SYSTEM AND METHOD FOR DETERMINING A STATE OF A PHOTOVOLTAIC PANEL

Abstract

A system for determining a state of a photovoltaic panel. The system includes a data acquisition device having a circuit and a communication module. The circuit is arranged to transmit excitation signals to a photovoltaic panel and detect response signals generated by the photovoltaic panel in response to the excitation signal, during normal operation of the photovoltaic panel. The communication module is arranged to communicate the response signals to a control device for analysis and determination of a state of the photovoltaic panel.

Description

TECHNICAL FIELD

The invention relates to a system and method for determining a state of a photovoltaic (PV) panel. Particularly, although not exclusively, the invention relates to a system and method for diagnosing health of a PV panel.

BACKGROUND

Solar power is one of the fastest growing clean energy sources. In a solar power plant, Photovoltaic (PV) panels or solar cells are usually used for converting light energy into electrical energy.

The output of photovoltaic (PV) panels declines over time. And different panels decline at different rates. This phenomenon is also called “PV panel degradation”, which is a complex nonlinear process. The degradation may be due to aging of components, utilization conditions, and environmental factors such as extreme weather conditions and physical damages or scratches. PV panel degradation has been found to be the one of the main cause of PV power generation failure.

There is a need to properly and timely determine the condition of PV panels in order to ensure proper, continuous, and effective power generation.

SUMMARY OF THE INVENTION

In accordance with a first aspect of the invention, there is provided a system for determining a state of a photovoltaic panel, comprising: a data acquisition device having: a circuit arranged to transmit excitation signals to a photovoltaic panel and detect response signals generated by the photovoltaic panel in response to the excitation signal, during normal operation of the photovoltaic panel, and a communication module arranged to communicate the response signals to a control device for analysis and determination of a state of the photovoltaic panel.

In one embodiment of the first aspect, the excitation signals one or more voltage signals each having a respective non-zero perturbation frequency. Each excitation signal may have a duration in the order of milliseconds, e.g., several or several tens of milliseconds. Preferably, the response signals comprise a terminal voltage across the photovoltaic panel and an output current of the photovoltaic panel. The voltage signals are preferably arranged to drive a terminal voltage of the photovoltaic panel from minimum voltage level to maximum voltage level.

In one embodiment of the first aspect, the data acquisition device further includes a sampler for sampling the detected response signals, and the communication module is arranged to communicate the sampled detected response signals to the control device. The sampler may be implemented with a circuit, a controller, or the like.

In one embodiment of the first aspect, the data acquisition device includes a memory for storing the detected response signals. The memory may include one or more volatile memory unit, non-volatile unit, or any of their combinations.

In one embodiment of the first aspect, the circuit is arranged to be connected with two photovoltaic panels for transmitting respective excitation signals and detecting respective response signals from the two photovoltaic panels.

In one embodiment of the first aspect, the circuit includes: a power converter with two switches; a driving circuit for providing gating signals to operate the two switches; a controller for controlling the driving circuit based on a difference between the detected signal of one of the two photovoltaic panels and a reference signal; the driving circuit is arranged to operate the two switches complementarily; and the controller is arranged to control the driving circuit so as to control the detected signal of the one of the two photovoltaic panels to follow the reference signal. The switches are preferably semiconductor switches. In one example, the reference signal comprises a reference voltage signal with a non-zero perturbation frequency, and the detected signal of one of the two photovoltaic panels comprises a terminal voltage across the one of the two photovoltaic panels. Preferably, the power converter comprises a DC-DC converter such as a buck-boost converter arranged to operate in continuous conduction mode.

In one embodiment of the first aspect, the communication module comprises a wireless communication module. For example, the wireless communication module may include a ZigBee communication module.

In one embodiment of the first aspect, the system further includes the control device that includes a communication module for communicating with the data acquisition device. The communication module of the control device may comprise a wireless communication module complementary to the wireless communication module of the data acquisition device (i.e., uses same type of wireless transmission protocol). For example, the wireless communication module of the control device may include a ZigBee communication module. Preferably, the data acquisition device and the control device are both provided on-site, e.g., at the solar power plant.

In one embodiment of the first aspect, the control device further comprises a processing unit arranged to process the response signals to determine one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the first aspect, the processing unit is arranged to process the response signals by matching the response signals of the photovoltaic panel with a predetermined model for determining the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the first aspect, the predetermined model comprises a dynamic single-diode model of a solar cell with: a current source providing a current based on incident light; a diode connected in parallel with the current source; a capacitor connected in parallel with the current source; a first resistor connected in parallel with the current source; and a second resistor connected in series with the first resistor.

In one embodiment of the first aspect, the processing unit includes: a current predictor for predicting a terminal current generated in the dynamic single-diode model based on the measured terminal voltage of the photovoltaic panel and a set of parameters in the dynamic single-diode model.

In one embodiment of the first aspect, the set of parameters in the dynamic single-diode model comprises one or more (and preferably all) of: a current I_ph provide by the current source in the dynamic single-diode model; a reverse saturation current I₀ in the dynamic single-diode model; a thermal voltage VT in the dynamic single-diode model; a resistance R_sh of the first resistor in the dynamic single-diode model, indicative of an intrinsic p-n junction resistance; a capacitance C_sh of the capacitor in the dynamic single-diode model, indicative of an intrinsic p-n junction capacitance; and a resistance R_s of the second resistor in the dynamic single-diode model.

In one embodiment of the first aspect, the processing unit further includes an optimization unit, operably connected with the current predictor, for determining one or more values of an objective function of the set of parameters in the dynamic single-diode model. Preferably, the one or more values of the objective function comprises the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the first aspect, the optimization unit is arranged to determine one or more values of the objective function of the set of parameters by iteratively reducing a difference between the predicted terminal current generated in the dynamic single-diode model and the detected output current of the photovoltaic panel.

In one embodiment of the first aspect, the optimization unit is arranged to determine one or more values of the one or more intrinsic parameters indicative of the state of the photovoltaic panel using a real jumping gene genetic algorithm-based method. In one example, the values of intrinsic parameters are generated using the real-jumping gene genetic method. In one example, the value of objective function is determined by the current predictor, which establishes the optimality or correctness of the intrinsic parameters.

In one embodiment of the first aspect, the control device further comprises a further communication module for communicating the one or more determined intrinsic parameters to a storage that is preferably remote from the control device. The further communication module may comprise a wireless communication module. The further communication module may be a Wi-Fi communication module. Preferably, the communication protocol of the further communication module and the communication protocol of the communication module are different.

In one embodiment of the first aspect, the system also includes the remote storage arranged to store the intrinsic parameters determined by the control device. The remote storage may be a server, such as a cloud computing server, implemented with any number of information handling systems.

In accordance with a second aspect of the invention, there is provided a data acquisition device in the system of the first aspect.

In accordance with a third aspect of the invention, there is provided a control device in the system of the first aspect.

In accordance with a fourth aspect of the invention, there is provided a method for determining a state of a photovoltaic panel, comprising: transmitting excitation signals to a photovoltaic panel during normal operation of the photovoltaic panel; detecting response signals generated by the photovoltaic panel in response to the excitation signal; and communicating the response signals to a control device for analysis and determination of a state of the photovoltaic panel.

In one embodiment of the fourth aspect, the excitation signals comprise a plurality of voltage signals each having a respective non-zero perturbation frequency; and the response signals comprise a terminal voltage across the photovoltaic panel and an output current of the photovoltaic panel.

In one embodiment of the fourth aspect, the method further includes sampling the detected response signals and the communicating step comprises communicating the sampled detected response signals to the control device.

In one embodiment of the fourth aspect, the method further includes storing the detected response signals.

In one embodiment of the fourth aspect, the method further includes processing the response signals to determine one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the fourth aspect, the step of processing comprises matching the response signals of the photovoltaic panel with a predetermined model for determining the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the fourth aspect, the predetermined model comprises a dynamic single-diode model of a solar cell with: a current source providing a current based on incident light; a diode connected in parallel with the current source; a capacitor connected in parallel with the current source; a first resistor connected in parallel with the current source; and a second resistor connected in series with the first resistor.

In one embodiment of the fourth aspect, the step of matching comprises: predicting a terminal current generated in the dynamic single-diode model based on the measured terminal voltage of the photovoltaic panel and a set of parameters in the dynamic single-diode model.

In one embodiment of the fourth aspect, the set of parameters in the dynamic single-diode model comprises one or more of: a current I_ph provide by the current source in the dynamic single-diode model; a reverse saturation current I₀ in the dynamic single-diode model; a thermal voltage ν_T in the dynamic single-diode model; a resistance R_sh of the first resistor in the dynamic single-diode model, indicative of an intrinsic p-n junction resistance; a capacitance C_sh of the capacitor in the dynamic single-diode model, indicative of an intrinsic p-n junction capacitance; and a resistance R_s of the second resistor in the dynamic single-diode model.

In one embodiment of the fourth aspect, the step of matching further comprises determining one or more values of an objective function of the set of parameters in the dynamic single-diode model, the one or more values of the objective function comprises the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

In one embodiment of the fourth aspect, the determination of the one or more values of the objective function of the set of parameters comprises iteratively reducing a difference between the predicted terminal current generated in the dynamic single-diode model and the detected output current of the photovoltaic panel.

In one embodiment of the fourth aspect, the determination of the one or more values of the one or more intrinsic parameters indicative of the state of the photovoltaic panel is performed using a real jumping gene genetic algorithm-based method.

In one embodiment of the fourth aspect, the method further includes communicating the one or more intrinsic parameters indicative of the state of the photovoltaic panel to a remote storage.

In one embodiment of the fourth aspect, the method further includes analyzing a time series of the one or more intrinsic parameters to determine a predicted state of the photovoltaic panel.

In one embodiment of the fourth aspect, the method may be implemented partly or wholly using the system of the first aspect.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings in which:

FIG. 1 is a schematic diagram illustrating a solar power plant connected with a system for determining a state of a PV panel in one embodiment of the invention;

FIG. 2 is a block diagram showing a method for determining a state of a PV panel implemented using the system of FIG. 1 in one embodiment of the invention;

FIG. 3 is a simplified circuit diagram of the data acquisition device in the system of FIG. 1 in one embodiment of the invention;

FIG. 4 is a graph showing key waveforms in the circuit of FIG. 3 during a period of perturbation;

FIG. 5 is a graph showing current-voltage (I-V) response of a PV panel in FIG. 1 under different perturbation frequencies;

FIG. 6 is a circuit diagram illustrating a dynamic model of a solar cell for modeling the PV panel in the system of FIG. 1 in one embodiment of the invention;

FIG. 7 is a functional block diagram illustrating interaction of current predictor and RJGGA-based optimization unit of the control device of FIG. 1 in one embodiment of the invention;

FIG. 8 is a flow chart illustrating a RJGGA-based optimization method implemented by the optimization unit of FIG. 7 in one embodiment of the invention;

FIG. 9 is a schematic diagram of a solver in an FPGA unit of the control device of FIG. 1 in one embodiment of the invention;

FIG. 10 is a functional block diagram illustrating an ARM processor and a FPGA unit of the control device of FIG. 1 in one embodiment of the invention;

FIG. 11 is a schematic diagram showing a test setup for simulation (a) and experiment (b) for verifying the performance of the system of FIG. 1 and the method of FIG. 2;

FIG. 12 is a graph showing test results of a change in the value of the objective function during optimization;

FIG. 13A is a picture showing a first PV panel used in the experiment for verifying the performance of the method in FIG. 2;

FIG. 13B is a picture showing a second PV panel used in the experiment for verifying the performance of the method in FIG. 2;

FIG. 13C is a picture showing a third PV panel used in the experiment for verifying the performance of the method in FIG. 2;

FIG. 13D is a picture showing a fourth PV panel used in the experiment for verifying the performance of the method in FIG. 2;

FIG. 14A is a graph showing measured voltage and current of the first PV panel in FIG. 13A;

FIG. 14B is a graph showing the measured dynamic current-voltage (I-V) response of the first PV panel in FIG. 13A;

FIG. 15A is a graph showing measured voltage, measured current, and predicted current of the first PV panel in FIG. 13A under a first perturbation;

FIG. 15B is a graph showing measured voltage, measured current, and predicted current of the first PV panel in FIG. 13A under a second perturbation;

FIG. 15C is a graph showing measured voltage, measured current, and predicted current of the first PV panel in FIG. 13A under a third perturbation;

FIG. 15D is a graph showing measured voltage, measured current, and predicted current of the first PV panel in FIG. 13A under a fourth perturbation;

FIG. 16A is a graph showing measured voltage, measured current, and predicted current of the second PV panel in FIG. 13B under a first perturbation;

FIG. 16B is a graph showing measured voltage, measured current, and predicted current of the second PV panel in FIG. 13B under a second perturbation;

FIG. 16C is a graph showing measured voltage, measured current, and predicted current of the second PV panel in FIG. 13B under a third perturbation;

FIG. 16D is a graph showing measured voltage, measured current, and predicted current of the second PV panel in FIG. 13B under a fourth perturbation; and

FIG. 17 is a block diagram of an exemplary information handling system arranged to implement at least part of the system in FIG. 1.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

PV Panel Health Diagnostic System

Referring to FIG. 1, the PV solar power plant 10 includes four series-connected PV panels and a grid-tie inverter. A system 100 for determining the state of the PV panel is connected to the PV panels. In this embodiment, the system 100 includes two data acquisition devices 102, a control device 104 arranged to communicate data and information with the data acquisition devices 102, and a storage 106 arranged to communicate data and information with the control device 104. In this embodiment, the data acquisition devices 102 and the control device 104 are deployed on-site at the solar power plant 10, and the storage 106 is arranged off-site remote from the solar power plant 10. Each data acquisition device 102 is arranged to be connected with two PV panels, with connections across each PV panel. In this embodiment, the data acquisition device 102 communicates with the control device 104 through a wireless communication link L₁. The control device 104 communicates with the storage 106 through a wireless communication link L₂.

FIG. 2 shows the general operation flow in the system 100 of FIG. 1. In this embodiment, the control device 104 controls the schedule of performing diagnosis. The control device 104 sends triggering signals to the data acquisition devices 102 to trigger them to perform measurement. The data acquisition devices 102, upon receiving the triggering signals, generate excitations signals and provide the excitation signals to the respective PV panel, during normal operation of the PV panel. The data acquisition devices 102 then detect a response signals generated by the PV panels in response to the excitation signal. The response signals may include signals indicative of dynamic current-voltage characteristics of the PV panel. The data acquisition devices 102 transmit the received response signals to the control device 104, to perform analysis and identification of intrinsic parameters indicative of the state of the PV panel, using a real jumping gene genetic algorithm (RJGGA) based method, as further described below. The control device 104 transmits the data, for example the response signals or the analysis results, to the storage 106. The data in the control device 104 or the data in the storage 106 can be used for analysis and evaluation of the state of degradation of PV panels and for predicting the degradation of PV panels.

The circuit topology of the data acquisition device 102 of FIG. 1 is illustrated in FIG. 3. As illustrated, the circuit topology is based on differential power processing architecture. This architecture facilitates integration of functionality of data acquisition devices into differential power processing devices, allowing the differential power processing devices to also conduct maximum power point tracking and diagnostics. As shown in FIG. 3, the data acquisition device 102 is connected to two PV panels and it serves two main functions. First, it injects large-signal perturbations into the connected PV panels. Second, it samples and records the terminal voltage and current, and then wirelessly transfers them to control device 104.

The data acquisition device 102 includes power converter, in the form of a DC-DC converter with a buck-boost-derived structure controlled to operate in continuous conduction mode. The converter includes two switches S₁ and S₂ controlled by a driving circuit. The converter also includes two capacitors C₁ and C₂ each connected across a respective PV panel and an inductor L connected with the junction of the two switches S₁ and S₂ and the junction of the two capacitors C₁ and C₂. The data acquisition device 102 also includes a current detector 102V1 arranged to detect an output current of the PV panel P₁, a voltage detector 102V1 arranged to detect a terminal voltage across the PV panel P₁, an output current 102I2 of the PV panel P₂, and a voltage detector 102V2 arranged to detect a terminal voltage across the PV panel P₂. Each voltage detector 102V1, 102V2 and current detector 102V1, 102I2 is connected with a respective sampler 102S for sampling the detected signals. The samplers 102S are connected to a multiplexer 102M1 and a memory 102X for storing the detected and sampled measurements. The sampled voltage signals are fed back into the power converter through a multiplexer 102M2 and a PI controller 102C for controlling the driving circuit 102D of the switches S₁ and S₂ in the power converter. In operation, the switches S₁ and S₂ are controlled to operate complementarily.

FIG. 4 shows the key waveforms in the circuit of FIG. 3, including the gate signals ν_GS1(t) and ν_GS2(t), and the respective terminal voltages of the panels ν_P1(t) for P₁ and ν_P2(t) for P₂. Since the power converter is controlled to operate in continuous conduction mode, it can be shown that ν_P1(t) and ν_P2(t) can be related by

$\begin{array}{cc}\frac{v_{P1}\left(t\right)}{v_{P2}\left(t\right)}=\frac{1-d\left(t\right)}{d\left(t\right)}& \left(1\right)\end{array}$

where d(t) is duty ratio of S₁.

As shown in FIG. 4, ν_P1(t) is controlled to follow the waveform of a sinusoidal reference ν_ref(t) with the perturbation frequency f. In particular, the sampled ν_P1(t) is firstly compared with viva) and their difference is then processed by a proportional-plus-integral (PI) controller to control the switching states of the two switches S₁ and S₂.

FIG. 5 shows the trajectories of the response signals, i.e., the terminal voltage and output current, under different perturbation frequencies f. When f=0, the trajectory is a static current-voltage characteristic. As f increases, a hysteresis loop appears due to the equivalent shunt capacitance in the PV panels. The hysteresis loop becomes more significant when the perturbation frequency or shunt capacitance increases.

During excitation and detection operation, the data acquisition devices sample the terminal voltage and current of the PV panel throughout the perturbation period. Generally, each perturbation last several or several tens of milliseconds. In one embodiment a total of four perturbation cycles are conducted. The perturbation process enforces the required voltage ratio between the two connected PV panels to drive the terminal voltage of the tested panel from minimum to maximum voltage level. The minimum and maximum voltages are between the nominal value of short circuit and open circuit voltage, as bounded by the current level of degradation and irradiation, or by the operation point of the grid-tie inverter, or even by both.

Having obtained the detected and sampled signals, it is necessary to process them to determine the state of the PV panels. Essentially, this diagnostic process can be translated into an optimization task. The objective of the optimization task is to match the measured I-V characteristic to the mathematical model prediction. Matching the model and measurements reveals intrinsic parameters governing the behavior of the PV panel. The model is based on the dynamic single-diode model of a solar cell shown in FIG. 6.

Referring to FIG. 6, the model includes a current source I_ph providing a current based on incident light, a diode D_sh connected in parallel with the current source, a capacitor C_sh connected in parallel with the current source, a first resistor R_sh connected in parallel with the current source, and a second resistor R_s connected in series with the first resistor R_sh. The voltage across the first resistor R_sh is denoted as ν_sh(t). The terminal output voltage and output current are denoted as ν(t) and i(t) respectively.

The general mechanism for determining the intrinsic parameters of PV panels is shown in FIG. 7. The mechanism includes two main components, a current predictor and an optimization unit arranged to implement a RJGGA-based optimization method. As shown in FIG. 7, the interface between the current predictor and the optimization unit is provided through parameters of the solar cell model in FIG. 6, described by a vector G:

$\begin{array}{cc}G=\left\{I_{ph},I_o,v_T,R_{\mathrm{sh}},C_{\mathrm{sh}},R_s\right\}& \left(2\right)\end{array}$

where I_ph is the current determined by the incident light, I₀ is the reverse saturation current, ν_T is the thermal voltage, and C_sh and resistor R_sh are used to model the p-n junction, respectively, and R_s is the series resistance of the circuit model for describing the electrical characteristics of solar cell.

The current predictor is used for predicting the generated terminal current by the solar cell model, given the terminal voltage ν(t) and set of solar cell parameters G. The predicted current i_p(t) is the basis for fitness evaluation necessary for RJGGA optimization. The time series of the panel voltage ν(t) and panel current i(t) contains samples. Let V and I be the two time series:

$\begin{array}{cc}V=\left\{v\left[0\right],v\left[1\right],\dots v\left[k\right],\dots v\left[N\right]\right\}& \left(3\right)\\ I=\left\{i\left[0\right],i\left[1\right],\dots i\left[k\right],\dots i\left[N\right]\right\}& \left(4\right)\end{array}$

The current predictor calculates voltage across C_sh as a time-series V_sh, defined as

$\begin{array}{cc}{V}_{\mathrm{sh}}=\left\{v\text{?}\left[0\right],v\text{?}\left[1\right],\dots v_{\mathrm{sh}}\left[k\right],\dots v_{\mathrm{sh}}\left[N\right]\right\}& \left(5\right)\\ \text{?}\text{indicates text missing or illegible when filed}& \end{array}$

The current predicted by the solar cell model is defined as

$\begin{array}{cc}{I}_p=\left\{i_p\left[0\right],i_p\left[1\right],\dots i_p\left[k\right],\dots i_p\left[N\right]\right\}& \left(6\right)\end{array}$

The current predictor iterates through the measured voltage ν[k], calculating ν_sh[k] and predicted current i_p[k] in the following steps.

Step (1): The initial condition assumes zero current passing through C_sh,

$\begin{array}{cc}{C}_{\mathrm{sh}}\frac{d_{v_{\mathrm{sh}}}}{\mathrm{dt}}\left(v_{\mathrm{sh}},v\right)=0& \left(7\right)\end{array}$

therefore ν_sh[0] can be determined from

$\begin{array}{cc}{I}_{ph}=I\text{?}\left(e\text{?}-1\right)+\frac{v_{\mathrm{sh}}\left[0\right]}{R_{\mathrm{sh}}}+\frac{v_{\mathrm{sh}}\left[0\right]-v\left[0\right]}{R\text{?}}& \left(8\right)\\ \text{?}\text{indicates text missing or illegible when filed}& \end{array}$

Step (2): Calculate predicted current at time step k

$\begin{array}{cc}{i}_p\left[k\right]=\frac{v_{\mathrm{sh}}\left[k\right]-v\left[k\right]}{R_s}& \left(9\right)\end{array}$

Step (3): Calculate the value of ν_sh[k+1]

$\begin{array}{cc}{v}_{\mathrm{sh}}\left[k+1\right]-v_{\mathrm{sh}}\left[k\right]=\frac{h}{2}\frac{d_{v_{\mathrm{sh}}}}{\mathrm{dt}}\left(v_{\mathrm{sh}}\left[k\right],v\left[k\right]\right)+\frac{h}{2}\frac{d_{v_{\mathrm{sh}}}}{\mathrm{dt}}\left(v_{\mathrm{sh}}\left[k+1\right],v\left[k+1\right]\right)& \left(10\right)\end{array}$

where

$\frac{d_{v_{\mathrm{sh}}}}{\mathrm{dt}}\left(v\text{?},v\right)$ $\text{?}\text{indicates text missing or illegible when filed}$

is defined by

$\begin{array}{cc}\frac{d_{v_{\mathrm{sh}}}}{\mathrm{dt}}\left(v_{\mathrm{sh}},v\right)=\frac{1}{C\text{?}}\left[I_{\mathrm{ph}}-I_0\left(e\text{?}-1\right)-\frac{v\text{?}}{R_{\mathrm{sh}}}-\frac{v_{\mathrm{sh}}-v}{R_s}\right]& \left(11\right)\\ \text{?}\text{indicates text missing or illegible when filed}& \end{array}$

Step (4): k is increased by 1.

Step (5): Steps (2) to (4) are repeated until k=N

In the above, equations (10) and (11) are solved by the trapezoidal rule.

In this embodiment, RJGGA is used to solve the optimization task by operating over a pool of individuals. In RJGGA, Every individual G is represented by a set of chromosomes, such that the chromosomes match one set of solar cell model parameters outlined in Equation (2).

The pool of individuals evolves iteratively in generations, with every generation striving to minimize the discrepancy between the predicted current in Equation (6) and the measured current in Equation (4). Identification of individuals with favorable chromosomes is based on

$\begin{array}{cc}{f}_{\mathrm{obj}}\left(G\right)=\frac{1}{N_2-N_1+1}\sum _{k=N_1}^{N_2}{\left(i_p\left[k\right]-i\left[k\right]\right)}^2& \left(12\right)\end{array}$

where N₁ and N₂ are the first and the last sample of i_p[k] and i[k] respectively, used for evaluation. For the samples k<N, of i_p[k], the current predictor is in a transitional state, due to the assumption about the initial current through C_sh according to Equation (7).

The fitness of individual is indirectly proportionate to its value of Equation (12). Therefore, the lower the value, the more fit solution the individual represents. The goal of the optimization task is then represented by the following

$\begin{array}{cc}{G}_0=\underset{G}{\arg \min }\left(f_{\mathrm{obj}}\left(G\right)\right)& \left(13\right)\end{array}$

The individuals evolve within a bounded space with upper and lower boundaries, G^MAX and G^MIN respectively

$\begin{array}{cc}{G}^{\mathrm{MAX}}=\left\{I_{\mathrm{ph}}^{\mathrm{MAX}},I_0^{\mathrm{MAX}},v_T^{\mathrm{MAX}},R_{sh}^{\mathrm{MAX}},C_{\mathrm{sh}}^{\mathrm{MAX}},R\text{?}\right\}& \left(14\right)\\ G^{\mathrm{MIN}}=\left\{I_{\mathrm{ph}}^{\mathrm{MIN}},I_0^{\mathrm{MIN}},v_T^{\mathrm{MIN}},R_{sh}^{\mathrm{MIN}},C_{\mathrm{sh}}^{\mathrm{MIN}},R\text{?}\right\}& \left(15\right)\\ \text{?}\text{indicates text missing or illegible when filed}& \end{array}$

The evolution process in the RJGGA-based method in the present embodiment is illustrated in FIG. 8 and is executed according to the following steps.

Step (1): Initialization of pool G^M with M individuals with random values of chromosomes.

Step (2): Fitness evaluation of individuals in G^M

Step (3): Selection of parent pool P^N, as n individuals from G^M, according to probability p_w, using Roulette wheel selection disclosed in G. Jones, “Genetic and Evolutionary Algorithms,” Encyclopedia of Computational Chemistry. John Wiley & Sons, Ltd, 15 Apr. 2002.

Step (4): For every Pⁿ∈P^N, and for every chromosome P_iⁿ∈Pⁿ, perform jumping operation according to probability p_j on currently iterated parent P¹∈P^N, and randomly selected parent P²∈P^N. A copy of P¹ is generated except the chromosome P_iⁿ, resulting in offspring individual O∈O^K, with modified according to:

• • If P¹≠P², cross-over operation is applied to P_iⁿ according to Equation (16).
  • If P¹=P², mutation operation is applied to P_iⁿ according to Equation (19).

Step (5a): Perform cross-over operation according to probability p_c, on arbitrarily coupled individuals P¹={p₁¹, . . . , p₁⁶}, P²={p₂¹, . . . , p₂⁶}∈P^N. Two off-spring individuals O¹={o₁¹, . . . , o₁⁶} and O²={o₂¹, . . . , o₂⁶}∈O^K are generated based on

$\begin{array}{cc}O\text{?}=\frac{1}{2}\left[\left(1-{\beta }_k\right)p_i^1+\left(1+{\beta }_k\right)p_i^2\right]& \left(16\right)\\ O\text{?}=\frac{1}{2}\left[\left(1+{\beta }_k\right)p_i^1+\left(1-{\beta }_k\right)p_i^2\right]& \left(17\right)\\ \text{?}\text{indicates text missing or illegible when filed}& \end{array}$

where β_k is a sample from distribution prescribed by

$\begin{array}{cc}\beta \left(u\right)=\{\begin{array}{cc}{\left(2u\right)}^{1/{\eta }_c+1}& \mathrm{if}u\le 0.5\\ {\left[2\left(1-u\right)\right]}^{-1/{\eta }_c+1}& \mathrm{otherwise}\end{array}& \left(18\right)\end{array}$

η_c is a parameter. The higher the value of η_c the higher the probability for the off-spring to resemble parents. The u is random number from interval [0,1].

Step (5b): Perform jumping operation according to the probability p_m, for every Pⁿ∈P^N and for its every chromosome P_iⁿ∈Pⁿ, to generate one off-spring individual O∈O^K as copy of Pⁿ, with p_iⁿ exchanged for y_i, defined by

$\begin{array}{cc}{y}_i=\{\begin{array}{cc}{p}_i^n+\left(G_i^{\mathrm{MAX}}-p_i^n\right)& \mathrm{if}\delta \ge 0\\ p_i^n+\left(p_i^n-G_i^{\mathrm{MIN}}\right)& \mathrm{otherwise}\end{array}& \left(19\right)\end{array}$

where δ is a sample from distribution prescribed by

$\begin{array}{cc}\delta \left(u\right)=\{\begin{array}{cc}{\left(2u\right)}^{1/{\eta }_m+1}-1& \mathrm{if}u\le 0.5\\ 1-{\left[2\left(1-u\right)\right]}^{-1/{\eta }_m+1}& \mathrm{otherwise}\end{array}& \left(20\right)\end{array}$

η_m is a parameter and u is a random number from interval [0,1].

Step (6): Fitness evaluation of O^K, where O^K is a pool of offspring individuals generated in Step 4) and 5).

Step (7): G^M is re-established by selecting the fittest M individuals from P^N and O^K, other individuals are discarded.

Step (8): Repeat the Steps (3) to (7) until the generation count reaches the predefined number of generations N_iter.

If any off-spring individual, after cross-over operation, has chromosome outside of the search space limit G^MIN or G^MAX, the chromosome is randomly generated within the limits.

One of the major bottleneck of the optimization process is the evaluation of Equations (9) to (11). In this embodiment, the RJGGA optimization unit can be split into hardware part using FPGA, and software part on ARM processor. Such implementation benefits from parallel acceleration by multiple parallel Solvers deployed in FPGA and ease of software programming.

FIG. 9 illustrates the internal structure of one Solver. The Solver includes input registers, output registers, and an interconnect connected between the input and output registers for controlling data flow. Various arithmetic modules are connected with the interconnect. The Solver is arranged to break down the equations into single mathematical operations. The Solver also has a state machine for controlling the timing and data flow through the interconnect. The Interconnect re-routes the connections among the input and output registers and the floating-point operation cores.

FIG. 10 illustrates the integration of Solvers into RJGGA optimization. As the group of Solvers share the same memory and are controlled by the same Control Logic, the implementation can be referred to as Hardware Accelerator.

The RJGGA interacts with the Hardware Accelerator through individuals G as previously described and illustrated in FIG. 7. Besides the individuals, the input to the Hardware Accelerator is the time series in Equations (3) and (4). The output is the value of the objective function in Equation (12) for evaluated individuals.

Experimental Results

The use of RJGGA-based method for solar cell model identification in the above embodiment is tested on data generated by PSIM model, following the configuration topology outlined in FIG. 11, part (a). The two solar cell models connected to the data acquisition devices are simulated with values listed in Table I.

TABLE I
Solar Cell Model Parameters
Parameter Value
Iph (A)   1
Io (A)    1 × 10−7
νT (V)    5
Rsh (Ω)   1000
Csh (F)   1 × 10−6
Rs (Ω)    1

The data acquisition devices' DC-DC converter is simulated using values C₁=C₂=1 μF and L=150 μH. The reference signal for data acquisition devices excitation process is set at 1 kHz sine wave with peak-to-peak value of 75 V and positive bias 42.5V, that is from 5V to 80V. The electronic load is represented by a constant resistance, as the perturbations are performed at speed, where grid-tie inverter is seen as a static load.

To prove the identification capabilities of the RJGGA-base method in the above embodiment, the search space limits are set to cover a large variety of PV panels. The limits of the RJGGA search space are listed in Table II. Table III states the parameters of both RJGGA and Current Predictor.

TABLE II
RJGGA Search Space Limits
	Parameter	GMIN	GMAX
	Iph (A)	0.1	6.0
	Io (A)	0	1 × 10−3
	νT (V)	3	100
	Rsh (Ω)	100	10000
	Csh (F)	1 × 10−7	1 × 10−5
	Rs (Ω)	0.1	10.0

TABLE III
RJGGA and Current Predictor Parameters
Parameter                   Value
M                           100
n                           50
pj                          0.05
ps                          0.8
pm                          0.2
ηc                          0
ηw                          4
Niter                       4000
trapezoidal rule error      1 × 10−5
trapezoidal rule iterations 50
N1                          100
N2                          300

The diagnostic process is performed 100 times on a single set of terminal voltage and current time series generated by the PSIM simulation. The value of objective function for the best individual within every generation is recorded and plotted in FIG. 12. Table IV lists the mean value μ and standard deviation σ of six solar cell parameters obtained as the result of RJGGA optimization. The value f_obj is included, for which the standard deviation σ shows the consistency of RJGGA optimization.

TABLE IV
Solar Cell Model Identification by RJGGA
	Parameter	μ	σ
	Iph(A)	0.99	7.63 × 10−6
	Io(A)	8.51 × 10−6	1.08 × 10−8
	νT(V)	4.93	5.34 × 10−4
	Rsh(Ω)	983.10	0.36
	Csh(F)	9.99 × 10−7	9.36 × 10−12
	Rs(Ω)	1.12	6.34 × 10−5
	fobj (—)	7.78 × 10−6	2.01 × 10−10

The convergence trend in FIG. 12 shows steady performance with the RJGGA as all 100 performed runs attain final value of objective function below 10⁻⁵. The RJGGA outperforms MPSO algorithm by steadily achieving one order of magnitude lower value of objective function in spite of the wider search space limits.

In A. Sangwongwanich, Y. Yang, D. Sera and F. Blaabjerg, “Lifetime Evaluation of Grid-Connected PV Inverters Considering Panel Degradation Rates and Installation Sites,” IEEE Trans. on Power Electronics, vol. 33, no. 2, pp. 1225-1236, February 2018., researchers examine environmental effects on PV panel degradation over 20 years of active duty. The power generation capabilities between the new and 20 years old PV panels differ up to 20%. The intrinsic parameters reflect the change in power generation output as the identification process is based on matching terminal voltage and current.

Since the degradation process occurs over the years, for practical purposes the tests are realized on a set of PV panels with different levels of power generation, as opposed to the collection of data from a single panel over years.

Two prototypes are developed for experimental verification of the above embodiment. The lists of components used in data acquisition devices and control device are given in Table V and Table VI respectively.

The configuration of the experimental setup is shown in FIG. 11, part (b). The tests are conducted with four different PV panels, P₁, P₂, P₃ and P₄ as shown in FIG. 13. The PV panels are divided into two groups according to their power generation output. The PV panels P₂ and P₃ exhibit relatively lower power output compared to P₁ and P₄, due to physical damage. The grid-tie inverter is represented by an electronic load. The list of devices used in experiments is shown in Table VII.

TABLE V
Components for DAM
Part	Manufacturer	Model number	Specification
Microcontroller	STM	STM32F405RG	168 MHz, 12-bit
			ADC
Op-Amp (V)	TI	INA826AIDR	1 MHz, 100 dB
Op-Amp (I)	TI	INA240A1	400 KHz, 120 dB
MOSFET	STM	STP26NM60N	600 V VDS
Inductor	Eaton	CTX150-5-52	150 μH, 7.7 A
Capacitor	Panasonic	ECQ-E6105KF	1 μF, 630 VDC
Gate driver	Infineon	IR21108PBF	up to 500 V, 2 A
Wireless Mod.	NXP	MC1322V	ZigBee

TABLE VI
Components for CM
Part	Manufacturer	Model number	Specification
ARM-FPGA SoC	Terasic	DE0-nano-S0C	Cyclone V
Wireless Module	NXP	MC1322V	ZigBee
WIFI Module	Ai-Thinker	ESP8266MOD	D1 ESP8266

TABLE VII
List of Devices used in Experimental Verification
Device	Manufacturer	Model no.	Specification
Solar Panel	Sungen	SG-NH80-GG	80 W, a-Si
Light	Philips	QVF137	1 kW
Elec. Load	HP	HP6050A/60503B	250 V, 10 A

An example of data acquired by data acquisition devices for PV panel P₁ is plotted in FIG. 14A. By plotting the current samples against the terminal voltage samples in FIG. 14B, the hysteresis loop is revealed. 16 sets of experimental data similar to those in FIGS. 14A and 14B are obtained for PV panels P₁ to P₄ with the data acquisition devices module.

The diagnostic process is performed 10 times on each of the 16 sets of the acquired data per PV panel. The result is 160 sets of intrinsic parameters for every PV panel. The mean value μ and standard deviation a is listed in Table VIII.

TABLE VIII
Intrinsic Parameter Identification by RJGGA
	P1	P2	P3	P4
Parameter	μ	σ	μ	σ	μ	σ	μ	σ
Iph (A)	1.21	8.10 × 10−3	0.59	5.32 × 10−3	0.45	7.14 × 10−3	1.17	0.01
Io (A)	4.13 × 10−6	4.32 × 10−6	9.83 × 10−10	2.46 × 10−9	1.32 × 10−2	1.56 × 10−7	2.61 × 10−6	3.11 × 10−6
νT (V)	5.72	0.41	4.61	0.38	4.83	0.33	6.13	0.53
Rsh (Ω)	265.98	16.09	332.49	21.14	524.84	35.01	412.02	32.61
Csh (F)	3.00 × 10−7	4.27 × 10−9	2.94 × 10−2	1.49 × 10−2	1.97 × 10−2	5.43 × 10−9	3.53 × 10−2	8.69 × 10−9
Rs (Ω)	5.66	1.00	7.99	1.75	10.75	1.69	8.82	1.64
fobj (—)	8.45 × 10−4	1.10 × 10−4	3.19 × 10−4	4.24 × 10−5	7.65 × 10−5	9.43 × 10−6	3.31 × 10−4	3.02 × 10−5

The standard deviation a in Table VIII shows high confidence for averaged values of the parameters I_ph, ν_T, R_sh, C_sh, and R_s. The I₀ exhibits large standard deviation σ, due to the nonlinearity of the diode D_sh in the solar cell model. The RJGGA may produce outliers by misidentifying the I₀ in case the current through diode i_p(t) is not exhibited enough in the measured data. In such case, the RJGGA identifies wider range of I₀ as fit solution, as the objective function is not able to reflect the difference.

As the verification results later show, the mean value of I₀ can be rectified by removing 10% of the values furthest from the mean. The rectified mean value μ for the parameter I₀ and the standard deviation a are listed in Table IX.

To validate the intrinsic parameters identified by RJGGA, another set of measurements is realized for PV panel P₁ and P₂, representing healthy and damaged PV panel respectively. The measurements differ in perturbation frequency and peak-to-peak amplitude. They are used as the input for averaged solar cell model represented by mean values of intrinsic parameters listed in Table VIII and rectified value of I₀ listed in Table IX.

TABLE IX
Rectified Value of Parameter Io
	P1	P2
	μ	σ	μ	σ
Io	2.99 × 10−6	2.41 × 10−6	4.51 × 10−10	4.41 × 10−10

Table X lists the value of objective function attained for verification measurements

TABLE X
Objective Function Value for Averaged Model Verification
	Measurement Settings		Objective Function Value
	Frequency	Amplitude	P1	P2
	1 kHz	Regular	10.05 × 10−4	7.30 × 10−4
	1 kHz	Low	7.46 × 10−4	5.27 × 10−4
	500 Hz	Regular	14.76 × 10−4	6.36 × 10−4
	500 Hz	Low	5.10 × 10−4	7.38 × 10−4

FIGS. 15A to 15D and 16A to 16D show the input terminal voltage and current and predicted current (following the measurement settings listed in Table X) for P₁ and P₂ respectively. Reversing the identification problem shows the diagnosed intrinsic PV panel parameters fully describe the tested PV panels. However, application of the averaged model to multiple different measurements under different settings would lead to some discrepancies, as expected. This is illustrated by comparing the value of objective function in Table VIII and Table X.

The above embodiments have provided a diagnostic system and method for photovoltaic solar farms, in particular, a system and method for determining a state or condition of a PV panel. The system and method utilizes non-invasive data acquisition by hardware based on DPP concepts. The stochastic algorithm RJGGA is employed on embedded platform, to demonstrate the practical implementation, providing diagnostics on-site at the photovoltaic solar plant. The system and method embodiments facilitate non-invasive, on-line data acquisition, and embedded application of CI methods, to provide effective and efficient diagnostics for PV panels.

The system in the above embodiments operates by acquiring the terminal voltage and current of the diagnosed PV panels under voltage perturbations. The perturbations are generated by attached device, at high frequency, lasting several or several tens of milliseconds. This prevents significant disruption to power generation process, rendering the data acquisition process transparent to regular operation of the photovoltaic solar plant. The diagnostics of the PV panel are formulated as optimization problem and solved by RJGGA on embedded platform. RJGGA is a non-elitist algorithm, with a reduced chance of being trapped into local optima. The outcome is a set of intrinsic parameters reporting the current state of the diagnosed PV panel. Relative comparison of these parameters over time reveals the undergoing degradation process and enables failure prediction and maintenance scheduling. Advantages of the system and method above are their discreet operation, practical applicability, scalability and potential to be integrated within DPP systems. The system and method also follows the trends of Industry 4.0, possibly augmenting existing solutions with advanced optimization techniques. This yields a multipurpose smart electronic system, not only generating power but self-diagnosing and forecasting its future development.

FIG. 17 shows a schematic diagram of exemplary information handling system 200 that can be used as a server or other information processing systems for implementing any part of the data acquisition device, any part of the control device, or any part of the storage. The information handling system 200 may have different configurations, and it generally comprises suitable components necessary to receive, store, and execute appropriate computer instructions, commands, or codes. The main components of the information handling system 200 are a processor 202 and a memory unit 204. The processor 202 may be formed by one or more CPU, MCU, controllers, logic circuits, Raspberry Pi chip, etc. The memory unit 204 may include one or more volatile memory unit (such as RAM, DRAM, SRAM), one or more non-volatile unit (such as ROM, PROM, EPROM, EEPROM, FRAM, MRAM, FLASH, SSD, NAND, and NVDIMM), or any of their combinations. Preferably, the information handling system 200 further includes one or more input devices 206 such as a keyboard, a mouse, a stylus, an image scanner, a microphone, a tactile input device (e.g., touch sensitive screen), and an image/video input device (e.g., camera). The information handling system 200 may further include one or more output devices 208 such as one or more displays (e.g., monitor), speakers, disk drives, headphones, earphones, printers, 3D printers, etc. The display may include a LCD display, a LED/OLED display, or any other suitable display that may or may not be touch sensitive. The information handling system 200 may further include one or more disk drives 212 which may encompass solid state drives, hard disk drives, optical drives, flash drives, and/or magnetic tape drives. A suitable operating system may be installed in the information handling system 200, e.g., on the disk drive 212 or in the memory unit 204. The memory unit 204 and the disk drive 212 may be operated by the processor 202. The information handling system 200 also preferably includes a communication module 210 for establishing one or more communication links (not shown) with one or more other computing devices such as servers, personal computers, terminals, tablets, phones, or other wireless or handheld computing devices. The communication module 210 may be a modem, a Network Interface Card (NIC), an integrated network interface, a radio frequency transceiver, an optical port, an infrared port, a USB connection, or other wired or wireless communication interfaces. The communication links may be wired or wireless for communicating commands, instructions, information and/or data. Preferably, the processor 202, the memory unit 204, and optionally the input devices 206, the output devices 208, the communication module 210 and the disk drives 212 are connected with each other through a bus, a Peripheral Component Interconnect (PCI) such as PCI Express, a Universal Serial Bus (USB), an optical bus, or other like bus structure. In one embodiment, some of these components may be connected through a network such as the Internet or a cloud computing network. A person skilled in the art would appreciate that the information handling system 200 shown in FIG. 17 is merely exemplary and different information handling systems 200 with different configurations may be applicable in the invention.

Although not required, the embodiments described with reference to the Figures can be implemented as an application programming interface (API) or as a series of libraries for use by a developer or can be included within another software application, such as a terminal or personal computer operating system or a portable computing device operating system. Generally, as program modules include routines, programs, objects, components and data files assisting in the performance of particular functions, the skilled person will understand that the functionality of the software application may be distributed across a number of routines, objects or components to achieve the same functionality desired herein.

It will also be appreciated that where the methods and systems of the invention are either wholly implemented by computing system or partly implemented by computing systems then any appropriate computing system architecture may be utilized. This will include stand-alone computers, network computers, dedicated or non-dedicated hardware devices. Where the terms “computing system” and “computing device” are used, these terms are intended to include any appropriate arrangement of computer or information processing hardware capable of implementing the function described.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the scope of the invention as defined in the claims. The described embodiments of the invention should therefore be considered in all respects as illustrative, not restrictive.

For example, in the system of FIG. 1, the number of data acquisition devices, control device, and storage may be different than the numbers shown. The system may include one control device for communicating with one or more data acquisition devices, or multiple control devices each for communicating with one or more respective data acquisition devices. The first communication link between the data acquisition device and the control device can be wired or wireless. Preferably, the first communication link is wireless, in the form of ZigBee, Bluetooth, NFC, RF, Wi-Fi, cellular, or the like. The second communication link between the control device and the storage can be wired or wireless. Preferably, the second communication link is also wireless, in the form of ZigBee, Bluetooth, NFC, RF, Wi-Fi, cellular, or the like. The communication between the data acquisition devices and the control device can occur periodically, intermittently or continuously. The communication between the control device and the storage can occur periodically, intermittently or continuously. The storage can alternatively be on-site, or implemented partly on-site and partly off-site. The control device can alternatively be off-site, or implemented partly on-site and partly off-site. The system of FIG. 1 can be used in other types of solar power plants.

The method of FIG. 2 can be modified. For example, the triggering of measurement may be initiated by the data acquisition devices. The analysis may be performed by one or more control devices. The data can be stored locally in the control devices. The step of communicating data to the storage may be omitted. The method may include further steps such as analyzing the time series of the intrinsic parameters to determine a past, present, or future state of the PV panels.

Various modifications can be made to the data acquisition devices, control devices, and the storage, without departing from the scope of the invention as defined by the claims. For example, the response signals need not be sampled for processing.

Claims

1. A system for determining a state of a photovoltaic panel, comprising: a data acquisition device having: a circuit arranged to transmit excitation signals to a photovoltaic panel and detect response signals generated by the photovoltaic panel in response to the excitation signal, during normal operation of the photovoltaic panel; anda communication module arranged to communicate the response signals to a control device for analysis and determination of a state of the photovoltaic panel.

2. The system of claim 1, wherein the excitation signals comprise a voltage signal with a non-zero perturbation frequency.

3. The system of claim 2, wherein each excitation signal has a duration in the order of milliseconds.

4. The system of claim 2, wherein the excitation signals comprise a plurality of voltage signals each having a respective non-zero perturbation frequency.

5. The system of claim 4, wherein the response signals comprise a terminal voltage across the photovoltaic panel and an output current of the photovoltaic panel.

6. The system of claim 5, wherein the excitation signals are arranged to drive a terminal voltage of the photovoltaic panel from minimum voltage level to maximum voltage level as bounded by at least one of: a current level of degradation or irradiation of the photovoltaic panel, andan operation point of a grid-tie inverter operably connected with the photovoltaic panel.

7. The system of claim 1, wherein the data acquisition device further includes a sampler for sampling the detected response signals and the communication module is arranged to communicate the sampled detected response signals to the control device.

8. The system of claim 1, wherein the data acquisition device includes a memory for storing the detected response signals.

9. The system of claim 5, wherein the circuit is arranged to be connected with two photovoltaic panels for transmitting respective excitation signals and detecting respective response signals from the two photovoltaic panels.

10. The system of claim 9, wherein the circuit includes: a power converter with two switches;a driving circuit for providing gating signals to operate the two switches;a controller for controlling the driving circuit based on a difference between the detected signal of one of the two photovoltaic panels and a reference signal;wherein the driving circuit is arranged to operate the two switches complementarily; andwherein the controller is arranged to control the driving circuit so as to control the detected signal of the one of the two photovoltaic panels to follow the reference signal.

11. The system of claim 10, wherein the reference signal comprises a reference voltage signal with a non-zero perturbation frequency, and the detected signal of one of the two photovoltaic panels comprises a terminal voltage across the one of the two photovoltaic panels.

12. The system of claim 11, wherein the power converter comprises a DC-DC converter.

13. The system of claim 12, wherein the DC-DC converter comprises a buck-boost converter arranged to operate in continuous conduction mode.

14. The system of claim 1, wherein the communication module comprises a wireless communication module.

15. The system of claim 5, further comprising the control device, wherein the control device includes a communication module for communicating with the data acquisition device.

16. The system of claim 15, wherein the communication module of the control device comprises a wireless communication module.

17. The system of claim 15, wherein the control device further comprises a processing unit arranged to process the response signals to determine one or more intrinsic parameters indicative of the state of the photovoltaic panel.

18. The system of claim 17, wherein the processing unit is arranged to process the response signals by: matching the response signals of the photovoltaic panel with a predetermined model for determining the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

19. The system of claim 18, wherein the predetermined model comprises a dynamic single-diode model of a solar cell with: a current source providing a current based on incident light;a diode connected in parallel with the current source;a capacitor connected in parallel with the current source;a first resistor connected in parallel with the current source; anda second resistor connected in series with the first resistor.

20. The system of claim 19, wherein the processing unit includes: a current predictor for predicting a terminal current generated in the dynamic single-diode model based on the measured terminal voltage of the photovoltaic panel and a set of parameters in the dynamic single-diode model.

21. The system of claim 20, wherein the set of parameters in the dynamic single-diode model comprises one or more of: a current Iph provide by the current source in the dynamic single-diode model;a reverse saturation current I0 in the dynamic single-diode model;a thermal voltage νT in the dynamic single-diode model;a resistance Rsh of the first resistor in the dynamic single-diode model, indicative of an intrinsic p-n junction resistance;a capacitance Csh of the capacitor in the dynamic single-diode model, indicative of an intrinsic p-n junction capacitance; anda resistance Rs of the second resistor in the dynamic single-diode model.

22. The system of claim 20, wherein the processing unit further includes an optimization unit, operably connected with the current predictor, for determining one or more values of an objective function of the set of parameters in the dynamic single-diode model.

23. The system of claim 22, wherein the one or more values of the objective function comprises the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

24. The system of claim 22, wherein the optimization unit is arranged to determine one or more values of the objective function of the set of parameters by iteratively reducing a difference between the predicted terminal current generated in the dynamic single-diode model and the detected output current of the photovoltaic panel.

25. The system of claim 22, wherein the optimization unit is arranged to determine one or more values of the one or more intrinsic parameters indicative of the state of the photovoltaic panel using a real jumping gene genetic algorithm-based method.

26. The system of claim 17, wherein the control device further comprises a further communication module for communicating the one or more determined intrinsic parameters to a remote storage.

27. The system of claim 26, wherein the further communication module comprises a wireless communication module.

28. The system of claim 26, further comprising the remote storage arranged to store the intrinsic parameters determined by the control device.

29. The system of claim 28, wherein the remote storage comprises a server.

30. A method for determining a state of a photovoltaic panel, comprising: transmitting excitation signals to a photovoltaic panel during normal operation of the photovoltaic panel;detecting response signals generated by the photovoltaic panel in response to the excitation signal; andcommunicating the response signals to a control device for analysis and determination of a state of the photovoltaic panel.

31. The method of claim 30, wherein the excitation signals comprise a plurality of voltage signals each having a respective non-zero perturbation frequency; and the response signals comprise a terminal voltage across the photovoltaic panel and an output current of the photovoltaic panel.

32. The method of claim 31, further comprising: sampling the detected response signals; andthe communicating step comprises communicating the sampled detected response signals to the control device.

33. The method of claim 32, further comprising: storing the detected response signals.

34. The method of claim 31, further comprising: processing the response signals to determine one or more intrinsic parameters indicative of the state of the photovoltaic panel.

35. The method of claim 34, wherein the step of processing comprises: matching the response signals of the photovoltaic panel with a predetermined model for determining the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

36. The method of claim 35, wherein the predetermined model comprises a dynamic single-diode model of a solar cell with: a current source providing a current based on incident light;a diode connected in parallel with the current source;a capacitor connected in parallel with the current source;a first resistor connected in parallel with the current source; anda second resistor connected in series with the first resistor.

37. The method of claim 34, wherein the step of matching comprises: predicting a terminal current generated in the dynamic single-diode model based on the measured terminal voltage of the photovoltaic panel and a set of parameters in the dynamic single-diode model.

38. The method of claim 37, wherein the set of parameters in the dynamic single-diode model comprises one or more of: a current Iph provide by the current source in the dynamic single-diode model;a reverse saturation current I0 in the dynamic single-diode model;a thermal voltage νT in the dynamic single-diode model;a resistance Rsh of the first resistor in the dynamic single-diode model, indicative of an intrinsic p-n junction resistance;a capacitance Csh of the capacitor in the dynamic single-diode model, indicative of an intrinsic p-n junction capacitance; anda resistance Rs of the second resistor in the dynamic single-diode model.

39. The method of claim 37, wherein the step of matching further comprises: determining one or more values of an objective function of the set of parameters in the dynamic single-diode model, wherein the one or more values of the objective function comprises the one or more intrinsic parameters indicative of the state of the photovoltaic panel.

40. The method of claim 39, wherein the determination of the one or more values of the objective function of the set of parameters comprises: iteratively reducing a difference between the predicted terminal current generated in the dynamic single-diode model and the detected output current of the photovoltaic panel.

41. The method of claim 39, wherein determination of the one or more intrinsic parameters indicative of the state of the photovoltaic panel is performed using a real jumping gene genetic algorithm-based method.

42. The method of claim 39, further comprising: communicating the one or more intrinsic parameters indicative of the state of the photovoltaic panel to a remote storage.

43. The method of claim 39, further comprising: analyzing a time series of the one or more intrinsic parameters to determine a predicted state of the photovoltaic panel.