Patent Application
20190252880
The present invention relates to a trading planning apparatus and a trading planning method for laying down an order plan to each of exchanges and trade connections for an energy business operator.
There is known a system for planning an order to an exchange to meet supply in response to supply destinations' demand that changes on a daily basis. A system described in Patent Document 1 is designed to determine a bid quantity and a bid price in an electricity market from a power generation plan created on the basis of demand estimate data and power supply data. It is thereby possible to conduct bidding uniquely assuming a price in the electricity market.
There is also known a system for conducting trading in a market in a constantly changing market condition. Patent Document 2 describes a system in which while a trading quantity is split into a plurality of slices and each slice is made to be associated with a time slot of trading time to automatically execute an order, the system being designed to determine whether an order quantity satisfies an order schedule change condition preset on the basis of market data and if the order quantity satisfies the order schedule change condition, to increase or reduce a planned execution rate by a predetermined value at each order timing for flexibly and automatically adjusting an order quantity in response to a market trend. It is thereby possible to conduct trading that satisfies a preset total trading quantity.
However, the system of Patent Document 1 gives no consideration to the fact that the price of energy to be traded on date and hour basis varies during a trading period; thus, with the system, it is impossible to conduct rational trading when the price of the energy to be traded during the trading period changes with a change in weather forecast data that influences demand and a change in prediction data about states of power generation facilities and electric transmission and distribution facilities.
Furthermore, with the system of Patent Document 2, it is impossible to appropriately place an order for trading for which a quantity demanded is not defined yet at a timing of start of the trading period. Moreover, the system of Patent Document 2 gives no consideration to changes in the demand and the market condition of the energy to be traded influenced by climate and operating statuses of the power generation facilities and the electric transmission facilities; thus, it is impossible to place an order responding to the changes in the demand and the market condition in the light of such physical influences thereon.
Therefore, any of the techniques as background art has a problem that it is impossible to appropriately place an order related to trading in response to the demand that is unlikely to be defined during the trading period due to the climate influence and the influence of operation of the power generation facilities and the electric transmission facilities, to the exchange or the trade connection.
The present invention has been proposed to solve the problems of background art described above, and an object of the present invention is to provide a trading planning apparatus and a trading planning method capable of dynamically changing an order condition depending on a market trend, a demand fluctuation, and an operational status of a trade connection. The present invention makes it possible to determine orders to exchanges and trade connections different in delivery timing or in price determination period and to determine an order quantity and order timing for each trading while the demand and the market trend gradually become clear.
(1) A trading planning apparatus according to the present invention includes an order quantity planning section including a trading position determination section receiving a trading cumulative quantity and estimated data about future quantities related to demand and a plurality of trade connections, and determining trading quantities with the plurality of trade connections.
Executing such a trading plan makes it possible to execute a plan of trading in response to constantly changing future quantities.
(2) The order quantity planning section of the trading planning apparatus according to (1) includes a data table in which positive and negative values can be taken as data related to each of a plurality of types of trading.
Executing such a trading plan makes it possible to simultaneously conduct sell trading and buy trading with respect to commodities having the same delivery period (delivery deadline or period for continuously providing a service). Preferably, buying energy delivered over a span of several hours in wholesale trading in response to peak load electricity supply and selling and providing a residual of the supply generated in time zones other than a peak load electricity supply time zone in wholesale trading makes it possible to accelerate the efficient use of energy.
(3) The trading planning apparatus according to the present invention includes a split-time-based split order planning section including a trading and ordering data determination section that generates trading and ordering data containing data about an order price (for example, a price of a buy bid or a price of a sell bid) or an order quantity related to trading and ordering in each of planned trading periods that are periods into which a trading period, during which trading can be conducted, is subdivided.
Executing such a trading plan makes it possible to execute a plan of efficient trading in each time section of the trading period.
(4) The trading planning apparatus according to the present invention includes a future quantity estimation section including a convergence estimation section estimating estimated quantities of data (for example, dispersion or likelihood values) related to errors in the future quantities estimated from actual record values.
Performing such future quantity estimation makes it possible to execute a plan of trading (for example, an order quantity plan and an order process time splitting plan) in response to a situation as to whether estimation accuracy of the future quantities is favorable or not favorable.
(5) The order quantity planning section of the trading planning apparatus according to the present invention includes the trading position determination section increasing or reducing the trading quantity of any of the trade connections (on the basis of an estimated value of a trading price or a tradable quantity with each of the trade connections) when a difference between a sum of the trading quantities of all of the trade connections and a future quantity demanded is equal to or greater than a predetermined value.
(6) Preferably, the trading position determination section increases a quota of trading with one certain trade connection in a case in which an estimated value of the trading price with the certain trade connection is lower than the trading price related to trading cumulation of all types of trading, and reduces the quota of the trading with the certain trade connection in a case in which the estimated value of the trading price with the certain trade connection is higher than the trading price related to the trading cumulation of all types of trading.
(7) More preferably, the trading position determination section reduces a trading quantity quota to any of the trade connections for which a value of data related to an error (dispersion or likelihood) in a trading price of each of the trade connections increases, and increases the trading quantity quota to any of the trade connections for which the data related to the error decreases.
Executing such a trading plan makes it possible to place an order that meets demand with economic rationality. Preferably, executing such a trading plan makes it possible to place an order for which a loss likely to be generated due to fluctuations in various future quantities is mitigated.
(8) The order quantity planning section of the trading planning apparatus according to the present invention includes a trading position determination section calculating a combination of trading on the basis of an efficient frontier calculated from expected returns by trading and a risk of the expected returns (for example, a value of dispersion of the expected returns), calculating a latter trading efficient frontier from data in a range in which future expected returns or a future expected return dispersion changes in a latter period of the trading period, and determining the combination of trading with the plurality of trade connections in such a manner that a portfolio (procurement and sales proportions of power generation and electricity commodities) in the vicinity of the efficient frontier can be changed to a portfolio in the vicinity of the latter efficient frontier.
Executing such a trading plan makes it possible to conduct trading throughout the trading period in a case in which it is predicted that estimation related to the future quantities (for example, an estimated quantity of a quantity demanded at appointed time on an appointed day at which supply is provided or of a marketing price of trading related to the time, or estimated quantities of dispersion values thereof (convergence estimated quantities)) differs between initial estimation and latter estimation, and in which a plan result, which is a target, changes from a result of an optimum trading plan at initial timing to a result of an optimum trading plan at latter timing.
More preferably, the trading planning apparatus according to the present invention defers order timing as evaluation values of the estimated values related to the future quantities in the time course of convergence are larger.
Executing such a trading plan makes it possible to conduct trading in accordance with a portfolio with the highest economic efficiency even in a case in which a change quantity of the latter efficient frontier is large.
(9) The trading planning apparatus according to the present invention includes the split-time-based split order planning section including the trading and ordering time splitting section that splits an order quantity to each of the trade connections into target values related to temporal transitions of an order during the trading period.
Executing such a trading plan makes it possible to execute a plan of trading that efficiently secures a quantity by which a sales business operator finally supplies commodities to customers by sequentially changing the order quantities to the trade connections even in a case in which the estimated values of the future quantities cannot be defined yet.
(10) Preferably, the trading planning apparatus according to the present invention includes a split-time-based split order planning section that increases or reduces a value of order data in each of the planned trading periods depending on a magnitude of an error (dispersion or likelihood) in future quantity estimation related to each planned trading period, with respect to orders in the planned trading periods into which the trading period is subdivided.
Executing such a trading plan makes it possible to preferentially execute trading with a minor error and to execute a trading plan that stably realizes a target trading quantity and target trading returns.
(11) Preferably, the trading planning apparatus according to the present invention includes the split-time-based split order planning section including a trading and ordering data determination section generating data about a target trading quantity in each of the planned trading periods at predetermined intervals generated on the basis of data about a target order transition (for example, data generated by splitting the trading period of commodities the trading period of which is 48 hours to 24 hours before time of delivery to set 10-minute planned trading periods, and setting a target value of the trading quantity to be completed in each of the trading periods), and creating the trading and ordering data containing a price obtained by performing weight addition between an estimated trading price and an estimated trading price error in response to a difference between the target trading quantity and a trading quantity actual record.
Determining such trading and ordering data makes it possible to execute a trading plan for completing trading of a necessary trading quantity.
For example, a bid price is determined by the present method when a procurement or sales quantity is to be secured with an eye on gate closure of the trading market. In a case of collecting the procurement quantity as a buyer, it is possible to determine the bid price in bidding based on the present method as follows. The bid price that enables a contract of the necessary quantity is determined on the basis of a statistical tendency that buy bidding at the bid price of a value P+σ(P), where P is an expected price and σ is a dispersion thereof, can lead to a successful bid at a probability of 95% and that bidding at the bid price of a value P−σ(P) can lead to a successful bid at a probability of 95%. More preferably, it is possible to appropriately change the bid price to a low value in the initial stage of trading in which a contracted quantity may be small, and to a high value such as the bid price P+σ in a stage in which the trading quantity is necessary. Furthermore, in a case of a power generation seller, it is possible to appropriately change the bid price to a relatively high value at timing at which the contracted quantity may be small and to a relatively low value such as P−σ with an aim of more successful bidding when the seller desires to completely sell power generated in accordance with a shutdown constraint on generators.
(12) The trading planning apparatus according to the present invention includes the trading and ordering data determination section generating data related to an amount of money of trading conducted in each of the planned trading periods at the predetermined intervals on the basis of the estimated data about the future quantities and creating the trading and ordering data.
Preferably, the trading planning apparatus according to the present invention includes a split-time-based split order planning section including the trading and ordering data determination section that generates the trading and ordering data from the data about the amount of money of the trading obtained by adding or subtracting a value, which is obtained by multiplying an estimated quantity of an error in estimated quantities of the future quantities by a coefficient, to or from the amount of money of the trading.
Such a trading plan makes it possible to execute a plan of trading that prevents an excessive payment amount by a trading cost leveling effect even in a case of generation of fluctuations in the future quantities (for example, a fluctuation in the trading price or a fluctuation in the quantity demanded) during the whole trading period.
More preferably, a plan is laid down such that a trading amount of money including an allowance for risk (value obtained by subtracting the value, which is obtained by multiplying the dispersion value p by the coefficient, from an expected trading amount of money Q) is made uniform among the planned trading periods as the amount of money of the trading conducted in each of the planned trading periods at the predetermined intervals. Such a trading plan makes it possible to execute a trading plan such that an amount of money to be used is set high when errors in the estimated quantities of the price and the quantity of trading are minor and the trading amount of money is set low when the errors are significant (market risk is high), and to conduct efficient trading irrespective of the market risk.
(13) The trading planning apparatus according to the present invention includes the order quantity planning section that determines buy order quantities of commodities overlapping in time of delivery (supply time) (to result in cancellation of actual supply) and a trading position in a range of opposite position limitation to limit a sell order quantity.
Imposing such limitation on the order quantities makes it possible to plan trading that can meet the actual supply without an unlimited increase in the trading quantity.
For example, it is possible to determine arbitrage trading and appropriate quantities thereof in a case in which the trade connections or the trading markets differ. In a case of conducting the arbitrage trading such that at a time of occurrence of a state in which prices of equivalent electricity commodities (for example, electricity commodities that can be used to be supplied in the same time zone) take different values (two prices for one commodity) in different exchange markets, the electricity commodity is procured in the low price market and sold in the high price market, conducting the arbitrage trading without price ceilings disturbs business operation because of a financial loss realized when the predicted price is a miss. According to the trading planning of the present invention, it is possible to execute a plan of trading that enables the realized financial loss to fall in an allowable range.
For example, in trading (cross trading) of buying 4-hour block power generation and selling a commodity in a 30-minute spot or intraday market, a ceiling set on the opposite position is limited to prevent an increase in the trading quantity without price ceilings by the cross trading. For example, evaluation is performed using returns including an allowance for risk. Alternatively, a ceiling of a predetermined value is set, or a value obtained by multiplying actual supply by predetermined N is set as a trading ceiling.
(14) The trading planning apparatus according to the present invention includes the trading and ordering data determination section calculating a weighted addition value between data about a price (for example, closing price or weighted average cost of capital (WACC)) during the trading period and data related to convergence, comparing the calculated weighted addition value with a seller bid price in a market to determine a magnitude relationship, and generating the trading and ordering data.
Such a trading plan makes it possible to execute a plan of trading with overall rationality by prompt response to an offer (bid) of buying commodities needed by the other business operator urgently at a high price and power interchange of residual commodities at a low price via the market and with economic rationality for business operators in charge of trading.
According to the present invention, it is possible to execute a trading plan in response to a market trend, a demand fluctuation, and an operational status of each trade connection, and to determine a quantity of an order to the trade connection and the market, and order timing as a result of execution of the trading plan.
A plurality of embodiments of a trading planning system according to the present invention will be described hereinafter with reference to the drawings.
The future quantity estimation section 10 acquires actual record data from a plurality of demand systems (1 to N) 1000, a power generation business operation system (1) 2000, a market A system (including sales business operation systems (2 to L) 4000 and power generation business operation systems (2 to K) 2000) 3000, and a market B system 3100, estimates future quantities related to demand and trade connections in each of the systems, and outputs estimated data to the order quantity planning section 20. In the present embodiment, in particular, the future quantity estimation section 10 includes the convergence estimation section 104 that estimates estimated quantities of data (for example, dispersion and likelihood values) related to errors in future quantities estimated from actual record values.
The order quantity planning section 20 includes the trading position determination section 201 receiving the estimated data about the future quantities related to the demand and the trade connections from the future quantity estimation section 10, determining trading quantities with a plurality of trade connections, and outputting determined contents to the split-time-based split order planning section 30, and the trading cumulative quantity storage section 202 storing a cumulative quantity of trading conducted so far on the basis of contract data from a market ordering terminal 5000, power generation plan data from a power generation ordering terminal 5100, and demand plan data from an aggregator ordering terminal 5200.
The split-time-based split order planning section 30 includes the trading and ordering time splitting section 301 generating planning-period-based data such as quantities of trading and ordering or amounts of money to be used in planned trading periods which are periods into which a trading period, during which trading can be conducted, is subdivided, a trading and ordering data determination section 302 generating order telegraphic messages (messages each containing data about a price and a quantity of a buy or sell order) to the markets or the trade connections, and that outputs the generated order telegraphic messages (order data) to the market ordering terminal 5000, a power generation plan processing section 303 that generates power generation plan data on the basis of the planning-period-based data generated by the trading and ordering time splitting section 301 and outputting the generated power generation plan data to the power generation ordering terminal 5100, the electric storage etc. demand plan processing section 304 generating demand plan data on the basis of the planning-period-based data generated by the trading and ordering time splitting section 301 and outputting the generated demand plan data to the aggregator ordering terminal 5200, and the target order transition table 305 storing the planning-period-based data generated by the trading and ordering time splitting section 301.
In
(Step S101) The demand fluctuation estimation section 101 estimates a future quantity demanded of electricity consumed by the customers to which the electricity is supplied. In this example, the demand fluctuation estimation section 101 splits a future period into 30-minute periods, and estimates the quantity demanded in each of the periods. The demand fluctuation estimation section 101 estimates the future quantity demanded on the basis of actual record data about past demand. For example, the demand fluctuation estimation section 101 may select a demand curve of similar demanded days similar in day of week, calendar day, and weather data, generate a multiple regression prediction model of a daytime maximum, daytime minimum, daytime average, or maximal/minimal quantity demanded from the weather data, predict a daytime maximum, daytime minimum, daytime average, or maximal/minimal quantity demanded from weather forecast data, and correct the demand curve so as to predict the future quantity demanded. Alternatively, the demand fluctuation estimation section 101 may use past demand temporal fluctuations as time series data and perform time series prediction using an autoregression model.
(Step S102) The supply fluctuation estimation section 102 estimates a future quantity of data related to power generation of the power generation business operator who supplies at wholesale the sales business operator with electricity which is supplied to the customers by the sales business operator. The present process includes a process for estimating a future quantity of a quantity supplied of photovoltaic power generation and wind power generation renewable energy. The supply fluctuation estimation section 102 estimates the future quantity of the quantity supplied on the basis of actual record data about past power generation. For example, the supply fluctuation estimation section 102 may select a power generation curve of similar demanded days similar in weather data, generate a multiple regression prediction model of daytime maximum, daytime minimum, daytime average, or maximal/minimal electricity generated from the weather data, predict daytime maximum, daytime minimum, daytime average, or maximal/minimal electricity generated from the weather forecast data, and correct the power generation curve so as to predict the future quantity of the quantity supplied. Alternatively, the supply fluctuation estimation section 102 may use past power generation temporal fluctuations as time series data and perform time series prediction using an autoregression model.
Preferably, the process performed by the supply fluctuation estimation section 102 according to the present embodiment includes a process for estimating future quantities of thermal power generation and pumped storage generation controllable quantities (electricity generated that can be increased and electricity generated that can be reduced within 30 minutes by issuing a control request) as data related to power generation data.
(Step S103) The market fluctuation prediction section 103 estimates future quantities of data about a market price, the number of bids, and bid electric energy in a wholesale market where the sales business operator procures the electricity to be supplied to the customers (these pieces of data are handled as continuous quantities or quantities of discrete values). The market fluctuation prediction section 103 estimates the future quantities of the data on the basis of past market data. For example, the market fluctuation prediction section 103 may select a price curve, a bid quantity curve, and a bid electric energy curve of similar days similar in data about a day of week, a calendar day, weather, an expected quantity demanded, and a capacity of generators in planned non-operation, generate a multiple regression prediction model related to a daytime maximum, daytime minimum, daytime average, or maximal/minimal value of each of the curves, calculate a multiple regression model of each of the curves on the basis of expectation data about the day of week, the calendar day, the weather, the expected quantity demanded, and the capacity of the generators in planned non-operation, and correct the respective curves so as to predict the future quantities of the data. Alternatively, the market fluctuation prediction section 103 may use past data temporal fluctuations as time series data and perform time series prediction using autoregression models.
Furthermore, the market fluctuation prediction section 103 according to the present embodiment preferably includes an estimation section that estimates future quantities of free capacities of electric transmission lines related to transmission of electricity procured from power plants, an estimation section that estimates future quantities of a price and a quantity of a reserve-related power generation right (reserve market trading) bought by the electric transmission business operator, an estimation section that estimates a future quantity of a price of the ancillary service (service for replenishing a difference between the electricity generated and the quantity demanded) for solving an imbalance by the electric transmission business operator, an estimation section that estimates a future quantity of negawatt power trading, and an estimation section that estimates future quantities of fuels (marketing prices and futures trading prices of liquefied natural gas (LNG) and crude oil), and performs processes for estimating these future quantities. It is thereby possible to lay down a trading plan including an electric transmission reservation right, provision of the power generation right to a reserve, use of the ancillary service, and the use of the negawatt power.
(Step S104) The convergence estimation section 104 estimates transitions of errors for the estimated quantities of the future quantities. In this example, the convergence estimation section 104 splits the future period into the 30-minute periods, and estimates errors in the estimated quantities in each of the periods. The convergence estimation section 104 estimates the errors on the basis of past actual record data. For example, the convergence estimation section 104 uses past actual record data temporal fluctuations as time series data and performs time series prediction using a dispersion autoregression model.
Preferably, the convergence estimation section 104 may subdivide the past actual record data into pieces of data at intervals of predetermined periods (for example, intervals of 24 hours, 48 hours, or one week), perform fast Fourier transform or wavelet transform on each of the subdivided pieces of data to classify resultant pieces of data by a period similar in feature amount as a periodic fluctuation, and extract a periodic fluctuation pattern of each of groups into which the piece of data is classified (calculate an average value of feature amounts by inverse transform). In addition, the convergence estimation section 104 may generate an identification tree for identifying conditions (attributes) for generating the patterns from attributes (the day of week, the calendar day, temperature, sunshine, other weather data, the number of generators in planned shutdown, the free electric transmission capacities, a demand predicted value, and the like) common to each of the groups into which the piece of data is classified using an identification algorithm such as CART (Classification and Regression Trees) or ID3 (Iterative Dichotomiser 3), estimate a plurality of pattern candidates to be generated in the future period from the identification tree, and combine the estimated patterns to obtain a frequency distribution of the future quantities to be generated. It is thereby possible to appropriately estimate the future quantities including an irregular fluctuation called, for example, a market price spike, and to lay down a trading plan in the light of occurrence of the spike (trading plan in the light of a change in the efficient frontier because of the presence of the spike).
It is assumed herein that a prediction error is reduced (a tail of the frequency distribution is made narrower to increase the likelihood) by the advancement of estimation execution timing from the time t1 in the phase Ph2 to the time t2 in the phase Ph3.
It is more preferable in the present embodiment to use values of the errors (dispersion values or likelihoods) as time series data and to predict a change (convergence) in the values.
In a process performed by the order quantity planning section 20 in Steps S201 and S202, the order quantity planning section 20 determines an order quantity, order destinations, and types of order commodities for electricity procurement necessary to supply electricity from data about the estimated values of data related to the demand, the markets, and various types of power generation (thermal power generation, hydro power generation, and photovoltaic power generation) operated by the sales business operator under relative contracts and data about the errors (dispersions and likelihoods) in the estimated values. Details of the process performed by the order quantity planning section 20 will next be described.
(Step S201) The trading position determination section 201 performs a process for determining a trading position that indicate values of trading quantities (quantities of orders placed or quantities of orders taken) with respect to the trade connections per 30-minute delivery time from the trading cumulative quantity and the estimated data about the future quantities related to the demand and the trade connections (the power generation business operators in charge of wholesale supply, the other business operators in charge of wholesale procurement, and the markets). The trading position determination section 201 determines the trading position from a portfolio that indicates planned allocation values of operation funds. The portfolio indicates proportions of the operation funds allocated to a plurality of risk-free assets and risk assets. In the present embodiment, the trading position determination section 201 plans a portfolio with the electricity generated under an agreement reached with trade connection power plants and electricity commodities traded in the markets. The efficient frontier (which is the efficient and best portfolio and also referred to as “efficient frontier”) means herein a feasible portfolio, which satisfies three conditions that (1) the portfolio is feasible satisfying constraints on the quantity supplied such as capacities of the generators and physical constraints in operational constraints such as startup time, shutdown time, minimum shutdown time, and minimum operation time, (2) the portfolio has a maximum evaluation value of an amount of periodic returns with the same risk value, and (3) an obtained amount of periodic returns is equal to or higher than an amount of periodic returns of a portfolio having a lower evaluation value of the risk.
It is assumed that a return evaluation period in the present embodiment is one week including a day of delivery of electricity to the customers, and portfolios that satisfy a feasible solution within the physical constraints (feasible solution to a generator startup/shutdown plan and output power allocation of one week) are evaluated.
The Pf1 denotes one portfolio that holds proportions of electricity by the contract generators in a total quantity supplied per day, 4-hour block electricity, 30-minute electricity by day-ahead trading, 30-minute electricity by intraday trading, and an operation reserve fund as 6:1:1:1:1 in the feasible solution obtained as trading per 30-minute unit. Likewise, the Pf2 allocates the operation funds per day thereto at proportions of 4:3:1:1:1, the Pf3 allocates the operation funds per day thereto at proportions of 1:6:1:1:1, and the Pf4 allocates the operation funds per day thereto at proportions of 1:2:2:4:1.
It is noted that the portfolio may hold data as allocation proportions related to power generation, power selling, and power purchase in the supply kWh as an alternative to the present embodiment.
In an example of
The trading position determination section 201 calls the trading position per 30 minutes that realizes the portfolio which the trading position determination section 201 is instructed to select from the simulation result described above, and stores the trading position in a data table T1 of
In the present embodiment, a simulation is performed with the position including the evaluation values as negative values. This signifies that sell bidding is conducted in the market. In the example of the data table T1 of
(Step S202) The trading cumulative quantity storage section 202 receives and records trading cumulative data contracted in the markets and data about a plan agreement result with the power generation business operators that are the trade connections and an aggregator business operator supplying the negawatt power, with respect to the trading position. A recording result is output to a user through the input/output interface 70.
The description given so far is the process performed by the order quantity planning section 20. As a result of this process, a target value of the electricity generated procured from the trade connections for supply time is determined per delivery time zone (per delivery unit) as data with a tag identifying the types of commodities, the trade connections, and the types of power generation.
In the present embodiment, Steps S201 and S202 are repeatedly executed in the Phases Ph1 to Ph6 depicted in
For example, by executing the process (Step S201) in the phase Ph3, a portfolio of
At this time, in Step S201, the process is not changed in a case in which the efficient frontier for which the portfolio is to be selected matches the efficient frontier at previous execution time. In a case in which the efficient frontier differs from that at the previous execution time (for example, the efficient frontier differs between
In an example depicted in
“A reduction in the market price resulting from an increase in the supply of the photovoltaic renewable energy and an increase in sell bidding of the thermal power generation by the power generation business operators that have had extra supply capabilities in a 30-minute commodity intraday market, and an increase in the dispersion value of the market price due to an increase in bidding of the renewable energy power generation highly dependent on the weather occur”; thus, changes in the estimated values of market-related future quantities occur. The portfolio that satisfies the condition as a portfolio on the efficient frontier is the Pf4 described above on the basis of new market-related estimated quantities.
Furthermore, in an example depicted in
The estimated values of the future quantities are updated being reflective of “an increase in the market price resulting from a reduction in the supply of the photovoltaic renewable energy and a reduction in the sell bidding by the power generation business operators that have reduced supply capabilities in the 30-minute spot or intraday market, and further, an increase in the dispersion value of the market price resulting from an increase in the frequency of occurrence of market segmentation caused by an increase in the bias of power generation locations and thereby occurrence of confusion of a power flow and an open position average up phenomenon in the wholesale market.”
The portfolio that satisfies the condition as the portfolio on the efficient frontier is the Pf1 described above on the basis of new market-related estimated quantities.
In the preferred embodiment of the present invention, a portfolio evaluation result is output as current and future evaluation values of the portfolio estimated from error estimated data in convergence estimation. Furthermore, an efficient frontier simulation result at each time (and in each phase of
In outputting from the input/output interface for selection of the portfolio, current evaluation of the portfolio (power generation configuration proportions and electricity commodity procurement configuration proportions) and evaluation of the portfolio generated in a future phase (expected occurrence of the efficient frontier in
Preferably in Step S201 in the present embodiment, the trading position determination section 201 increases or reduces the trading quantity of any of the trade connections when a difference between a sum of the trading quantities with all the trade connections and a future quantity demanded is equal to or greater than a predetermined value.
Furthermore, the trading position determination section 201 performs an update process for increasing a quota of trading with one certain trade connection in a case in which an estimated value of a trading price with the certain trade connection is lower than a trading price related to trading cumulation of all types of trading, and for reducing the quota of trading with the certain trade connection in a case in which the estimated value of the trading price with the certain trade connection is higher than the trading price related to trading cumulation of all trades.
Furthermore, the trading position determination section 201 performs an update process for reducing a trading quantity quota to any of the trade connections for which a value of data related to an error in the trading price of each of the trade connections increases, and for increasing the trading quantity quota to any of the trade connections for which the data related to the error decreases.
Moreover, the trading position determination section 201 calculates a combination of trading on the basis of the efficient frontier calculated from expected returns by trading and a risk of the expected returns, calculates a latter trading efficient frontier from data in a range in which future expected returns or future expected return dispersion changes in a latter period of the trading period, and determines the combination of trading with the plurality of trade connections in such a manner that the portfolio (procurement and sales proportions of the power generation and electricity commodities) in the vicinity of the efficient frontier can be changed to the portfolio in the vicinity of the latter efficient frontier. Alternatively, the trading position determination section 201 determines the combination of trading with the plurality of trade connections on the basis of data about an intersecting point between the efficient frontier and the latter efficient frontier or a predetermined point in the vicinity of the intersecting point. Determining the trading position particularly from the portfolio in the vicinity of the intersecting point makes it possible to obtain a plan effective for both a case in which various fluctuations occur and a case in which the various fluctuations do not occur. For example, in a trading plan for a day of supply at which a probability of occurrence of an extreme fluctuation in the market price is high, a trading plan effective for both a case in which the price fluctuation occurs and a case in which the price fluctuation does not occur can be laid down.
In the process performed by the split-time-based split order planning section 30 in Steps S301 to S304, the split-time-based split order planning section 30 generates trading and ordering data related to orders placed on the trade connections and an instruction of the orders during a trading period during which the sales business operator can conduct electricity trading with the trade connections and the markets (it is noted that since the electricity is supplied to the customers continuously and consecutively, the electricity trading is conducted while subdividing the trading period into time zones at predetermined intervals (for example, intervals of 30 minutes or four hours), trading of the power generation and the negawatt power to be provided for the supply of the electricity in each of the time zones is conducted on the assumption that a predetermined period before the time zone in which the electricity is actually supplied as the trading period (for example, period from 17:00 one day before the day of delivery until one hour before the time of delivery, from 48 hours to 24 hours before the time of delivery, from 24 hours to one hour before the time of delivery, from ten days to three days before the day of delivery). The orders are placed stepwise in a plurality of planned trading periods into which the trading period is subdivided, so that it is possible to lay down an economical order plan responding to the fluctuation in the market price and the like during the trading period. Details of the process in Steps S301 to S304 will next be described.
(Step S301) The trading and ordering time splitting section 301 receives a value of each trading position described above, and splits the trading quantity in such a manner that the trading quantity of commodities for each position (contracted electric energy (kW)) is made uniform as a target quantity for procurement (finishing position) of commodities by the trade connections and the markets during the trading period.
The split values are stored in a target order transition table T2 depicted in
The trading and ordering time splitting section 301 may perform splitting such that the trading amount of money is made uniform as an alternative to the present embodiment. In another alternative, the trading and ordering time splitting section 301 may perform splitting such that a weighted sum of the trading quantity and the dispersion value of the trading quantity is made uniform. In yet another alternative, the trading and ordering time splitting section 301 may perform splitting such that a weighted sum of the trading amount of money and the dispersion value of the trading amount of money is made uniform.
Preferably, with respect to the orders in the planned trading periods into which the trading period is subdivided, the trading and ordering time splitting section 301 corrects a value of order data in each of the planned trading periods to increase or reduce the value thereof depending on a magnitude of an error in future quantity estimation that is a value of convergence estimation related to the planned trading period. It is thereby possible to preferentially execute trading in the planned trading period with a minor error and to execute a trading plan that can stably realize a target trading quantity and target trading returns.
Furthermore, the trading and ordering time splitting section 301 preferably performs a process for calculating a degree of mismatch among the efficient frontiers (or trading positions) at each point in time of the trading period (for example, the degree of mismatch with respect to distances between centers of gravity of points (portfolios) contained in a set of each efficient frontier) calculated in Step S201, and for reducing the trading quantities in the planned trading periods in ascending order of time in a case in which the degree of mismatch is high. It is thereby possible to obtain a trading result that avoids occurrence of a situation in which procurement configurations of power generation and electricity commodities cannot be changed, resulting in uneconomical trading when the efficient frontier fluctuates in the latter period of the plan.
(Step S302) The trading and ordering data determination section 302 performs a process, which is related to orders in the planned trading periods, for referring to the target order transition table T2, determining the value of the trading and ordering data about the procurement of commodities to the markets in each of the planned trading periods, and transmitting the data to the market ordering terminal 5000. When a target value M of the trading quantity to be contracted is given, price data X and data about a bid quantity 0 are generated as expressed by (Equations 1).
In Equations 1, M is a variable designating the bid quantity (electric energy serving as the open position), R is a variable designating a target contract rate, and P and σ are variables indicating the market condition (condition), where P is an expected value of a trading price and σ is a variable indicating a dispersion of the trading price. Function f is a function for giving the price X that is a price at which trading is completed (which can be contracted) at the rate R from a confidence interval defined from a contracted price distribution by referring to a numerical table (for example, t-distribution table) indicating market-related statistical nature (for example, the confidence interval for a contract at a probability of 95% is (P±1σ) and a contract can be expected at 95% by designating a price of P+σ in a case of buy trading). R is a value into which a user's designated value is substituted through the input/output interface, for example, 0.1 to 0.9.
(Step S303) The power generation plan processing section 303 performs a process for simulating a power generation operation plan known as a generator startup/shutdown plan and a load allocation plan to the contracted power generation business operators. In a case of obtaining a result indicating that the plan is operational by the simulation, the power generation plan processing section 303 transmits data to the power generation ordering terminal 5100. In a case of a simulation result indicating that the plan is not operational, the power generation plan processing section 303 outputs an alert indication to the input/output interface 70.
(Step S304) The electric storage etc. demand planning processing section 304 performs a process for simulating a power conservation plan in a contracted DR (demand response) aggregator. In a case of obtaining a result indicating that the plan is operational by the simulation, the electric storage etc. demand planning processing section 304 transmits data to the aggregator ordering terminal 5200. In a case of a simulation result indicating that the plan is not operational, the electric storage etc. demand planning processing section 304 outputs an alert indication to the input/output interface 70.
It is noted that contract/contracted means that a trade is completed for bidding (transmission of the telegraphic messages) to the markets (also referred to as “markets”), that the sales business operator has reached an agreement in the offer of delivery of the electricity with the power generation business operators or the negawatt power supply business operator (aggregator), or that the sales business operator has reached an agreement in the offer of commercial trading.
In the conventional case (dotted lines), orders are fixedly placed from an initial stage of the overall trading without consideration to occurrence of multimodal errors as depicted in
According to the present embodiment, it is possible to execute the trading plan in response to the market trend, the demand fluctuation, and the operational statuses of the trade connections, and to determine the quantity of orders placed on the trade connections and the markets and the order timing as a result of the execution of the trading plan.
The trading planning apparatus 1 according to the present embodiment is provided with the future quantity estimation section 10 including a demand control quantity estimation section that generates a quantity related to a change in the occurrence of demand or an estimated quantity of data about time, and the order quantity planning section 20 including a demand control limiting section that performs addition or subtraction of the data about the trading positions (indicating values of the trading positions different in time of delivery) on the basis of a value of a demand control quantity.
Such a trading plan makes it possible to execute a plan of trading that gives consideration to demand induction by the demand response or electric storage control.
Furthermore, it is possible to increase the demand (such as the electric storage) at low price time and reduce the demand at high price time. It is noted, however, that a tradable quantity can be limited to a quantity by which the electricity can be absorbed by charge and discharge of a storage battery and to a duration of shift time.
The trading planning apparatus 1 according to the present invention is provided with the future quantity estimation section 10 including a supply estimation section that estimates a future quantity of a quantity supplied by the renewable energy power generation and a demand estimation section that estimates future quantities of quantities demanded by the contracted customers, and the order quantity planning section 20 including the trading position determination section 201 that sells the power generation in response to a quantity by which the electricity generated by the renewable energy surpasses the quantities demanded.
Such a trading plan makes it possible to execute a trading plan capable of committing the generated electricity to the supply.
Furthermore, the trading planning apparatus 1 according to the present embodiment is provided with the future quantity estimation section 20 including an estimation section that estimates future quantities of free capacities of the electric transmission lines and the interconnection lines, and the split-time-based split order planning section 30 including an electric transmission line utilization planning section that lays down an electric transmission line utilization plan.
Such a trading plan makes it possible to execute a trading plan capable of avoiding the occurrence of a situation of shutting down the power generation by the renewable energy due to the saturation of the electric transmission lines and to commit the generated electricity to the supply through the electric transmission lines and the interconnection lines. It is thereby possible to achieve trading also reflective of an economic value of a reduction in a sunk cost generated by stopping the renewable energy.
The trading planning apparatus 1 according to the present embodiment is provided with the future quantity estimation section 10 including an estimation section which estimates estimated quantities of future quantities related to reserve market trading (for example, estimated quantities of a price and a trading quantity of the reserve market trading), and the order quantity planning section 20 including the trading position determination section 201 which determines a quantity provided to a reserve, updates data related to the quantity demanded in response to the quantity provided to the reserve, and calculates the trading position.
Such a trading plan makes it possible to execute a plan of trading including reserve-related market trading guaranteed to be provided when needed by external business operators and difficult to buy back.
Furthermore, the trading planning apparatus 1 according to the present embodiment can calculate the value of the convergence estimation of estimating estimated quantities of the data (for example, dispersion and likelihood values) related to the errors in the future quantities estimated from the actual record values, and lay down an order quantity plan or a time splitting plan.
It is noted that the present invention is not limited to the embodiments described above but encompasses various modifications. In addition, the configuration of a certain embodiment can be partially replaced by the configuration of the other embodiment or the configuration of the other embodiment can be added to the configuration of the certain embodiment. Moreover, for a part of the configuration of each embodiment, addition, deletion, and/or replacement of the other configuration can be made.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2017-060224 | Mar 2017 | JP | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/JP2018/010904 | 3/19/2018 | WO | 00 |
