JP2005352802A - Electric power trading determining device, electric power trading method reply system, program, and storage medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a determining method for electric power trading allowing quantitative grasping of a trade-off relationship between profit and risk for maximizing utility of a power generation company. <P>SOLUTION: A utility function generating device is provided for generating a utility function of the power generation company, a constraint equation generating device is provided for generating a constraint equation of a mathematical programming problem, an objective function generating device is provided for generating an objective function of the mathematical programming problem on the basis of the utility function and the constraint equation, and an optimum consumer portfolio solution device is provided for deriving a set of consumers and a distribution ratio to each consumer optimizing the objective function. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a power transaction determination device, a power transaction method answering system, a program, and a storage medium.

  Electric utilities need to accurately predict parameters that affect revenue in order to maximize revenue from power transactions. Specifically, demand characteristics of multiple consumers receiving power supply, power generation costs by generators or power procurement costs by procurement from power sources such as backup power sources, that is, parameters such as fuel costs that affect power procurement Etc. need to be predicted accurately.

  These parameters are not necessarily determined and will vary depending on uncertain events at a future time. For example, when a general household is targeted as a consumer, it is considered that the demand amount for an air conditioner or a heater increases depending on the temperature and season. Regarding power generation costs, fluctuations in fuel prices due to trends in exchange rates and the conditions of crude oil suppliers are considered to have an impact. However, it is not easy for electric utilities to grasp the changes in temperature or fuel prices at a future time.

  Therefore, in order to predict the profits generated by the power generation company through power transactions, a method of performing a simulation assuming a scenario that may occur in the future based on past customer demand data, etc. It is disclosed in 2003-70164. A simulation method that assumes fluctuations in power generation costs is also disclosed in Japanese Patent Laid-Open No. 2001-86645.

JP 2003-70164 A JP 2001-86645 A

  The conventional simulation method reflects future uncertainties that affect the profits of power generation companies. However, these methods cannot quantitatively grasp the degree of the influence of the uncertain factors on the profit of the power generation company. In addition, in general, when the revenue of a power generation company is in a trade-off relationship with risk, and the power procurement method changes depending on the amount of customer demand, the issue is to quantitatively grasp the trade-off relationship between revenue and risk. there were.

  In this way, when the earnings of a power generation company fluctuate due to uncertain factors, in order to maximize the utility of the power generation company, quantitatively grasp the influence of the uncertain factors on the profits of the power generation company. In addition, it is necessary to construct a power transaction determination method that can quantitatively grasp the trade-off relationship between profit and risk.

  One feature of the present invention is that, in the power transaction determination device, the customer data related to the power demand characteristics of the consumer, the business data related to the profit of the electric power company, and the fluctuation range of the business profit and the profit of the electric power company Utility parameters that correlate the risks to be generated, constraint parameters related to constraints that the electric utility considers, and the amount of power that the electric utility should supply and the amount of power that can be generated by the company or the amount that can be procured from other entities An input device that receives power procurement method parameters related to a power procurement method according to the relationship, a storage device that stores each parameter and data received in the input device, the utility parameter stored in the storage device, and the A utility function creation device for creating a utility function of an electric power company based on a power procurement method parameter; A constraint formula creation device that creates a constraint formula of a mathematical programming problem based on the stored constraint parameter and the power procurement method parameter, and an objective function of the mathematical programming problem based on the utility function and the constraint formula An objective function creation device to be created, an optimal customer portfolio solving device for deriving a set of consumers and an allocation ratio to each customer that optimize the objective function, and an optimum derived by the optimal customer portfolio solving device And a customer portfolio output device that outputs a distribution ratio to the consumers.

  By quantitatively grasping the trade-off relationship between risk and return, it becomes possible to construct a power consumer portfolio so as to reduce the risk of electric utilities and increase profits.

  The electric power transaction determination method of the present invention is performed by a mathematical programming method such as a linear programming method or a quadratic programming method by an electric power company that supplies electric power to a consumer using a computer. Specifically, the supply ratio to the power supply destination or a demand group such as the same industry composed of a plurality of power supply destinations is determined so that the business profit of the electric power supplier can be maximized. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  In an embodiment of the present invention, an example of a power transaction determination problem in which an electric power company generates power generated in-house or a consumer who is a supply destination of power procured from another company is taken as an example. This problem is a problem of finding a supplier capable of maximizing the utility function set by the operator under given constraints. However, the method for procuring electric power, that is, the method for determining the proportion of in-house power generation or procurement from other companies shall be determined so as to maximize the profit for the electric power company. The utility function is represented by the profit of the electric power company, the risk such as the standard deviation indicating the degree of fluctuation of the profit, and the degree of risk avoidance as follows.

Utility function = revenue-risk aversion x risk or return-λ x (risk squared)
(1)
The risk avoidance degree represents a parameter indicating which of the maximization of profit and the minimization of risk is more important in the trade-off relationship between profit and risk. The risk may be defined by a parameter other than the fluctuation amount of the profit, that is, the fluctuation amount of the demand of the customer or the fluctuation amount of the fuel cost.

  Revenue is defined as follows by subtracting the cost required for power generation or the cost of procurement from other power sources from the revenue obtained by selling power to the supplier.

Revenue = ΣΣω (i) (d (i, t) * p (i, t) + fp (i, t))
−ΣΣ [a (j) × {Gn (j, t)} ^ 2 + b × {Gn (j, t)} + c]
−ΣBu (k, t) * Bexp (k, t) −d (2)
In the above formula, the first term is the income from the supply of power, the second term is the relationship between the amount of power generation and the amount of petroleum fuel required for power generation, that is, the relationship between the amount of power generation and the fuel cost, and the third term is from other power sources. The relationship between the amount of procurement and the cost for procurement, and the fourth term represents the fixed cost incurred during power generation.

  As parameters used in the first term, d (i, t) is a demand amount of the customer i at a predetermined time t, p (i, t) is a metered rate of electric power imposed on the customer i at the predetermined time t, fp (i, t) means a fixed charge of electric power imposed on the customer i at a predetermined time t. Further, ω (i) indicates the distribution ratio to the customer i, and is standardized so that the total sum of all the consumers becomes 1. Actually, the contracted power of the consumer is fixed, so the electric utility decides whether to supply power to each consumer (the amount required by the consumer) or not at all. Although necessary, an embodiment of this method will be described later.

  As parameters used in the second term, Gn (j, t) is the output amount of the generator j at a predetermined time t, a (j) and b (j) and c (j) are the power generation amount and fuel cost of the generator j. It means a characteristic coefficient that represents the relationship. In addition, although the relational expression between the power generation amount and the fuel cost uses a model approximated by a quadratic expression, it can be replaced by a model approximated by a linear expression. As parameters used in the third term, Bu (k, t) is the procurement amount from the supplier k at the predetermined time t, and Bexp (k, t) is the procurement cost unit price required for procurement from the supplier k at the predetermined time t. Means. The fourth term represents the cost required in addition to the fuel cost, and it is possible to use a model that takes into account factors such as time and power generation amount, but in the embodiment of the present invention, a fixed constant d is used. It shall be expressed as

  The risk of revenue can be expressed by the variance or standard deviation of revenue expressed above. Specifically, the standard deviation of the demand amount at each time point for each consumer who is a candidate for the supply destination, the standard deviation of fuel cost in the future time point, and the like are used as parameters. In addition, since the demand fluctuation of a consumer cannot always be expressed by a normal distribution, it may be considered that it is not appropriate to use a standard deviation as a risk index. In that case, it is possible to adopt lower half variance, Value-at-Risk, Conditional-Value-at-Risk (expected shortfall), etc. as an index of risk, but in explaining this embodiment, Standard deviation shall be used as a risk of revenue.

  FIG. 1 shows a schematic configuration of a power transaction determination apparatus according to the present invention. The power transaction determination device includes a parameter input device 101, a parameter storage device 102 for storing parameters input from the parameter input device, a utility function creation device 103, an objective function creation device 104, a constraint equation creation device 105, It comprises an optimum customer portfolio deriving device 106 and a customer portfolio output device 107.

  In the parameter input device 101, parameters of customer data, business operator data, utility parameters, constraint parameters, and power procurement method parameters are input. Regarding customer data, data such as demand characteristics of each customer at a predetermined point in time are stored, and will be exemplified with reference to FIGS. FIG. 2 shows the expected value of the demand amount at each customer's future time point, and FIG. 3 shows the standard deviation of the demand amount at each customer's future time point. FIG. 4 shows data related to a contract such as a metered charge and contract power at each consumer.

With respect to the company data, data relating to the generators owned by the electric company, data on the amount of power that can be procured from other companies at each time point, data on the procurement unit price, and the like are stored. FIG. 5 shows data related to the generator, and FIG. 6 shows an example of data related to power procurement from other companies. FIG.
(A) shows the value of the characteristic coefficient of the generator, and FIG. 5 (b) shows the number of operating generators at each time point. FIG. 5A shows that the amount of fuel and the amount of heat used for power generation are approximated by a quadratic equation, and the effective output amount is shown as the rated output. FIG. 5B shows that the number of operating generators is adjusted according to day / night and season. It should be noted that the data relating to the charge structure of each consumer shown in FIG. 4 can also be regarded as a category of business data.

  Regarding utility parameters, one or a plurality of risk aversion degrees for determining an electric power trading method having an optimal risk / return trade-off relationship for a power generation company is set. FIG. 6 shows an example of parameters when a plurality of parameters are set. Generally, the higher the risk aversion, the lower the risk and the lower return portfolio, and the lower the risk aversion, the higher the risk and return portfolio.

  With respect to the constraint parameters, the constraint conditions to be taken into consideration when the electric power company decides the optimum power trading method are input. FIG. 7 shows an example of data. FIG. 7A shows the upper and lower limit constraints of the distribution ratio, and FIG. 7B shows the supply amount constraint at a specific time point.

As for the power procurement method parameters, information such as the priority order when the electric power company performs power procurement and the preconditions regarding the procurement fee is input. In particular,
・ During the daytime, we will procure the shortage from other companies, giving priority to our own power generation from other companies.
・ During night hours, procurement from other companies will be given priority over in-house power generation, and the shortage will be procured through in-house power generation. However, our generator will generate the minimum output regardless of the amount required for supply.
Starting with priorities for power procurement methods such as
・ During the daytime hours, if more than a predetermined amount of power is required, purchase power at a unit price that is higher than usual, and during the nighttime hours, if less than a predetermined amount of power is required , Purchase electricity at a unit price discounted than usual.
Information such as preconditions regarding procurement fees such as

  In the utility function creation device 103, the objective function is composed of profit and loss, return, and penalty term based on the information input in the power procurement method input device. A specific method will be described later.

  The objective function creation device 104 includes utility functions and penalty terms created by the utility function creation device. A specific method will be described later.

  The constraint formula creation device 105 creates a constraint formula that can be processed by mathematical programming based on the information of the power procurement method parameters in addition to the constraints stored in the parameter storage device 102. A specific method will be described later.

  The optimum customer portfolio deriving device 106 derives a customer portfolio that maximizes the objective function created by the objective function creating device 104 under the constraint formula created by the constraint formula creating device 105. As the optimization method, if the utility function is a quadratic expression, a typical main dual interior point method or effective constraint method of the quadratic programming problem is used. If the utility function is a linear expression, a typical linear programming problem is represented. The simple unit method or the main dual interior point method is generally used.

  The customer portfolio output device 107 displays information on the portfolio calculated by the optimum customer portfolio deriving device 106. An example is shown in FIG. FIG. 8A shows an optimal customer portfolio according to the degree of risk avoidance, and FIG. 8B shows a frontier curve showing the relationship between risk and profit in each portfolio.

  Next, the procedure of the process in determining the optimum portfolio executed by the apparatus of FIG. 1 will be described using the flowchart of FIG.

  In step 901, data relating to demand characteristics of the customer is input. The data relating to the demand characteristic has been described above and is as shown in FIGS.

  In step 902, data relating to the characteristics of the electric utility is entered. The data relating to the characteristics of the electric power company has been described above and is as shown in FIG.

In step 903, information such as the order of priority when the electric power company performs power procurement and preconditions regarding the procurement fee is input. In particular,
・ During the daytime, we will procure the shortage from other companies, giving priority to our own power generation from other companies.
・ During night hours, procurement from other companies will be given priority over in-house power generation, and the shortage will be procured through in-house power generation. However, our generator will generate the minimum output regardless of the amount required for supply.
Starting with priorities regarding power procurement methods such as the above, during the daytime hours, when more than a predetermined amount of power is required, purchase power at a unit price that is higher than usual. When less than a certain amount of power is required, the power is purchased at a unit price discounted from the normal price.

  Information such as preconditions regarding procurement fees such as

  In step 904, based on the data input in step 901 and step 902, utility parameters of the electric utility are defined and a utility function is formulated. For utility parameters, the electric utility sets the risk aversion degree so that a desired trade-off relationship between risk and return is obtained.

Regarding the return, the above-mentioned formula (2),
Revenue = ΣΣω (i) (d (i, t) * p (i, t) + fp (i, t))
−ΣΣ [a (j) × {Gn (j, t)} ^ 2 + b × {Gn (j, t)} + c]
−ΣBu (k, t) * Bexp (k, t) −d (2)
As described above, it is considered that a general method is defined by revenues obtained by subtracting costs required for power procurement such as power generation from revenues received by electric utilities from customers. Regarding the cost of power procurement, in addition to the cost generated by power generation, methods such as purchase from other power generation companies are also conceivable.

As for risk, the degree to which the return defined in (2) above fluctuates due to uncertain factors such as fluctuations in future customer demand, generator operating conditions, and fuel costs generated during power generation. It is quantified. Risk can be defined by the standard deviation of returns. However, if the return distribution does not follow the normal distribution, the risk may be underestimated. There are several ways to avoid underestimating risk: Value-at-Risk or
It is conceivable to define risk by Conditional Value-at-Risk.

  The utility function representing the utility of the electric utility is defined as follows using the risk and the return, for example.

Utility function = revenue-risk aversion x risk or return-λ x (risk squared)
(1)
λ represents the degree of risk avoidance, and it is known that a low-risk low-return customer portfolio is obtained when λ is large, and a high-risk high-return customer portfolio is obtained when λ is small. The relationship between λ and the optimal customer portfolio is as shown in FIG.

  In step 905, based on the parameters input in steps 901 and 902, information on the power procurement method input in step 903, and the utility function formulated in step 904, an objective function is applied to apply mathematical programming. And set constraint expression.

In general, electric utilities need to change the amount of power generated by their own generators and the amount of procurement from other companies according to the amount of demand at a specific point in time. Here, for the sake of simplicity, it is assumed that the following preconditions are input in step 903, and the utility of the electric utility is maximized in the power supply business to the consumer.
(A) In the daytime, prioritize procurement from power generation by our own generators, and cover the shortfall by procurement from other companies.
(B) At night, the procurement from other companies is given priority, and the shortage is covered by power generation by our own generator.

  Based on this assumption, the constraints to consider are as follows:

Σω (i) d (i, dt) −ΣGnmax (k, dt) = x (dt, 1) −x (dt, 2) (3)
Σω (i) d (i, nt) −ΣBumax (1, nt) = y (nt, 1) −y (nt, 2) (4)
In the above formula (3), the first term is the required amount of power supply to all consumers at the time dt in the daytime, and the second term is the output upper limit of the generator owned by the electric power company at the time in the daytime. It means sum.

That is,
(3)-(a) When the demand of all customers can be covered by the power generation of the operator, that is, when the first term ≦ the second term, x (dt, 1) = 0 and x (dt, 2) ≧ 0 Must be
(3)-(b) When supply from other companies (or organizations) is required in addition to the power generation of the operator, that is, when first term> second term, x (dt, 1)> 0 and x (dt, 2) = 0 needs to be satisfied.

  In equation (4), the first term is the required amount of power supply to all customers at a certain time t, and the second term is the upper limit of the amount of power procured from multiple other companies (or organizations) at a certain time Means the sum of

That is,
(4)-(a) When the demand of all consumers can be covered by procurement from the power source, that is, when first term ≦ second term, y (nt, 1) = 0 and y (nt, 2) ≧ 0 Must be
(4)-(b) When supply from other companies is required in addition to the power generation of the operator, that is, when first term> second term, y (nt, 1)> 0 and y (nt, 2) = 0 It is necessary to become.

From the above examination, it is necessary that at least one of x (dt, 1) and x (dt, 2) and at least one of y (nt, 1) and y (nt, 2) should be 0, respectively. is there.

  However, in addition to the constraints (3) and (4), x (dt, 1) * x (dt, 2) = 0 and y (nt, 1) * y (nt, 2) = 0 are constraints. In this case, since it is a nonlinear equation, it becomes difficult to handle it as a mathematical programming problem.

Therefore, in the mathematical programming problem, the objective function to be maximized is the objective function = utility function (= revenue−risk avoidance × risk)
Instead of, the positive constants Penalty1 and Penalty2 are introduced, and the objective function to be maximized is configured as follows.

Objective function = Utility function−Penalty1 * Σ (x (dt, 1) + x (dt, 2))
-Penalty2 * Σ (y (nt, 1) + y (nt, 2)) (5)
Furthermore, as a constraint condition, in addition to the above (3) and (4),
x (dt, 1), x (dt, 2), y (nt, 1), y (nt, 2) ≧ 0 (6)
If the non-negative constraint is taken into consideration, the constraint equations are all linear.

  If the objective function is maximized by the above handling, either x (dt, 1) or x (dt, 2) and either y (nt, 1) or y (nt, 2) become 0. Can be.

  Next, the reason why either x (dt, 1) or x (dt, 2) is 0 when the objective function is maximized under the constraint (6) is proved below (y ( The same applies to nt, 1) and y (nt, 2)). When the objective function is maximized, Penalty1 is a positive constant, so x (dt, 1) + x (dt, 2) is minimized. Therefore, when x (t, 1) + x (t, 2) is minimized under constraint (6), either x (dt, 1) or x (dt, 2) is set to 0. You can show that

The value of x (dt, 1) −x (dt, 2) is calculated from the equation (3) and the amount of power necessary for supply to the consumer (first term = Σω (i) d (i, dt)) It is defined by the difference in the amount of power that can be generated by the operator (second term = ΣGnmax (k, dt)). In the following, cases are classified according to the magnitude relationship between the first term and the second term, and either x (dt, 1) or x (dt, 2) is 0 under the constraint (6). .

In the formula (3), when the first term is not larger than the second term, that is,
Σω (i) d (i, dt) ≦ ΣGnmax (k, dt)
In FIG. 11 (a) showing the situation, the left side and the right side of the above equation are non-negative regions (based on the constraint (6)), and in the allowable region represented by the half line that satisfies the equation, the optimum points indicate This indicates that the portion to be minimized is a point that minimizes x (dt, 1) + x (dt, 2). At the optimum point, x (dt, 1) is 0, and x (dt, 2) is ΣGnmax (k, dt) −Σω (i) d (i, dt) (≧ 0). I understand that.

Similarly, in the formula (3), when the first term is larger than the second term, that is,
Σω (i) d (i, dt)> ΣGnmax (k, dt)
In FIG. 11 (b) showing the situation, the left side and the right side of the above equation are non-negative regions (based on the constraint (6)), and in the allowable region represented by the half line that satisfies the equation, the optimum points indicate This indicates that the portion to be minimized is a point that minimizes x (dt, 1) + x (dt, 2). At the optimum point, x (dt, 2) is 0, and x (dt, 1) is Σω (i) d (i, dt) −ΣGnmax (k, dt) (> 0). I understand that.

  As described above, it has been shown that when the objective function (5) is maximized under the constraint (6), at least one of x (dt, 1) and x (dt, 2) is zero.

The prerequisites discussed so far are
(A) 'During the daytime, priority will be given to procurement from power generation by its own generator, and the shortage will be covered by procurement from other companies.
(B) ′ At night, the procurement from other companies is given priority, and the shortage is covered by the power generated by its own generator.
If nt is used instead of dt and dt is used instead of nt, the same approach can be used. Specifically, as a constraint equation, instead of equations (3) and (4),
Σω (i) d (i, nt) −ΣGnmax (k, nt) = x (nt, 1) −x (nt, 2)
(3) ′
[Sigma] [omega] (i) d (i, dt)-[Sigma] Bumax (l, dt) = y (dt, 1) -y (dt, 2)
(4) '
As an objective function, instead of equation (5),
Objective function = Utility function−Penalty1 * Σ (x (nt, 1) + x (nt, 2))
−Penalty2 * Σ (y (dt, 1) + y (dt, 2)) (5) ′
Should be considered. Furthermore, as a constraint condition, in addition to the above (3) (4) ',
x (dt, 1), x (dt, 2), y (nt, 1), y (nt, 2) ≧ 0 (6) ′
Should be considered.

  In step 906, an optimal customer portfolio is obtained by applying an optimization technique. As an optimization method, the interior point method or the effective constraint method is generally used when the utility function is quadratic, and the simplex method or the interior point method is generally used when it is linear. As an example of the present invention, an outline of the interior point method will be described.

In the interior point method, a point inside a constrained region is used as an initial point, and thereafter, a search direction for searching for the next repetitive point within the constrained region is obtained, and the work for obtaining the repetitive point is repeated sequentially. This is a method for obtaining the optimum point. The feature of the interior point method is that the inside of the constraint area is an object to be updated, and it is empirically known that the optimal point can be reached in several tens of times regardless of the scale of the problem. For this reason, it is often superior to other methods in relatively large-scale problems. FIG. 12 shows a conceptual diagram of the interior point method. Next, an outline of the interior point method processing will be described with reference to the flowchart of FIG.

  In step 1301, an arbitrary point inside the restricted area is obtained and set as an initial point.

  In step 1302, the direction in which the optimum solution is searched is obtained. In calculating the search direction, a simultaneous linear equation constructed based on information such as the current iteration point, the utility function to be optimized, and the coefficient of the constraint equation is solved.

  In step 1303, along the search direction, the iteration points that are considered advantageous for the next iteration are calculated. Specific methods include a linear search that searches the iteration point that optimizes the objective function along the search direction, or a maximum that can maintain the non-negativeity of the variable to be optimized by minimizing the load in the search. A method of limited movement is conceivable.

  In step 1304, it is checked whether the current iteration point satisfies the optimality condition. The optimality condition is generally called the Kuen-Tucker condition, and is known to occupy an important position in nonlinear programming. If the optimality condition is satisfied, the series of processes as the optimization technique application in step 906 is terminated. If not satisfied, the process returns to step 1302.

  In step 907, information about the customer portfolio obtained in step 906 is output. FIGS. 14A and 14B show output examples. The example output in FIG. 14A shows the distribution ratio, but in the contract with each customer, it is necessary to decide whether to supply power or not, so the distribution ratio itself is final. Is meaningless. Therefore, in step 908, it is necessary to correct the power transaction method based on the information of FIG. In addition, when the power supply destination is configured by a plurality of consumer groups as shown in FIG. 14B, the distribution ratio is significant when supplying power.

  In step 908, based on the information output in step 907, the electric utility determines a consumer to be supplied with power. As an example of a method for determining a customer, priorities are set in the order of closer distribution ratio and customer demand, and customers are selected in accordance with the priorities within an allowable range such as the power generation equipment's installed capacity. A method is conceivable. An output example by the former determination method is shown in FIG.

  FIG. 16 shows an example of the system configuration of the power transaction determining apparatus according to the present invention. A computer device for presenting a consumer portfolio determined by a power transaction determination method to an electric power company is constituted by a personal computer. In calculating the customer portfolio, a database for storing information on the demand characteristics of the customer and information on parameters affecting the business profit of the electric utility is required. Based on these databases, electric utilities derive customer portfolios, select customers based on the amount of contracted electric power according to each customer's demand, etc., and simulate business revenues. Application software for displaying the results is required.

  In FIG. 16, a plurality of customer computers are connected to a computer network. The server is installed with an application for constructing the database, and stores three databases connected to the server. Here, the four databases are information on customer demand characteristics, information on parameters affecting business profits of electric utilities, and data on risk aversion that defines the trade-off relationship between risks and returns for electric utilities. The data for making it into each is stored.

  In the central processing unit, application software for performing calculation of optimal customer portfolio and simulation of business revenue when a specific customer is selected and a program for displaying the simulation result to the user are installed. A simulation is performed to calculate an optimal portfolio based on data input from two databases. Data relating to the optimum customer portfolio calculated in the central processing unit is transferred to the client computer on the customer side via the computer network.

  The client computer on the customer side receives information on the optimum customer portfolio calculated by the computer on the server side, and displays the customer portfolio. In the client computer, it is only necessary to install an application program for displaying the customer portfolio and an application program for inputting data relating to optimization guidelines for customers.

  Thus, by constructing the system configuration of the power transaction determining apparatus of the present invention, it is possible to construct an optimal customer portfolio.

  As described above, according to the present invention, the trade-off relationship between risk and return can be quantitatively grasped. As a result, we formulate an optimization problem with the trade-off relationship between risk and return as an objective function, and by solving this, build a power consumer portfolio to reduce the risk of electric utilities and increase profits. It becomes possible.

The block diagram which illustrates the composition of the power transaction determination device of the present invention. Example 1 of demand characteristic data stored in the parameter storage device. Example 2 of demand characteristic data stored in the parameter storage device. The example of the electric utility company characteristic data memorize | stored in the parameter memory | storage device. An example of data related to a contract between an operator and a customer. Examples of set values related to risk aversion and assumed risks and returns. The example of the data input in a constraint parameter input device. The example of the data output by the consumer portfolio output apparatus. The flowchart which illustrates the outline | summary of the process in the determination method of the electric power supplier in an electric power transaction determination apparatus. Examples of risk indicators. The conceptual diagram relevant to utility function formulation. The conceptual diagram of the interior point method which is a kind of optimization method. The flowchart which illustrates the outline | summary of the process in the interior point method which is a kind of optimization method. The example of the optimal customer portfolio output in the electric power transaction determination apparatus. The example of the electric power supplier data determined in the electric power transaction determination apparatus. The block diagram which shows an example of the system of an electric power transaction determination apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Parameter input device, 102 ... Parameter storage device, 103 ... Utility function creation device, 104 ... Objective function creation device, 105 ... Restriction formula creation device, 106 ... Optimal customer portfolio derivation device, 107 ... Customer portfolio output device

Claims (11)

From the multiple consumers receiving power supply, the electric utility supplies the electric power so as to optimize the utility of the electric utility that sells the electric power to the customer under the predetermined constraints. In the power transaction determination device that determines the distribution ratio to the consumers to be supplied and the power generation amount of the company and the power procurement from other entities,
Consumer data on consumer demand characteristics,
Business data on the profits of electric power companies,
Utility parameters that relate the business revenue of the electric utility to the risk that represents the fluctuation range of the revenue,
Constraint parameters related to constraints considered by the utility,
An input device that receives a power procurement method parameter related to a power procurement method according to a relationship between an amount of power to be supplied by an electric power company and an amount of power that can be generated by the company or a quantity procurable from another business entity;
A storage device for storing each parameter and data received in the input device;
An utility function creation device for creating an utility function of an electric power company based on the utility parameter and the power procurement method parameter stored in the storage device;
A constraint equation creation device that creates a constraint equation of a mathematical programming problem based on the constraint parameter and the power procurement method parameter stored in the storage device;
An objective function creation device for creating an objective function of a mathematical programming problem based on the utility function and the constraint equation;
An optimal customer portfolio solving apparatus for deriving a set of consumers that optimizes the objective function and a distribution ratio to each customer;
An electric power transaction determination device comprising an optimal customer derived in an optimal customer portfolio solution device and a customer portfolio output device that outputs a distribution ratio to the customer. In claim 1,
The consumer data is data relating to a predicted value of power demand and a fluctuation amount of a predicted value of power demand, and data relating to a power charge obtained from a consumer by an electric power company. In claim 1,
The business operator data is generator characteristic data owned by an electric power company and data relating to a power supplier, and is a power transaction determination device. In claim 1,
The business profit of the electric utility is defined by the difference between the income obtained from the consumer and the cost represented by the sum of the procurement cost of power from other power sources and the output cost of the generator by the electric utility. A power transaction determination device characterized by the above. In claim 1,
The risk is defined by the amount of fluctuation of at least one item among the income obtained from consumers, the procurement cost from other power sources, and the output cost of the generator by the electric power company. Decision device. In claim 1,
The utility parameters are
It is a risk avoidance degree showing which of the business profit of the said electric power supplier and the said risk is preferred, The electric power transaction determination apparatus characterized by the above-mentioned. In claim 1,
The constraint parameter is
Distribution ratio to the consumer, or income obtained from the consumer, or the risk, or an upper limit value or a lower limit value of the electric energy that can be generated by the electric utility, or an upper limit value of the electric power that can be procured by the electric utility or An electric power transaction determination device characterized by being a restriction on a lower limit value. In claim 1,
The power procurement method parameters are:
At least one of the amount of electric power that can be generated by an electric utility and the amount of electric power that can be procured from other entities;
An electric power transaction determining apparatus, characterized in that it is a method relating to electric power procurement according to the relationship with the amount of electric power required by a consumer. In the power transaction method answering system that connects a server and multiple client computers via a network,
The server
Parameters related to the power demand characteristics of consumers, such as the predicted value of power demand and the amount of fluctuation in the predicted value of power demand;
A database for storing constraint parameters constituting constraint conditions for optimizing an objective function composed of profits and risks for electric utilities;
A central processing unit that accepts data stored in the database and derives a set of consumers and a distribution ratio to each consumer so as to maximize the objective function;
Based on a predetermined restriction requested from the client computer, the distribution unit to the customer and the distribution ratio to the consumer derived by the central processing unit is configured to be transmitted to the requested client computer. A power transaction method answering system. On the computer,
Consumer data on consumer demand characteristics,
Business data on the profits of electric power companies,
Utility parameters that relate the business revenue of the electric utility to the risk that represents the fluctuation range of the revenue,
Constraint parameters related to constraints considered by the utility,
An input procedure for receiving power procurement method parameters relating to a power procurement method according to the relationship between the amount of power to be supplied by an electric power company and the amount of power that can be generated by the company or the amount that can be procured from another business entity,
A storage procedure for storing each parameter and data received in the input device in a storage device;
A utility function creating procedure for creating a utility function of an electric power company based on the utility parameter and the power procurement method parameter stored in the storage device;
A constraint formula creation procedure for creating a constraint formula of a mathematical programming problem based on the constraint parameter and the power procurement method parameter stored in the storage device;
An objective function creation procedure for creating an objective function of a mathematical programming problem based on the utility function and the constraint equation,
An optimal customer portfolio solution procedure for deriving a set of consumers that optimize the objective function and a distribution ratio to each customer;
A program for executing an output procedure for outputting an optimum consumer and a distribution ratio to the consumer derived in the optimum consumer portfolio solving apparatus.   A computer-readable recording medium storing the program according to claim 10.

Images