JP2009026092A - Facility planning supporting device for cogeneration system, supporting method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce an operation time in deciding a cogeneration system optimizing facility configuration by the mixed integer linear programming without degrading accuracy of a solution. <P>SOLUTION: When cogeneration facility information A of a cogeneration model 20, parameters B of electric power load and thermal load patterns required by the cogeneration model, optimization purpose information C of operation cost or an amount of CO<SB>2</SB>emission of the cogeneration model, and constraints D of the cogeneration model are given, an optimizing facility configuration decision device 10 solves an optimization problem of a facility configuration by a branch and bound method while setting the information as variables. The number of operated cogeneration facilities is set as an integer variable for each of time sections obtained by roughly classifying an electric power load and a thermal load, and a series of time sections belonging to the same category with respect to both of the electric power and the thermal load are assumed as one time section. For time section setting, one of four classes of annual maximal load ratio, five classes of annual maximum load ratio, four classes of seasonal maximum load ratio, and five classes of seasonal maximum load ratio is used. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an apparatus and method for supporting an equipment plan of an electric heat cogeneration system, and more particularly to an apparatus and a method for determining an optimum equipment configuration based on an equipment planning model.

  There is a method of determining an optimal equipment configuration by linear programming (LP) or nonlinear programming (NLP) with respect to the equipment planning model when constructing the combined electric and heat system.

For example, in order to determine the optimal equipment configuration of a combined heat and power system for a power / heat load pattern, the problem is modeled as a mixed integer linear program, in which an electric heat generator (generator, generator, Cogeneration, boilers, etc.) and other electric heating facilities are controlled, and the optimal equipment configuration is obtained by calculating which equipment is installed and how it is operated to minimize costs or CO ₂ emissions There exists a method (for example, refer patent document 1).

As another method, instead of representing the combined electric and heat system as a linear programming model, the equipment configuration is obtained by nonlinear mixed integer programming that allows detailed settings such as consideration of partial load, control of multiple units, and economic constraints. There is a method (for example, refer nonpatent literature 1).
JP 2004-317049 A Ishida et al., “Establishment of an optimal CGS introduction decision support system for business buildings considering economic constraints and partial load characteristics of equipment” IEEJ Transaction B, Vol. 125, No. 4, 005, p. 373-380

  In the calculation to obtain an optimized equipment plan by nonlinear mixed integer programming (Non-Patent Document 1), a large number of variables are set for each equipment in the equipment planning model, and the combination of variables becomes enormous and practical. There is a problem that a solution cannot be obtained in a long calculation time.

  For example, in order to express that there is a lower limit in the output in the operation of the electric heat generator and it is not possible to operate less than the minimum output, the state of the generator at each time is represented by an integer variable representing the number of operations and a continuous variable representing the operation output. It is expressed with. For this reason, it is necessary to prepare a large number of integer variables, and the calculation time for obtaining the optimized equipment configuration becomes enormous.

  Similarly, when the branch and bound method, which is an exact solution, is used as a solution of the mixed integer linear programming (Patent Document 1), if there are too many integer variables, the operation time increases exponentially and the solution takes a practical time. Is not required. In order to cope with this, if the time interval for handling the number of operations is set to “2 categories of day / night” or “6 categories per day”, the accuracy of the solution is lowered and it is not sufficient as an optimal solution.

  There is an approximate solution method such as the tab search method as another solution method of mixed integer linear programming, but there is no method to measure the accuracy of the solution that came out (difference in target value from the true optimal solution), so the optimal solution is guaranteed. I can't.

  An object of the present invention is to provide an equipment plan support apparatus, a support method, and a program for an electric heat cogeneration system that can reduce the calculation time without reducing the accuracy of the solution to the determination of the optimized equipment configuration of the electric cogeneration system by mixed integer linear programming. It is to provide.

In order to solve the above-mentioned problems, the present invention is provided with electric facility information of an electric heat model, load pattern, optimization objective information of cost and CO ₂ emission amount, and constraints of the electric heat model. It is equipped with an optimized equipment configuration determination device that solves the optimization problem of equipment configuration by the branch and bound method, sets the number of operating electric heating equipment as a variable for each time segment roughly classifying power load and heat load, By reducing the number of integer variables by making consecutive times belonging to the same classification for both the power load and the heat load into one time segment, an appropriate solution can be obtained with a practical calculation time. , Features methods and programs.

(Invention of the device)
(1) Electric heat equipment information of the electric heat model expressing the equipment plan of the cogeneration system, electric load / heat load pattern required for the electric heat model, optimization objective information of the operating cost of the electric heat model or CO ₂ emissions In the facility planning support device of the cogeneration system equipped with the optimized facility configuration determination device, which is given the constraints of the electric heat model, sets the information as variables and solves the optimization problem of the facility configuration by the branch and bound method,
For each time segment roughly classifying the power load and heat load, the number of operating electric heating facilities is set as an integer variable, and the time segment is defined as one time segment for consecutive times belonging to the same category for both power load and heat load. Dynamic time segment setting means for setting in the optimized equipment configuration determining device is provided.

(2) The dynamic time segment setting means includes:
The annual maximum load ratio 4 classification means set by classifying into 4 time segments from the ratio to the maximum load for each year of electricity and heat,
The annual maximum load ratio 5 classification means to classify and set in 5 time segments from the ratio to the maximum load for each year of electricity and heat,
The seasonal maximum load ratio 4 classification means set by classifying into 4 time segments from the ratio to the maximum load in each season for each electric power and heat,
5 seasonally maximum load ratio classification means set by classifying into 5 time segments from the ratio of maximum load in each season for each of electric power and heat,
Among them, any one means is provided.

(Invention of method)
(3) Electric heat equipment information of the electric heat model expressing the equipment plan of the combined electric heat system, electric load / heat load pattern required for the electric heat model, optimization cost information of the operating cost of the electric heat model or CO ₂ emissions In the facility planning support method of the cogeneration system with the optimized facility configuration determination device, which is given the constraint conditions of the electric heat model, sets the information as variables and solves the optimization problem of the facility configuration by the branch and bound method,
For each time segment roughly classifying the power load and heat load, the number of operating electric heating facilities is set as an integer variable, and the time segment is defined as one time segment for consecutive times belonging to the same category for both power load and heat load. A dynamic time segment setting procedure for setting in the optimized equipment configuration determining apparatus is provided.

(4) The dynamic time segment setting procedure is as follows:
The annual maximum load ratio 4 classification procedure set by classifying into 4 time segments from the ratio to the maximum load for each year for power and heat,
The annual maximum load ratio 5 classification procedure to set by classifying into 5 time segments from the ratio to the maximum load for each year of electricity and heat,
Four seasonal maximum load ratio classification procedures set by classifying into four time segments from the ratio of maximum load in each season for power and heat,
The seasonal maximum load ratio 5 classification procedure to set and classify into 5 time categories from the ratio to the maximum load in each season for each electric power and heat,
Among these, it has any one procedure.

(Invention of the program)
(5) The processing procedure in the facility planning support method for the combined electric and heat system according to claims 3 and 4 is configured to be executable by a computer.

As described above, according to the present invention, the electric facility information of the electric heating model, the load pattern, the cost and the optimization objective information of the CO ₂ emission amount, and the constraints of the electric heating model are given, and these information are set as variables. Equipped with an optimized equipment configuration determination device that solves the optimization problem of equipment configuration by the branch and bound method, and sets the number of operating electric heating equipment as an integer variable for each time segment roughly classifying power load and heat load. Since the number of integer variables has been reduced by making consecutive times that belong to the same category for both load and heat load, the number of integer variables has been reduced. Can be determined.

  Specifically, the integer variable representing the number of units in operation is constrained so that the number becomes substantially 1/3 to 1/2. The Since the part that is originally restricted is the same value, the target value of the optimization result is close enough when there is no restriction, and the equipment configuration is almost the same.

FIG. 1 shows an equipment plan support apparatus for an electric and heat cogeneration system targeted in the present embodiment, and shows an apparatus for determining an optimum equipment configuration based on an electric heat model. The optimum facility configuration determining apparatus 10 uses a computer resource and a software configuration using the computer resource, and obtains the optimum facility configuration by a branch and bound method that is an exact solution of the mixed integer linear programming. Basically, it is optimal for the optimum equipment configuration determining apparatus 10 from the equipment information A of the electric heat model 20, the parameter B of the power / thermal load pattern, the optimization objective information C of the cost and CO ₂ emission amount, and the constraint condition D of the electric heat model. The optimal equipment configuration is determined by constructing and solving this problem.

  An example of the electric heat model 20 of the electric heat combined supply system is shown in FIG. An electric heat generator (generator, cogeneration facility, boiler, etc.) 1 is a source of electric energy and thermal energy, an electric power purchase (purchasing electric power from the grid) 2 is an electric energy source, and a storage battery 3 is an electric energy source. As a storage source, the power converter 4 is a conversion device from power energy to heat energy. The power energy generated by these is supplied to the power load 5 and the heat energy is supplied to the heat load 6.

In the facility information A, the rated power / heat output set for each facility of the electric heat model 20, the operating cost, the CO2 emission amount, and the like are set as variables. The facility information A can also include an installation area necessary for installing each facility. As the parameter B, a load pattern assumed for each load of the electric heat model 20 (a state of use of power and heat for each season and each time) is set as a variable. As the optimization objective information C, economic efficiency (cost) associated with operation of each facility, a target value of CO ₂ emission amount, and the like are set. As the constraint condition D of the electric heating model, a constraint condition of energy supply and demand between each facility of the electric heating model is set.

  Hereinafter, the facility information A, the load pattern B, and the constraint condition D will be described in detail. Moreover, the improvement of the calculation time in the optimal installation configuration determination apparatus 10 is demonstrated based on these settings and constraint conditions.

(1) Electrothermal model constraint D
As energy supply and demand constraints, electricity requires that supply and demand always match, and heat requires that supply always exceed demand.

  In the electric heating model of FIG. 2, the power supply consists of the power purchase 2, the electric power output of the electric heat generator 1, and the discharge output of the storage battery 3, and the electric power demand is the conversion of the electric power load 5, the charging power of the storage battery 3 and the electric power converter 4. Since it consists of electricity, the power supply and demand constraints are as follows.

  The heat supply is composed of the heat output of the electric heat generator 1 and the output of the power converter 4, and the heat demand is composed of the heat load 6, so the heat supply and demand constraints are as follows.

(2) Parameter B of power / heat load pattern
For each of the electric power load and the heat load, a daily load pattern is given by hourly data, and these are prepared for three seasons.

  In the power purchase, for each power purchase contract condition, the contract power is set as an integer variable, and the amount of power purchased per hour is set as a continuous variable. A power purchase contract is subject to minimum contract power and maximum contract power. In addition, the amount of power purchased at each hour is less than the contract power. In addition, since power purchase selects only one contract from a plurality of contract conditions, a binary variable is prepared for each contract condition to give a restriction that only one contract can be selected. When these are combined, the following equations (3) to (5) are obtained.

(3) Facility information A
The cost of purchasing electricity is the basic monthly charge proportional to the contracted power, and the amount of electricity purchased, proportional to the unit price depending on the season (from 8:00 to 21:00), night (22:00 to 7:00) and the season. It consists of a fee, and the CO ₂ emission is a value proportional to the amount of power purchased.

Each of the electric heat generator 1, the storage battery 3, and the power converter 4 prepares specifications related to a plurality of facilities, and expresses the number of introductions of each facility as an integer variable. The introduction cost of each facility and the CO ₂ emissions accompanying the introduction are given as specifications.

  The electric heat generator 1 sets an integer variable for setting the number of units to be operated and a continuous variable for setting the operation level every hour, and restricts the operation level as in the following formulas (6) and (7). It expresses that operation below the output is not possible (see FIG. 3).

Also, for each of the minimum operation and maximum operation, the power output, heat output, cost, and CO ₂ emissions from one hour of operation per unit are given as specifications, and the values in between are shown in equation (8). As shown in FIG. 3, linear interpolation is performed (see FIG. 3).

The heat output, cost, and CO ₂ emission amount are similarly interpolated. Although the efficiency is not explicitly expressed in the electrothermal model, since the fuel consumption is reflected in the cost, it is interpolated as a ratio of approximately two linear quantities. Usually, a curve of an upper chord like a broken line in FIG. 3 is drawn.

  The storage battery 3 and the power converter 4 are simplified models, and the charge level, discharge level, and operation level of the power converter are set as continuous variables. For the battery model, the maximum charging power, maximum discharging power, charging efficiency, and charging capacity are given as specifications, and charging (discharging) is performed at or below the maximum charging (discharging) power according to the level of charging (discharging). Sometimes charging power × charging efficiency is charged. Although charging and discharging cannot be performed simultaneously with one storage battery, this is indicated by the fact that the total number of charging / discharging levels cannot exceed the number of introductions as shown in equation (9).

  Further, as shown in the following equation (10), it is imposed that the accumulated charge amount per day is equal to or less than the charge capacity and the accumulated discharge amount per day is equal to or less than the accumulated charge amount per day.

  The power converter 4 gives the used power and heat output at the time of maximum operation as specifications, and operates linearly according to the operation level as in the following (11) to (13).

In addition, the cost by the operation of the storage battery 3 and the power converter 4 and the CO ₂ emission amount are not taken into account. In addition, the cost and CO ₂ emissions of the storage battery 3 and power converter 4 are integrated with what is required when installing each facility and what is required when operating them according to the load pattern for the specified number of years. To do. When accumulating, the number of days in each season is also taken into account.

(4) Improvement of calculation time 4.1 Problem of calculation time The calculation of the optimum equipment configuration by the determining device 10 based on the above settings and the constraint conditions is performed by setting the number of operation units and the operation level of the electric heat generator 1 to 1 Although it can be set every hour, the number of integer variables for setting the number of operating units increases. This number is 72 variables (24 hours × 3 seasons) per electric heat generator model, and reaches 360 variables in five models. In general, such a problem can be practically solved because it is estimated that there are up to about 150 integer variables. Therefore, depending on the data to be set, it is predicted that it cannot be practically solved. Example occurs.

  In order to remedy this problem, the operating unit variable is substantially reduced. Specifically, a new time section in which several times are collected is prepared, and this is used as a unit of time for controlling the number of operating units, and a restriction is added to match the number of operating units in each time belonging to the same time section. This is equivalent to having one operating unit variable for each time segment.

  Considering that the number of integer variables is limited to about 150, the time division is considered to be about 25 to 30 divisions per model, assuming that there are five candidate models for the electric heat generator. Each season has 8 to 10 divisions, but it is not necessary to have the same time division for all seasons.

  Of course, the introduction of new time divisions will reduce the number of operating points, so the calculation result of the optimized equipment configuration will change (deteriorate). However, the difference can be reduced by devising the time division. I think I can.

4.2 Time division method The number of operating units and the operating output are controlled according to the load, but it is natural to respond to small load fluctuations with operating output and large load fluctuations with operating units. For this reason, if the time division for controlling the number of operating units is created based on the result of roughly classifying the load values, it is possible to reduce the variables without degrading the results in line with the fact that the number of operating units should actually change. Conceivable. Therefore, we decided to dynamically generate time segments from the classification of load pattern values.

  Specifically, time is classified according to the values of power load and heat load, and as shown in the example in FIG. 4, continuous time belonging to the same classification for both power load and heat load is set as one time section, and the same section In the time to enter, the number of operating electric heat generators, etc. is the same, and the integer variables are substantially reduced.

  The time division according to this method generally varies from season to season. The following four methods are used as the value classification method.

(A) Annual maximum load ratio 4 classification Looking at the ratio to the maximum load for each year of power and heat, less than 1/4, 1/4 or more and less than 1/2, 1/2 or more and less than 3/4, 3 / Classify into 4 or more.

  When the load value moves cleanly between the respective classes, the time classification of 6 classes for each season is obtained for each load from the 4 classes. When power and heat are combined, each season has 6 to 12 divisions, but load fluctuations of power and heat tend to be similar, so it can be expected that the time divisions will fall within a reasonable number.

(B) Annual maximum load ratio 5 classification For each of power and heat, looking at the ratio of maximum annual load, less than 1/8, 1/8 or more, less than 1/4, 1/4 or more, less than 1/2, / 2 or more, less than 3/4, and 3 or more.

  Since the output of the electric heat generator cannot be reduced below the minimum value, the control of the number of operating units at low load is further detailed. Ideally, there are 8 time segments for each season for each load. When the power and heat are combined, each season has 8 to 16 divisions. However, if the load fluctuations of power and heat overlap, the time division can be expected to fall within a reasonable number.

(C) Seasonal maximum load ratio 4 classification Looking at the ratio to the maximum load in each season for each of electric power and heat, less than 1/4, 1/4 or more and less than 1/2, 1/2 or more and less than 3/4, 3 / Classify 4 or more.

  This is because the time division for each season is made equal since the time division is reduced in the low load season when the difference in load due to the season is large based on the maximum annual load.

(D) Seasonal maximum load ratio of 5 categories Looking at the ratio to the maximum load in each season for power and heat, less than 1/8, 1/8 or more, less than 1/4, 1/4 or more, less than 1/2, / 2 or more, less than 3/4, and 3 or more.

(5) Trial Calculation of Optimal Equipment Configuration This trial calculation was performed by building a model with the modeling language << SIMPLE >> using a commercially available mathematical programming package (Mathematical System (NUOPT)).

In addition, the following data was prepared for calculation. Since optimization was performed with the total cost of operation for 10 years, data on CO ₂ emissions is omitted.

5.1 Load pattern The load pattern is based on public information and the power and heat (hot water supply, cooling, heating) of standard office (5000 m ² ), store (10000 m ² ), hotel (30000 m ² ) as a consumer. Total) 24 hour December load data was created and reorganized in the summer (July-September), winter (March-March) and interim periods taking into account the number of days in each season.

  As a feature of the load, the load varies greatly with time and season in the office. The store's load changes greatly with time, but there is little seasonal variation. The hotel has a certain load even when the load is low.

5.2 Power purchase model The power purchase model prepared two contracts (BP1, BP2). The conditions are shown in Table 1.

5.3 Electric heat generator model Two types of boiler models (BR1, BR2) and three cogeneration models (GE1, GE2, GE3) were prepared as electric heat generator models. The specifications are shown in Table 2.

<5.4> Storage battery model Three types of storage battery models (ST1, ST2, ST3) were prepared. The specifications are shown in Table 3.

5.5 Power converter model Two models (EH1, EH2) of power converter models were prepared. The specifications are shown in Table 4. The output energy of the power converter exceeds the input energy because a heat pump is assumed.

5.6 Trial calculation Using each of the above data, a cost optimization decision was calculated using an optimal equipment planning model. Regarding the time division, in addition to the hourly division, (a) the four annual maximum load ratio classification categories, (b) the five annual maximum load ratio five classification categories, (c) the seasonal maximum load ratio four classification categories, (d ) Calculations were made for five classification categories with seasonal maximum load ratios.

  First, when calculating the time division for controlling the number of operating units from the load data, the number of time divisions is as shown in Table 5. In the hotel load pattern, the time classification is the same for the 4 classifications and the 5 classifications. For this reason, five classifications of hotels are omitted in the optimization calculation.

  Table 6 shows the result of the optimization calculation. However, the optimization of the hourly segment in the hotel load pattern did not produce a result even if it was calculated for more than 4 days, so the lower bound of cost at this point was described for reference. Note that the upper bound at this point was omitted because it was larger than the results of the four categories of seasonal maximum load ratios.

  In the trial calculation results shown in Table 6, the number of time segments is almost reasonable in the trial calculation load pattern. In addition, it was found that even if the time division for controlling the number of operating units was set by classification based on the load value, the effect on the optimization result was suppressed to a few percent. In particular, in the 5 classifications, in an office or a store, the equipment configuration is the same as every hour (the contracted power of the office is different) and the cost is only about 0.1%. . As for the hotel, since the optimal solution for each hour is not obtained, the equipment configuration cannot be compared, but the cost is within 2% even when compared with the provisional lower bound value, which is considered to be sufficiently close to the optimal solution.

  As for the calculation time, the office and the store could be processed in a short time of less than 1 minute every hour, and there was no significant difference. However, in a hotel, the calculation that is not expected to be completed every hour can be greatly shortened to 5 to 6 minutes.

(6) Summary As described above, the electric heat optimum equipment plan is modeled as a mixed integer linear programming problem, and the optimum equipment configuration can be calculated by a branch and bound method for the prepared equipment list and a given load pattern.

  In addition, it becomes a model with many integer variables to express the partial load characteristics of the electric heat generator, and depending on the data, it takes a lot of calculation time, but by creating the operation time division dynamically from the load pattern, the result is The number of substantial integer variables could be reduced while reducing the effect, and an appropriate solution could be obtained in a practical calculation time.

  The present invention is a development support method having a part or all of the processing procedure of the equipment plan support device of the combined electric and heating system shown in the above embodiment, or the processing procedure of this method can be executed by a computer. Can be provided as a program.

The equipment plan assistance apparatus of the electric heat combined supply system which shows embodiment of this invention. An example of an electric heating model. Example of partial load characteristics. Examples of load patterns and time divisions.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Electric heat generator 2 Purchased electricity 3 Storage battery 4 Power converter 5 Electric power load 6 Thermal load 10 Optimal equipment configuration determination apparatus 20 Electric heat model

Claims (5)

Electric heat equipment information of the electric heat model that expresses the equipment plan of the combined electric heat system, electric load / heat load pattern required for the electric heat model, optimizing information on the operating cost or CO ₂ emission amount of the electric heat model, and electric heat model In the facility planning support device of the combined heat and power system with the optimized facility configuration determination device that solves the facility configuration optimization problem by the branch and bound method by setting these information as variables,
For each time segment roughly classifying the power load and heat load, the number of operating electric heating facilities is set as an integer variable, and the time segment is defined as one time segment for consecutive times belonging to the same category for both power load and heat load. An equipment plan support apparatus for an electric and heat cogeneration system, comprising dynamic time section setting means for setting in the optimized equipment configuration determining apparatus. The dynamic time segment setting means includes:
The annual maximum load ratio 4 classification means set by classifying into 4 time segments from the ratio to the maximum load for each year of electricity and heat,
The annual maximum load ratio 5 classification means to classify and set in 5 time segments from the ratio to the maximum load for each year of electricity and heat,
The seasonal maximum load ratio 4 classification means set by classifying into 4 time segments from the ratio to the maximum load in each season for each electric power and heat,
5 seasonally maximum load ratio classification means set by classifying into 5 time segments from the ratio of maximum load in each season for each of electric power and heat,
The facility planning support device for the combined electric and heat system according to claim 1, comprising any one of the means. Electric heat equipment information of the electric heat model that expresses the equipment plan of the combined electric heat system, electric load / heat load pattern required for the electric heat model, optimizing information on the operating cost or CO ₂ emission amount of the electric heat model, and electric heat model In the facility planning support method of the combined heat and power system with the optimized facility configuration determination device that solves the facility configuration optimization problem by the branch and bound method by setting these information as variables,
For each time segment roughly classifying the power load and heat load, the number of operating electric heating facilities is set as an integer variable, and the time segment is defined as one time segment for consecutive times belonging to the same category for both power load and heat load. A facility planning support method for a combined electric and heat system comprising a dynamic time segment setting procedure for setting in the optimized facility configuration determining device. The dynamic time segment setting procedure includes:
The annual maximum load ratio 4 classification procedure set by classifying into 4 time segments from the ratio to the maximum load for each year for power and heat,
The annual maximum load ratio 5 classification procedure to set by classifying into 5 time segments from the ratio to the maximum load for each year of electricity and heat,
Four seasonal maximum load ratio classification procedures set by classifying into four time segments from the ratio of maximum load in each season for power and heat,
5 seasonal maximum load ratio classification procedures set by classifying into 5 time segments from the ratio of maximum load in each season for each electric power and heat,
4. The facility planning support method for the combined electric and heat system according to claim 3, comprising any one of the procedures.   The program which comprised so that the processing procedure in the equipment plan assistance method of the electric-heat combined supply system of Claims 3 and 4 was executable with a computer.

Images