WO2021161625A1 - Dispositif de calcul de quantité physique, dispositif de calcul de quantité de fonctionnement, procédé de calcul de quantité physique, programme, et système d'optimisation du fonctionnement - Google Patents

Dispositif de calcul de quantité physique, dispositif de calcul de quantité de fonctionnement, procédé de calcul de quantité physique, programme, et système d'optimisation du fonctionnement Download PDF

Info

Publication number
WO2021161625A1
WO2021161625A1 PCT/JP2020/044743 JP2020044743W WO2021161625A1 WO 2021161625 A1 WO2021161625 A1 WO 2021161625A1 JP 2020044743 W JP2020044743 W JP 2020044743W WO 2021161625 A1 WO2021161625 A1 WO 2021161625A1
Authority
WO
WIPO (PCT)
Prior art keywords
physical quantity
amount
fluid
flow
quantity calculation
Prior art date
Application number
PCT/JP2020/044743
Other languages
English (en)
Japanese (ja)
Inventor
麗子 川上
一幸 若杉
哲平 手島
Original Assignee
三菱重工業株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from PCT/JP2020/030986 external-priority patent/WO2021161559A1/fr
Application filed by 三菱重工業株式会社 filed Critical 三菱重工業株式会社
Priority to JP2022500238A priority Critical patent/JP7411061B2/ja
Publication of WO2021161625A1 publication Critical patent/WO2021161625A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators

Definitions

  • This disclosure relates to a physical quantity calculation device, an operation amount calculation device, a physical quantity calculation method, a program, and an operation optimization system.
  • This application claims priority based on Japanese Patent Application No. 2020-021857 filed in Japan on February 12, 2020 and PCT / JP2020 / 030986 filed internationally on August 17, 2020. Is used here.
  • Patent Document 1 discloses a system that derives plant loss based on a physical base model, a statistical model, a heuristic model, etc. of equipment components, works on equipment control based on a cost function, and enhances plant efficiency. ing.
  • the cost function calculates the cost of operating the equipment in the plant. Since the loss derived in the system described in Patent Document 1 includes physical loss such as pressure loss, heat loss, and vibration loss, it is necessary to construct a large number of physical models and the like in order to derive the loss in the system. It becomes. In addition, when considering aging, it is necessary to reconstruct the physical model and the like.
  • Patent Document 1 In the system described in Patent Document 1, it is necessary to construct and reconstruct a complicated model in order to perform optimum process control, and there is a problem that the configuration becomes complicated.
  • the present disclosure has been made in view of the above circumstances, and an object of the present disclosure is to provide a physical quantity calculation device, a manipulated quantity calculation device, a physical quantity calculation method, and a program capable of calculating a predetermined physical quantity with a simple configuration. ..
  • the physical quantity calculation device targets each of the plurality of consuming devices consuming the fluid and the plurality of channels for flowing the fluid through the consuming devices.
  • a physical quantity calculation unit that calculates each physical quantity by using the flow model and the supply amount of the fluid substance and the consumption amount of the fluid substance as input parameters.
  • the physical quantity calculation method targets a plurality of consuming devices consuming a fluid substance and a plurality of flow paths for flowing the fluid substance in each of the consuming devices, and only the flow rate of the fluid substance flowing in each of the flow paths.
  • a flow model including a balance calculation formula for calculating the balance according to the above and a physical quantity calculation formula for calculating a predetermined physical quantity of the fluid in the flow path based on each flow rate of the fluid in the flow path. It includes a step of calculating each physical quantity by using the supply amount of the fluid substance and the consumption amount of the fluid substance as input parameters.
  • the program according to the present disclosure is intended for a plurality of consuming devices consuming a fluid substance and a plurality of flow paths for flowing the fluid substance in each of the consuming devices, and relates only to a flow rate of the fluid substance flowing in each of the flow paths.
  • the fluid using a flow model including a balance calculation formula for calculating the balance and a physical quantity calculation formula for calculating a predetermined physical amount of the fluid in the flow path based on each flow rate of the fluid in the flow path.
  • the computer is made to perform the step of calculating each physical amount.
  • the operation optimization system inputs an operation amount for an equipment (for example, a plant), simulates the state of the equipment according to the operation amount and a disturbance factor, and calculates a physical quantity indicating the state of the equipment.
  • a physical quantity calculation unit an optimization unit that obtains the manipulated quantity so that the evaluation function indicating whether or not the state of the equipment is appropriate with the physical quantity as a variable are maximized or minimized, and the disturbance element are at least. It includes a prediction unit that outputs to the physical quantity calculation unit as a prediction value predicted by using machine learning using an actual value as an explanatory variable.
  • a predetermined physical quantity can be calculated with a simple configuration.
  • FIG. 6 It is a block diagram which shows the structural example of the operation amount calculation apparatus which concerns on 1st Embodiment of this disclosure. It is a schematic diagram which shows the structural example of the flow model 6 shown in FIG. It is a schematic diagram which shows the structural example of the mathematical expression which represents the flow model 6 shown in FIG. It is a schematic diagram which shows the structural example of the mathematical expression which represents the flow model 6 shown in FIG. It is a schematic diagram which shows the other structural example of the flow model 6 shown in FIG. It is a schematic diagram which shows the structural example of the mathematical expression which represents the flow model 6a shown in FIG. It is a schematic diagram which shows the operation example of the operation amount calculation apparatus 1 shown in FIG.
  • FIG. 1 is a block diagram showing a configuration example of an operation amount calculation device according to the first embodiment of the present disclosure.
  • the operation amount calculation device 1 shown in FIG. 1 is configured by using, for example, a computer such as a server, a personal computer, or a tablet terminal, a peripheral device of the computer, or the like, and hardware such as a computer and software such as a program executed by the computer.
  • a functional configuration composed of a combination with and, a physical quantity calculation unit 2, an optimization unit 3, a prediction unit 4, and a storage unit 5 are provided. Further, the storage unit 5 stores the flow model 6, the actual value 7, and the operation plan value 8.
  • the operation quantity calculation device 1 also has an aspect as a physical quantity calculation device including a physical quantity calculation unit 2.
  • the operation quantity calculation device 1 is configured to include a physical quantity calculation device including a physical quantity calculation unit 2, an optimization unit 3, and the like.
  • the generated amount determines the generated amount (amount of generated material).
  • the optimum additional supply amount (amount to be additionally supplied in addition to the generated amount) is calculated (simulated) as an operation amount using the flow model 6.
  • the generated amount corresponds to, for example, the amount of generated matter generated outside the target range of optimization such as outside the plant 100, outside the factory provided with the plant 100, and other processes of the plant 100.
  • the generated amount may include, for example, the amount of by-products produced as a by-product in the plant 100 or the like.
  • the additional supply amount (operation amount) is an amount (amount to be changed) operated by the optimization unit 3 in the optimization process, and corresponds to an amount additionally supplied within the target range of optimization.
  • the optimum additional supply amount means an additional supply amount having a relatively high degree of appropriateness among a plurality of candidates.
  • the generated product to be calculated by the operation amount calculation device 1 is a fluid substance such as a gas or a liquid, for example, a simple substance such as hydrogen (H2) or nitrogen (N2), carbon dioxide (CO2), or water. Alternatively, it is a compound such as steam (H2O), methane (CH4), ammonia (NH3), or a mixture thereof.
  • FIG. 1 shows an example in which the operation amount calculation device 1 is provided inside the plant 100
  • the operation amount calculation device 1 may be provided outside the plant 100.
  • FIG. 2 is a schematic diagram showing a configuration example of the flow model 6 shown in FIG. 3 and 4 are schematic views showing a configuration example of a mathematical formula representing the flow model 6 shown in FIG.
  • FIG. 5 is a schematic view showing another configuration example of the flow model 6 shown in FIG.
  • FIG. 6 is a schematic diagram showing a configuration example of a mathematical formula representing the flow model 6a shown in FIG.
  • the flow model 6 is a model that simulates the main flow of the fluid in the process of the plant 100.
  • the plant 100 includes a plurality of consuming devices that consume the fluid, a plurality of flow paths for flowing the fluid through each consuming device, and an operation amount calculation device 1 shown in FIG.
  • the flow model 6 includes a predicted generation amount 10, an additional supply amount 20, consumption devices 31 to 34, fuel 35, flare 36, flow paths 51 to 53, and flow paths 61 to 71 as elements.
  • the flow model 6 is constructed as a model for a plurality of consuming devices that consume the fluid substance and a plurality of flow paths for flowing the fluid substance to each consuming device.
  • control conditions for ensuring operational quality and safety in the process are set as constraints.
  • the constraint condition 1 is in the flow path 63 (inlet of the consumer device 31)
  • the constraint condition 2 is in the flow path 67 (outlet of the consumer device 32)
  • the constraint condition 3 is in the flow path 70 (inlet of the fuel 35).
  • Constraints 1 to 3 are, for example, that a predetermined physical quantity such as concentration, partial pressure, and flow rate of a fluid is within a range of a predetermined lower limit value or more, a predetermined upper limit value or less, and a predetermined upper and lower limit values. be.
  • a plurality of constraint conditions may be set in one flow path, or a plurality of physical quantities may be targeted under one condition.
  • the flow paths 51 to 53 are elements that are connected to three or more flow paths and need to confirm the balance.
  • the predicted generated amount (generated amount) 10 is the flow rate of a generated product (generated fluid) that is predicted to be generated in the future in, for example, a reformer or a reactor.
  • the additional supply amount 20 is the flow rate of the fluid material to be additionally supplied when the predicted consumption amount exceeds the predicted generation amount.
  • the additional supply amount 20 was manufactured by, for example, the flow rate of the fluid material flowing in from the tank for storing the fluid material, the flow rate of the fluid material flowing in from the pipeline through which the fluid material flows, and the fluid material manufacturing equipment in the plant 100.
  • the total amount of the predicted generation amount 10 and the additional supply amount 20 is the supply amount of the fluid in the flow model 6.
  • Consumer devices 31 to 34 are devices that consume fluids, such as desulfurizers.
  • the fuel 35 is a fluid that is consumed as fuel in equipment such as a gas turbine.
  • the flare 36 is a fluid that is guided by a flare stack that burns the fluid.
  • the predicted generation amount 10 is represented by ⁇ 10
  • the additional supply amount 20 is represented by ⁇ 20
  • the flow rate of the fluid material predicted to be consumed by the consuming devices 31 to 34 is represented by the predicted consumption amounts ⁇ 1 to ⁇ 4, respectively.
  • the flow rate of the fluid in the flow path 61 is F01
  • the flow rate in the flow path 62 is F02
  • the flow rates in the flow paths 63 to 65 are F11 to F13
  • the flow rates in the flow paths 66 and 67 are F21 and the flow rate F22
  • the flow path is represented by F31
  • the flow rate of the flow path 69 is represented by F41
  • the flow rates of the flow paths 70 and 71 are represented by flow rates F51 and F52, respectively.
  • the predicted generation amount 10 is supplied to the flow path 51 through the flow path 61, and the additional supply amount 20 is supplied to the flow path 51 through the flow path 62.
  • the flow path 51 is connected to each inlet of the consumer devices 31 to 33 via the flow paths 63 to 65.
  • the outlets of the consumer devices 31 and 32 are connected to the flow path 52 via the flow paths 66 and 67.
  • the inlet of the consumer device 34 is connected to the flow path 52 via the flow path 68.
  • the outlet of the consumer device 34 is connected to the flow path 53 via the flow path 69.
  • a fluid is supplied from the flow path 53 to the fuel 35 via the flow path 70 or to the flare 36 via the flow path 71.
  • the flow model 6 can be represented by, for example, a plurality of balance calculation formulas 6-1 shown in FIG. 3 and a plurality of (or 1) physical quantity calculation formulas 6-2 shown in FIG.
  • Each balance calculation formula 6-1 is a formula for calculating the balance related to the flow rate of the fluid flowing in each flow path.
  • the first balance calculation formula in FIG. 3 represents that the flow rate F01 of the flow path 61 is the flow rate ⁇ 10 (predicted generation amount 10).
  • the second balance calculation formula represents that the flow rate F02 of the flow path 62 is the flow rate ⁇ 20 (additional supply amount 20).
  • the third balance calculation formula represents that the flow rate entering the flow path 51 is (flow rate F01 + flow rate F02) and the flow rate exiting the flow path 51 is (flow rate F11 + flow rate F12 + flow rate F13).
  • the fourth balance calculation formula indicates that the flow rate F13 is the predicted consumption amount ⁇ 3.
  • the fifth balance calculation formula indicates that the flow rate F21 is (flow rate F11-predicted consumption ⁇ 1).
  • the sixth balance calculation formula indicates that the flow rate F22 is (flow rate F12-predicted consumption ⁇ 2).
  • the seventh balance calculation formula indicates that the flow rate entering the flow path 52 is (flow rate F21 + flow rate F22) and the flow rate exiting the flow path 52 is the flow rate F31.
  • the eighth balance calculation formula indicates that the flow rate F41 is (flow rate F31-predicted consumption ⁇ 4).
  • the ninth balance calculation formula indicates that the flow rate entering the flow path 53 is the flow rate F41 and the flow rate exiting the flow path 53 is (flow rate F51 + flow rate F52).
  • each physical quantity calculation formula 6-2 is a formula for calculating a predetermined physical quantity of a fluid substance in each flow path based on each flow rate of the fluid substance in the flow path.
  • the predetermined physical quantity of the fluid is, for example, partial pressure, concentration, flow rate, and the like.
  • the physical quantity calculation formula related to the constraint condition 1 set in the flow path 63 when the physical quantity of the management target (control target) is P1, the physical quantity P1 sets the flow rate F11 of the flow path 63. It means that it is calculated by the function R1 which is a variable.
  • the physical quantity calculation formula related to the constraint condition 2 set in the flow path 67 is calculated by the function R2 in which the physical quantity P2 is a variable of the flow rate F22 of the flow path 67 when the physical quantity to be managed is P2. Represents that.
  • the physical quantity calculation formula related to the constraint condition 3 set in the flow path 70 is calculated by the function R3 in which the physical quantity P3 has the flow rate F51 of the flow path 70 as a variable when the physical quantity to be managed is P3. Represents that.
  • the function R1 is a mathematical formula based on the physical relationship between the flow rate F11 and the physical quantity P1. Physical quantities related to the characteristics of fluids such as partial pressure and concentration often have a physical relationship proportional to the flow rate in plants such as chemical plants. Therefore, for example, by statistically obtaining the physical relationship based on the actual values of the physical quantity and the flow rate, the function R1 can be determined based on the obtained physical relationship. In the example shown in FIG.
  • each physical quantity calculation formula 6-2 is set with the flow rates F11, F22 and F51 of the flow paths 63, 67 and 70 in which the constraints 1, 2 and 3 are set as variables.
  • the functions are R1 (F11), R2 (F22) and R3 (F51).
  • the variable in each physical quantity calculation formula 6-2 under each constraint condition is not limited to the flow rate of the flow path in which the constraint condition is set, and the flow rate can be a variable.
  • the flow rate at the inlet and the flow rate at the outlet of the consumer device can be included in the variables.
  • the physical quantity targeted by the constraint condition has a physical relationship with the amount or ratio of the additional supply amount and the predicted generation amount, or the amount or ratio of the generation amount and the circulation amount among the additional supply amount, the additional supply amount,
  • the flow rate corresponding to the predicted generated amount, generated amount, circulating amount, etc. can be included in the variable.
  • the wording of the balance calculation formula 6-1 includes the meaning of one balance calculation formula 6-1 and the meaning of a plurality of balance calculation formulas 6-1 and the wording of the physical quantity calculation formula 6-2. Includes the meaning of one physical quantity calculation formula 6-2 and the meaning of a plurality of physical quantity calculation formulas 6-2.
  • the additional supply amount 20 ( ⁇ 20) can be obtained by using the balance calculation formula 6-1.
  • the additional supply amount 20 ( ⁇ 20) obtained based on the predicted generation amount 10 ( ⁇ 10) and the predicted consumption amounts ⁇ 1 to ⁇ 4 is the predicted minimum necessary additional supply amount, and the constraint condition is not necessarily required. I'm not satisfied. Therefore, in the present embodiment, the flow model 6 is used to perform the optimization process as described later, and the optimum additional supply amount 20 ( ⁇ 20) that satisfies the constraints 1 to 3 is calculated.
  • the flow model 6a of another example shown in FIG. 5 differs from the flow model 6 shown in FIG. 2 in that the additional supply amount ⁇ 20 is composed of the production amount 20a ( ⁇ 20a) and the circulation amount 80 ( ⁇ 80). ..
  • the production amount 20a is the flow rate of the fluid material generated by the fluid product manufacturing equipment, and is supplied to the flow path 51 via the flow path 62.
  • the circulation amount 80 is a flow rate supplied from the flow path 53 via the flow path 72 (flow rate F53), and the circulation amount 80 is supplied to the flow path 51 via the flow path 73 (flow rate F03).
  • FIG. 6 represents the balance calculation formula 6a-1 representing the flow model 6a shown in FIG.
  • the physical quantity calculation formula 6-2 shown in FIG. 4 is common to the flow model 6 shown in FIG. 2 and the flow model 6a shown in FIG. According to the flow model 6a, the additional supply amount ⁇ 20 can be divided into a production amount 20a ( ⁇ 20a) and a circulation amount 80 ( ⁇ 80) to calculate the balance and physical quantity.
  • the physical quantity calculation unit 2 shown in FIG. 1 calculates one or a plurality of physical quantities targeted by the constraint condition using the flow model 6 stored in the storage unit 5. That is, the physical quantity calculation unit 2 has a plurality of flow rates of the fluid flowing in each flow path in the plant 10 including a plurality of consuming devices consuming the fluid and a plurality of flow paths for flowing the fluid through each consuming device.
  • One or more physical quantities to calculate one or more predetermined physical quantities of a fluid in one or more channels based on multiple balance formulas to calculate the balance and each flow rate of the fluid in one or more channels.
  • each physical quantity is calculated with the supply amount of the fluid substance (the total amount of the generated amount of the fluid substance and the additional supply amount of the fluid substance) and the consumption amount of the fluid substance as input parameters.
  • the optimization unit 3 has a maximum evaluation function based on a predetermined evaluation value that changes according to the operation amount and the achievement degree of one or a plurality of constraints related to one or a plurality of physical quantities, using the additional supply amount as the operation amount.
  • the operation amount is calculated so as to be the minimum.
  • the predetermined evaluation value can be, for example, a value representing the cost (cost) required to obtain the operation amount (additional supply amount), a value representing the exhaust gas reduction effect in the plant 100, or the like.
  • the degree of achievement of the constraint condition is set to a binary value indicating whether or not the constraint condition is satisfied, or a value indicating how much margin is given to the threshold value when the constraint condition is satisfied, or when the constraint condition is not satisfied. It can be a value that indicates how close it was to the key value.
  • An example of the evaluation function is shown below.
  • Evaluation function cost + constraint expression 1 + constraint expression 2 + constraint expression 3 + ...
  • the constraint condition expressions 1, 2, 3, ... are designed so that the values increase when the constraint conditions are deviated.
  • Let the additional supply amount cost, and substitute the physical quantity calculated by the physical quantity calculation unit 2 into the constraint condition expression.
  • the optimization unit 3 repeatedly performs the optimization calculation so that the value of the calculated evaluation function is minimized. Gradient descent method is used for optimization calculation.
  • the physical quantity calculation unit 2 can calculate each physical quantity by using the additional supply amount obtained as the operation amount by the optimization unit 3 as an input parameter.
  • the prediction unit 4 calculates predicted values of the amount of fluid generated and the amount of consumption, which are input parameters of the physical quantity calculation unit 2. For example, the prediction unit 4 calculates the predicted value of the generated amount and the predicted value of the consumption amount by using machine learning using the operation plan information, the past actual value, and the like as explanatory variables. In addition to the actual value, future information such as a planned value such as an operation plan and a forecast value of the weather can be used as the explanatory variable.
  • the machine learning algorithm is not limited, but a random forest or the like can be used, for example.
  • the physical quantity calculation unit 2 can calculate the predicted value of each physical quantity by using the amount of fluid generated and the amount of fluid consumed predicted by the prediction unit 4 as input parameters. For example, the prediction unit 4 can learn the tendency of the plant by learning the actual data of the latest process value, and can predict the process value several hours to several days ahead according to the operation plan.
  • the flow model 6 stored in the storage unit 5 includes, for example, information representing the balance calculation formulas 6-1 and 6a-1 as shown in FIGS. 3 and 6 and the physical quantity calculation formula 6-2 as shown in FIG. , Information indicating the correspondence between the actual equipment, piping, etc. in the plant 100 and the equipment, the flow path, etc. in the flow model 6.
  • the flow model 6 may include information representing the balance calculation formula and the physical quantity calculation formula as, for example, a program for solving the formula.
  • the actual value 7 (also referred to as the process actual value) stored in the storage unit 5 is, for example, the generated amount, the additional supply amount (generated amount, circulating amount), each consumption amount, and each flow rate defined in the flow model 6. Includes time-series data of measured values measured for each unit time in the past of each physical quantity, other time-series process data (data managed by the process) not defined in the flow model 6. Further, the operation plan value 8 stored in the storage unit 5 is information representing planned values such as the amount of generation and the amount of consumption based on the operation plan of the plant 100.
  • FIG. 7 is a schematic view showing an operation example of the operation amount calculation device 1 shown in FIG.
  • FIG. 8 is a flowchart showing an operation example of the operation amount calculation device 1 shown in FIG. 9 and 10
  • FIGS. 10A to 10C are schematic views for explaining an operation example of the operation amount calculation device 1 shown in FIG.
  • an operation example for the flow model 6 described with reference to FIGS. 2 to 4 will be described.
  • FIG. 7 shows an outline of the optimization process of the additional supply amount by the operation amount calculation device 1.
  • the prediction unit 4 predicts the predicted amount ⁇ 10 based on the actual value 7 and the like (actual value 7, operation plan value 8, predicted value predicted by the prediction unit 4 and the like).
  • Predicted consumption ⁇ 1, ⁇ 2, ⁇ 3, ⁇ 4, etc. are predicted and output.
  • the physical quantity calculation unit 2 inputs the additional supply amount ⁇ 20 calculated by the optimization unit 3 as the operation amount, and the disturbance elements such as the generated amount ⁇ 10, the predicted consumption amount ⁇ 1, ⁇ 2, ⁇ 3, and ⁇ 4 calculated by the prediction unit 4.
  • the physical quantities P1, P2 and P3 are output as calculation results.
  • the optimization unit 3 receives the calculation result calculated by the physical quantity calculation unit 2, calculates the evaluation function described above, for example, reviews the operation amount (additional supply amount) so that the evaluation function is minimized, and the physical quantity calculation unit 2 Enter in.
  • the process shown in FIG. 8 is a process of calculating the optimum value of the additional supply amount after T time.
  • the prediction unit 4 first acquires the operation plan value and the process actual value (step S11).
  • the prediction unit 4 uses the acquired operation plan value and process actual value as explanatory variables, and predicts the amount of generation and the amount of consumption after T time by a machine learning method (step S12).
  • the physical quantity calculation unit 2 inputs the generated amount and the consumption amount predicted by the prediction unit 4 and the initial value of the additional supply amount, and calculates the physical quantity (step S13).
  • the initial value of the additional supply amount for example, the first value may be the actual value, and the subsequent and subsequent values may be the values determined in step S15.
  • the optimization unit 3 calculates the evaluation function from the calculation result of the physical quantity and the additional supply amount (step S14).
  • the optimization unit 3 adjusts (optimizes) the additional supply amount in the direction in which the value of the evaluation function becomes the minimum (step S15).
  • the optimization unit 3 determines whether or not the convergence condition is satisfied (step S16).
  • the convergence conditions are, for example, that the number of repetitions is equal to or greater than a predetermined upper limit, and that the slope of the change in the value of the evaluation function is less than a predetermined threshold value. If the convergence condition is not satisfied (“NO” in step S16), the process returns to step S13, and the physical quantity calculation unit 2 calculates the physical quantity again (step S13). On the other hand, when the convergence condition is satisfied (when "YES” in step S16), the optimization unit 3 outputs the optimum additional supply amount (step S17). In step S17, the optimization unit 3 displays, for example, information indicating the optimum additional supply amount on a predetermined display unit, or notifies information indicating the optimum additional supply amount to a control device (not shown) in the plant 100. To do.
  • FIG. 9 shows an example of the result of performing the optimization process using the operation amount calculation device 1 with the actual value of the additional supply amount as the initial value.
  • the horizontal axis is the number of optimization calculations (number of repetitions), and the vertical axis is the amount of additional supply. In the example shown in FIG. 9, the improvement effect of reducing the additional supply amount by about 15% was obtained.
  • FIG. 10 shows an example of the result of performing the optimization processing on the time series of the actual values.
  • the horizontal axis is the time and the vertical axis is the additional supply amount.
  • FIGS. 10A to 10C show examples of values before and after optimization of physical quantities (flow rate, pressure and concentration) confirmed in the optimization process.
  • the horizontal axis is time, and the vertical axis is flow rate, pressure and concentration.
  • Each physical quantity shown in FIGS. 10A to 10C has a value equal to or higher than each threshold value indicated by a broken line defined as a constraint condition.
  • the process balance calculation is performed by the flow model 6 simulating the flow based on the flow rate of the fluid in the process, and the evaluation function consisting of the cost and the control condition of the equipment is appropriate.
  • the additional supply can be calculated.
  • a predetermined physical quantity can be calculated without constructing a physical model such as an instrument or a chemical reaction. That is, since only the parameters of the elements related to the process balance equation, flow rate, and control conditions are included in the target range of the simulation, it is not necessary to construct a physical model such as an instrument or a chemical reaction, and it is easy to construct a simulation model.
  • the optimum additional supply amount that satisfies the management conditions related to plant quality and safety and minimizes the cost can be obtained. can get.
  • a predetermined physical quantity can be calculated with a simple configuration, and an optimum operation quantity can be calculated with a simple configuration.
  • FIG. 11 is a schematic diagram showing a configuration example of a mathematical formula representing the flow model 6 according to the second embodiment of the present disclosure.
  • FIG. 11 shows the balance calculation formula included in the flow model 6 shown in FIG. 1 as the balance calculation formula 6-1b.
  • the balance calculation formula 6-1b shown in FIG. 11 is different from the balance calculation formula 6-1 of the first embodiment shown in FIG. 3 in the sixth to eighth balance calculation formulas.
  • the physical quantity calculation formula 6-2 shown in FIG. 4 is common to the first embodiment and the second embodiment.
  • the flow model 6 shown in FIG. 2 is a model that simulates the main flow by the main pipe or the like, and a part of the flow of the sub pipe, the branch pipe, etc. may be omitted (a case is allowed).
  • the difference between the calculation result using the model 6 and the actual value may be large.
  • one or more balance calculation formulas 6-1b include a calculation term for correcting the difference from the actual value.
  • the correction constant C1 is added as a calculation term for correction by using the operator "+" on the right side of the sixth balance calculation formula 6-1b.
  • the first calculation term (first term) is F21 * C2 and the second calculation term (second term) by multiplying the flow rates F21 and F22 on the left side by the correction coefficient C2.
  • Is F22 * C2 and the first calculation term and the second calculation term are correction calculation terms.
  • the correction constant C4 is added as a calculation term for correction by using the operator "+" on the left side of the eighth balance calculation formula 6-1b, and the flow rate F31 on the left side is multiplied by the correction coefficient C3. 1
  • the calculation term is F31 * C3, and the calculation term is for correction.
  • the correction constants C1 and C4 and the correction coefficients C2 and C3 shown in FIG. 11 aggregate the difference between the process actual value and the calculated value when there is a difference between the calculated value using the flow model 6 and the actual value of the actual plant.
  • Differences and distances can be calculated by statistical methods and included in the balance calculation formula.
  • the calculation term for correction defined by the statistical method in the balance calculation formula, it is easy to construct the model, and the change due to the aged deterioration of the equipment of the actual plant, the change in the activity of the catalyst, etc. Can be followed at any time.
  • the calculation term for correction can be set as follows, for example.
  • the balance of the flow path that cannot be taken into consideration when correcting the balance of the flow path that cannot be taken into consideration, + correction, correction of missing information (temperature and pressure) for normal conversion of flow rate, missing concentration information, etc. If you do, you can correct x. Further, if the flow rate of the flow path that cannot be taken into consideration is too large to be ignored, it may be corrected by using the consumption amount in the device.
  • the difference from the actual value corrected by the correction calculation term included in the balance calculation formula is, for example, the measurement data of the state quantity of the fluid (for example, the physical quantity that describes the physical system in thermal equilibrium such as temperature and pressure). The balance does not match due to the lack of Corresponds to at least one of the things.
  • the flow model 6 shown in FIG. 2 may include a device that basically does not consume a fluid such as a device that removes impurities, in addition to the device that consumes the device. Further, in that case, the operating amount of the device other than the consumer device may be included in the operating amount.
  • FIG. 12 is a schematic block diagram showing the configuration of a computer according to at least one embodiment.
  • the computer 90 includes a processor 91, a main memory 92, a storage 93, and an interface 94.
  • the operation amount calculation device 1 described above is mounted on the computer 90.
  • the operation of each processing unit described above is stored in the storage 93 in the form of a program.
  • the processor 91 reads a program from the storage 93, expands it into the main memory 92, and executes the above processing according to the program. Further, the processor 91 secures a storage area corresponding to each of the above-mentioned storage units in the main memory 92 according to the program.
  • the program may be for realizing a part of the functions exerted on the computer 90.
  • the program may exert its function in combination with another program already stored in the storage or in combination with another program mounted on another device.
  • the computer may include a custom LSI (Large Scale Integrated Circuit) such as a PLD (Programmable Logic Device) in addition to or instead of the above configuration.
  • PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array).
  • PLDs Programmable Integrated Circuit
  • PAL Programmable Array Logic
  • GAL Generic Array Logic
  • CPLD Complex Programmable Logic Device
  • FPGA Field Programmable Gate Array
  • Examples of the storage 93 include HDD (Hard Disk Drive), SSD (Solid State Drive), magnetic disk, optical magnetic disk, CD-ROM (Compact Disc Read Only Memory), and DVD-ROM (Digital Versatile Disc Read Only Memory). , Semiconductor memory and the like.
  • the storage 93 may be internal media directly connected to the bus of the computer 90, or external media connected to the computer 90 via the interface 94 or a communication line. When this program is distributed to the computer 90 via a communication line, the distributed computer 90 may expand the program in the main memory 92 and execute the above processing.
  • the storage 93 is a non-temporary tangible storage medium.
  • the physical quantity calculation device (one aspect of the operation amount calculation device 1) according to the first aspect is a plurality of consumer devices 31 to 34 that consume the fluid substance and a plurality of consumer devices 31 to 34 that flow the fluid substance to each of the consumer devices 31 to 34.
  • the balance calculation formula 6-1 for calculating the balance related only to the flow rate of the fluid flowing in each flow path, and the flow path based on each flow rate of the fluid in the flow path.
  • the supply amount of the fluid substance and the consumption amount of the fluid substance are set as input parameters, and each physical quantity is set.
  • a physical quantity calculation unit 2 for calculation is provided. According to the first aspect, a predetermined physical quantity can be calculated with a simple configuration.
  • the physical quantity calculation device is the physical quantity calculation device of (1), and the balance calculation formula includes a calculation term for correcting the difference from the actual value. According to the second aspect, if there is a discrepancy between the calculated value and the actual value of the actual plant, it can be corrected.
  • the physical quantity calculation device is the physical quantity calculation device of (2), and the difference from the actual value is due to the lack of measurement data of the state quantity of the fluid, and the balance is balanced. It corresponds to at least one of the fact that there is no such balance, that the balance does not match due to an error in modeling the flow path, and that the balance does not match due to aged deterioration or loss.
  • the physical quantity calculation device is the physical quantity calculation device of (1) to (3), and the supply amount of the fluid is the generated amount of the fluid and the additional supply of the fluid. Is the total amount of.
  • the physical quantity calculation device is the physical quantity calculation device of (4), and the additional supply amount includes the production amount generated by the fluid production device and the circulation amount of the fluid material. ..
  • the additional supply amount ⁇ 20 can be divided into a production amount 20a ( ⁇ 20a) and a circulation amount 80 ( ⁇ 80) to calculate the balance and physical quantity.
  • the physical quantity calculation device is the physical quantity calculation device according to (4) or (5), and performs machine learning in which the generated amount and the consumed amount each use at least an actual value as an explanatory variable. It is a predicted value predicted using. According to the sixth aspect, for example, the tendency of the plant can be learned by learning the actual data of the latest process value, and the process value of several hours to several days ahead according to the operation plan can be predicted.
  • the manipulated quantity calculation device 1 is a predetermined evaluation value and a physical quantity that change according to the manipulated quantity, with the physical quantity calculating apparatus of (4) to (6) and the additional supply quantity as the manipulated quantity. It is provided with an optimization unit 3 for obtaining a manipulated variable so that the evaluation function based on the degree of achievement of the constraint condition according to the above is maximized or minimized. According to the seventh aspect, the optimum operation amount can be calculated with a simple configuration.
  • the operation amount calculation device 1 described in the first embodiment and the second embodiment described above is configured as a plant operation optimization system (equipment operation optimization system) 9 including a prediction unit 4 as shown in FIG. Will be done. That is, the plant operation optimization system 9 inputs an operation quantity for the plant (equipment) 100, simulates the state of the plant 100 according to the operation amount and the disturbance element, and calculates a physical quantity indicating the state of the plant 100.
  • the physical quantity calculation unit 2 to be calculated, the optimization unit 3 for obtaining the manipulated variable so that the evaluation function indicating whether or not the state of the plant 100 is appropriate, using the physical quantity as a variable, and the disturbance element are included. It includes at least a prediction unit 4 that outputs to the physical quantity calculation unit 2 as a predicted value predicted by using machine learning using the actual value as an explanatory variable.
  • the "operation amount with respect to the plant 100” is the additional supply amount 20 ( ⁇ 20) of the fluid in the first and second embodiments.
  • the "physical quantity indicating the state of the plant 100” is the above-mentioned physical quantities P1, P2, and P3 in the first and second embodiments.
  • the “evaluation function indicating the degree of whether the state of the plant 100 is appropriate” takes into consideration the above-mentioned cost (additional supply amount) and the degree of achievement of the constraint conditions. It is an evaluation function.
  • the “disturbance element” refers to the above-mentioned generated amount ⁇ 10, predicted consumption amount ⁇ 1, ⁇ 2, ⁇ 3, ⁇ 4, etc. in the first and second embodiments.
  • the disturbance element is an element that cannot be obtained from the state quantity of the plant, and is calculated based on the state of the equipment different from that of the plant 100.
  • FIGS. 14 and 15 are diagrams showing an example of the configuration of the plant operation optimization system, respectively.
  • the above-mentioned plant operation optimization system 9 is not limited to the first and second embodiments, and can be applied, for example, as in the embodiments shown in FIGS. 14 and 15, respectively.
  • the plant operation optimization system 9 shown in FIG. 14 predicts energy demand inside and outside the company for the purpose of minimizing power costs, and considers various fuel costs and profits from electric power transactions, and then boilers and turbines. By controlling the operation of the turbine, the operation is optimized. Optimizing the amount of fuel stored is the main evaluation index for minimizing the power cost because it has a large cost impact.
  • the physical quantity calculation unit 2 is a statistical model and a physical model of a boiler, a turbine, and a plant. As shown in FIG. 14, the amount of fuel used from the tank and the boiler / turbine operation parameters are input to the physical quantity calculation unit 2 as the operation amount.
  • the prediction unit 4 shown in FIG. 14 outputs the predicted value of the steam demand in addition to the predicted value of the power demand as a disturbance factor. Further, in the present embodiment, the prediction unit 4 further outputs predicted values such as by-product fuel inflow / unit price, power purchase unit price, and fuel unit price. The prediction unit 4 inputs the above predicted values to the physical quantity calculation unit 2 as a disturbance element.
  • the physical quantity calculation unit 2 simulates (simulates) the operation of the plant 100 according to the input manipulated variable (fuel consumption, boiler / turbine operating parameters) and disturbance factors (predicted values such as power demand and steam demand), and calculates the operation. As a result, the remaining fuel amount, the amount of power purchased, and the flow rate of externally procured fuel are output.
  • the input manipulated variable fuel consumption, boiler / turbine operating parameters
  • disturbance factors predicted values such as power demand and steam demand
  • the optimization unit 3 substitutes the calculation results (remaining amount of fuel, amount of electricity purchased, and flow rate of externally procured fuel) of the physical quantity calculation unit 2 into an evaluation function created in consideration of the remaining amount of fuel having a large cost impact. Calculate the evaluation value of plant operation.
  • the optimization unit 3 performs an optimization calculation by a gradient descent method or the like, and outputs an optimum operation amount that minimizes the evaluation value.
  • the plant operation optimization system 9 shown in FIG. 15 predicts energy demand inside and outside the company for the purpose of maximizing profits in the energy supply business, and considers various fuel costs and profits from electric power transactions, and then boilers. And by controlling the operation of the turbine, optimization is performed. Revenue is expressed as "steam + electricity sales-fuel cost", and the impact of selling surplus electricity in the electricity market is particularly large in maximizing this profit, so boiler operation for that purpose The quantity is the main evaluation index.
  • the physical quantity calculation unit 2 is a statistical model and a physical model of a boiler, a turbine, and a plant. As shown in FIG. 15, the boiler main steam flow rate is input to the physical quantity calculation unit 2 as the operation amount.
  • the prediction unit 4 outputs predicted values such as steam demand, JPEX (Japan Electric Power eXchange) price, fuel unit price, fuel intensity, etc. as disturbance factors.
  • the prediction unit 4 inputs these predicted values as a disturbance element to the physical quantity calculation unit 2.
  • the physical quantity calculation unit 2 simulates (simulates) the operation of the plant 100 according to the input operation amount (boiler main steam flow rate) and disturbance factors (predicted values such as steam demand), and as the calculation result, the generator output and the inside of the facility. Outputs auxiliary power, amount of steam in the facility, bleed valve opening, and exhaust pressure.
  • the optimization unit 3 maximizes the profit (that is, the sale of surplus power) by maximizing the calculation result (generator output, in-house auxiliary power, in-house steam amount, bleed valve opening, exhaust pressure) of the physical quantity calculation unit 2.
  • the evaluation value of the plant operation is calculated by substituting the boiler operation quantity for the purpose into the evaluation function as the evaluation index.
  • the optimization unit 3 performs an optimization calculation by a gradient descent method or the like, and outputs an optimum operation amount that minimizes the evaluation value.
  • the plant operation optimization system 9 combines the predicted value of the disturbance factor with the manipulated variable to the plant, and provides "guidance” to present the optimum manipulated variable. Then, in the factory equipment or the like, "control” may be performed to automatically operate with the guidance value presented by the plant operation optimization system 9. More specifically, the plant operation optimization system 9 models a process such as heat generation in the plant 100 as a simulator (physical quantity calculation unit 2), and the actual manipulated amount that can be operated in the plant 100 and that cannot be operated in the plant 100. Enter the disturbance element into the simulator.
  • the predicted value by machine learning is used as the external factor value.
  • the simulation result is evaluated by the evaluation function, the optimization calculation is performed through the evaluation function so as to match the KPI set in the plant 100, for example, and the optimum solution of the operation amount is displayed in the guidance.
  • a predetermined physical quantity can be calculated with a simple configuration.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

Selon la présente invention, un dispositif de calcul de quantité physique comprend : un modèle d'écoulement comprenant une formule de calcul de bilan pour calculer, en ce qui concerne une pluralité de dispositifs consommateurs qui consomment un fluide et une pluralité de passages d'écoulement à travers lesquels le fluide s'écoule vers chacun des dispositifs consommateurs, un bilan relatif uniquement au volume d'écoulement du fluide s'écoulant dans chacun des passages d'écoulement, et d'une formule de calcul de quantité physique pour calculer une quantité physique prescrite du fluide dans les passages d'écoulement, sur la base des volumes d'écoulement individuels du fluide dans les passages d'écoulement ; et une unité de calcul de quantité physique qui utilise le modèle d'écoulement et prend, en tant que paramètres d'entrée, la quantité du fluide fourni et la quantité du fluide consommé pour calculer chaque quantité physique.
PCT/JP2020/044743 2020-02-12 2020-12-01 Dispositif de calcul de quantité physique, dispositif de calcul de quantité de fonctionnement, procédé de calcul de quantité physique, programme, et système d'optimisation du fonctionnement WO2021161625A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2022500238A JP7411061B2 (ja) 2020-02-12 2020-12-01 物理量計算装置、操作量計算装置、物理量計算方法、プログラムおよび運転最適化システム

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
JP2020021857 2020-02-12
JP2020-021857 2020-02-12
PCT/JP2020/030986 WO2021161559A1 (fr) 2020-02-12 2020-08-17 Dispositif de calcul de quantité physique, dispositif de calcul de quantité d'exploitation, procédé de calcul de quantité physique, et programme
JPPCT/JP2020/030986 2020-08-17

Publications (1)

Publication Number Publication Date
WO2021161625A1 true WO2021161625A1 (fr) 2021-08-19

Family

ID=77291735

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/044743 WO2021161625A1 (fr) 2020-02-12 2020-12-01 Dispositif de calcul de quantité physique, dispositif de calcul de quantité de fonctionnement, procédé de calcul de quantité physique, programme, et système d'optimisation du fonctionnement

Country Status (2)

Country Link
JP (1) JP7411061B2 (fr)
WO (1) WO2021161625A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6166808A (ja) * 1984-09-11 1986-04-05 ウエスチングハウス　エレクトリック　コ−ポレ−ション 蒸気および電力の同時発生装置
JPH06230829A (ja) * 1992-05-29 1994-08-19 Hitachi Ltd 上水道運用計画方法
JP2003150233A (ja) * 2001-11-12 2003-05-23 Hitachi Ltd プラントの性能劣化状態評価方法と性能劣化状態評価装置
WO2018230645A1 (fr) * 2017-06-14 2018-12-20 株式会社東芝 Dispositif de détection d'anomalie, procédé de détection d'anomalie et programme
JP2019185468A (ja) * 2018-04-12 2019-10-24 株式会社日立産機システム 異常予測制御システム

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4004170B2 (ja) 1999-01-08 2007-11-07 大阪瓦斯株式会社 熱源設備
JP5215736B2 (ja) 2008-06-02 2013-06-19 株式会社ダイヘン 流体供給装置
US9864383B2 (en) 2015-04-02 2018-01-09 Belimo Holding Ag Method and system for determining characteristic parameters of a hydraulic network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS6166808A (ja) * 1984-09-11 1986-04-05 ウエスチングハウス　エレクトリック　コ−ポレ−ション 蒸気および電力の同時発生装置
JPH06230829A (ja) * 1992-05-29 1994-08-19 Hitachi Ltd 上水道運用計画方法
JP2003150233A (ja) * 2001-11-12 2003-05-23 Hitachi Ltd プラントの性能劣化状態評価方法と性能劣化状態評価装置
WO2018230645A1 (fr) * 2017-06-14 2018-12-20 株式会社東芝 Dispositif de détection d'anomalie, procédé de détection d'anomalie et programme
JP2019185468A (ja) * 2018-04-12 2019-10-24 株式会社日立産機システム 異常予測制御システム

Also Published As

Publication number Publication date
JP7411061B2 (ja) 2024-01-10
JPWO2021161625A1 (fr) 2021-08-19

Similar Documents

Publication Publication Date Title
Shen et al. Large-scale industrial energy systems optimization under uncertainty: A data-driven robust optimization approach
US7974826B2 (en) Energy system modeling apparatus and methods
US8874242B2 (en) Graphical language for optimization and use
Liew et al. Centralised utility system planning for a Total Site Heat Integration network
CA2745357C (fr) Systeme de gestion de l'energie
US8897900B2 (en) Graphical language for optimization and use
US8615325B2 (en) Advanced control system for steam hydrocarbon reforming furnaces
EP2500787A1 (fr) Modèles transparents pour une optimisation et une commande à grande échelle
Sarabia et al. Data reconciliation and optimal management of hydrogen networks in a petrol refinery
US8364327B2 (en) Systems, program product, and methods for targeting optimal process conditions that render an optimal heat exchanger network design under varying conditions
Cheng et al. Dantzig–Wolfe decomposition and plant-wide MPC coordination
Andiappan et al. Synthesis and optimisation of biomass-based tri-generation systems with reliability aspects
Galan et al. Implementation of RTO in a large hydrogen network considering uncertainty
Chen et al. Optimal design and operation of refinery hydrogen systems under multi-scale uncertainties
JP2015125643A (ja) 分散型エネルギーシステムの設備計画方法、プログラムおよび装置
Taccari et al. Short-term planning of cogeneration power plants: a comparison between MINLP and piecewise-linear MILP formulations
Chen et al. Techno-economic evaluation and optimization of CCGT power Plant: A multi-criteria decision support system
Dehghan et al. A new hybrid stochastic‐robust optimization approach for self‐scheduling of generation companies
Papalexandri et al. Modelling and optimization aspects in energy management and plant operation with variable energy demands-application to industrial problems
JP7233964B2 (ja) 運転指標提示装置、運転指標提示方法、およびプログラム
JP2016143336A (ja) 分散型エネルギーシステムの構成最適化方法及び装置
WO2021161625A1 (fr) Dispositif de calcul de quantité physique, dispositif de calcul de quantité de fonctionnement, procédé de calcul de quantité physique, programme, et système d'optimisation du fonctionnement
CN114266165A (zh) 考虑碳排放的蒸汽动力系统中蒸汽透平布局优化方法
WO2021161559A1 (fr) Dispositif de calcul de quantité physique, dispositif de calcul de quantité d'exploitation, procédé de calcul de quantité physique, et programme
Fuhrmann et al. Model-predictive energy management system for thermal batch production processes using online load prediction

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20918444

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022500238

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20918444

Country of ref document: EP

Kind code of ref document: A1