Operation support device, operation support method, and program

Abstract

To automatically correct parameters of a simulator for use in creation of an operation plan of a plant.SOLUTION: An operational support apparatus for creating an operation plan of a plant includes a prediction part for predicting a future demand of the plant by using operation performance data including activation states, input values and output values of respective devices constituting the plant, an optimization part for creating the operation plan representing at least a plan for future activation states and input values of the respective devices constituting the plant by using a simulator for simulating the operation of the plant to optimize a prescribed objective function with the demand and characteristics of the respective devices constituting the plat defined as constraints, and a correction part for correcting parameters of the simulator using a difference between output values calculated by the simulator when the operation plan is created and output values when the plant is actually operated.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an operation support device, an operation support method, and a program. [Background technology]

[0002] A technique for creating an optimal operation plan for an energy plant composed of equipment such as a gas turbine and a chiller by using a steady-state plant simulator has been known for some time. For example, Patent Document 1 discloses a technique for determining an operation method that satisfies the energy supply (supply and demand balance) for the energy load prediction value and minimizes the energy cost by calculating the output value of each equipment by using a steady-state plant simulator after performing an energy load prediction. Note that Patent Document 1 uses a neural network for the load prediction and a method called Particle Swarm Optimization (PSO) for determining the operation method. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-84805 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in conventional technology, when the deviation between the actual output value of the equipment and the output value calculated by the simulator becomes large due to deterioration of the equipment over time or the like (i.e., when the accuracy of the simulator decreases), it is necessary to correct the simulator parameters (e.g., parameters representing the output characteristics of the equipment) each time.

[0005] An embodiment of the present invention has been made in view of the above points, and has an object to automatically correct parameters of a simulator used to create a plant operation plan. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, an operations support device according to one embodiment is an operations support device that creates an operations plan for a plant, and includes: a prediction unit that predicts future demand for the plant using operational performance data including the startup states, input values, and output values ​​of each piece of equipment that constitutes the plant; an optimization unit that creates the operations plan, which represents at least a plan for the future startup states and input values ​​of each piece of equipment that constitutes the plant, by optimizing a predetermined objective function using a simulator that simulates the operation of the plant, with the demand and characteristics of each piece of equipment that constitutes the plant as constraints; and a correction unit that corrects parameters of the simulator using the difference between output values ​​calculated by the simulator when the operations plan was created and output values ​​when the plant is actually operated. Effect of the Invention

[0007] The parameters of the simulator used to create the plant operation plan can be automatically corrected. [Brief description of the drawings]

[0008] [Figure 1] FIG. 2 is a diagram illustrating an example of a hardware configuration of an operation support device according to the present embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a functional configuration of an operation support device according to the present embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of a supply and demand system. [Figure 4] 13 is a flowchart illustrating an example of an operation plan creation process according to the present embodiment. [Diagram 5] 10 is a flowchart illustrating an example of a simulator correction process according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] An embodiment of the present invention will be described below. In this embodiment, an operation support device 10 will be described that can create an optimal operation plan for a plant and automatically correct parameters of a steady-state plant simulator used to create the operation plan, for the purpose of supporting the operation of the plant.

[0010] <Hardware configuration of the operation support device 10> The hardware configuration of an operation support device 10 according to this embodiment is shown in Fig. 1. As shown in Fig. 1, the operation support device 10 according to this embodiment has an input device 101, a display device 102, an external I / F 103, a communication I / F 104, a processor 105, and a memory device 106. Each of these pieces of hardware is connected to each other via a bus 107 so as to be able to communicate with each other.

[0011] The input device 101 is, for example, a keyboard, a mouse, a touch panel, various buttons, etc. The display device 102 is, for example, a display, a display panel, etc. Note that the operation support device 10 does not necessarily have to have at least one of the input device 101 and the display device 102.

[0012] The external I / F 103 is an interface with an external device such as a recording medium 103a. The operation support device 10 can read and write data from and to the recording medium 103a via the external I / F 103. Examples of the recording medium 103a include a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), and a USB (Universal Serial Bus) memory card.

[0013] The communication I / F 104 is an interface for connecting the operation support device 10 to a communication network. The processor 105 is, for example, various arithmetic devices such as a central processing unit (CPU) or a graphics processing unit (GPU). The memory device 106 is, for example, various storage devices such as a hard disk drive (HDD), a solid state drive (SSD), a random access memory (RAM), a read only memory (ROM), and a flash memory.

[0014] The operation support device 10 according to this embodiment can realize various processes described below by having the hardware configuration shown in Fig. 1. Note that the hardware configuration shown in Fig. 1 is an example, and the operation support device 10 may have, for example, multiple processors 105, multiple memory devices 106, or various other hardware.

[0015] <Functional configuration of the operation support device 10> A functional configuration of the operation support device 10 according to this embodiment is shown in Fig. 2. As shown in Fig. 2, the operation support device 10 according to this embodiment includes a data acquisition unit 201, a demand prediction unit 202, an optimization unit 203, a simulation unit 204, and a simulator correction unit 205. Each of these units is realized, for example, by a process executed by the processor 105 of one or more programs installed in the operation support device 10. In particular, the simulation unit 204 is realized by a process executed by the processor 105 of a program that functions as a steady-state plant simulator. Note that the steady-state plant simulator is a program or software for simulating the steady state of a plant.

[0016] Moreover, the operation support device 10 according to this embodiment has an operation record storage unit 206. The operation record storage unit 206 is realized by, for example, an auxiliary storage device such as an HDD or SSD. Note that the operation record storage unit 206 may also be realized by, for example, a storage device (such as a database server) connected to the operation support device 10 via a communication network.

[0017] The operation performance storage unit 206 stores past operation performance data of the plant 20 (e.g., a supply and demand system, etc.). Here, it is assumed that the operation performance storage unit 206 stores at least operation performance data for past times s=1, . . . , S, where S is the number of pieces of operation performance data used for demand forecasting of the plant 20. Note that the time is a time index that indicates the control time of the plant 20 (i.e., the time having a time span equal to the control period).

[0018] Here, for example, each device constituting the plant 20 is represented by i, and the start-up state of the device i at time s is represented by r is ∈{0,1}, input value is x is , the output value is y is Then, the performance data for the past time s is {(r is ,x is ,y is )|i=1, ,I}, where I is the total number of devices that make up the plant 20. Also, r is = 0 indicates that the device i is stopped, and r is =1 indicates that device i is up and running.

[0019] Note that the input value x is and the output value y is may be a scalar or a vector, and may differ depending on the type of equipment i. For example, if equipment i is a gas turbine, the input value is a scalar representing fuel, and the output value is a vector whose elements are power and steam. On the other hand, if equipment i is a chiller, the input value is a scalar representing power, and the output value is a scalar representing heat (amount of cold).

[0020] Also, in general, r isIf x = 0 (i.e., device i is down), then is =0,y is = 0, but for example, due to residual heat, etc. is = 0 even if y is It could also be >0.

[0021] The operation performance data stored in the operation performance storage unit 206 is collected, for example, from various sensors that sense the start-up state, input values, and output values ​​of each device i constituting the plant 20.

[0022] The data acquisition unit 201 acquires past operation performance data {(r is ,x is ,y is ) |i=1, . . . , I, s=1, . . . , S} is obtained from the performance record storage unit 206.

[0023] The demand forecasting unit 202 predicts the demand (load) for each energy type in the entire plant 20 at future time t=1, . . . , T using the past operation performance data acquired by the data acquisition unit 201. That is, the energy type is k, and the demand for the energy type k at future time t is L t k Then, the demand forecasting unit 202 calculates {(L t 1 ,···,L t K )|t=1, ,T}, where T is the final time of prediction and K is the number of energy types.

[0024] The demand prediction unit 202 may predict the demand for each energy type in the entire plant 20 using a known method (e.g., deep learning, neural network, PLS (Partial Least Squares) regression, etc.). For example, the above-mentioned Patent Document 1 describes a method for predicting the demand for each energy type using a neural network, and the demand prediction unit 202 may predict the demand for each energy type in the entire plant 20 using this method. In addition, the demand for each energy type may be predicted using a demand (load) prediction method described in, for example, JP 2000-132203 A, JP 2004-171548 A, JP 2015-125593 A, etc. Note that examples of the energy types include electricity, heat, steam, etc.

[0025] The optimization unit 203 creates an optimal operation plan for the plant 20 at future times t=1,...,T' using the demand predicted by the demand prediction unit 202. Here, T' is the number of future times desired to be created as an operation plan, and any integer value can be set as long as 1 ≤ T' ≤ T.

[0026] That is, the optimization unit 203 sets the supply and demand balance of the demand (demand forecast for each energy type) predicted by the demand forecasting unit 202 and the characteristics of each device constituting the plant 20 as constraint conditions, and sets the start-up state and input value of each device i constituting the plant 20 as state variables {(r it ,x it )|t=1, ,T'}, the plant optimization problem that minimizes a given objective function is formulated as a nonlinear mixed integer programming problem, a linear mixed integer programming problem, etc. The characteristics of the equipment include, for example, the input / output limits (upper input / output limits) of the equipment, start / stop times, etc.

[0027] Then, the optimization unit 203 generates the initial values ​​of the above-mentioned state variable values, creates an optimal operation plan by the following steps 1 to 4, and transmits it to the operator terminal 30. Note that the initial values ​​of the state variable values ​​may be generated randomly, or may be provided by a user based on operational knowledge of the plant 20, etc.

[0028] Step 1: Input the state variable values ​​to the simulation unit 204. As a result, the simulation unit 204 calculates the output value {y it |t=1, ,T'} is calculated.

[0029] Step 2: The output value {y it |t=1, ,T'} is used to calculate the objective function value.

[0030] Step 3: Update the state variable values ​​using a specified optimization method so that the objective function value becomes smaller.

[0031] Step 4: Determine whether a predetermined termination condition is met. If the termination condition is not met, return to step 1 above. On the other hand, if the termination condition is met, set the current state variable value to {(r it ,x it )|t=1, ,T'} is the optimal operation plan. Note that the current state variable values ​​and the corresponding output values ​​{(r it ,x it ,y it )|t=1, ,T'} may be the optimal operation plan.

[0032] Here, examples of the objective function value include the operating cost (energy cost) of the entire plant 20, CO2 emissions, and maintenance time. Examples of the optimization method include metaheuristics such as Brain Storm Optimization, Modified Brain Storm Optimizaiton, and Brain Storm Optimization with Differential Evolution in the case of a nonlinear mixed integer programming problem. On the other hand, examples of the optimization method include linear programming in the case of a linear mixed integer programming problem.

[0033] Furthermore, examples of the termination condition include when steps 1 to 3 are repeated a predetermined number of times, when the objective function value converges, and so on.

[0034] The reason why the output value is required to calculate the objective function value in step 2 above is that, for example, when the energy cost of the entire plant 20 is considered as the objective function value, the cost of energy type k is expressed as c k Then the objective function E is E=Σc k and c k is expressed as the product of the output value of energy type k in the entire plant 20 from time t=1 to t=T' and its unit cost.

[0035] The simulation unit 204 receives the state variable value {(r it ,x it ,y it )|t=1, ,T'} is input, a simulation of the plant 20 (operation simulation in a steady state) is performed to obtain the output value {y it |t=1, ,T'}. At this time, the simulation unit 204 calculates r it = 0 if y it = 0, r = 1 for y it =f i (x it ;α i ) to calculate the output value of device i. Here, f i is the simulation parameter α i and represents the output characteristic of the device i. In this embodiment, for simplicity, the function f i f i (x;α i )=α i (a i x+b i ) is a linear function. i and b i is a parameter determined according to the output characteristics of device i. i is a linear function, but depending on the type of device i, f i may be a nonlinear function.

[0036] The simulator correction unit 205 corrects the output value {y it|t=1, ,T'} and the difference between the output value of each device when the plant 20 is actually operated based on the optimal operation plan, and a simulation parameter α i For example, the actual output value of each device i at time t=1, . . . , T' during the actual operation of the plant 20 is corrected as y' it Then, for each i=1, . . . , I, the simulator correction unit 205 calculates |y it -y' it The simulation parameter α is set to minimize the sum of | i Correct the following.

[0037] <Processing Executed by Operation Support Device 10> Hereinafter, an operation plan creation process and a simulator correction process will be described as processes executed by the operation support device 10. The operation plan creation process is a process for creating an optimal operation plan for the plant 20, and the simulator correction process is a process for correcting parameters of a steady-state simulator when there is a discrepancy between the optimal operation plan and the actual operation of the plant 20.

[0038] In the following, the supply and demand system shown in Fig. 3 is mainly assumed as plant 20. The supply and demand system shown in Fig. 3 is an energy plant made up of a gas turbine 301 that inputs fuel and outputs electric power and steam, a boiler 302 that inputs fuel and outputs steam, turbo chillers 303 and 304 that input electric power and output heat, and steam absorption chillers 305 and 306 that input steam and output heat, and has inputs of purchased electric power and purchased fuel and outputs of electric power load, heat load, and steam load. These electric power load, heat load, and steam load are required by various facilities (e.g., production facilities, etc.).

[0039] <Operation plan creation process> The operation plan creation process according to this embodiment will be described with reference to FIG.

[0040] First, the data acquisition unit 201 acquires past operation performance data used for demand forecasting of the plant 20 (i.e., forecasting the power load, heat load, and steam load of the supply and demand system shown in FIG. 3) from the operation performance storage unit 206 (step S101).

[0041] Next, the demand forecasting unit 202 uses the past operation performance data acquired in the above step S101 to forecast the demand for each energy type in the entire plant 20 at future times t=1, . . . , T (step S102). That is, the demand forecasting unit 202 forecasts the demand for each energy type in the entire plant 20 at future times t=1, . . . , T (step S102). t 1 ,L t 2 ,L t 3 )|t=1, ,T}, where L t 1 is the predicted value of the power load at time t, L t 2 is the predicted value of the heat load at time t, L t 3 is the predicted steam load at time t.

[0042] Next, the optimization unit 203 creates an optimal operation plan for the plant 20 at future times t=1, . . . , T' using the demand predicted in the above step S102 (step S103).

[0043] That is, the optimization unit 203 calculates the {(L t 1 ,L t 2 ,L t 3 )|t=1, ,T'} and the characteristics of each device are the constraints, and the start-up state and input value of each device i constituting the plant 20 are the state variables {(r it ,x it)|t=1, ,T'}, and a plant optimization problem for minimizing a predetermined objective function is formulated as a nonlinear mixed integer programming problem, a linear mixed integer programming problem, or the like. Note that the devices i constituting the plant 20 correspond to the gas turbine 301, the boiler 302, the turbo chiller 303, the turbo chiller 304, the steam absorption chiller 305, and the steam absorption chiller 306, respectively.

[0044] Then, the optimization unit 203 generates the initial values ​​of the above state variables, and creates an optimal operation plan by the above steps 1 to 4. As described above, in the above step 1, the simulation unit 204 calculates the output value {y it |t=1, ,T'} is calculated.

[0045] Next, the optimization unit 203 transmits the optimal operation plan created in the above step S103 to the operator terminal 30 (step S104). This allows the operator to control the operation of the plant 20 by referring to the optimal operation plan at each time t=1, . . . , T'.

[0046] <Simulator correction process> The simulator correction process according to this embodiment will be described with reference to FIG. 5. In the following, it is assumed that an optimal operation plan for time t=1,...,T' is transmitted to the operator terminal 30, and the operation of the plant 20 is controlled at each time t=1,...,T' with reference to this optimal operation plan. Therefore, the operation performance storage unit 206 stores the operation performance data {(r' it ,x' it ,y' it )|i=1, ,I,t=1, ,T'} is stored. it is the actual startup state of device i at time t, and x' it is the actual input value of device i at time t, y' it represents the actual output value of device i at time t.

[0047] First, the simulator correction unit 205 calculates the actual output value {y' it ,I,t=1,...,T'} from the operation result storage unit 206 (step S201). That is, the simulator correction unit 205 obtains the output value {y' it Obtain |i=1,···,I,t=1,···,T'}. Next, the simulator correction unit 205 corrects the output value {y it |t=1, ,T'} and the output value {y' it |i=1,···,I,t=1,···,T'}, the simulation parameter α' that minimizes this difference i (Step S202). That is, the simulator correction unit 205 calculates |y i1 -y' i1 |+···+|y iT' -y' iT' The simulation parameter α' that minimizes | i This is calculated by |y i1 -y' i1 |+···+|y iT' -y' iT' It is possible to calculate | as an objective function using a known optimization method.

[0048] Next, the simulator correction unit 205 calculates the simulation parameter α′ calculated in step S202 for each i=1, . . . , I. i The simulator correction unit 205 determines whether or not to reflect |α i -α' i If | is greater than a certain threshold, the simulation parameter α' i is reflected in the steady-state plant simulator, while |α i -α' iIf | is less than or equal to the threshold, the simulation parameter α' i It is decided not to reflect in the steady-state plant simulator. i are the current simulation parameters for device i.

[0049] Then, the simulator correction unit 205 corrects the simulation parameter α′ determined to be reflected in the steady-state plant simulator in step S203. i In other words, the simulator correction unit 205 reflects α i ←α' i and the simulation parameter α i This updates the simulation parameter α i is corrected.

[0050] As described above, the operation support device 10 according to the present embodiment calculates the simulation parameter α i In order to correct this, it becomes possible to create an optimal operation plan that also takes into account changes in characteristics due to aging and other factors of the equipment that constitutes the plant 20.

[0051] <Modification> A modification of this embodiment will now be described.

[0052] <Variation 1> In step S204 in Fig. 4, the optimal operation plan was transmitted to the operator terminal 30. However, for example, when the steady-state plant simulator accurately simulates the actual operation of the operator, the operation support device 10 may control the operation of the plant 20 according to the optimal operation plan. Note that the steady-state plant simulator accurately simulates the actual operation of the operator means, for example, a case in which |α i -α' i | is equal to or smaller than a predetermined threshold.

[0053] <<Variation 2>> In step S204 of FIG. 5, the simulation parameter α i α i ←α' i However, it may be updated by a weighted sum. For example, 0<γ i <1 is a predetermined parameter, and α i ←γ i α i +(1-γ i )α' i In addition, at this time, γ i may be the same for all i or may be different. By using this modification 2, it becomes possible to perform correction while taking into account the current simulation parameter values.

[0054] The present invention is not limited to the above-described embodiments specifically disclosed, and various modifications, changes, combinations with known technologies, etc. are possible without departing from the scope of the claims. [Explanation of symbols]

[0055] 10 Operation support equipment 20 Plant 30 Operator terminal 101 Input Device 102 Display device 103 External I / F 103a Recording media 104 Communication I / F 105 processors 106 Memory Device 107 Bus 201 Data Acquisition Department 202 Demand Forecasting Department 203 Optimization Department 204 Simulation Department 205 Simulator correction section 206 Operational Performance Memory Unit

Claims

1. An operation support device that creates a plant operation plan, a prediction unit that predicts future demand for the plant using operation performance data including the start-up state, input values, and output values ​​of each device that constitutes the plant; an optimization unit that creates the operation plan, which represents at least a plan of a future start-up state and an input value of each device that configures the plant, by optimizing a predetermined objective function using a simulator that simulates the operation of the plant and setting the demand and characteristics of each device that configures the plant as constraint conditions; a correction unit that corrects parameters of the simulator using a difference between an output value calculated by the simulator when the operation plan is created and an output value when the plant is actually operated; An operational support device having the above configuration.

2. The prediction unit is forecasting the demand at future times t=1, ..., T; The optimization unit is The startup state r of each device i at future times t=1, ..., T' (where 1 ≤ T' ≤ T) ｉｔ and the input value x ｉｔ and creating the operational plan, The correction unit is The start-up state r included in the operation plan ｉｔ and the input value x ｉｔ The output value y calculated by inputting ｉｔ and the output value y′ of the device i when the plant is actually operated at time t. ｉｔ The operation support device according to claim 1 , wherein the parameter is corrected using a difference between and so as to minimize the difference.

3. The parameter for device i is expressed as α ｉ As, The simulator comprises: The startup state r ｉｔ If y is a value indicating that the device i is not activated, ｉｔ = 0, The startup state r ｉｔ is a value indicating that the device i is activated, the parameter α ｉ A function f ｉ By ｉｔ = f ｉ (x ｉｔ ; α ｉ 3. The operation support device according to claim 2, wherein the calculation is as follows:

4. The correction unit is The value α' of the parameter that minimizes the difference ｉ Calculate |α ｉ -α' ｉ If | is greater than a predetermined threshold, then α' ｉ Using the above parameter α ｉ The operation support device according to claim 3 , further comprising:

5. The correction unit is α ｉ ←α' ｉ The parameter α ｉ The operation support device according to claim 4 , further comprising:

6. The correction unit is α ｉ ←γ ｉ α ｉ + (1-γ ｉ ) α' ｉ (However, 0 < γ ｉ <1) by the parameter α ｉ The operation support device according to claim 4 , further comprising:

7. An operation support device that creates a plant operation plan, a prediction step of predicting future demand for the plant using operation result data including the start-up state, input values, and output values ​​of each device constituting the plant; an optimization procedure for creating the operation plan, which represents at least a plan of a future start-up state and an input value of each device constituting the plant, by optimizing a predetermined objective function using a simulator that simulates the operation of the plant and setting the demand and characteristics of each device constituting the plant as constraint conditions; a correction step of correcting parameters of the simulator using a difference between an output value calculated by the simulator when the operation plan is created and an output value when the plant is actually operated; An operational support method implemented by a computer.

8. The operation support device that creates the plant operation plan a prediction step of predicting future demand for the plant using operation result data including the start-up state, input values, and output values ​​of each device constituting the plant; an optimization procedure for creating the operation plan, which represents at least a plan of a future start-up state and an input value of each device constituting the plant, by optimizing a predetermined objective function using a simulator that simulates the operation of the plant and setting the demand and characteristics of each device constituting the plant as constraint conditions; a correction step of correcting parameters of the simulator using a difference between an output value calculated by the simulator when the operation plan is created and an output value when the plant is actually operated; A program that causes a computer to execute the following.

Images