Control design support method and control design support device

Abstract

To generate control logic for an energy supply system equipped with multiple energy facilities in a short period and at low cost.SOLUTION: A control design support method for assisting in designing control logic for an actual system, has a first step for creating a first control logic within a control simulator that determines load distribution for each facility by importing specification data for each facility constituting an actual system, a second step for determining control parameters within the first control logic by analyzing the variation characteristics of prediction errors of constraint conditions, and a third step for simulating, using a control simulator, the load distribution of each facility based on operational plans and constraint condition data, which may include past data or mock data, and a fourth step for constructing a second control logic for determining the load distribution of each facility in the actual system by setting specification data and control parameters validated in the third step into the control device of the actual system.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a control design support method and control design support device that constructs a control logic for adjusting the actual energy load distribution in real time based on various actual measurement values ​​in an energy supply system that includes multiple energy facilities such as electric power facilities that supply electric power, thermal facilities that supply heat, and thermoelectric power facilities that supply both electric power and heat, in relation to the energy load distribution plan for each facility. [Background technology]

[0002] The recent trend toward decarbonization has created a demand for energy supply systems that can meet demand for electricity and heat while suppressing carbon dioxide emissions. This type of energy supply system is typically realized by combining a variety of energy facilities. Known examples of power generation facilities incorporated into energy supply systems include power generation facilities that utilize renewable energy sources such as solar and wind power, as well as storage and discharge facilities such as storage batteries that are often used in conjunction with such power generation facilities. Known examples of heat and power generation or heat generation facilities incorporated into energy supply systems include gas engines and boilers that emit carbon dioxide, and hydrogen-mixed combustion engines, hydrogen boilers, and fuel cells that do not emit carbon dioxide. Known examples of equipment often used in conjunction with hydrogen-consuming heat and power generation facilities include water electrolysis systems that consume electricity to produce hydrogen from water.

[0003] In addition, customer demands for the combination of energy equipment that makes up an energy supply system are becoming more diverse, and incorporating various types of equipment with different characteristics also complicates the design of the control logic for controlling the energy supply system.

[0004] Here, Patent Document 1 is known as a document that illustrates the control logic of an energy supply system. For example, the abstract of the document states that the problem to be solved is to provide power utilization equipment that can respond to various supply and demand requirements for power exchanged with an external power system of the power utilization equipment using a simple configuration without considering the responsiveness of the power output device, and as a solution, states that "the power utilization equipment includes at least one power output device connected to the external power system via a predetermined first connection point, a power converter connected to a second connection point that is closer to the external power system than the first connection point, a battery connected to the power converter, and a controller. The controller includes a first command generation unit that generates a first command to increase or decrease the output power from the power output device based on the first power deviation, and a second command generation unit that generates a second command to adjust the output power based on a second power deviation obtained by subtracting the exchanged power from the exchange power target value. The second command generation unit generates a charge command to charge the battery if the second power deviation exceeds an upper limit value determined based on the power exchange request value, and generates a discharge command to discharge the battery if the second power deviation is below a lower limit value."

[0005] In this way, in Patent Document 1, the control logic is such that the power output device is instructed to increase or decrease the output power based on the first power deviation, the storage battery is instructed to charge if the second power deviation exceeds a predetermined upper limit value, and the storage battery is instructed to discharge if the second power deviation is below a predetermined lower limit value, thereby allowing the power exchanged with the external power system of the power utilization facility to respond to various supply and demand requirements without taking into consideration the responsiveness of the power output device. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2023-13243 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in Patent Document 1, the upper and lower load limit settings when determining the load allocation of multiple power facilities are determined only based on the fluctuation range, such as the maximum and minimum values ​​of power demand, and do not take into account the fluctuation characteristics of the power supply side when renewable energy power generation facilities are included, or the fluctuation characteristics of power demand.As a result, the upper and lower limit settings had to be set with more leeway than necessary, which resulted in a narrowing of the load range that each power facility could assume.Furthermore, Patent Document 1 did not particularly disclose any ideas for speeding up the design of a power system consisting of multiple power facilities.

[0008] Therefore, an object of the present invention is to provide a control design support method and a control design support device that generate control logic for an energy supply system equipped with multiple energy facilities in a short period of time and at low cost by taking into account the fluctuation characteristics of the energy supply and demand sides and optimizing the upper and lower limit settings of the load distribution of each energy facility. [Means for solving the problem]

[0009] In order to solve the above problems, the control design support method of the present invention supports the design of a control logic for an actual system that supplies load according to demand, and includes the following steps: a first step of creating a first control logic in a control simulator that determines the load allocation of each piece of equipment by importing specification data of each piece of equipment that constitutes the actual system; a second step of determining control parameters in the first control logic by analyzing the fluctuation characteristics of prediction errors of constraint conditions; a third step of simulating, using the control simulator, the load allocation of each piece of equipment based on past data or simulated data of the constraint conditions and an operation plan; and a fourth step of constructing a second control logic that determines the load allocation of each piece of equipment of the actual system by setting, in a control device of the actual system, the specification data and the control parameters whose validity has been verified in the third step. [Effects of the Invention]

[0010] According to the control design support method and control design support device of the present invention, it is possible to generate control logic for an energy supply system equipped with a plurality of energy facilities in a short period of time and at low cost. [Brief explanation of the drawings]

[0011] [Figure 1] Schematic diagram of the energy supply system. [Figure 2] 10 is an example of an operation plan optimized by the planning device 2. [Figure 3] FIG. 1 is a diagram showing the flow of data in the control logic. [Figure 4] FIG. 1 is a diagram showing a control design support device according to a first embodiment and its peripheral devices. [Figure 5] FIG. 2 is a diagram showing a network that defines the connection relationships of the facilities in the first embodiment. [Figure 6] FIG. 2 is a diagram showing data defining a gas engine module. [Figure 7] FIG. 4 is a diagram showing data defining a storage battery module. [Figure 8] FIG. 1 shows data defining a PV module. [Figure 9] FIG. 10 is a diagram showing time-series data of PV power generation, which is a constraint condition for the investment simulator. [Figure 10] FIG. 2 is a diagram showing data defining a grid power module. [Figure 11] FIG. 2 illustrates data defining a power demand module. [Figure 12] FIG. 10 is a diagram showing time-series data of electricity demand, which is a constraint condition of the investment simulator. [Figure 13] FIG. 10 is a diagram showing data defining the conditions of an investment simulator. [Figure 14] FIG. 10 is a diagram showing the analysis results of an investment simulator. [Figure 15] FIG. 10 is a diagram showing data defining analysis conditions for the planning simulator. [Figure 16] FIG. 10 is a diagram showing time-series data of PV power generation amount, which is a constraint condition of the planning simulator. [Figure 17]FIG. 10 is a diagram showing time-series data of power demand that serves as a constraint condition for the planning simulator. [Figure 18] FIG. 2 is a diagram showing the configuration of a control simulator. [Figure 19] FIG. 10 is a diagram showing analysis conditions of a control simulator. [Figure 20] FIG. 10 is a diagram showing time-series data of PV power generation amount, which is a constraint condition of the control simulator. [Figure 21] FIG. 10 is a diagram showing time-series data of power demand that serves as a constraint condition for the control simulator. [Figure 22] FIG. 10 is a diagram showing the simulation results of a control simulator. [Figure 23] FIG. 10 is a graph showing the simulation results of a control simulator. [Figure 24] FIG. 10 is a diagram showing a method for evaluating a margin when determining a control parameter. [Figure 25] FIG. 10 is a diagram showing a network that defines the connection relationships of equipment in the second embodiment. [Figure 26] FIG. 2 is a diagram showing data defining a gas engine module. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of a control design support device according to the present invention will be described with reference to the drawings. [Example]

[0013] <Energy supply system> First, the configuration of an energy supply system 100, which is a practical system, will be described with reference to Figures 1 to 3. In this embodiment, electric power is also referred to as a load.

[0014] 1 is a configuration diagram of an energy supply system 100. As shown here, the energy supply system 100 includes a control device 1, a planning device 2, a gas engine 3, a storage battery 4, a photovoltaic power generation facility 5 (hereinafter referred to as "PV 5"), and sensors 6 and 7. Each of these will be explained in turn below.

[0015] <<Power equipment>> The gas engine 3, storage battery 4, and PV 5 are power facilities that work in conjunction with the grid power and supply power to various demand facilities via shared power lines. Sensor 6 measures the amount of power taken in from the grid, and sensor 7 measures the total power consumption (power demand) of the demand facilities.

[0016] <<Planning Device 2>> The planning device 2 is a device for transmitting, as an operation plan to the control device 1, the load distribution of each power facility optimized based on a target function f designated in advance by a designer or the like, using predicted values ​​of power demand and PV power generation as constraint conditions (inputs). Specifically, the planning device 2 is a computer equipped with hardware such as a calculation device such as a CPU, a storage device such as a semiconductor memory, and a communication device. The calculation device executes predetermined programs to realize each function, but the following description will omit such well-known techniques as appropriate.

[0017] Here, the target function f is a function that takes into account the cost function, the CO2 emission function, or both, and is a function that finds the load allocation of each power facility over time that minimizes cost and CO2 emissions.The reason that only the power generation amount of PV5 among the power facilities is treated as a constraint is that PV power generation is dependent on the weather, and the upper limit of power generation cannot be controlled arbitrarily.

[0018] Figure 2 is an example of an operation plan optimized by the planning device 2, and is a table that expresses the load of each power facility for one day, January 1, 2023, as time-series planned value data at 30-minute intervals. For example, the data row from 12:00 to 12:30 in the operation plan specifies an operation plan in which the grid power outputs 200 kW of power, the gas engine 3 outputs 300 kW of power, 500 kW of power is input (charged) to the storage battery 4, and the PV 5 outputs 1500 kW of power. This data row indicates that the planning device 2 predicts that if the operation of each power facility is controlled according to each planned value, costs and CO2 emissions for that time period can be minimized.

[0019] <<Control device 1>> The control device 1 is a device for controlling each piece of power equipment that constitutes the actual system based on the operation plan from the planning device 2. However, the operation plan is merely prediction data and does not guarantee the minimization of costs, etc., so in order to minimize costs, etc. in the actual environment, it is necessary to adjust the contents of the operation plan according to the current situation.

[0020] Therefore, the control logic L1 implemented in the control device 1 calculates the load distribution of each piece of power equipment in real time based on the operation plan (planned values) drawn up by the planning device 2 and the actual measured values ​​of power demand measured by the sensor 7, and controls each piece of power equipment so as to minimize actual costs and CO2 emissions by outputting load command values ​​according to each load distribution to each piece of power equipment. Note that the control device 1 having such functions is specifically a computer similar to the planning device 2.

[0021] Fig. 3 is a diagram showing the flow of data during load distribution adjustment in the control logic L1 of the control device 1. The control logic L1 illustrated here is configured with four blocks B1 to B4, the same number as the number of power sources, as it controls the energy supply system 100 of Fig. 1, which is equipped with four types of power sources.

[0022] Block B1 is a control logic corresponding to grid power, which is the power source with the highest usage priority (priority P1). In this block B1, the power value a obtained by subtracting the sum of the power generation plan value of the gas engine 3 and storage battery 4 at the current time calculated from the operation plan and the actual power generation value of the PV 5 from the actual measured value of the demand power measured by the sensor 7 is input to the range limiter for grid power. This range limiter is configured to detect when the power value a is lower than the lower limit E L and upper limit E U If the power value a is within the range of the lower limit E, the power value a is output as is. L If it is less than or equal to the lower limit E L The power value a is output when the upper limit E U If it is equal to or greater than the upper limit E UNote that the output of the range limiter of block B1 is formally a power generation output request value for the grid power, but since the grid power is outside the control of the control device 1, this output is actually useless data.

[0023] Furthermore, block B1 outputs the deviation amount when the power value a falls outside the range of the range limiter as a correction value to block B2. Note that the correction value (deviation amount from the range limiter) output by block B1 is the upper limit value E U Max(aE U ,0) and the lower limit value E of the range limiter L Min(aE L ,0).

[0024] Block B2 is a control logic corresponding to the storage battery 4, which is the power source with the second highest usage priority (priority P2). This block B2 calculates the power value a obtained by adding together the power generation plan value of the storage battery 4 at the current time calculated from the operation plan and the correction value output by block B1, and calculates the power value a by adjusting the lower limit E of the range limiter of the storage battery 4. L and upper limit E U Within this range, block B2 outputs the power generation output request value for the storage battery 4 (i.e., the load allocation of the storage battery 4). Furthermore, when the power value a falls outside the range of the range limiter, block B2 outputs the deviation amount to block B3 as a correction value.

[0025] Block B3 is a control logic corresponding to the gas engine 3, which is a power source with a usage priority (priority P3) second only to priority P2. This block B3 first generates a power value a by adding the power generation plan value of the gas engine 3 at the current time calculated from the operation plan and the correction value output by block B2. Next, the power storage limiter monitoring unit of block B3 monitors the power storage amount of the storage battery 4 and performs the following process.

[0026] That is, if the amount of stored power in the storage battery 4 is less than a predetermined amount, the storage amount limiter monitoring unit outputs the power value a to the range limiter and outputs the power value 0 to the block B4. Then, the range limiter, by the above-mentioned mechanism, outputs the lower limit value E L and upper limit E U Within this range, block B3 outputs the power generation output requirement value for the gas engine 3 (i.e., the load allocation of the gas engine 3). At this time, block B3 outputs the deviation amount when the power value a falls outside the range limiter range as a correction value to block B4.

[0027] On the other hand, if the amount of stored power in the storage battery 4 is equal to or greater than the predetermined amount, the storage amount limiter monitoring unit outputs a power value of 0 to the range limiter and a power value of a to the block B4. L and upper limit E U The block B3 outputs the power generation output requirement value for the gas engine 3 within the range of the range limiter. At this time, the block B3 outputs the deviation amount when the power value 0 is outside the range limiter and the sum of the power value a as a correction value to the block B4.

[0028] Block B4 is a control logic corresponding to PV5, which is a power source with the second highest usage priority (priority P4) after priority P3. This block B4 calculates the power value a obtained by adding the actual power output measurement value output by the sensor in PV5 and the correction value output by block B3, and calculates the power value a by adjusting it to the lower limit E of the range limiter of PV5. L and upper limit E U The power generation output demand value for PV5 (i.e., the load allocation for PV5) is output within the range of 1. Even if a power generation output demand value greater than the actual measured power generation output value of PV5 is generated, the amount of power generated by PV5 cannot be increased in response, so the power generation output demand value for PV5 is intended to be used solely as a command to suppress the actual measured power generation output value of PV5.

[0029] By processing each block described above, the load distribution of multiple facilities is adjusted according to the error between the environment assumed when generating the operation plan in the planning device 2 and the actual environment, thereby minimizing the cost and CO2 emissions of the energy supply system 100 in the actual environment.

[0030] As shown in Figure 2, the data rows of the operation plan are at 30-minute intervals, so if the processing cycle of the control logic L1 is set to 1 minute, the data row for 0:00 should be taken for the period from 0:00 to 0:29, and the data row for 0:30 should be taken for the period from 0:30 to 0:59.

[0031] <Control design support device 10, planning simulator 20, investment simulator 30> Next, the control design support device 10, the plan simulator 20, and the investment simulator 30 of the present invention will be described in detail.

[0032] FIG. 4 shows the configuration of the control device 1 and the control design support device 10, and the connection with peripheral devices (planning simulator 20, investment simulator 30) that cooperate with them.

[0033] The control design support device 10 is a device for supporting the design of the control logic L1 to be implemented in the control device 1, and is composed of a control simulator 11 and a control parameter determination unit 12. The control logic L11 implemented in the control simulator 11 is logic that executes the same processing (see FIG. 3) as the control logic L1 to be implemented in the control device 1. The control parameter determination unit 12 is a functional unit that determines the control parameters of both control logics.

[0034] The plan simulator 20 is a simulator that executes the same processing as the planning device 2, and generates an operation plan based on desk calculations that minimizes short-term operation costs and the like.

[0035] The relationship between the control simulator 11 (after the control parameters have been determined) and the planning simulator 20 in Fig. 4 is equivalent to the relationship between the control device 1 (after the control parameters have been determined) and the planning device 2 in Fig. 1. Therefore, the simulation results obtained by the control simulator 11 importing the operation plan optimized by the planning simulator 20 are equivalent to the operation of the actual system obtained by the control device 1 importing the operation plan optimized by the planning device 2.

[0036] Next, we will explain the investment simulator 30. The above-mentioned plan simulator 20 is a simulator that generates an operation plan that minimizes short-term operation costs, etc., but the investment simulator 30 is a simulator that calculates long-term operation costs or CO2 emissions and determines equipment specifications based on these calculation results, and is a functional unit that does not exist in the actual system shown in Figure 1.

[0037] Here, the equipment specifications determined by the investment simulator 30 refer to the combination of the type of power equipment (gas engine, storage battery, PV, etc.) selected and the capacity of each power equipment (load range: minimum output to rated output, amount of stored power, etc.). In other words, in the case of power equipment specifications, the investment simulator 30 calculates the cost and CO2 emission reduction amount for the required annual pattern of power demand, and determines the type and capacity of the power equipment accordingly, thereby identifying the equipment specifications.

[0038] The calculation itself in the investment simulator 30 is basically the same as the optimization processing in the planning device 2 and the plan simulator 20. In other words, constraints such as power demand and PV power generation amount are given to the investment simulator 30, and an operation pattern (time-series load distribution) that minimizes the target function defined by cost and CO2 emissions is determined by optimization processing. However, there are the following differences between the processing in the plan simulator 20 and the investment simulator 30. (1) The input power demand and PV power generation amount in the plan simulator 20 are short-term (e.g., one day) time series data, whereas in the investment simulator 30 they are long-term (e.g., one year) time series data. (2) The costs and CO2 emissions calculated by the plan simulator 20 are short-term (e.g., one day) costs, whereas the costs and CO2 emissions calculated by the investment simulator 30 are long-term (e.g., one year) costs. (3) The costs calculated by the planning simulator 20 are only operating costs (OPEX), whereas the costs calculated by the investment simulator 30 also include capital expenditures (CAPEX). (4) The planning simulator 20 does not have the function of determining equipment specifications, but the investment simulator 30 has the function of verifying each equipment specification according to the calculated value of the objective function obtained when the equipment specifications are changed, and determining the optimal equipment specifications.

[0039] Fig. 5 shows a network N that defines the connection relationships of the power equipment used during simulation in the investment simulator 30. The network N illustrated here corresponds to the actual system in Fig. 1, and is composed of a gas engine module M3 corresponding to the gas engine 3, a storage battery module M4 corresponding to the storage battery 4, a PV module M5 corresponding to the PV 5, a grid power module M6 corresponding to the grid power measured by the sensor 6, a power demand module M7 corresponding to the power demand measured by the sensor 7, and a power connection module M1 that relays input and output between each module.

[0040] If such modules for each facility are prepared, any energy supply system 100 can be expressed in network form by combining them appropriately with the type and number of facilities. Details of the data that is set for each module, such as input / output characteristics and upper and lower output limits, will be explained below.

[0041] <<Data for gas engine module>> FIG. 6 is an example of data D3 set in the gas engine module M3. As shown in FIG. 5, the gas engine module M3 determines the amount of gas (Nm 3) is input, and the power (kWh), cost (yen), and CO2 emissions (tons) are output. Therefore, the following information is registered in data D3.

[0042] The input / output characteristics of the output D31 (power) are Y=5X, and the gas is 1Nm 3 The relationship is set such that 5 kWh of electricity is output (generated) per hour. The upper limit of output D31 (electricity) is set to one hour of the rated output (1000 kW).

[0043] The input / output characteristics of output D32 (cost) is Y=0.08X, and gas 1Nm 3 The relationship is set such that the unit price is 0.08 yen.

[0044] The input / output characteristics of output D33 (CO2 emissions) are Y = 0.002X, and the gas 3 The relationship is set to 0.002 tonnes of CO2 per unit of energy consumed.

[0045] The data D3 also includes the equipment cost of gas engine 3, which is 100 million yen.

[0046] <<Data for battery module>> Fig. 7 shows an example of data D4 set in the storage battery module M4. As shown in Fig. 5, the storage battery module M4 inputs and outputs electric power (kWh). Therefore, the following information is registered in the data D4:

[0047] The output characteristic of output D41 (power) is Y = 0.9X, and the relationship is set such that the efficiency during charging and discharging is 90%. In addition, 500kW is set as the upper limit corresponding to the discharge rated value, and -500kW is set as the lower limit corresponding to the charge rated value. Discharging is expressed as a positive value because it supplies power to demand like other power facilities, while charging is expressed as a negative value because it takes in other power.

[0048] The difference between the storage battery 4 and other power facilities is that it has the function of storing power. That is, in the storage battery 4, the stored power amount is added by the charged amount, and subtracted by the discharged amount. In this data D4, the upper limit value of the stored power amount is set to 4000 kWh, and the storage battery 4 is controlled so as not to increase the stored power amount beyond this value. Also, in the data D4, the equipment cost of the storage battery 4 is set to 50 M￥.

[0049] <<Data for PV module>> Figure 8 shows an example of the data D5 set in the PV module M5. As shown in Figure 5, the PV module M5 outputs electric power (kWh). As described above, PV5 is a power facility that cannot be controlled, and it is a constraint condition when obtaining an operation plan through optimization. Therefore, in the item of the input / output characteristics of the data D5, this constraint condition is given in a file format (for example, the file pv.csv) instead of a mathematical formula. Also, for PV5, the equipment cost is set.

[0050] Figure 9 illustrates the content of the file pv.csv that describes the constraint conditions. In this example, in the CSV (comma separated values) format, the time-series data of the PV power generation amount every 30 minutes is described. As is obvious from Figure 9, in the file pv.csv, the data in the first column is the date and time data, and the data in the second column is the power generation amount data. Also, in this example, the data for one day is regarded as one unit, and a data set for one year (365 days) is regarded as one file. [[ID=,12]]

[0051] <<Data for grid power module>> [[ID=,16]]Figure 10 shows an example of the data D6 set in the grid power module M6. As shown in Figure 5, the grid power module M6 receives electric power (kWh) as input and outputs electric power (kWh), cost (￥), and CO₂ emission amount (ton). Therefore, the following information is registered in the data D6.

[0052] The input / output characteristics of output D61 (power) are Y=X, which sets a relationship in which input and output are equal. The upper limit of output D61 (power) is set to 2000 kW, and the lower limit is set to 0 kW. The upper and lower limits are set to the upper limit of the grid power, that is, one hour's worth of the physical capacity of the power receiving equipment, 2000 kW.

[0053] The input / output characteristics of output D62 (cost) vary depending on the time of day, with Y = 20X during the 7:00-22:00 time period and Y = 15X during other time periods. These input / output characteristics indicate that the electricity rate during the day is 20 yen / kWh and the electricity rate at night is 15 yen / kWh.

[0054] The input characteristic of output D63 (CO2 emissions) is Y=0.45X, which indicates that 0.45 tons of CO2 are emitted per 1 kWh of input electricity.

[0055] <<Data for the Power Demand Module>> Figure 11 shows data D7 set in the power demand module M7. As shown in Figure 5, power (kWh) is input to the power demand module M7. As with the PV5 described above, power demand is also uncontrollable and is a constraint when determining an operation plan through optimization. Therefore, in the input / output characteristics section of data D7, this constraint is given in file format (for example, file demand.csv) rather than as a formula.

[0056] FIG. 12 shows an example of the contents of the file demand.csv, which describes constraints. In this example, time-series data for power demand every 30 minutes is written in CSV format. As is clear from FIG. 12, in the file demand.csv, the data in the first column is date and time data, and the data in the second column is power demand data. In this example, one day's worth of data is considered one unit, and one year's worth of data (365 days) is considered one file.

[0057] The above is the definition of each module shown in Figure 5. Because the input / output characteristics of each module are expressed in linear form, any output variable can be defined for any equipment. By adopting such a general-purpose method, it becomes possible to perform analysis under conditions of equipment configuration and equipment specifications (capacity, etc.) that meet customer requirements. The reason for using linear forms for input / output characteristics is to apply linear programming to the optimization process of operation plans, which will be described later.

[0058] <<Processing by Investment Simulator 30>> Next, the analysis method by the investment simulator 30 will be described in detail with reference to FIGS.

[0059] 13 is an example of definition data D8 that defines the conditions of the investment simulator 30. In this definition data D8, the sum of the output D62 (cost) of the grid power module M6 and the output D32 (cost) of the gas engine module M3 is defined as the objective function of the optimization process. This allows the investment simulator 30 to obtain an operation plan that minimizes the total cost of electricity and gas through optimization on a daily basis. Then, by repeating this process for 365 days, an operation plan for one year can be obtained.

[0060] Fig. 14 is an example of an analysis result D9 output by the investment simulator 30. Compared to the operation plan of Fig. 2 output by the planning device 2, the analysis result D9 of Fig. 14 has the following characteristics: cost data is also output; and the results for one year (365 days if it is not a leap year, and 366 days if it is a leap year) are output. Therefore, by aggregating this analysis result D9, the operating cost (total value of gas and electricity costs) for any one year can be calculated.

[0061] Furthermore, the module data (data D3, D4, and D5) for each piece of power equipment also defines the equipment costs, so if the power equipment to be used can be identified, the equipment costs can also be obtained. Therefore, the investment simulator 30 can determine conditions that optimize operating costs and equipment costs while changing the type and capacity of the equipment. Changing the type of equipment simply requires changing the definition data for the equipment configuration shown in Figure 5, and changing the specifications of each piece of equipment, such as capacity, simply requires changing the setting data for each module.

[0062] One way to use the investment simulator 30 is to present to the customer the changes in equipment costs and annual operating costs when the equipment conditions are changed. In this case, the customer can select the equipment configuration and equipment specifications that best suit their needs based on this cost information. In designing the planning device 2 and the control device 1, the efficiency of the design work can be improved by utilizing the planning / control simulator based on this data.

[0063] <<Processing by the plan simulator 20>> Next, an analysis by the planning simulator 20 and the control simulator 11, which is carried out after selecting the equipment configuration and equipment specifications of the actual system based on the simulation results of the investment simulator 30, will be described.

[0064] The internal processing in the planning simulator 20 is the same as that in the above-described investment simulator 30. Therefore, the planning device 2, the planning simulator 20, and the investment simulator 30 implement the same optimization processing.

[0065] Fig. 15 shows definition data D10 that defines the analysis conditions of the plan simulator 20. The objective function in this definition data D10 is the sum of the cost of grid power and the cost of the gas engine, similar to the definition data D8 in Fig. 13 that defines the conditions of the investment simulator 30. Furthermore, in the definition data D10, an upper limit of 1900 kW and a lower limit of 100 kW are set for the grid power. On the other hand, in the definition data D6 for the grid power module M6 shown in Fig. 10, an upper limit of 2000 kW and a lower limit of 0 kW are set.

[0066] The difference between the two is that the upper and lower limits in Fig. 10 are physical values ​​determined by equipment specifications, while the upper and lower limits in Fig. 15 are values ​​for obtaining an operation plan that realizes system operation with a margin of error. In other words, the plan simulator 20 using the definition data D10 in Fig. 15 generates an operation plan for grid power within the range of upper and lower limits (100 to 1900 kW) that provides a margin of 100 kW relative to the physical upper and lower limits (0 to 2000 kW) of grid power. The purpose of this is to avoid various problems caused by high-frequency fluctuations contained in the actual values ​​of power demand and PV power generation.

[0067] In other words, if an operation plan is created using a load close to the upper or lower physical limits of the equipment, the load of the equipment may reach the upper or lower limits in the event of a sudden change in the constraint conditions, which may cause a disruption to the power supply (for example, a power outage).In order to prevent such a situation, in this embodiment, a margin is provided in the load range used when creating the operation plan.The same applies to the capacity of the storage battery 4, and an operation plan is created by providing a margin of 100 kWh for the physical capacity of 4000 kWh shown in data D4 in Figure 7.

[0068] 16 and 17 show examples of time-series data D11 of PV power generation and time-series data D12 of power demand, in which the PV power generation and power demand for one day are defined at 30-minute intervals.

[0069] The plan simulator 20 uses the above-described conditions to calculate the load of each facility so as to minimize costs under the environment of the time-series data D11 of power demand and the time-series data D12 of PV power generation, and outputs the calculated load as an operation plan. The operation plan by the plan simulator 20 is also output in the same format as the operation plan by the planning device 2, as shown in FIG.

[0070] <<Processing by Control Simulator 11>> Next, the processing of the control simulator 11 will be described. Fig. 18 is a diagram showing the details of the control simulator 11. As shown here, the control simulator 11 is implemented with a control logic L11 that performs the same processing as the control logic L1 in Fig. 3. However, a difference from the configuration of the control device 1 is that the control simulator 11 is implemented with equipment models (a grid power model 11a, a gas engine model 11b, a storage battery model 11c, and a PV model 11d) that simulate the control operation of each piece of equipment.

[0071] The actual control device 1 incorporated in the energy supply system 100 is connected to the control unit of each power facility and operates each power facility by outputting a control signal to each control unit. The control simulator 11, which simulates this, outputs a control signal to each facility model instead of the actual power facility, and each facility model calculates a control operation in response to the control signal. Each facility model then obtains upper and lower limit values ​​for the load from the facility specification data defined during analysis by the investment simulator 30, and outputs a load within that range. Just as upper and lower limit values ​​are set for each facility model, upper and lower limit values ​​for each facility are also set in the control logic L11 of the control simulator 11.

[0072] Here, the control logic L11 of the control simulator 11 has the same function as the control logic L1 of the control device 1 illustrated in Fig. 3. That is, in the blocks B1 to B4 of the control logic L11, the range limiters in each block are set with values ​​defined in the equipment specification data of the investment simulator 30 as upper and lower limit values.

[0073] Furthermore, parameters not present in the control logic L1 are additionally set in the control logic L11 of the control simulator 11. Definition data D13 that defines the analysis conditions of the control simulator 11 is shown in FIG.

[0074] As shown here, in the definition data D13, a control priority P (P1 to P4 in this example) is set for each piece of power equipment. This defines the order of each block in the control logic L11. Other parameters that are additionally set in the definition data D13 include upper and lower limit values ​​for the grid power module M6. This is a load range that takes into account a margin, similar to the condition settings of the planning simulator 20 described above. For example, for the physical upper and lower limit values ​​of the grid power of 0-2000 kW, a margin of 100 kW is taken into account in the control operation, and the load range is set to 100-1900 kW.

[0075] As mentioned above, the default values ​​of the upper and lower limits of each block of the control logic L1 illustrated in Fig. 3 are the same as those in the equipment specification data of the investment simulator 30. However, if upper and lower limits are set in the definition data D13 as shown in Fig. 19, they are overwritten with those values.

[0076] In addition, in the definition data D13 of Fig. 19, the start-up time and time constant are set for the gas engine module M3. Therefore, the control simulator 11 can perform an analysis that also takes into account the response delay of the equipment. Here, the start-up time and the delay time constant are set in the equipment model of the gas engine 3 as parameters related to the response time, and are reflected in the response calculation. Note that, since there is no response delay to control signals in equipment other than the gas engine, these parameters are not set.

[0077] 20 and 21 show the time series data D14 of PV power generation amount and the time series data D15 of power demand, which are input to the control simulator 11. The difference between the time series data used in the plan simulator 20 and the investment simulator 30 (see FIGS. 9, 12, 16, and 17) and the time series data used in the control simulator 11 is that the time intervals of the former are long, at 30 minutes, while the time intervals of the latter are short, at 1 second.

[0078] The purpose of the analysis by the control simulator 11 is to verify whether the network N equivalent to the actual system will supply the power demand without excess or shortage even when time-series data of constraint conditions (PV power generation amount, power demand) including instantaneous fluctuations that may actually occur is input, in other words, whether the set facility specifications, parameters, etc. are effective in avoiding a power system interruption (power outage) caused by an instantaneous gap between demand and supply. Therefore, by using actually measured past data or simulated data created to approximate actual fluctuations as the PV power generation amount time-series data D14 and the power demand time-series data D15, the control robustness of the energy supply system 100 in an environment where the weather suddenly changes and the PV power generation amount suddenly changes is verified.

[0079] FIG. 22 is an example of an operation plan for each facility output by the control simulator 11. The difference from the operation plan by the planning device 2 illustrated in FIG. 2 is that the time intervals are shorter, at one second. The time intervals of the operation plan in FIG. 22 correspond to the time intervals of the time series data (FIGS. 20 and 21) input to the control simulator 11. The structure of the control logic L11 of the control simulator 11 is similar to the control logic L1 of the control device 1 illustrated in FIG. 3, and the time series data of the PV power generation amount and power demand shown in FIG. 20 and FIG. 21 are input here to calculate a control signal for each power facility.

[0080] An example output from the control simulator 11 is shown in Figure 23. In the figure, the dotted line is the operation plan imported from the plan simulator 20, and the solid line is the result of control calculations performed by the control simulator 11. However, the solid lines for (a) power demand and (e) PV power generation amount do not show calculated values, but rather the data (constraint conditions) given as input. For other equipment, the results are those of control signals calculated by the control logic L11 and responses calculated by the equipment model (as mentioned above, the gas engine 3 has a response delay).

[0081] In Figure 23, when a sudden fluctuation in (a) power demand or (e) PV power generation occurs, the system first responds by adjusting (b) grid power, which has priority P1. However, if it is determined that (b) grid power will exceed the upper or lower limit according to the planned value of (a) power demand, the system adjusts (c) the charge / discharge amount of the storage battery 4 from the planned value, which has priority P2. Furthermore, if it is determined that the charge / discharge amount of the storage battery 4 also exceeds the upper or lower limit, or if it is determined that the upper or lower limit of the storage capacity is reached, the system adjusts (d) the load of the gas engine 3 from the planned value, which has priority P3. Through the above process, the load of the equipment is adjusted from the planned value so that the respective upper and lower limits are satisfied. Note that in Figure 23, the graph (f) of the supply / demand gap shows the difference between power demand and power supply (total for each equipment). This graph shows that the gap is zero throughout the entire time period, confirming that the control operation for this condition can be performed without interrupting the power grid (power outage).

[0082] <<Processing by Control Parameter Determination Unit 12>> Finally, the processing performed by the control parameter determination unit 12 included in the control design support device 10 will be described with reference to FIGS. 4 and 24. As mentioned above, the upper and lower limits set in the control logic L11 of the control simulator 11 are determined by taking into account a margin relative to the physically operable load range in preparation for fluctuations in power demand and PV power generation. For example, if the control period of the control device 1 is set to one minute, if a sudden fluctuation occurs and the upper and lower limits of the equipment with priority P1 are exceeded within one minute, load adjustment will no longer be possible, and a power outage will occur due to the power supply-demand gap. Therefore, to be able to respond to sudden fluctuations, equipment with priority P1 in particular is operated with a margin, avoiding operation at a load close to the upper and lower limits. Below, a method for determining this margin in the control parameter determination unit 12 will be described.

[0083] When the planning device 2 creates an operation plan, it sets predicted values ​​of power demand and PV power generation as constraints. If there is no difference between the predicted values ​​and the actual values, no supply-demand gap will occur if the equipment load is operated according to the operation plan. However, in reality, there will be a prediction error, which will require adjustment from the operation plan. In other words, an appropriate margin amount can be determined from the characteristics of the prediction error.

[0084] Figure 24 shows the process for determining the margin (control parameter) from the prediction error, using PV power generation as an example. When there is a (a) predicted value and a (b) measured value of PV power generation, the difference between them is the (c) prediction error. Next, a frequency filter is used to decompose the fluctuation characteristics of the (c) prediction error into (d) high-frequency components and (e) low-frequency components. The (d) high-frequency components are short-period fluctuation characteristics, and the (e) low-frequency components are long-period fluctuation characteristics. The time interval for the (a) predicted value of PV power generation is determined by the time interval of the weather information used as input for the prediction process, which is usually between 30 minutes and 1 hour. Therefore, changes in PV power generation shorter than this interval will appear as a prediction error. Furthermore, if the weather information used as input differs from the actual value, it will naturally appear as a prediction error longer than the 30-minute to 1-hour interval.

[0085] (e) The supply-demand gap caused by the low-frequency (long-cycle) prediction error can be addressed by load adjustment by the control device 1, so there is no problem. On the other hand, (d) the supply-demand gap caused by the high-frequency (short-cycle) prediction error may not be addressed by load adjustment by the control device 1, and must be addressed by a margin determined by the control parameter determination unit 12. In other words, the control parameter determination unit 12 determines the margin based on the fluctuation range of the high-frequency prediction error (d). In the example of FIG. 24, the high-frequency prediction error (d) has a fluctuation range of ±50 kW, so the control parameter determination unit 12 determines the margin to be, for example, double that, or 100 kW, in consideration of some margin.

[0086] For example, in the definition data D13 in Figure 19, the upper and lower limits of the grid power are set to 100 to 1900 kW, which is a value that takes into account a margin of 100 kW relative to the physical capacity of the power receiving equipment, which is 2000 kW. As a result, when block B1 of control logic L1 shown in Figure 3 determines that the grid power load will exceed the upper limit of 1900 kW, it increases the discharge amount of storage battery 4, which is equipment with priority P2 or higher, or increases the load on gas engine 3, thereby adjusting the power supply from the grid power to be below 1900 kW. The above processing enables system operation with a margin that prevents the grid power from approaching the physical upper limit of 2000 kW.

[0087] Using the method described above, control parameters such as upper and lower limits are set in the control logic L11 of the control simulator 11. The same control logic is used for the control simulator 11 and the control device 1. Therefore, the control parameters are set after checking the control operation under the same conditions as the actual system using the planning simulator 20 and the control simulator 11. In other words, the design of the control device 1 consists only of adjusting and setting the control parameters, which simplifies the design work. This reduces design costs and shortens the work time. [Example]

[0088] Next, a second embodiment of the present invention will be described with reference to Figures 25 and 26. Only the differences from the first embodiment will be described here. Note that the load in this embodiment also includes heat.

[0089] FIG. 25 shows a network N in the second embodiment. The difference from the network N in the first embodiment described in FIG. 5 is that heat is also controlled in addition to electricity. The gas engine module M3a here is assumed to be a cogeneration type facility that outputs heat in addition to electricity. Therefore, compared to the gas engine module M3 in FIG. 5, heat has been added to the output. A boiler module M8 has also been added as a facility that outputs heat. Furthermore, a heat connection module M9 and a heat demand module M10 have also been added.

[0090] Figure 26 shows data D3a for the gas engine module M3a, which defines the equipment specifications for the gas engine 3 of this embodiment. The difference from data D3 (Figure 6) for the gas engine module M3 of embodiment 1 is that data on heat is defined as output D34. As described above, the definition of equipment specifications is general-purpose, and can be defined in the same format for both power and heat.

[0091] The same applies to the boiler module M8; it is only necessary to define the linear expression representing the input / output characteristics and the upper and lower limit values ​​according to the boiler specifications. Also, for control, the control logic L1 in Figure 3 of the first embodiment can be used as is.

[0092] As described above, in the system according to the present invention, the simulation condition definition and control logic are general-purpose, and the same processing can be applied to both electric power and heat. [Explanation of symbols]

[0093] 100: Energy supply system 1: control device, 2: Planning device, 3: Gas engine, 4: storage battery, 5: Solar power generation equipment, 6, 7: Sensor, 10: Control design support device, 11: Control simulator, 12: Control parameter determination unit; 20: Planning simulator, 30: Investment Simulator

Claims

1. A control design support method for supporting the design of control logic for an actual system that supplies load according to demand, comprising: a first step of creating, in a control simulator, a first control logic for determining load allocation of each piece of equipment by importing specification data of each piece of equipment constituting the actual system; a second step of determining control parameters in the first control logic by analyzing fluctuation characteristics of the prediction error of the constraint; a third step of simulating, using the control simulator, load distribution of each facility based on the past data or simulated data of the constraint conditions and the operation plan; a fourth step of constructing a second control logic for determining load distribution of each facility of the real system by setting the specification data and the control parameters whose validity has been verified in the third step in a control device of the real system; and A control design support method comprising:

2. 2. The control design support method according to claim 1, The control design support method is characterized in that the specification data is data indicating a load range or capacity range in which each piece of equipment can physically operate.

3. 3. The control design support method according to claim 2, a control design support method characterized in that in the second step, a prediction error of the constraint condition is decomposed into high-frequency components and low-frequency components using a frequency filter, and the control parameters are determined based on the amplitude of the high-frequency components.

4. 4. The control design support method according to claim 3, A control design support method characterized in that the load is electric power, each facility is electric power facility that supplies electric power, the constraint conditions are electric power demand and the amount of power generated by a solar power generation facility, the effectiveness is that no power outages occur, and the control parameters are data on the amount of margin for specification data of each facility.

5. A control design support device that supports the design of control logic for an actual system that supplies load according to demand, a control simulator that implements a first control logic that determines load allocation for each piece of equipment based on specification data for each piece of equipment that constitutes the actual system; a control parameter determination unit that determines a control parameter in the first control logic by analyzing a fluctuation characteristic of a prediction error of a constraint condition, the control simulator simulates load distribution of each facility based on past data or simulated data of the constraint conditions and an operation plan; The control parameter determination unit constructs a second control logic for determining load allocation of each facility of the real system by setting the specification data and the control parameters whose validity has been verified in the simulation in a control device of the real system.

Images