_pre represents introduction of a resilience preference coefficient; C_L^INV represents an investment cost for line hardening; C_ES^INV represents an investment cost for energy storage resource configuration; c_ij^l represents the unit hardening investment cost of the line; s_ij represents the line length; y_ij represents the binary decision variable for line hardening; β_l represents a coefficient for converting a total investment cost of the line into a one-year cost based on a planned period; Ω_L represents a line set of an AC/DC distribution network; c^e and c^p represent cost coefficients; β_es represents a coefficient for converting a total investment cost of the ESS into a one-year cost based on a planned period; Ω_B represents a node set of the AC/DC distribution network; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; c_i^LS represents the unit LS cost after the load weight is considered; ρ_i,t represents the LS ratio of node i at time t; p_i,t^D represents the load of the node at the corresponding time; T represents duration of an extreme event; and c_g^G, C_i^ch, and c_i^dis represent the power output of the DG, the unit cost coefficient for charging of the ESS, and the unit cost coefficient for discharging of the ESS respectively.

Another aspect of the present disclosure provides a resilience enhancement-oriented energy storage control system for a DG in a distribution network. The system includes:

• • an objective function determining unit configured to determine an objective function, where the objective function includes a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize an LS cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level;
  • an outer-level constraint determining unit configured to determine an outer-level constraint based on a quantity of configured ESSs, a quantity of hardened lines, and a rated power and capacity configuration of the ESS;
  • a middle-level constraint determining unit configured to determine a middle-level constraint based on a line failure;
  • an inner-level constraint determining unit configured to determine an inner-level constraint based on node power of a distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS;
  • a parameter determining unit configured to determine, based on the outer-level constraint and the first sub-objective function, a unit hardening investment cost c_ij^l of a line, a line length s_ij, a binary decision variable y_ij for line hardening, a first cost coefficient c^e, a second cost coefficient c^p, and a capacity E_i^R and rated power P_i^R of an ESS configured on node i; and determine a power output c_g^G of a DG, a unit cost coefficient c_i^ch of charging of the ESS, a unit cost coefficient c_i^dis of discharging of the ESS, a unit LS cost c_i^LS after a load weight is considered, an LS ratio ρ_i,t of node i, and load p_i,t^D of node i based on the middle-level constraint, the inner-level constraint, and the second sub-objective function; and
  • an energy storage control unit configured to control energy storage of the DG in the distribution network based on the obtained parameters.

Preferably, the outer-level constraint includes: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS;

• • the sub-constraint on the quantity of configured ESSs includes:

$\sum _{i\in {\Omega }_B}{\sigma }_i\le N_{\mathrm{ES}}^{\mathrm{INV}}$

• • the sub-constraint on the quantity of hardened lines includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}{y}_{\mathrm{ij}}\le N_L^{\mathrm{INV}};$

and

• • the sub-constraint on the rated power and capacity configuration of the ESS includes:

$0\le P_i^R\le {\sigma }_i{P}_i^{R,\max },\forall i\in {\Omega }_B$ $0\le E_i^R\le {\sigma }_i{E}_i^{R,\max },\forall i\in {\Omega }_B$

• • where σ_i represents a binary decision variable for ESS configuration, where when a value of the variable is 1, node i is configured with the ESS, otherwise, the node is not configured with the ESS; N_ES^INV represents a maximum allowable quantity of configured ESSs; N_L^INV represents a maximum quantity of hardened lines; P_i^R,max and E_i^R,max respectively represent upper limits of configurable rated power and a configurable capacity of the ESS; when σ_i=0, node i is not configured with the ESS, which is equivalent to setting both the rated power and the capacity of the ESS to 0; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; Ω_B represents the node set of the AC/DC distribution network; Ω_L represents the line set of the AC/DC distribution network; and y_ij represents the binary decision variable for line hardening.

Preferably, the middle-level constraint includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}\left(-{\log }_2{p}_{\mathrm{ij}}^{\mathrm{dam}}\right){𝓏}_{\mathrm{ij},0}\le W_{\max }$

• • where p_ij^dam represents a line failure rate; _ij,0 represents a binary variable, indicating whether a failure occurs on the line, and a value of the binary variable is 1 when a failure occurs and 0 when no failure occurs; W_max represents an upper limit of an uncertain entropy value of the line failure in the distribution network; and Ω_L represents a line set of an AC/DC distribution network.

Preferably, the inner-level constraint includes: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint;

• • the state sub-constraint of the ESS includes:

$S_i\left(t\right)=S_i\left(t-1\right)+p_i^{\mathrm{ch}}\left(t\right){\eta }_{\mathrm{ch}}\Delta t-\frac{p_i^{\mathrm{dis}}\left(t\right)\Delta t}{{\eta }_{\mathrm{dis}}},\forall i\in {\Omega }_B,\forall t\in T$ $\mathrm{SO}{C}_{i,\min }\le \frac{S_i\left(t\right)}{E_i^R}\le \mathrm{SO}{C}_{i,\max },\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{\mathrm{ch}}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{\mathrm{dis}}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T$ $p_i^{\mathrm{ch}}\left(t\right)·p_i^{\mathrm{dis}}\left(t\right)=0,\forall i\in {\Omega }_B,\forall t\in T;$

and

• • the LS sub-constraint includes:

$0\le {\rho }_{i,t}\le 1,\forall i\in {\Omega }_B,\forall t\in T$

• • where S_i(t) represents a remaining electricity quantity at time t; η_ch, η_dis, represent charging efficiency and discharging efficiency respectively; Δt represents a time interval; p_i^ch(t) represents charging power; p_i^dis(t) represents discharging power; and P_i^R represents an upper charging/discharging limit.

Preferably, the first sub-objective function includes:

$\underset{y\in Y}{\min }{C}^{\mathrm{INV}}$ $C^{\mathrm{INV}}=C_L^{\mathrm{INV}}+C_{\mathrm{ES}}^{\mathrm{INV}}$ $C_L^{\mathrm{INV}}={\beta }_l\sum _{\left(i,j\right)\in {\Omega }_L}{c}_{\mathrm{ij}}^l{s}_{\mathrm{ij}}{y}_{\mathrm{ij}}$ $C_{\mathrm{ES}}^{\mathrm{INV}}={\beta }_{\mathrm{es}}\left(\sum _{i\in {\Omega }_B}{c}^e{E}_i^R+c^p{P}_i^R\right);$

and

• • the second sub-objective function includes:

${\rho }_{\mathrm{pre}}\tau \underset{u\in U}{\max }\underset{x\in F\left(y,u\right)}{\min }\left(C^{\mathrm{OPE}}+C^{\mathrm{LS}}\right)$ $C^{\mathrm{OPE}}=\sum _{t\in T}\left[\sum _{g\in {\Omega }_G}{c}_g^G{p}_{g,t}^G+\sum _{i\in {\Omega }_B}\left(c_i^{\mathrm{dis}}{p}_{i,t}^{\mathrm{dis}}+c_i^{\mathrm{ch}}{p}_{i,t}^{\mathrm{ch}}\right)\right]$ $C^{\mathrm{LS}}=\sum _{i\in {\Omega }_B,t\in T}{c}_i^{\mathrm{LS}}{\rho }_{i,t}{p}_{i,t}^D$

• • where Y, U, F represent variable sets of tri-level decision-making respectively; C_INV represents an investment cost; C^OPE and C^LS represent the post-disaster operation cost and the LS cost respectively; ρ_pre represents introduction of a resilience preference coefficient; C_L^INV represents an investment cost for line hardening; C_ES^INV represents an investment cost for energy storage resource configuration; c_ij^l represents the unit hardening investment cost of the line; s_ij represents the line length; y_ij represents the binary decision variable for line hardening; β_l represents a coefficient for converting a total investment cost of the line into a one-year cost based on a planned period; Ω_L represents a line set of an AC/DC distribution network; c^e and c^p represent cost coefficients; β_es represents a coefficient for converting a total investment cost of the ESS into a one-year cost based on a planned period; Ω_B represents a node set of the AC/DC distribution network; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; c_i^LS represents the unit LS cost after the load weight is considered; ρ_i,t represents the LS ratio of node i at time t; p_i,t^D represents the load of the node at the corresponding time; T represents duration of an extreme event; and c_g^G, c_i^ch, and c_i^dis represent the power output of the DG, the unit cost coefficient for charging of the ESS, and the unit cost coefficient for discharging of the ESS respectively.

The present disclosure provides a resilience enhancement-oriented energy storage control method and system for a DG in a distribution network, to study a line hardening strategy and an ESS configuration strategy in a planning period based on a pre-disaster defense stage, construct tri-level constraints and corresponding objective functions, and control energy storage of a DG in a distributed network based on solved parameters. Based on an impact of a predicted failure caused by an extreme event to the distribution network on safe operation of the system, the present disclosure establishes an energy storage planning model that is of the DG in the distribution network and considers resilience enhancement. The method in the present disclosure can weigh a relationship between a resilience enhancement cost of the distribution network and an improvement of an ability to resist the extreme event.

BRIEF DESCRIPTION OF THE DRAWINGS

Exemplary implementations of the present disclosure can be more completely understood with reference to the following accompanying drawings.

FIG. 1 is a flowchart of a resilience enhancement-oriented energy storage control method 100 for a DG in a distribution network according to an implementation of the present disclosure;

FIG. 2 is a schematic diagram of an optimization model according to an implementation of the present disclosure;

FIG. 3 is an example topological graph of a distribution network according to an implementation of the present disclosure;

FIG. 4 is an example load curve according to an implementation of the present disclosure;

FIG. 5 is an example power output curve of a photovoltaic unit according to an implementation of the present disclosure; and

FIG. 6 is a schematic structural diagram of a resilience enhancement-oriented energy storage control system 600 for a DG in a distribution network according to an implementation of the present disclosure.

DETAILED DESCRIPTION

Exemplary implementations of the present disclosure are described below with reference to the accompanying drawings. However, the present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. The embodiments are provided to thoroughly and completely understand the present disclosure, and fully convey the scope of the present disclosure to a person skilled in the art. The terms in the exemplary implementations in the accompanying drawings are not a limitation of the present disclosure. In the accompanying drawings, the same units/elements are designated by same reference numerals.

Unless otherwise specified, the terms (including technical and scientific terms) used herein have the same meanings as those commonly understood by those skilled in the art. In addition, it can be understood that the terms defined in commonly used dictionaries will be interpreted as having the same meanings as the contextual meanings in the related technical field, and will not be interpreted as having idealized or overly formal meanings.

FIG. 1 is a flowchart of a resilience enhancement-oriented energy storage control method 100 for a DG in a distribution network according to an implementation of the present disclosure. As shown in FIG. 1, this implementation of the present disclosure provides a resilience enhancement-oriented energy storage control method for a DG in a distribution network, to study an impact of a predicted failure caused by an extreme event to a distribution network on safe operation of a system, and establish an energy storage planning model that is of a DG in the distribution network and considers resilience enhancement, so as to weigh a relationship between a resilience enhancement cost of the distribution network and an improvement of an ability to resist the extreme event. The resilience enhancement-oriented energy storage control method 100 for a DG in a distribution network provided in this implementation of the present disclosure starts from step 101. In the step 101, an objective function is determined, where the objective function includes a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize an LS cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level.

Preferably, the first sub-objective function includes:

$\underset{y\in Y}{\min }{C}^{\mathrm{INV}}$ $C^{\mathrm{INV}}=C_L^{\mathrm{INV}}+C_{\mathrm{ES}}^{\mathrm{INV}}$ $C_L^{\mathrm{INV}}={\beta }_l\sum _{\left(i,j\right)\in {\Omega }_L}{c}_{\mathrm{ij}}^l{s}_{\mathrm{ij}}{y}_{\mathrm{ij}}$ $C_{\mathrm{ES}}^{\mathrm{INV}}={\beta }_{\mathrm{es}}\left(\sum _{i\in {\Omega }_B}{c}^e{E}_i^R+c^p{P}_i^R\right);$

and

• • the second sub-objective function includes:

${\rho }_{\mathrm{pre}}\tau \underset{u\in U}{\max }\underset{x\in F\left(y,u\right)}{\min }\left(C^{\mathrm{OPE}}+C^{\mathrm{LS}}\right)$ $C^{\mathrm{OPE}}=\sum _{t\in T}\left[\sum _{g\in {\Omega }_G}{c}_g^G{p}_{g,t}^G+\sum _{i\in {\Omega }_B}\left(c_i^{\mathrm{dis}}{p}_{i,t}^{\mathrm{dis}}+c_i^{\mathrm{ch}}{p}_{i,t}^{\mathrm{ch}}\right)\right]$ $C^{\mathrm{LS}}=\sum _{i\in {\Omega }_B,t\in T}{c}_i^{\mathrm{LS}}{\rho }_{i,t}{p}_{i,t}^D$

• • where Y, U, F represent variable sets of tri-level decision-making respectively; C^INV represents an investment cost; C^OPE and C^LS represent the post-disaster operation cost and the LS cost respectively; ρ_pre represents introduction of a resilience preference coefficient; C^INV represents an investment cost for line hardening; C_INV represents an investment cost for energy storage resource configuration; c_ij^l represents a unit hardening investment cost of a line; s_ij represents a line length; y_ij represents a binary decision variable for line hardening; β_l represents a coefficient for converting a total investment cost of the line into a one-year cost based on a planned period; Ω_L represents a line set of an AC/DC distribution network; c^e and c^p represent cost coefficients; β_es represents a coefficient for converting a total investment cost of an ESS into a one-year cost based on a planned period; Ω_B represents a node set of the AC/DC distribution network; E_i^R and P_i^R respectively represent a capacity and rated power of an ESS configured on node i; c_i^LS represents a unit LS cost after a load weight is considered; ρ_i,t represents an LS ratio of node i at time t; p_i,t^D represents load of the node at the corresponding time; T represents duration of an extreme event; and c_g^G, c_i^ch, and c_i^dis represent a power output of a DG, a unit cost coefficient for charging of the ESS, and a unit cost coefficient for discharging of the ESS respectively.

An operation state is the best test and verification of planning and configuration schemes. Therefore, this patent gives consideration to a response state of the distribution network for the extreme event when determining a pre-disaster resilience enhancement strategy, and constructs a two-stage optimization model. In a second stage, an uncertain failure set is generated. Stochastic optimization and robust optimization are most widely used to process failure uncertainty. For the following two reasons, this patent adopts robust optimization to study the operation state of the distribution network under a most serious failure set:

• • 1) In a planning stage, various complex situations of actual operation shall be fully considered, and accordingly, planning behaviors shall ensure the stable supply of an important load in the face of all extreme events.
  • 2) It is difficult to apply stochastic optimization to two-stage problems. A random variable in the second stage will be affected by a decision in a first stage, and its probability distribution is also uncertain. Moreover, this uncertainty is fused in an optimization problem, making it difficult to establish a model independently. Therefore, feasibility of stochastic optimization in practical operations is low.

Therefore, the present disclosure proposes a two-stage robust optimization model, as shown in FIG. 2. Therefore, a total objective function of the two-stage robust optimization model can be expressed as:

$\begin{array}{cc}\underset{y\in Y}{\min }{C}^{\mathrm{INV}}+{\rho }_{\mathrm{pre}}\tau \underset{u\in U}{\max }\underset{x\in F\left(y,u\right)}{\min }\left(C^{\mathrm{OPE}}+C^{\mathrm{LS}}\right)& \left(1\right)\end{array}$

The formula (1) is essentially a tri-level defender-attacker-defender model. In the tri-level defender-attacker-defender model, the outermost level reflects an active behavior of the distribution network and determines line hardening and ESS configuration strategies; the middle level reflects a passive behavior and determines the most serious failure set after the extreme event; and the inner level reflects the active behavior of the distribution network, and determines failure response and dispatching operation strategies of the distribution network. Y, U, F represent the variable sets of tri-level decision-making respectively. C^INV represents the investment cost, and C^OPE and C^LS represent the post-disaster operation cost and the LS cost respectively. Considering that an investment is not only used to enhance resilience, the resilience preference coefficient ρ_pre is introduced to reflect a resilience tendency in a planning period. τ represents a quantity of extreme events in the planning period.

The investment cost includes a line hardening cost and an ESS configuration cost, and its expression is shown in the following formula (2). The line hardening strategy includes overhead line cabling, tower upgrading, line expansion, line material selection upgrading, and the like. The present disclosure selects a most direct hardening mode, namely, line material selection upgrading, to conveniently analyze a change of a line failure rate before and after hardening. A most widely used lithium battery ESS is selected as the ESS.

$\begin{array}{cc}{C}^{\mathrm{INV}}=C_L^{\mathrm{INV}}+C_{\mathrm{ES}}^{\mathrm{INV}}& \left(2\right)\end{array}$

In the above formula, C_L^INV represents the investment cost for line hardening, and C_ES^INV represents the investment cost for energy storage resource configuration. An investment period of line hardening is usually provided by a manufacturer. A service life of the ESS is related to its charging and discharging depths, and an operation state of the ESS needs to be constrained based on a specified service life expectation. As the planned periods of the line and the ESS are different, investment costs of the line and the ESS need to be converted into one-year costs. Accordingly, τ in the formula (1) also represents an annual average quantity of extreme events.

$\begin{array}{cc}{C}_L^{\mathrm{INV}}={\beta }_l\sum _{\left(i,j\right)\in {\Omega }_L}{c}_{\mathrm{ij}}^l{s}_{\mathrm{ij}}{y}_{\mathrm{ij}}& \left(3\right)\end{array}$ $\begin{array}{cc}{C}_{\mathrm{ES}}^{\mathrm{INV}}={\beta }_{\mathrm{es}}\left(\sum _{i\in {\Omega }_B}{c}^e{E}_i^R+c^p{P}_i^R\right)& \left(4\right)\end{array}$

In the above formulas, c_ij^l represents the unit hardening investment cost of the line; s_ij represents the line length; and y_ij represents the binary decision variable for line hardening, and line hardening is performed when the variable is set to 1, and no hardening is performed when the variable is set to 0. β_l represents the coefficient for converting the total investment cost of the line into the one-year cost based on the planned period. Ω_L represents the line set of the AC/DC distribution network. It should be noted that a tie line of a voltage source converter (VSC) is not considered for line hardening. The ESS configuration cost can be obtained according to the formula (4), and is linear with a total capacity and rated power of the ESS. c^e and c^p represent the cost coefficients, and β_es represent the coefficient for converting the total investment cost of the ESS into the one-year cost based on the planned period. Ω_B represents the node set of the AC/DC distribution network, and E_i^R and P_i^R represents the capacity and the rated power of the ESS configured on node i.

The post-disaster operation cost C^OPE includes a power output cost of the DG and an operation cost of the ESS, and can be expressed as follows:

$\begin{array}{cc}{C}^{\mathrm{OPE}}=\sum _{t\in T}\left[\sum _{i\in {\Omega }_G}{c}_i^G{p}_{i,t}^G+\sum _{i\in {\Omega }_B}\left(c_i^{\mathrm{dis}}{p}_{i,t}^{\mathrm{dis}}+c_i^{\mathrm{ch}}{p}_{i,t}^{\mathrm{ch}}\right)\right]& \left(5\right)\end{array}$

In the above formula, T represents the duration of the extreme event, and c_g^G, C_i^ch, and c_i^dis represent the power output of the DG, the unit cost coefficient for charging of the ESS, and the unit cost coefficient for discharging of the ESS respectively.

The LS cost C^LS is expressed as the following formula (6):

$\begin{array}{cc}{C}^{{ }_{}\mathrm{LS}}=\sum _{i\in {\Omega }_B,t\in T}{c}_i^{{ }_{}\mathrm{LS}}{\rho }_{i,t}{p}_{i,t}^{{ }_{}D}& \left(6\right)\end{array}$

In the above formula, c_i^LS represents the unit LS cost after the load weight is considered; ρ_i,t represents the LS ratio of node i at time t; and p_i,t^D represents the load of the node at the corresponding time.

In step 102, an outer-level constraint is determined based on a quantity of configured ESSs, a quantity of hardened lines, and a rated power and capacity configuration of the ESS.

Preferably, the outer-level constraint includes: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS;

• • the sub-constraint on the quantity of configured ESSs includes:

$\sum _{i\in {\Omega }_B}{\sigma }_i\le N_{\mathrm{ES}}^{{ }_{}\mathrm{INV}}$

• • the sub-constraint on the quantity of hardened lines includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}{y}_{\mathrm{ij}}\le N_L^{{ }_{}\mathrm{INV}};$

and

• • the sub-constraint on the rated power and capacity configuration of the ESS includes:

$0\le P_i^{{ }_{}R}\le {\sigma }_i{P}_i^{{ }_{}R,\max },\forall i\in {\Omega }_B$ $0\le E_i^{{ }_{}R}\le {\sigma }_i{E}_i^{{ }_{}R,\max },\forall i\in {\Omega }_B$

• • where σ_i represents a binary decision variable for ESS configuration, where when a value of the variable is 1, node i is configured with the ESS, otherwise, the node is not configured with the ESS; N_ES^INV represents a maximum allowable quantity of configured ESSs; N_L^NV represents a maximum quantity of hardened lines; P_i^R,max and E_i^R,max respectively represent upper limits of configurable rated power and a configurable capacity of the ESS; when σ_i=0, node i is not configured with the ESS, which is equivalent to setting both the rated power and the capacity of the ESS to 0; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; Ω_B represents the node set of the AC/DC distribution network; Ω_L represents the line set of the AC/DC distribution network; and y_ij represents the binary decision variable for line hardening.

In the present disclosure, an outer-level decision variable is Y={y_ij, σ_i, E_i^R, P_i^R}. Considering that the investment cost is an optimization item of the objective function, the investment cost is not directly limited in the constraints, but the quantity of hardened lines and the quantity of configured ESSs are limited to enhance investment feasibility.

(1) Constraint on the Quantity of Configured ESSs

$\begin{array}{cc}\sum _{i\in {\Omega }_B}{\sigma }_i\le N_{\mathrm{ES}}^{{ }_{}\mathrm{INV}}& \left(7\right)\end{array}$

In the above formula, σ_i represents the binary decision variable for ESS configuration. When a value of the variable is 1, node i is configured with the ESS, otherwise, the node is not configured with the ESS. N_ES^INV represents the maximum allowable quantity of configured ESSs.

(2) Constraint on the Quantity of Hardened Lines

$\begin{array}{cc}\sum _{\left(i,j\right)\in {\Omega }_L}{y}_{\mathrm{ij}}\le N_L^{{ }_{}\mathrm{INV}}& \left(8\right)\end{array}$

In the above formula, N_L^INV represents the maximum quantity of hardened lines.

(3) Constraint on the Rated Power and Capacity Configuration of the ESS

$\begin{array}{cc}0\le P_i^{{ }_{}R}\le {\sigma }_i{P}_i^{{ }_{}R,\max },\forall i\in {\Omega }_B& \left(9\right)\end{array}$ $0\le E_i^{{ }_{}R}\le {\sigma }_i{E}_i^{{ }_{}R,\max },\forall i\in {\Omega }_B$

In the above formula, P_i^R,max and E_i^R,max respectively represent the upper limits of the configurable rated power and the configurable capacity of the ESS. When σ_i=0, node i is not configured with the ESS, which is equivalent to setting both the rated power and the capacity of the ESS to 0.

In step 103, a middle-level constraint is determined based on a line failure.

Preferably, the middle-level constraint includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}\left(-{\log }_2{p}_{\mathrm{ij}}^{{ }_{}\mathrm{dam}}\right){𝓏}_{\mathrm{ij},0}\le W_{\max }$

• • where p_ij^dam represents the line failure rate; _ij,0 represents a binary variable, indicating whether a failure occurs on the line, and a value of the binary variable is 1 when a failure occurs and 0 when no failure occurs; W_max represents an upper limit of an uncertain entropy value of the line failure in the distribution network; and Ω_L represents the line set of the AC/DC distribution network.

In the present disclosure, the most serious failure set is determined at the middle level, to limit the uncertain entropy value of the line failure in the distribution network by referring to the Shannon information theory:

$\begin{array}{cc}\sum _{\left(i,j\right)\in {\Omega }_L}\left(-{\log }_2{p}_{\mathrm{ij}}^{{ }_{}\mathrm{dam}}\right){𝓏}_{\mathrm{ij},0}\le W_{\max }& \left(10\right)\end{array}$

In the above formula, p_ij^dam represents the line failure rate; and z_ij,0 represents the binary variable, indicating whether the failure occurs on the line, and the value of the binary variable is 1 when the failure occurs and 0 when no failure occurs. W_max represents the upper limit of the uncertain entropy value of the line failure in the distribution network. Obviously, if the line failure rate is 0, log₂ p_ij^dam tends to be positive infinity. Due to the limitation of W_max, _ij,0 is bound to be 0. If the line failure rate is 1, log₂ p_ij^dam=0, which means that the line failure does not occupy any uncertain share. Therefore, due to a limitation of a robust layer, _ij,0 is bound to be 1. With an increase in the line failure rate, a share of the uncertain entropy value occupied by the line failure is smaller, and a failure probability is larger. Compared with a limitation on a quantity of broken lines caused by the failure, the formula (10) is very suitable for a failure scenario in which a failure rate of each line is different.

It should be noted that spatio-temporal evolution processes of the extreme event and a failure set are huge topics. Most of existing studies focus on a certain type of extreme event (such as typhoons), divide the distribution network into a plurality of sub-regions, and then map evolution stages of the extreme event to the sub-regions. Compared with a single-region model of a single failure stage, complexity and a simulation degree are improved, but essential methods are not obviously different. Moreover, the present disclosure studies a planning strategy, and it is difficult to predict a type and evolution of the extreme event in a planning period. Therefore, the present disclosure does not consider space-time evolution characteristics of the extreme event.

It is assumed that a failure occurs at time t₀, a faulty line is repaired at time t₀+T, and a superior power supply is resumed. A state of the line keeps unchanged during the failure. In this case, the following formula is obtained:

$\begin{array}{cc}{u}_{\mathrm{ij},t}={𝓏}_{\mathrm{ij},0},\forall \left(i,j\right)\in {\Omega }_L,\forall t\in T.& \left(11\right)\end{array}$

In the above formula, u_ij,t represents a faulty state of the line at time t. A value 1 indicates that the line is in the faulty state; otherwise, the line is not faulty. A middle-level decision variable is U={, u_ij,t}.

The line failure rate is associated with the hardening strategy in the first stage, which leads to mutual restriction of the decision variables in the two stages. Therefore, a stage correlation constraint needs to be supplemented.

Line hardening can significantly reduce the failure rate. However, the failure rate usually is not reduced to 0, and a failure rate of a line not hardened may not be 1 in the face of the extreme event. It is assumed that the failure rate of the line is p_ij⁰ before hardening and p_ij¹ after hardening. The following formula is obtained:

$\begin{array}{cc}{\log }_2{p}_{\mathrm{ij}}^{{ }_{}\mathrm{dam}}=\left({\log }_2\frac{p_{\mathrm{ij}}^{{ }_{}1}}{p_{\mathrm{ij}}^{{ }_{}0}}\right)y_{\mathrm{ij}}+{\log }_2{p}_{\mathrm{ij}}^{{ }_{}0}& \left(12\right)\end{array}$

In step 104, an inner-level constraint is determined based on node power of the distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS.

Preferably, the inner-level constraint includes: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint;

• • the state sub-constraint of the ESS includes:

$S_i\left(t\right)=S_i\left(t-1\right)+p_i^{{ }_{}\mathrm{ch}}\left(t\right){\eta }_{\mathrm{ch}}\Delta t-\frac{p_i^{{ }_{}\mathrm{dis}}\left(t\right)\Delta t}{{\eta }_{\mathrm{dis}}},\forall i\in {\Omega }_B,\forall t\in T$ ${\mathrm{SOC}}_{i,\min }\le \frac{S_i\left(t\right)}{E_i^{{ }_{}R}}\le {\mathrm{SOC}}_{i,\max },\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{{ }_{}\mathrm{ch}}\left(t\right)\le P_i^{{ }_{}R},\forall i\in {\Omega }_B,\forall t\in T$ $0\le p_i^{{ }_{}\mathrm{dis}}\left(t\right)\le P_i^{{ }_{}R},\forall i\in {\Omega }_B,\forall t\in T$ $p_i^{{ }_{}\mathrm{ch}}\left(t\right)·p_i^{{ }_{}\mathrm{dis}}\left(t\right)=0,\forall i\in {\Omega }_B,\forall t\in T;$

• • the LS sub-constraint includes:

$0\le {\rho }_{i,t}\le 1,\forall i\in {\Omega }_B,\forall t\in T$

• • where S_i(t) represents a remaining electricity quantity at time t; η_ch, η_dis represent charging efficiency and discharging efficiency respectively; Δt represents a time interval; p_i^ch(t) represents charging power; p_i^dis(t) represents discharging power; and P_i^R represents an upper charging/discharging limit.

In the present disclosure, a dispatching operation strategy of the AC distribution network disconnected from the main network and a disaster response strategy during the failure are determined at the inner level, and a corresponding decision variable is x∈{p_i,t^G, q_i,t^G, p_i,t^dis, p_i,t^ch, v_i,t, p_s,t, p_d,t, q_s,t, M_t, δ_t, ρ_i,t}. The following constraints are included:

(1) Sub-Constraint on Node Power Balance of the Distribution Network:

$\begin{array}{cc}{P}_{i,t}^{{ }_{}G}+p_{i,t}^{{ }_{}\mathrm{dis}}-p_{i,t}^{{ }_{}\mathrm{ch}}=\sum _{\left(i,j\right)\in {\Omega }_L^{\mathrm{AC}}}{p}_{\mathrm{ij},t}+\left(1-{\rho }_{i,t}\right)p_{i,t}^{{ }_{}D}+\left(\sum _{\left(i,j\right)\in {\Omega }_L^{\mathrm{AC}}}{G}_{\mathrm{ij}}\right)V_i^{{ }_{}2},& \left(13\right)\end{array}$ $\forall i\in {\Omega }_B^{{ }_{}\mathrm{AC}},\forall t\in T$ $\begin{array}{cc}{q}_{i,t}^{{ }_{}G}=\sum _{\left(i,j\right)\in {\Omega }_L^{\mathrm{AC}}}{q}_{\mathrm{ij},t}+\left(1-{\rho }_{i,t}\right)q_{i,t}^{{ }_{}D}-\left(\sum _{\left(i,j\right)\in {\Omega }_L^{\mathrm{AC}}}{B}_{\mathrm{ij}}\right)V_i^{{ }_{}2},\forall i\in {\Omega }_B^{{ }_{}\mathrm{AC}},\forall t\in T& \left(14\right)\end{array}$ $\begin{array}{cc}{P}_{i,t}^{{ }_{}G}+p_{i,t}^{{ }_{}\mathrm{dis}}-p_{i,t}^{{ }_{}\mathrm{ch}}=\sum _{\left(i,j\right)\in {\Omega }_L^{\mathrm{DC}}}{p}_{\mathrm{ij},t}+\left(1-{\rho }_{i,t}\right)p_{i,t}^{{ }_{}D},\forall i\in {\Omega }_B^{{ }_{}\mathrm{DC}},\forall t\in T& \left(15\right)\end{array}$

In the above formulas, Ω_B^AC represents the node set of the AC distribution network, and Ω_L^AC represents the line set of the AC distribution network. p_i^G and q_i^D respectively represent a sum of active power outputs of a power supply of node i connected to the AC distribution network and a sum of reactive power outputs of the power supply of node i connected to the AC distribution network. p_i^D and q_i^D respectively represent active load and reactive load connected to node i. G and B represent node admittance matrices of the AC distribution network. v_i represents a voltage amplitude of node i.

(2) Sub-Constraint on the Line Load Capacity:

$\begin{array}{cc}{p}_{\mathrm{ij},t}^{{ }_{}2}+q_{\mathrm{ij},t}^{{ }_{}2}\le \left(1-u_{\mathrm{ij},t}\right)·{\left(S_{\mathrm{ij}}^{{ }_{}\max }\right)}^2,\forall \left(i,j\right)\in {\Omega }_L^{{ }_{}\mathrm{AC}},\forall t\in T& \left(16\right)\end{array}$ $\begin{array}{cc}-P_{\mathrm{ij}}^{{ }_{}\max }\le p_{\mathrm{ij},t}\le P_{\mathrm{ij}}^{{ }_{}\max },\forall \left(i,j\right)\in {\Omega }_L^{{ }_{}\mathrm{DC}},\forall t\in T& \left(17\right)\end{array}$

In the above formulas, S_ij^max represents an upper limit of apparent power transmitted by the line.

(3) Sub-Constraint on the Line Power Flow:

$\begin{array}{cc}-M_1{u}_{\mathrm{ij},t}\le p_{\mathrm{ij},t}+G_{\mathrm{ij}}{v}_{i,t}^2-v_{i,t}{v}_{j,t}\left(G_{\mathrm{ij}}\cos {\theta }_{\mathrm{ij},t}+B_{\mathrm{ij}}\sin {\theta }_{\mathrm{ij},t}\right)\le M_1{u}_{\mathrm{ij},t},\forall \left(i,j\right)\in {\Omega }_L^{AC},\forall t\in T& \left(18\right)\end{array}$ $\begin{array}{cc}-M_2{u}_{\mathrm{ij},t}\le q_{\mathrm{ij},t}-B_{\mathrm{ij}}{v}_{i,t}^2-v_{i,t}{v}_{j,t}\left(G_{\mathrm{ij}}\sin {\theta }_{\mathrm{ij},t}-B_{\mathrm{ij}}\cos {\theta }_{\mathrm{ij},t}\right)\le M_2{u}_{\mathrm{ij},t},\forall \left(i,j\right)\in {\Omega }_L^{AC},\forall t\in T& \left(19\right)\end{array}$ $\begin{array}{cc}-M_3{u}_{\mathrm{ij},t}\le p_{\mathrm{ij},t}-\frac{1}{r_{\mathrm{ij}}}\left(v_{i,t}^2-v_{i,t}{v}_{j,t}\right)\le M_3{u}_{\mathrm{ij},t},\forall \left(i,j\right)\in {\Omega }_L^{DC},\forall t\in T& \left(20\right)\end{array}$

In the above formulas, p_ij and q_ij respectively represent active power and reactive power transmitted by the line. A default positive direction is from node i to node j. θ_ij represents a difference between voltage phase angles of node i and node j. M₁, M₂, M₃ represent large enough positive constants, and they only need to be slightly greater than an upper limit of an absolute value of the corresponding power flow.

(4) Power Output Sub-Constraint of the Unit:

$\begin{array}{cc}{P}_i^{G,\min }\le p_{i,t}^G\le P_i^{G,\max },\forall i\in {\Omega }_G,\forall t\in T& \left(21\right)\end{array}$ $\begin{array}{cc}{Q}_i^{G,\min }\le q_{i,t}^G\le Q_i^{G,\max },\forall i\in {\Omega }_G^{AC},\forall t\in T& \left(22\right)\end{array}$

In the above formulas, P_i^G,max, P_i^G,min respectively represent upper and lower limits of an active power output of the unit. Upper and lower limits of a reactive power output of the unit are similar. Ω_G^AC represents a DG set of the AC distribution network.

(5) Node Voltage Sub-Constraint:

$\begin{array}{cc}{V}_i^{\min }\le v_{i,t}\le V_i^{\max },\forall i\in {\Omega }_B,\forall t\in T& \left(23\right)\end{array}$ $\begin{array}{cc}-2\pi \le {\theta }_{i,t}\le 2\pi ,\forall i\in {\Omega }_B^{AC},\forall t\in T.& \left(24\right)\end{array}$

In the above formulas, V_i^max, V_i^min respectively represent upper and lower limits of the voltage amplitude of node i.

(6) Climbing Sub-Constraint of the Unit:

$\begin{array}{cc}\Delta P_i^{G,\min }\le p_{i,t}^G-p_{i,t-1}^G\le \Delta P_i^{G,\max },\forall i\in {\Omega }_G,\forall t\in T& \left(25\right)\end{array}$ $\begin{array}{cc}\Delta Q_i^{G,\min }\le q_{i,t}^G-q_{i,t-1}^G\le \Delta Q_i^{G,\max },\forall i\in {\Omega }_G^{AC},\forall t\in T& \left(26\right)\end{array}$

In the above formulas, ΔP_i^G,min and ΔP_i^G,max respectively represent slipping and climbing limitations of unit i. In addition, it is necessary to set a power output state of the unit at a time point before the failure to constrain a power output of the unit when the failure occurs.

(7) State Sub-Constraint of the ESS:

$\begin{array}{cc}{S}_i\left(t\right)=S_i\left(t-1\right)+p_i^{ch}\left(t\right){\eta }_{ch}\Delta t-\frac{p_i^{\mathrm{dis}}\left(t\right)\Delta t}{{\eta }_{\mathrm{dis}}},\forall i\in {\Omega }_B,\forall t\in T& \left(27\right)\end{array}$ $\begin{array}{cc}\mathrm{SO}{C}_{i,\min }\le \frac{S_i\left(t\right)}{E_i^R}\le SO{C}_{i,\max },\forall i\in {\Omega }_B,\forall t\in T& \left(28\right)\end{array}$ $\begin{array}{cc}\begin{array}{c}0\le p_i^{ch}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T\\ 0\le p_i^{\mathrm{dis}}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T\end{array}& \left(29\right)\end{array}$ $\begin{array}{cc}{p}_i^{ch}\left(t\right)·p_i^{\mathrm{dis}}\left(t\right)=0,\forall i\in {\Omega }_B,\forall t\in T& \left(30\right)\end{array}$

In the above formulas, the formula (27) represents a constraint on a power balance equation of the ESS, where S_i(t) represents the remaining electricity quantity of the ESS at time t, and η_ch, η_dis respectively represent the charging efficiency and discharging efficiency of the ESS. The formula (28) represents a constraint on a state of charge (SOC) of the ESS. The formula (29) represents a constraint on the charging and discharging power of the ESS, and the formula (30) represents that the ESS can only be in a charging or discharging state at a time point. Δt represents the time interval, which is uniformly set to one hour in the present disclosure.

(8) LS Sub-Constraint:

$\begin{array}{cc}0\le {\rho }_{i,t}\le 1,\forall i\in {\Omega }_B,\forall t\in T& \left(31\right)\end{array}$

In step 105, the unit hardening investment cost c_ij^l of the line, the line length s_ij, the binary decision variable y_ij for line hardening, a first cost coefficient C^e, a second cost coefficient c^p, and the capacity E_i^R and the rated power P_i^R of the ESS configured on node i are determined based on the outer-level constraint and the first sub-objective function; and the power output c_g^G of the DG, the unit cost coefficient c_i^ch of charging of the ESS, the unit cost coefficient c_i^dis of discharging of the ESS, the unit LS cost c_i^LS after the load weight is considered, the LS ratio ρ_i,t of node i, and the load p_i,t^D of node i are determined based on the middle-level constraint, the inner-level constraint, and the second sub-objective function.

With reference to FIG. 2, the unit hardening investment cost c_ij^l of the line, the line length s_ij, the binary decision variable y_ij for line hardening, the first cost coefficient C^e, the second cost coefficient c^p, and the capacity E_i^R and the rated power P_i^R of the ESS configured on node i are determined based on the outer-level constraint and the first sub-objective function; and then the power output c_g^G of the DG, the unit cost coefficient c_i^ch of charging of the ESS, the unit cost coefficient c_i^dis of discharging of the ESS, the unit LS cost c_i^LS after the load weight is considered, the LS ratio ρ_i,t of node i, and the load p_i,t^D of node i are determined based on the middle-level constraint, the inner-level constraint, and the second sub-objective function.

In step 106, energy storage of the DG in the distribution network is controlled based on the obtained parameters.

In an implementation of the present disclosure, an improved AC distribution network system composed of IEEE-33 nodes is used to conduct example analysis of the proposed resilience enhancement strategy. A schematic topological graph is shown in FIG. 3. A voltage class of the AC distribution network is 12.66 kV. Four lines are added to a radial distribution network to increase its ring redundancy. All DGs connected to the AC distribution network are schedulable gas turbines. A new energy unit is only connected to the DC distribution network. Table 1 shows a type and an access location of each unit. FIG. 4 and FIG. 5 show a load curve and a power output curve of new energy. It is assumed that at most three lines are hardened, a hardening cost of a single line is 1000 CNY/m, and all the lines are 1 km long. Table 2 shows technical parameters of the ESS. The cost has been converted into a one-year cost. It is assumed that a distribution network operator has received failure warning information before an extreme event occurs, and has taken a countermeasure in advance. Therefore, an initial SOC of the ESS is set to 0.85.

After the failure occurs, the distribution network is split and disconnected from a tie line of the superior power supply. After 5 hours, the line is repaired and the superior power supply is resumed. Table 3 shows the LS cost. Considering that large users and important loads are more connected to the AC distribution network, a load weight of the AC distribution network is greater than that of the DC distribution network.

Five extreme events occur in a year, the line failure rate is 0.9 before hardening and 0.1 after hardening. It is assumed that the extreme event will cause disconnection of a tie line between the main network and the AC distribution network.

For example, four different scenarios are designed to analyze the pre-disaster resilience enhancement strategy of the distribution network. In scenario 1, the pre-disaster resilience enhancement strategy is not adopted. In scenario 2, only the line hardening strategy is adopted. In scenario 3, only the ESS configuration strategy is adopted. In scenario 4, a pre-disaster defense strategy combining line hardening and ESS configuration is adopted.

Table 4 compares the resilience enhancement strategies and corresponding effects in the four scenarios. Table 5 analyzes LS under the different scenarios. Table 6 and Table 7 list the ESS configuration strategies in scenario 3 and scenario 4 respectively. Since the load weight of the DC distribution network is low, the hardening strategy and failure are more specific to the AC distribution network.

From the analysis of the above data, it can be seen that if no pre-disaster defense strategy is adopted, the extreme event will directly lead to shedding of 57% of first-level loads of the distribution network, thus causing a huge economic loss. Although this is a result of a most serious failure, it is of very important practical significance. The extreme events include not only a natural disaster, but also a man-made attack. The natural disaster is accidental, but a malicious attack often has strong directivity. Therefore, an actual probability of a worst case is considerable and cannot be ignored. Analyzing the pre-disaster defense strategy under the most serious failure is also seeking an inevitable result from a contingency.

In scenario 2, the line hardening strategy is adopted, and the objective function is 10.7% lower than that in scenario 1. In this case, 43% of the first-level loads are still shed, and a resilience enhancement effect is limited. This is because the distribution network is already in a power vacancy state when it is disconnected, and a power shortage cannot be compensated for only through line hardening. In addition, if an important node is topologically far away from the DG, the whole line needs to be hardened to ensure power supply of an important load, which is obviously infeasible. Breaking through a budget constraint and increasing the maximum quantity of hardened lines may not enhance resilience of the distribution network. In this example, if the maximum quantity of hardened lines is set to 4, the optimal hardening strategy still keeps unchanged.

In scenario 3, only the ESS configuration strategy is adopted, and the objective function is 44.1% lower than that in scenario 1. In this case, an access point of the ESS coincides with the first-level load, and the first-level load is not shed. However, this strategy still has a large room for improvement, because ensuring a load supply of the important load by only relying on ESS configuration needs to be achieved in a point-to-point manner, which changes an original power supply mode of the distribution network. If the DG configured in the distribution network can be guided as far as possible to supply the important load, the investment cost can be further reduced. In addition, ensuring power supply for the first-level load is at the cost of large-scale LS of other loads in the AC distribution network. On one hand, due to an ESS configuration cost, the ESS does not supply electric energy to a load other than the first-level load. On the other hand, the ring redundancy of the distribution network is insufficient. If a state of a feeder switch can be remotely scheduled, and an optimal topology of the distribution network can be reconstructed in response to a disaster, large-scale LS can be further avoided. However, remote automation of the distribution network is not yet mature, a condition for configuring a remote switch on most lines is not yet available, and a large quantity of circuit breakers need to be turned on/off in reconstruction of the distribution network, which may lead to a new failure. A large quantity of binary variables also need to be introduced to configure and schedule the feeder switch, and algorithm efficiency is greatly affected by the binary variables. The present disclosure is mainly intended to ensure uninterruptible power supply of the first-level load, so the configuration and scheduling of the feeder switch are not within the scope of the present disclosure.

In scenario 4, the pre-disaster defense strategy combining line hardening and ESS configuration is adopted, and the objective function is 59.1% lower than that in scenario 1. Through analysis of investment results of scenarios 1 to 3, it can be concluded that if the DG is topologically close to the first-level load or a distance between first-level loads is short, and the DG has a sufficient capacity, the line hardening strategy is more appropriate. On the contrary, the ESS configuration strategy should be preferred. It is not difficult to know from FIG. 3 that node 4 is adjacent to DG1, and a cost of configuring the ESS on node 4 is greater than that of hardening lines 4 and 5. However, nodes 10 and 17 are far away from the DG, and line hardening costs are high. In this case, it is more appropriate to configure the ESS on the nodes. Therefore, the defense strategy combining ESS configuration and line hardening can reduce investment and planning costs while ensuring the resilience enhancement effect.

The resilience of the distribution network under the four scenarios is comprehensively evaluated based on resilience evaluation indexes. A calculation result of each index is shown in Table 8. The first column in the table lists each index weight obtained based on an entropy weight method. The last row in the table gives comprehensive resilience evaluation results obtained based on the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS). The results in the table show that a comprehensive resilience evaluation result in scenario 4 is significantly greater than comprehensive resilience evaluation results in the other three scenarios, which further verifies an effect of the planning strategy combining line hardening and ESS configuration on resilience enhancement of the distribution network.

TABLE 1
Parameters of units of the AC distribution network
	Bus	Minimum	Maximum
Unit	access	active power	active power	Cost
No.	node	output (MW)	output (MW)	(CNY/kWh)
DG1	5	0	500	0.5
DG2	20	0	500	0.5
DG3	24	0	500	0.5

TABLE 2
Parameters of the ESS
                                 Parameter
Parameter name                   value
Upper limit PR, max of the configurable rated power (kW) 400
Upper limit ER, max of the configurable capacity (kWh) 100
Unit power configuration cost cp (CNY/kW) 1050
Unit capacity configuration cost ce (CNY/kWh) 3000
Charging efficiency ηch          0.9
Discharging efficiency ηdis      0.9
Unit charging cost cch (CNY/kWh) 0.2
Unit discharging cost cch (CNY/kWh) 0.2
Maximum quantity of configured ESSs 5
Initial SOC                      0.85
Upper SOC limit                  0.9
Lower SOC limit                  20

TABLE 3
LS cost
Load type                        LS cost (CNY/kWh)
First-level load                 2500
Other loads in the AC distribution network 100
Load in the DC distribution network 20

TABLE 4
Comparison of resilience enhancement strategies under different scenarios
			Node on which ESS	Objective
		Line hardening	configuration	function value
Scenario	Failure set	strategy	is performed	(10000 CNY)
1	19-20, 4-5, 6-26, 29-30,	/	/	1965.7
	30-31, 2-19
2	19-20, 3-4, 17-18, 16-17,	4-5, 29-30, 28-29	/	1765.3
	9-10, 10-11
3	19-20, 3-4, 30-31, 15-16,	/	4, 10, 17, 28, 29	1098.7
	5-6, 12-13
4	19-20, 3-4, 30-31, 15-16,	4-5, 29-30, 28-29	10, 17	804
	5-6, 27-28

TABLE 5
LS analysis under different scenarios
		Amount shed for the other
	Amount shed for the	loads in the AC distribution
Scenario	first-level load (kWh)	network (kWh)
1	1137.6	10791.6
2	858.2	7323.3
3	0	15719.8
4	0	13156.9

TABLE 6
ESS configuration strategy in scenario 3
	Node on which ESS
	configuration is	PR	ER	Configuration cost
ESS No.	performed	(kW)	(kWh)	(10000 CNY)
1	4	88.5	588.7	185.9
2	10	90.7	613.2	193.5
3	17	83.5	561.9	177.3
4	28	83.1	545.9	172.5
5	29	98.2	660.1	208.3

TABLE 7
ESS configuration strategy in scenario 4
	Node on which ESS
	configuration is	PR	ER	Configuration cost
ESS No.	performed	(kW)	(kWh)	(10000 CNY)
1	10	90.8	614.1	193.8
2	17	83.5	561.9	177.3

TABLE 8
Resilience index evaluation results
	Index
Index name	weight	Scenario 1	Scenario 2	Scenario 3	Scenario 4
Proportion of a schedulable	0.117	0.883	0.883	0.905	0.893
emergency resource
Emergency resource supply rate	0.151	1	1	1	1
Configuration rate of a remote	0.133	0	0	0	0
feeder switch
AC/DC tie line supply rate	0.021	1	1	1	1
Expected quantity of faulty lines	0.152	6	6	6	6
Power loss time of the important	0.089	5	5	0	0
load (h)
Power loss rate of the important	0.155	0.570	0.430	0	0
load
Total LS loss of the distribution	0.182	393.97	294.53	159.04	134.98
network (10000 CNY)
Total resilience score	/	0	0.267	0.951	0.995

FIG. 6 is a schematic structural diagram of a resilience enhancement-oriented energy storage control system 600 for a DG in a distribution network according to an implementation of the present disclosure. As shown in FIG. 6, the resilience enhancement-oriented energy storage control system 600 for a DG in a distribution network according to this implementation of the present disclosure includes: an objective function determining unit 601, an outer-level constraint determining unit 602, a middle-level constraint determining unit 603, an inner-level constraint determining unit 604, a parameter determining unit 605, and an energy storage control unit 606.

Preferably, the objective function determining unit 601 is configured to determine an objective function, where the objective function includes a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize an LS cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level.

Preferably, the first sub-objective function includes:

$\underset{y\in Y}{\min }{C}^{\mathrm{INV}}$ $C^{\mathrm{INV}}=C_L^{\mathrm{INV}}+C_{ES}^{\mathrm{INV}}$ $C_L^{\mathrm{INV}}={\beta }_l\sum _{\left(i,j\right)\in {\Omega }_L}{c}_{\mathrm{ij}}^l{s}_{\mathrm{ij}}{y}_{\mathrm{ij}}$ $C_{ES}^{\mathrm{INV}}={\beta }_{es}\left(\sum _{i\in {\Omega }_B}{c}^e{E}_i^R+c^p{P}_i^R\right);$

and

• • the second sub-objective function includes:

${\rho }_{\mathrm{pre}}\tau \underset{u\in U}{\max }\underset{x\in F\left(y,u\right)}{\min }\left(C^{\mathrm{OPE}}+C^{\mathrm{LS}}\right)$ $C^{OPE}=\sum _{t\in T}\left[\sum _{g\in {\Omega }_G}{c}_g^G{p}_{g,t}^G+\sum _{i\in {\Omega }_B}\left(c_i^{dis}{p}_{i,t}^{dis}+c_i^{ch}{p}_{i,t}^{ch}\right)\right]$ $C^{LS}=\sum _{i\in {\Omega }_B,t\in T}{c}_i^{LS}{\rho }_{i,t}{p}_{i,t}^D$

• • where Y, U, F represent variable sets of tri-level decision-making respectively; C^INV represents an investment cost; C^OPE and C^LS represent the post-disaster operation cost and the LS cost respectively; ρ_pre represents introduction of a resilience preference coefficient; C_L^INV represents an investment cost for line hardening; C_ES^INV represents an investment cost for energy storage resource configuration; c_ij^l represents a unit hardening investment cost of a line; s_ij represents a line length; y_jj represents a binary decision variable for line hardening; β_l represents a coefficient for converting a total investment cost of the line into a one-year cost based on a planned period; Ω_L represents a line set of an AC/DC distribution network; c^e and c^p represent cost coefficients; β_es represents a coefficient for converting a total investment cost of an ESS into a one-year cost based on a planned period; Ω_B represents a node set of the AC/DC distribution network; E_i^R and P_i^R respectively represent a capacity and rated power of the ESS configured on node i; c_i^LS represents a unit LS cost after a load weight is considered; ρ_i,t represents an LS ratio of node i at time t; p_i,t^D represents load of the node at the corresponding time; T represents duration of an extreme event; and c_g^G, c_i^ch, and c_i^dis represent a power output of a DG, a unit cost coefficient for charging of the ESS, and a unit cost coefficient for discharging of the ESS respectively.

Preferably, the outer-level constraint determining unit 602 is configured to determine an outer-level constraint based on a quantity of configured ESSs, a quantity of hardened lines, and a rated power and capacity configuration of the ESS.

Preferably, the outer-level constraint includes: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS;

• • the sub-constraint on the quantity of configured ESSs includes:

$\sum _{i\in {\Omega }_B}{\sigma }_i\le N_{ES}^{INV}$

• • the sub-constraint on the quantity of hardened lines includes:and

$\sum _{\left(i,j\right)\in {\Omega }_L}{y}_{\mathrm{ij}}\le N_L^{INV};$

• • the sub-constraint on the rated power and capacity configuration of the ESS includes:

$\begin{array}{c}0\le P_i^R\le {\sigma }_i{P}_i^{R,\max },\forall i\in {\Omega }_B\\ 0\le E_i^R\le {\sigma }_i{E}_i^{R,\max },\forall i\in {\Omega }_B\end{array}$

• • where σ_i represents a binary decision variable for ESS configuration, where when a value of the variable is 1, node i is configured with the ESS, otherwise, the node is not configured with the ESS; N_ES^INV represents a maximum allowable quantity of configured ESSs; N_L^INV represents a maximum quantity of hardened lines; P_i^R,max and E_i^R,max respectively represent upper limits of configurable rated power and a configurable capacity of the ESS; when σ_i=0, node i is not configured with the ESS, which is equivalent to setting both the rated power and the capacity of the ESS to 0; E_i^R and P_i^R respectively represent the capacity and the rated power of the ESS configured on node i; Ω_B represents the node set of the AC/DC distribution network; Ω_L represents the line set of the AC/DC distribution network; and y_ij; represents the binary decision variable for line hardening.

Preferably, the middle-level constraint determining unit 603 is configured to determine a middle-level constraint based on a line failure.

Preferably, the middle-level constraint includes:

$\sum _{\left(i,j\right)\in {\Omega }_L}\left(-{\log }_2{p}_{\mathrm{ij}}^{dam}\right){𝓏}_{\mathrm{ij},0}\le W_{\max }$

• • where p_ij^dam represents a line failure rate; _ij,0 represents a binary variable, indicating whether a failure occurs on the line, and a value of the binary variable is 1 when a failure occurs and 0 when no failure occurs; W_max represents an upper limit of an uncertain entropy value of the line failure in a distribution network; and Ω_L represents the line set of the AC/DC distribution network.

Preferably, the inner-level constraint determining unit 604 is configured to determine an inner-level constraint based on node power of the distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS.

Preferably, the inner-level constraint includes: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint;

• • the state sub-constraint of the ESS includes:

$S_i\left(t\right)=S_i\left(t-1\right)+p_i^{ch}\left(t\right){\eta }_{ch}\Delta t-\frac{p_i^{\mathrm{dis}}\left(t\right)\Delta t}{{\eta }_{\mathrm{dis}}},\forall i\in {\Omega }_B,\forall t\in T$ $\mathrm{SO}{C}_{i,\min }\le \frac{S_i\left(t\right)}{E_i^R}\le SO{C}_{i,\max },\forall i\in {\Omega }_B,\forall t\in T$ $\begin{array}{c}0\le p_i^{ch}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T\\ 0\le p_i^{\mathrm{dis}}\left(t\right)\le P_i^R,\forall i\in {\Omega }_B,\forall t\in T\end{array}$ $p_i^{ch}\left(t\right)·p_i^{\mathrm{dis}}\left(t\right)=0,\forall i\in {\Omega }_B,\forall t\in T;$

and

• • the LS sub-constraint includes:

$0\le {\rho }_{i,t}\le 1,\forall i\in {\Omega }_B,\forall t\in T$

• • where S_i(t) represents a remaining electricity quantity at time t; η_ch, η_dis represent charging efficiency and discharging efficiency respectively; Δt represents a time interval; p_i^ch(t) represents charging power; p_i^dis(t) represents discharging power; and P_i^R represents an upper charging/discharging limit.

Preferably, the parameter determining unit 605 is configured to determine, based on the outer-level constraint and the first sub-objective function, the unit hardening investment cost c_ij^l of the line, the line length s_ij, the binary decision variable y_ij; for line hardening, a first cost coefficient c^e a second cost coefficient c^p, and the capacity E_i^R and the rated power P_i^R of the ESS configured on node i; and determine the power output c_g^G of the DG, the unit cost coefficient c_i^ch of charging of the ESS, the unit cost coefficient c_i^dis of discharging of the ESS, the unit LS cost c_i^LS after the load weight is considered, the LS ratio ρ_i,t of node i, and the load p_i,t^D of node i based on the middle-level constraint, the inner-level constraint, and the second sub-objective function.

Preferably, the energy storage control unit 606 is configured to control energy storage of the DG in the distribution network based on the obtained parameters.

The resilience enhancement-oriented energy storage control system 600 for a DG in a distribution network according to this embodiment of the present disclosure corresponds to the resilience enhancement-oriented energy storage control method 100 for a DG in a distribution network according to another embodiment of the present disclosure, and details are not be described herein again.

In this embodiment of the present disclosure, the objective function determining unit 601, the outer-level constraint determining unit 602, the middle-level constraint determining unit 603, the inner-level constraint determining unit 604, the parameter determining unit 605, and the energy storage control unit 606 each may be one or more processors or controllers that each have a communication interface, can realize a communication protocol, and may further include a memory, a related interface and system transmission bus, and the like if necessary. The processor or controller executes program-related code to realize a corresponding function. In an alternative solution, the objective function determining unit 601, the outer-level constraint determining unit 602, the middle-level constraint determining unit 603, the inner-level constraint determining unit 604, the parameter determining unit 605, and the energy storage control unit 606 share an integrated chip or share devices such as a processor or a controller and a memory. The shared processor, controller or chip executes program-related codes to implement a corresponding function.

The present disclosure has been described above with reference to a few implementations. However, it is well known to those skilled in the art that, as defined by the accompanying patent claims, besides those disclosed above, other embodiments of the present disclosure are equally within the scope of the present disclosure.

Generally, all terms used in the claims are interpreted based on their usual meanings in the technical field, unless otherwise explicitly defined therein. All references to “one/the said/the [apparatus, component, or the like]” are openly interpreted as at least one instance of the said apparatus, component, or the like, unless otherwise explicitly stated. The steps of any method disclosed herein need not be performed in the exact order disclosed unless explicitly stated.

Those skilled in the art should understand that the embodiments of the present disclosure may be provided as a method, a system, or a computer program product. Therefore, the present disclosure may use a form of hardware only embodiments, software only embodiments, or embodiments with a combination of software and hardware. Moreover, the present disclosure may be in a form of a computer program product that is implemented on one or more computer-usable storage media (including but not limited to a magnetic disk memory, a CD-ROM, an optical memory, and the like) that include computer-usable program code.

The present disclosure is described with reference to the flowcharts and/or block diagrams of the method, the device (system), and the computer program product according to the embodiments of the present disclosure. It should be understood that computer program instructions may be used to implement each process and/or each block in the flowcharts and/or the block diagrams and a combination of a process and/or a block in the flowcharts and/or the block diagrams. These computer program instructions may be provided for a general-purpose computer, a dedicated computer, an embedded processor, or a processor of another programmable data processing device to generate a machine, such that the instructions executed by a computer or a processor of another programmable data processing device generate an apparatus for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

These computer program instructions may be stored in a computer-readable memory that can instruct a computer or another programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate an artifact that includes an instruction apparatus. The instruction apparatus implements a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

These computer program instructions may be loaded onto a computer or another programmable data processing device, such that a series of operations and steps are performed on the computer or the another programmable device, thereby generating computer-implemented processing. Therefore, the instructions executed on the computer or the another programmable device provide steps for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

Finally, it should be noted that: the above embodiments are merely intended to describe the technical solutions of the present disclosure, rather than to limit thereto; although the present disclosure is described in detail with reference to the above embodiments, it is to be appreciated by a person of ordinary skill in the art that modifications or equivalent substitutions may still be made to the specific implementations of the present disclosure, and any modifications or equivalent substitutions made without departing from the spirit and scope of the present disclosure shall fall within the protection scope of the claims of the present disclosure.

Claims

1. A resilience enhancement-oriented energy storage control method for a distributed generator (DG) in a distribution network, comprising: determining an objective function, wherein the objective function comprises a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize a load shedding (LS) cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level;determining an outer-level constraint based on a quantity of configured energy storage systems (ESSs), a quantity of hardened lines, and a rated power and capacity configuration of an ESS;determining a middle-level constraint based on a line failure;determining an inner-level constraint based on node power of a distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS;determining, based on the outer-level constraint and the first sub-objective function, a unit hardening investment cost cijl of a line, a line length Sij, a binary decision variable yij for line hardening, a first cost coefficient ce, a second cost coefficient cp, and a capacity EiR and rated power PiR of an ESS configured on node i; and determining a power output cgG of a DG, a unit cost coefficient cich of charging of the ESS, a unit cost coefficient cidis of discharging of the ESS, a unit LS cost ciLS after a load weight is considered, an LS ratio ρi,t of node i, and load pi,tD of node i based on the middle-level constraint, the inner-level constraint, and the second sub-objective function; andcontrolling energy storage of the DG in the distribution network based on the obtained parameters.

2. The method according to claim 1, wherein the outer-level constraint comprises: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS; the sub-constraint on the quantity of configured ESSs comprises:

3. The method according to claim 1, wherein the middle-level constraint comprises:

4. The method according to claim 1, wherein the inner-level constraint comprises: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint; the state sub-constraint of the ESS comprises:

5. The method according to claim 1, wherein the first sub-objective function comprises:

6. A resilience enhancement-oriented energy storage control system for a DG in a distribution network, comprising: an objective function determining unit configured to determine an objective function, wherein the objective function comprises a first sub-objective function and a second sub-objective function, the first sub-objective function aims to minimize an investment cost at an outer level, and the second sub-objective function first aims to maximize an LS cost and a post-disaster operation cost at a middle level, and then minimize an LS cost and a post-disaster operation cost at an inner level;an outer-level constraint determining unit configured to determine an outer-level constraint based on a quantity of configured ESSs, a quantity of hardened lines, and a rated power and capacity configuration of an ESS;a middle-level constraint determining unit configured to determine a middle-level constraint based on a line failure;an inner-level constraint determining unit configured to determine an inner-level constraint based on node power of a distribution network, a line load capacity, a line power flow, a power output of a unit, a node voltage, climbing of the unit, operation of the ESS, and LS;a parameter determining unit configured to determine, based on the outer-level constraint and the first sub-objective function, a unit hardening investment cost cijl of a line, a line length Sij, a binary decision variable yij for line hardening, a first cost coefficient Ce, a second cost coefficient cp, and a capacity EiR and rated power PiR of an ESS configured on node i; and determine a power output cgG of a DG, a unit cost coefficient cich of charging of the ESS, a unit cost coefficient cidis of discharging of the ESS, a unit LS cost ciLS after a load weight is considered, an LS ratio ρi,t of node i, and load pi,tD of node i based on the middle-level constraint, the inner-level constraint, and the second sub-objective function; andan energy storage control unit configured to control energy storage of the DG in the distribution network based on the obtained parameters.

7. The system according to claim 6, wherein the outer-level constraint comprises: a sub-constraint on the quantity of configured ESSs, a sub-constraint on the quantity of hardened lines, and a sub-constraint on the rated power and capacity configuration of the ESS; the sub-constraint on the quantity of configured ESSs comprises:

8. The system according to claim 6, wherein the middle-level constraint comprises:

9. The system according to claim 6, wherein the inner-level constraint comprises: a sub-constraint on the node power of the distribution network, a sub-constraint on the line load capacity, a sub-constraint on the line power flow, a power output sub-constraint of the unit, a node voltage sub-constraint, a climbing sub-constraint of the unit, a state sub-constraint of the ESS, and an LS sub-constraint; the state sub-constraint of the ESS comprises:

10. The system according to claim 6, wherein the first sub-objective function comprises: