Apparatus, remote monitoring system, apparatus control method, and remote monitoring system control method

Abstract

To provide a device capable of achieving satisfactory control accuracy by efficiently selecting learning data to be used to reconstruct a learning model, a remote monitoring system, and a control method.SOLUTION: This device is a device for executing control of a plant on the basis of a prediction result using a learning model. The device selects data having large deviation from learning data used to construct a learning model from operation data as additional learning data when the prediction result using the learning model satisfies a prescribed condition. Then, the learning data and new learning data including the additional learning data are used to reconstruct a learning model.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an apparatus, a remote monitoring system, a method for controlling an apparatus, and a method for controlling a remote monitoring system. [Background technology]

[0002] In a wet flue gas desulfurization plant, which is one example of such a plant, exhaust gas generated in a combustion device such as a boiler is introduced into the absorption tower of the desulfurization plant and brought into gas-liquid contact with the absorbing liquid circulating in the absorption tower. During the gas-liquid contact process, the absorbent in the absorbing liquid (e.g., calcium carbonate) and the sulfur dioxide (SO ) in the exhaust gas are mixed together. 2 ) reacts with the SO 2 is absorbed in the absorbing solution, and SO is removed from the exhaust gas. 2 is removed (exhaust gas is desulfurized). 2 The absorbent that has absorbed the sulfur falls and is stored in a storage tank below the absorption tower. The storage tank is supplied with absorbent, and the absorbent that has restored its absorption capacity is supplied to the top of the absorption tower by a circulation pump, where it comes into gas-liquid contact with the exhaust gas (SO 2 (absorption of

[0003] In such wet flue gas desulfurization equipment, changes in the circulation flow rate of the absorbent and the absorbent concentration affect the amount of SO in the flue gas. 2 Therefore, in the control of wet flue gas desulfurization equipment, the control parameters such as the circulation flow rate of the absorbent and the absorbent concentration, and the concentration of SO in the flue gas are 2 The relationship between the concentration and the concentration of SO in exhaust gas is constructed as a learning model by machine learning, and the relationship between the concentration and the concentration of SO in exhaust gas is predicted by the learning model. 2 Based on the concentration, the control target value of the circulating flow rate of the absorbent and the absorbent concentration is determined, and the SO 2 For example, in Patent Document 1, in the control of such a wet flue gas desulfurization device, the amount of circulation of the absorption liquid and the amount of SO in the exhaust gas are controlled. 2 The control target value for the circulation rate of the absorbent is calculated using the first learning model that shows the relationship between the absorbent concentration in the absorbent and the SO concentration in the exhaust gas. 2A method is disclosed for controlling the circulation flow rate of the absorbing solution and the absorbent concentration by determining a control target value related to the absorbent concentration in the absorbing solution using a second learning model that represents the relationship with the concentration. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-11163 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the plant control described above, a learning model for predicting the control target value is prepared in advance. Such a learning model is constructed by machine learning using learning data selected from the plant's operation data. However, since the operation data has some variation, it may not be possible to obtain a sufficient prediction error depending on how the learning data is selected. In addition, even if the prediction error of the learning model is initially small enough, the prediction error may become large later due to changes in the plant's operation conditions from the time the learning model was constructed.

[0006] In this way, when the prediction error of the learning model is insufficient, it is necessary to reconstruct the learning model. The reconstruction of the learning model is performed, for example, using the learning data to which new data has been added. However, since the driving data contains a certain amount of variability, the prediction error may become large depending on how the added data is selected. In addition, since the driving data that is the source of the learning data is enormous, it is necessary to efficiently select data to reduce the prediction error by reconstructing the learning model.

[0007] At least one embodiment of the present disclosure has been made in consideration of the above-mentioned circumstances, and aims to provide an apparatus, a remote monitoring system, a control method for the apparatus, and a control method for the remote monitoring system that are capable of achieving good control accuracy by efficiently selecting learning data to be used to reconstruct a learning model. [Means for solving the problem]

[0008] In order to solve the above problems, at least one embodiment of the present disclosure provides an apparatus, An apparatus for executing a process related to plant control based on a prediction result using a learning model, an additional learning data selection unit for selecting, when a prediction result using the learning model satisfies a predetermined condition, data from the driving data that has a large deviation from the learning data used to construct the learning model as additional learning data; a learning model construction unit for reconstructing the learning model by using new learning data including the learning data and the additional learning data; Equipped with.

[0009] In order to solve the above problem, a remote monitoring system according to at least one embodiment of the present disclosure includes: A remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, The apparatus comprises: an additional learning data selection unit for selecting, when a prediction result using the learning model satisfies a predetermined condition, from the driving data, data having a large deviation from the learning data used to construct the learning model, as additional learning data, in response to a request from the terminal; a learning model construction unit for reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with.

[0010] In order to solve the above problem, a method for controlling an apparatus according to at least one embodiment of the present disclosure includes: A method for controlling an apparatus for executing a process related to plant control based on a prediction result using a learning model, comprising: an additional learning data selection step for selecting, when a prediction result using the learning model satisfies a predetermined condition, data from the driving data that has a large deviation from the learning data used to construct the learning model as additional learning data; a learning model construction step for reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with.

[0011] In order to solve the above problem, a control method for a remote monitoring system according to at least one embodiment of the present disclosure includes: A control method for a remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, an additional learning data selection step of selecting, when a prediction result using the learning model satisfies a predetermined condition, data having a large deviation from the learning data used to construct the learning model from the driving data, as additional learning data, in response to a request from the terminal; a learning model construction step of reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with. Effect of the Invention

[0012] According to at least one embodiment of the present disclosure, by efficiently selecting learning data to be used for reconstructing a learning model, it is possible to provide an apparatus, a remote monitoring system, a control method for an apparatus, and a control method for a remote monitoring system that can achieve good control accuracy. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a configuration diagram of a flue gas desulfurization device according to an embodiment. [Diagram 2] 1 is a schematic diagram illustrating a configuration of a remote monitoring system according to an embodiment. [Diagram 3] 4 is a flowchart showing a basic control of a wet flue gas desulfurization apparatus according to an embodiment. [Figure 4] 1 is a graph showing the trends in the predicted value of the SO2 concentration in the outflow gas, the measured value of the SO2 concentration by a gas analyzer, and the true value of the predicted value of the SO2 concentration. [Diagram 5] FIG. 2 is a diagram illustrating an example of a first relationship table created in basic control of a wet flue gas desulfurization apparatus according to an embodiment. [Figure 6] FIG. 4 is a diagram illustrating an example of a second relationship table created in basic control of a wet flue gas desulfurization apparatus according to an embodiment. [Figure 7] 3 is a flowchart showing a plant control method according to an embodiment. [Figure 8] 8 is a flowchart showing a method for selecting additional learning data in step S104 of FIG. 7. [Figure 9A] FIG. 9 is a diagram showing a process of selecting additional learning data in step S204 of FIG. 8. [Figure 9B] FIG. 9 is a diagram showing a process of selecting additional learning data in step S204 of FIG. 8. [Figure 10] A figure showing the distribution of learning data used to reconstruct the first learning model for each number of reconstructions performed. [Figure 11] FIG. 11 is a diagram showing the transition of the predicted value of the first learning model reconstructed using each of the learning data shown in FIG. [Figure 12] 1 is a configuration diagram of an information processing system according to an embodiment. [Figure 13] 13 is a diagram showing the internal configuration of the information processing device in FIG. 12 together with a control device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, several embodiments of the present invention will be described with reference to the drawings. However, the scope of the present invention is not limited to the following embodiments. The dimensions, materials, shapes, relative positions, etc. of the components described in the following embodiments are merely illustrative examples and are not intended to limit the scope of the present invention thereto.

[0015] First, a configuration of a wet flue gas desulfurization system 10, which is an example of a plant to be controlled by a plant control device according to at least one embodiment of the present disclosure, will be described with reference to Fig. 1. Fig. 1 is a configuration diagram of the wet flue gas desulfurization system 10 according to one embodiment.

[0016] In the following explanation, a wet flue gas desulfurization system 10 will be described as an example of a plant; however, the controlled object is not limited to the wet flue gas desulfurization system 10 and can broadly include plants in which control parameters are controlled based on control target values ​​predicted using a learning model.

[0017] The wet flue gas desulfurization system 10 is a plant facility for desulfurizing exhaust gas generated in a combustion device 1. The combustion device 1 is, for example, a boiler for generating steam, and is configured as a part of a power generation plant capable of generating electricity by supplying the steam generated in the combustion device 1 to a generator 5. The wet flue gas desulfurization system 10 includes an absorption tower 11 that communicates with the combustion device 1 via a pipe 2, a plurality of circulation pumps 12a, 12b, and 12c (collectively referred to as "circulation pumps 12" as appropriate) provided in a circulation pipe 3 for the absorption liquid circulating within the absorption tower 11, and calcium carbonate (CaCO 3 The absorber 11 is provided with an absorbent slurry supply section 13 for supplying a slurry (absorbent slurry) of the absorbent slurry into the absorber 11, and a gypsum recovery section 14 for recovering the gypsum in the absorption liquid. The absorber 11 is provided with an outlet pipe 16 for allowing the exhaust gas desulfurized in an operation described below to flow out of the absorber 11 as an outflow gas. The outlet pipe 16 is provided with a gypsum recovery section 14 for recovering the gypsum in the outflow gas. 2 A gas analyzer 17 is provided to measure the concentration.

[0018] The absorbent slurry supply unit 13 includes an absorbent slurry production equipment 21 for producing an absorbent slurry, an absorbent slurry supply pipe 22 connecting the absorbent slurry production equipment 21 and the absorption tower 11, and an absorbent slurry supply amount control valve 23 for controlling the flow rate of the absorbent slurry flowing through the absorbent slurry supply pipe 22. The gypsum recovery unit 14 includes a gypsum separator 25, a gypsum slurry extraction pipe 26 connecting the gypsum separator 25 and the absorption tower 11, and a gypsum slurry extraction pump 27 provided in the gypsum slurry extraction pipe 26.

[0019] The wet flue gas desulfurization system 10 is provided with a control device 15 which is a device for controlling a plant of at least one embodiment of the present disclosure. The control device 15 includes an operation data receiving unit 30 electrically connected to an operation data acquiring unit 20 including various detectors for acquiring various operation data (e.g., temperatures and pressures at various locations, flow rates of various fluids, etc.) of the combustion device 1 and the wet flue gas desulfurization system 10. The operation data acquiring unit 20 includes a gas analyzer 17.

[0020] The control device 15 includes a first learning model construction unit 38 electrically connected to the operation data receiving unit 30, a first relationship table creation unit 31 electrically connected to the first learning model construction unit 38, a circulation flow rate determination unit 32 electrically connected to the first relationship table creation unit 31, and a circulation pump adjustment unit 33 electrically connected to the circulation flow rate determination unit 32. The circulation pump adjustment unit 33 is electrically connected to each of the circulation pumps 12a, 12b, and 12c.

[0021] The control device 15 further includes a second learning model construction unit 39 electrically connected to the operation data receiving unit 30, a second relationship table creation unit 35 electrically connected to the second learning model construction unit 39, an absorbent slurry supply amount determination unit 36 ​​electrically connected to the second relationship table creation unit 35, and an absorbent slurry supply amount determination unit 36. The absorbent slurry supply control unit 37 is electrically connected to the absorbent slurry supply amount control valve 23.

[0022] The control device 15 further includes a prediction error calculation unit 40 electrically connected to the first learning model construction unit 38 and the second learning model construction unit 39, and an additional learning data selection unit 42 electrically connected to the prediction error calculation unit 40.

[0023] 2 shows the configuration of a remote monitoring system 44 for remotely monitoring the control state of the wet flue gas desulfurization apparatus 10 (see FIG. 1). The remote monitoring system 44 includes a distributed control system (DCS) 46 for each device constituting the combustion apparatus 1 (see FIG. 1) and the wet flue gas desulfurization apparatus 10 (see FIG. 1), an edge server 48 electrically connected to the DCS 46 and equipped with the control device 15, and a remote monitoring device 50 such as a desktop personal computer or tablet computer electrically connected to the edge server 48 via the cloud or a virtual private network (VPN). The control state of the wet flue gas desulfurization apparatus 10 can be remotely monitored by the remote monitoring device 50, which is usually located away from the edge server 48.

[0024] Next, the operation of the wet flue gas desulfurization apparatus 10 for desulfurizing the exhaust gas generated in the combustion apparatus 1 will be described. As shown in Fig. 1, exhaust gas generated in a combustion device 1 flows through a pipe 2 and enters an absorption tower 11, and rises within the absorption tower 11. By operating at least one of circulation pumps 12, the absorbing liquid flows through a circulation pipe 3 and enters the absorption tower 11, and the absorbing liquid flows downward within the absorption tower 11. The absorbing liquid that flows downward within the absorption tower 11 accumulates within the absorption tower 11, is caused to flow out of the absorption tower 11 by the circulation pump 12, and flows through the circulation pipe 3. In this manner, the absorbing liquid circulates within the absorption tower 11.

[0025] In the absorption tower 11, the rising flue gas and the falling absorbing liquid are in gas-liquid contact. 2 is expressed by the following reaction equation: SO 2 +CaCO 3 +2H 2 O+1 / 2O 2 →CaSO 4 2H 2 O+CO2 As shown in the figure, the CaCO 3 It reacts with gypsum (CaSO 4 2H 2 O) precipitates in the absorption liquid.

[0026] In this way, the SO in the exhaust gas 2 Since a part of the SO in the effluent gas flowing out from the absorber 11 through the outlet pipe 16 is removed as gypsum in the absorption liquid, that is, the exhaust gas is desulfurized, 2 The concentration is the SO in the exhaust gas flowing into the absorption tower 11 via the pipe 2. 2 The effluent gas from the absorption tower 11 flows through the outflow pipe 16 and is released into the atmosphere. 2 The concentration is measured, and the measurement result is transmitted to the operation data receiving unit 30 of the control device 15.

[0027] SO in effluent gas 2 The concentration is CaCO in the absorption liquid. 3 If there is no large fluctuation in concentration, the concentration tends to decrease as the circulation flow rate of the absorbing liquid circulating in the absorption tower 11 increases. The control device 15 controls the number of operating circulating pumps 12 and thereby the circulation flow rate by a control method described later, thereby reducing the concentration of SO in the outflow gas. 2 Controlling the concentration, for example, of SO in the effluent gas so that it is below a preset value. 2 The concentration can be controlled.

[0028] The gypsum precipitated in the absorption liquid in the absorption tower 11 is discharged as a gypsum slurry from the absorption tower 11 by a gypsum slurry discharge pump 27, and the gypsum slurry flows through a gypsum slurry discharge piping 26 and flows into a gypsum separator 25. In the gypsum separator 25, the gypsum and water are separated, the gypsum is recovered, and the water is sent to a drainage facility not shown.

[0029] CaCO in the absorption solution 3 SO 2 As the exhaust gas is desulfurized, the CaCO3 The control device 15 controls the aperture of the absorbent slurry supply amount control valve 23 by a control method described later, and the absorbent slurry produced in the absorbent slurry production facility 21 is supplied into the absorption tower 11 through the absorbent slurry supply pipe 22. As a result, the CaCO 3 The concentration falls within a preset range, and the CaCO 3 Large fluctuations in concentration are suppressed.

[0030] Next, a description will be given of basic control of the wet flue gas desulfurization system 10 by the control device 15. Fig. 3 is a flowchart showing basic control of the wet flue gas desulfurization system 10 performed by the control device 15 of Fig. 1.

[0031] In the basic control, first, in step S1, various operation data of the combustion device 1 and the wet flue gas desulfurization device 10 are collected, and then, in step S2, the various operation data and the future SO 2 A first learning model is constructed by machine learning for the relationship between the concentration and the concentration. Next, in step S3, a first relationship table, which will be described later, is created using the constructed first learning model. In the following step S4, the SO concentration in the outflow gas is calculated based on the first relationship table. 2 The circulation flow rate of the absorbing solution at which the concentration is equal to or lower than a preset value is determined, and in step S5, the operating conditions of the circulation pump 12 are adjusted based on the determined circulation flow rate. As a result, the SO 2 concentration in the outflow gas is adjusted to be equal to or lower than the preset value. 2 The concentration is controlled.

[0032] After step S1, in addition to steps S2 to S5, in step S12, various operational data and the future CaCO 3 A second learning model is constructed by machine learning for the relationship with the concentration. Next, in step S13, a second relationship table, which will be described later, is created using the constructed second learning model. In the following step S14, a second relationship table, which will be described later, is created based on the second relationship table. 3The supply amount of the absorbent slurry that brings the concentration into a preset range is determined, and in step S15, the absorbent slurry supply unit 13 is controlled, that is, the aperture of the absorbent slurry supply amount control valve 23 is controlled, so that the absorbent slurry is supplied into the absorption tower 11 at the determined supply amount. 3 The concentration falls within a preset range, and the CaCO 3 Large fluctuations in concentration are suppressed.

[0033] Next, each step of the method for controlling the wet flue gas desulfurization system 10 by the control device 15 will be described in detail. In step S1, as shown in Fig. 1, the operation data acquisition unit 20 acquires various operation data of the combustion device 1 and the wet flue gas desulfurization device 10, and the acquired various operation data are transmitted to the control device 15 and received by the operation data receiving unit 30, so that the control device 15 collects various operation data. As described above, since the operation data acquisition unit 20 includes the gas analyzer 17, the various operation data are 2 Contains concentration.

[0034] In step S2, the first learning model construction unit 38 calculates the various operations collected by the control device 15 and the future SO 2 In step S3, the first relationship table creating unit 31 creates a first model by machine learning for the relationship between the circulation flow rate of the absorption liquid at the first time and the SO concentration in the outflow gas at the second time, which is a time in the future from the first time, using the created first learning model. 2 A first relationship table is created, which is a correlation with the predicted value of concentration. Since the first relationship table is created using the first learning model constructed by machine learning, the first relationship table can be created quickly.

[0035] In the first relationship table, the circulating flow rate of the absorbent and the SO in the outflow gas 2 The time is different from the predicted concentration value. If the circulation flow rate of the absorbent is set to the current value, the SO 2 The predicted concentration is, for example, SO2 Therefore, various operational data includes at least the concentration of SO in the outflow gas at any given time. 2 The concentration of SO in the effluent gas at any time and the circulation flow rate of the absorption liquid at a time interval prior to the arbitrary time, the time interval being the time interval obtained by subtracting the first time from the second time. 2 The future SO concentration is calculated from actual operation data including the concentration of SO and the circulation flow rate of the absorbent at a time prior to the arbitrary time by a time interval obtained by subtracting the first time from the second time. 2 Since the concentration is predicted directly, future SO 2 The shorter the interval between the first and second times, the higher the prediction accuracy of the future SO concentration. 2 Therefore, the prediction performance of the concentration is improved. Therefore, the interval between the first and second times is determined based on the change in the concentration of SO in the outflow gas due to the change in the circulation flow rate of the absorption liquid. 2 The time it takes for the concentration to change and the gas analyzer 17 to detect the SO 2 It is preferable to add the time required to measure the concentration.

[0036] FIG. 4 shows the time interval between the first and second times, which is the time when the amount of SO in the outflow gas changes due to the change in the circulation flow rate of the absorption liquid. 2 The time it takes for the concentration to change and the gas analyzer 17 to detect the SO 2 The time required to measure the concentration of SO 2 Trends in predicted concentration (a) and SO2 concentration measured by gas analyzer 17 2 Changes in concentration measurements (b) and SO 2 The graphs show the trends in the true concentration of SO2 (a) and (b) of the gas that was detected by the gas analyzer 17. In each graph, the values ​​to the right are older, and the values ​​to the left are the most recent. 2 The most recent measured concentration was at Hour 1, and the SO 2 The latest predicted value of the concentration is the value at the second hour. 2 The latest concentration measurements and SO 2 The interval (i) between the true concentration and the latest value is the interval (i) between the true concentration and the latest value. 2 This corresponds to the time required to measure the concentration, i.e., the measurement delay. 2The latest true concentration value and SO 2 The interval (ii) between the predicted concentration value and the latest value is due to the change in the SO concentration in the outflow gas caused by the change in the circulation flow rate of the absorption liquid. 2 It corresponds to the time it takes for the concentration to change.

[0037] An example of the first relationship table is shown in Fig. 5. In this embodiment, the first relationship table has the SO 2 In the graph shown in FIG. 1, the predicted concentration is plotted on the vertical axis against the circulation flow rate of the absorbing liquid, but this does not necessarily have to be the case and may be in the form of a matrix or a formula, etc. In step S4, the circulation flow rate determination unit 32 calculates the future SO concentration in the outflow gas based on this first relationship table. 2 The circulation flow rate Q (control target value) of the absorbing solution is determined so that the concentration becomes a preset value SV.

[0038] In step S5, as shown in FIG. 1, the circulation pump adjustment unit 33 determines the number of circulation pumps 12a to 12c to be operated so that the circulation flow rate is equal to or greater than the determined circulation flow rate Q (control target value), and operates the determined number of circulation pumps. For example, if the supply amount of each of the three circulation pumps 12a to 12c during operation is the same, the circulation flow rate can be adjusted in three stages. If the number of circulation pumps is increased, the circulation flow rate can be adjusted more finely. Also, for example, if the supply amount of each of the three circulation pumps 12a to 12c during operation is different from each other, the circulation flow rate can be adjusted in up to six stages by combining the circulation pumps to be operated. Furthermore, for example, if the supply amount of each of the three circulation pumps 12a to 12c can be adjusted, the circulation flow rate can be adjusted more finely.

[0039] The adjustment of the circulation flow rate is not limited to being performed by controlling the number of circulation pumps. A single circulation pump with an adjustable supply rate may be used, and the supply rate of the circulation pump may be adjusted to the circulation flow rate determined by the circulation flow rate determination unit 32.

[0040] In this way, by adjusting the circulation flow rate of the absorbing liquid circulating in the absorption tower 11, the SO 2 The concentration can be controlled to be equal to or lower than a preset value, but for this purpose, the CaCO 3 It is necessary that there is no large fluctuation in the concentration. For this reason, in this embodiment, as described above, in addition to steps S2 to S5, steps S12 to S15 are performed to measure the CaCO 3 The density is controlled so as to fall within a preset range. Next, steps S12 to S15 will be described in detail.

[0041] In step S12, the second learning model construction unit 39 calculates the future CaCO 3 In step S13, the second relationship table creation unit 35 uses the constructed second learning model to create a relationship between the amount of absorbent slurry supplied to the absorption tower 11 at the third time and the amount of CaCO at the fourth time, which is a time in the future from the third time. 3 A second relationship table is created, which is a correlation with the predicted value of concentration. Since the second relationship table is created using the second learning model constructed by machine learning, the second relationship table can be created quickly.

[0042] In the second relationship table, the amount of absorbent slurry supplied to the absorption tower 11 and CaCO 3 The time is different from the predicted concentration value, and if the supply amount of absorbent slurry is the current value, CaCO 3 The predicted concentration is, for example, CaCO several minutes from now. 3 For this reason, various operational data includes at least the concentration of CaCO 3 The concentration of CaCO at any time and the amount of absorbent slurry delivered at a time interval prior to the arbitrary time (the time interval being the fourth time minus the third time). 3and the amount of absorbent slurry supplied at a time interval preceding said arbitrary time by a time interval equal to the fourth hour minus the third hour. 3 Since the concentration is predicted directly, the future CaCO 3 The prediction performance of concentration can be improved.

[0043] In this embodiment, CaCO at any time 3 The concentration is calculated using a simulation model based on mass balance calculations. 3 Since a sensor for detecting the concentration is generally expensive, the provision of such a sensor increases the cost of the wet flue gas desulfurization system 10. However, the concentration of CaCO 3 Calculating the concentration eliminates the need for expensive sensors, making it possible to suppress increases in the cost of the wet flue gas desulfurization system 10.

[0044] The shorter the interval between the third and fourth hours, the greater the future CaCO 3 The prediction performance of the concentration is improved. Therefore, the interval between the third and fourth hours is due to the change in the supply rate of the absorbent slurry. 3 The time required for the concentration to change is preferably set as the time required for the concentration to change. 2 The relationship is similar to that of the predicted value (a) and the true value (c) of the concentration. 3 The concentration was calculated using a simulation model based on mass balance calculations. 3 When the concentration is measured by a sensor, the transition of the predicted value of the supply amount of absorbent slurry, the transition of the measured value by the sensor, and the transition of the true value are respectively shown in SO 2 The relationship is similar to that of the various concentration trends (a) to (c).

[0045] In general, the SO in the effluent gas flowing out from the absorption tower 11 2 The number of steps required to change the concentration is 3The number of steps required to change the concentration is large, so CaCO 3 SO concentration control 2 The delay in controlling the concentration is large. Therefore, by making the time from the 3rd hour to the 4th hour shorter than the time from the 1st hour to the 2nd hour, the effect of the control delay can be properly considered, so that the future CaCO 3 The prediction performance of the concentration can be further improved.

[0046] An example of the second relationship table is shown in FIG. 6. In this embodiment, the second relationship table has CaCO 3 In step S14, the absorbent slurry supply amount determination unit 36 ​​determines the future CaCO concentration based on the second relationship table. 3 A supply amount F (control target value) of the absorbent slurry that keeps the concentration within a preset range R is determined.

[0047] In step S15, as shown in Fig. 1, the absorbent slurry supply control unit 37 controls the aperture of the absorbent slurry supply amount control valve 23 so that the supply amount of the absorbent slurry supplied into the absorption tower 11 through the absorbent slurry supply pipe 22 approaches the determined absorbent slurry supply amount F (control target value). In this way, by adjusting the supply amount of the absorbent slurry to the absorption tower 11, the future CaCO 3 The concentration can be controlled to be within a preset range.

[0048] In this way, from the operation data of the combustion device 1 and the operation data including the circulation flow rate of the absorbing liquid of the wet flue gas desulfurization device 10, the circulation flow rate of the absorbing liquid at the first time and the SO in the effluent gas flowing out from the absorption tower 11 at the second time, which is a time in the future from the first time, are calculated. 2 By creating a first relationship table between the concentration and actual operation data, future SO 2 Since the concentration is predicted directly, future SO 2A first relationship table having improved prediction performance of the concentration can be obtained, and based on this first relationship table, the SO 2 concentration in the outflow gas at the second time can be calculated. 2 The circulation flow rate of the absorption liquid in the first hour is determined so that the concentration becomes equal to or lower than a preset value, and the operating conditions of the circulation pumps 12a to 12c are adjusted based on the determined circulation flow rate in the first hour, so that the operating conditions of the circulation pumps 12a to 12c can be appropriately adjusted.

[0049] In this embodiment, steps S12 to S15 are carried out to remove CaCO 3 The concentration is set to be within a preset range. For example, CaCO 3 By measuring the concentration by a sensor and adjusting the supply amount of absorbent slurry to the absorption tower 11 as needed based on the measured value, steps S12 to S15 can be made unnecessary. In this case, the control device 15 does not need to include the second learning model construction unit 39, the second relationship table creation unit 35, the absorbent slurry supply amount determination unit 36, and the absorbent slurry supply control unit 37.

[0050] Next, a description will be given of a plant control method according to an embodiment, which is carried out by the control device 15 in addition to the basic control shown in Fig. 3. Fig. 7 is a flowchart showing a plant control method according to an embodiment.

[0051] In this plant control, a prediction value of the first learning model is calculated in step S100, apart from steps S2 to S5 shown in FIG. 3. Next, in step S101, an analysis result by the gas analyzer 17 is acquired. In the following step S102, the prediction value calculated in step S100 and the analysis result acquired in step S101 are compared to calculate a prediction result of the first learning model, that is, a prediction error. In the following step S103, when the prediction error calculated in step S102 satisfies a predetermined condition, for example, it is determined whether the prediction error is greater than a threshold value. If the prediction error is greater than the threshold value (step S103: YES), additional learning data is selected in the following step S104, and the first learning model is reconstructed in step S105 using the additional learning data selected in step S104. Then, in step S106, a prediction error is calculated for the first learning model reconstructed in step S105, and in step S107, it is determined whether the prediction error is equal to or less than a threshold value. If the prediction error calculated in step S106 is still greater than the threshold (step S107: NO), the process returns to step S104, and the selection of additional learning data and the reconstruction of the learning model are repeated. Such a repetitive process is performed until the prediction value of the reconstructed first learning model becomes equal to or less than the threshold. If the prediction error is equal to or smaller than the threshold in step S103 (step S103: NO), the process ends. However, the series of processes shown in FIG. 7 may be repeatedly performed at a predetermined timing.

[0052] Next, each step in FIG. 7 will be described in detail. In step S100, the first learning model constructed by the first learning model construction unit 38 is used to calculate the SO 2 The calculation of the predicted value by the first learning model in step S100 is carried out based on the SO concentration in the outflow gas in order to create the first relationship table in the above-mentioned step S3. 2 The predicted value of the concentration is the same as that of the first learning model. For the circulation flow rate of the absorption liquid at the first time, which is input to the first learning model, the concentration of SO in the outflow gas at the second time, which is a future time from the first time, is calculated. 2A predicted concentration is calculated.

[0053] In step S101, the SO 2 The actual measured value of the concentration of SO in the outflow gas calculated in step S100 is obtained. 2 The actual SO in the effluent gas at the second hour corresponds to the predicted concentration. 2 Concentration.

[0054] In step S102, the prediction error calculation unit 40 calculates the SO 2 The predicted concentration and the SO in the outflow gas obtained in step S101 2 A prediction error is calculated as the difference from the actual measured value of the concentration. This prediction error is an error corresponding to the prediction accuracy of the first learning model constructed by the first learning model construction unit 38, and includes various factors. For example, since the operation data received by the operation data receiving unit 30 has a certain degree of variability, the first learning model constructed by machine learning using the operation data as learning data has a learning error caused by the variability. In addition, the prediction error may become larger later due to a change in the operating conditions of the plant from the time the model was constructed.

[0055] In step S103, it is determined whether such a prediction error is greater than a preset threshold value ε. The success / failure determination in step S103 may be performed so that the success is established when the prediction error continues to be greater than the threshold value ε for a predetermined time or more. The magnitude of the prediction error may vary depending on the operating state of the wet flue gas desulfurization system 10. If the success determination is performed in step S103 based on a short-term determination, the first learning model may be reconstructed frequently, which may increase the burden of model management. Therefore, in step S103, if the state in which the prediction error is greater than the threshold value ε continues for a predetermined time or more, the success determination is performed in step S103, whereby the first learning model is appropriately reconstructed and efficient model management is possible.

[0056] In step S104, if the determination of establishment is made in step S103, the additional learning data selection unit 42 selects additional learning data included in the learning data used to reconstruct the first learning model. The learning data used for reconstruction includes new additional learning data in addition to old initial learning data used in constructing the previous first learning model (i.e., learning data used for reconstruction = initial learning data + additional learning data). The driving data receiving unit 30 continuously receives driving data, and appropriate additional learning data is selected from the driving data received after the previous first learning model was constructed.

[0057] The selection of additional learning data in step S104 may be performed on operation data acquired during steady operation of the plant. For example, operation data acquired during non-steady operation such as when an abnormality occurs in the plant, when the plant is started up, or when the plant is stopped is excluded from the selection of additional learning data. If the operation data includes data acquired during such non-steady operation, the data may be excluded by performing preprocessing on the operation data.

[0058] A method for selecting additional learning data by the additional learning data selection unit 42 will now be described in detail with reference to Fig. 8. Fig. 8 is a flow chart showing the method for selecting additional learning data in step S104 of Fig. 7.

[0059] In step S200, first, at least one explanatory variable of the first learning model is selected from a plurality of parameters included in the driving data by analyzing the driving data received by the driving data receiving unit 30. Such an explanatory variable is selected, for example, by selecting the SO in the outflow gas, which is the objective variable of the first learning model, for each of a plurality of driving data included in the driving data. 2The contribution degree may be calculated for each concentration by a method such as multiple regression, and the selection may be performed based on the contribution degree. For example, Z parameters may be selected as explanatory variables in descending order of contribution degree. By selecting some of the multiple parameters included in the driving data as explanatory variables of the first learning model in this way, the amount of calculation during learning can be effectively reduced while suppressing a decrease in learning accuracy, compared to the case where all parameters included in the driving data are used as learning targets.

[0060] In step S201, the explanatory variables selected in step S200 from the learning data (driving data) used in the previous construction of the first learning model are selected as the initial learning data. At this time, an average value over W time for V items selected from the learning data (driving data) used in the previous construction of the first learning model may be used as the initial learning data. In this case, by using the average value over a predetermined time for a specific parameter included in the driving data as the learning data, it is possible to effectively reduce the amount of calculation during learning while suppressing a decrease in learning accuracy.

[0061] In step S202, for the explanatory variables selected in step S200, additional learning data candidates are selected from the driving data received by the driving data receiving unit 30. The additional learning data candidates are selected from new driving data received by the driving data receiving unit 30 from the time when the first learning model was previously constructed to the present, and include parameters corresponding to the initial learning data selected in step S201. For example, when an average value over W hours is used as the initial learning data as described above, an average value over W hours is also used as the additional learning data candidates.

[0062] In step S203, the deviation is calculated between the initial learning data selected in step S201 and the additional learning data candidates selected in step S202. The deviation can be calculated using various methods for evaluating deviation, such as the k-nearest neighbor method and Mahalanobis distance. Then, in step S204, additional learning data to be added to the learning data is selected based on the deviation calculated in step S203.

[0063] 9A and 9B are diagrams showing the process of selecting additional learning data in step S204 of FIG.

[0064] 9A, in a space defined by arbitrary variable 1 and variable 2 included in the explanatory variables of the first learning model, multiple additional learning data candidates Dc1, Dc2, Dc3, ... are shown for a certain initial learning data Ds, and distances indicating the degree of deviation between the initial learning data Ds and each of the additional learning data candidates Dc1, Dc2, Dc3, ... are calculated. In this example, the additional learning data selection unit 42 selects, as the additional learning data, the additional learning data candidate Dc5 having the largest distance from among the multiple additional learning data candidates.

[0065] In the embodiment of FIG. 9B, in a space defined by any variable 1 and variable 2 included in the explanatory variables of the first learning model, multiple initial learning data Ds1, Ds2, ... selected in step S201 are shown for multiple additional learning data candidates Dc1, Dc2, Dc3, ... selected in step S202. Then, for each additional learning data candidate Dc1, Dc2, Dc3, ..., a distance to the closest initial learning data is calculated. The additional learning data selection unit 42 selects the additional learning data candidate with the maximum distance as the additional learning data from among the multiple additional learning data candidates. In FIG. 9A and FIG. 9B, the number of variables used in the determination of additional learning data is two, but this does not limit the scope of the present invention, and may be one or three or more in practice.

[0066] The additional learning data selection unit 42 calculates the degree of deviation between the initial learning data and the additional learning data candidates in this way, and selects learning data candidates to be added to the learning data for reconstructing the first learning model based on the degree of deviation. The number of additional learning data to be newly added may be arbitrary, and for example, by selecting additional learning data whose degree of deviation is equal to or greater than a predetermined value from the driving data, a predetermined number (A pieces) of additional learning data can be selected in order of the degree of deviation.

[0067] In this embodiment, the learning data used in the previous construction of the first learning model is treated as the initial learning data on the assumption that the first learning model has already been constructed by the first learning model construction unit 38, but if there is no construction history of the first learning model (for example, when the first learning model is constructed for the first time), one or more parameters arbitrarily selected from the driving data may be treated as the initial learning data. In this case, even when the first learning model is constructed for the first time, it is possible to construct a learning model with a small prediction error.

[0068] 7, in step S105, new learning data is created by adding the additional learning data selected in step S104 to the initial learning data, and the first learning model is reconstructed. This makes it possible to reconstruct the first learning model using new learning data that is the initial learning data used when the first learning model was previously constructed, plus the additional learning data selected from driving data obtained thereafter.

[0069] Then, in step S106, a prediction error is calculated using the first learning model reconstructed in step S105. The calculation of the prediction error in step S106 is similar to that in step S102 described above.

[0070] In step S107, as in step S103, it is determined whether the prediction error calculated in step S106 is equal to or less than the threshold ε. That is, it is determined whether the prediction error of the first learning model has been sufficiently improved by the reconstruction. As a result, if the prediction error of the first learning model has been improved to equal to or less than the threshold ε, it is determined that the prediction accuracy of the first learning model has been improved, and the process is terminated. On the other hand, if the prediction error of the first learning model is still greater than the threshold ε (step S107: NO), the process is returned to step S104. That is, if the prediction error of the first learning model is not sufficiently improved even by the reconstruction, additional learning data is selected again in step S104, the learning data is reviewed, and the construction of the first learning model is repeated. Such reconstruction of the first learning model is repeated until the prediction error becomes equal to or less than the threshold ε in step S107.

[0071] Here, we will specifically explain the change in the predicted value of the first learning model according to the number of times reconstruction is performed. Figure 10 shows the learning data (SO 2 FIG. 11 is a diagram showing the distribution of the learning data (the relationship between the concentration and the explanatory variable X used in the learning model) for each reconstruction performed, and FIG. 12 is a diagram showing the transition of the predicted value of the first learning model reconstructed using each of the learning data shown in FIG.

[0072] 10 shows that as the number of reconstructions increases, new additional learning data is selected in step S104, and the number of data included in the learning data increases. The prediction error of the first learning model reconstructed using such learning data decreases as the number of reconstructions increases, as shown in FIG 11. This indicates that the prediction accuracy of the first learning model is improved by appropriately selecting additional learning data each time reconstruction is performed.

[0073] As the number of reconstructions increases, the prediction error of the first learning model converges to a predetermined value (near 0.7 in the example of FIG. 11). Therefore, in step S107, in addition to or instead of the prediction value of the first learning model becoming equal to or less than a threshold, the end of the iterative process from step S104 onward may be determined based on whether the prediction error has sufficiently converged.

[0074] In this way, new learning data is created by selecting and adding additional learning data to the learning data for constructing the first learning model, and the first learning model is reconstructed using the selected learning data. At this time, the prediction error of the first learning model can be effectively reduced by appropriately selecting the additional learning data to be added to the learning data based on the degree of deviation from the initial learning data that has been conventionally included in the learning data. As a result, even if the prediction error of the first learning model is reduced due to some factor, good prediction accuracy can be obtained by reconstructing the first learning model.

[0075] The control device 15 can accurately set a control target value for the circulation flow rate based on the predicted value of the first learning model by reconstructing the first learning model with improved prediction accuracy in the first learning model construction unit 38. As a result, the circulation pump adjustment unit 33 can suitably control the circulation flow rate by adjusting the number of circulation pumps 12 based on the control target value.

[0076] The reduction in prediction error by such reconstruction of the learning model can be performed similarly for the second learning model handled by the second learning model construction unit 39. That is, when the prediction error of the second learning model calculated by the prediction error calculation unit 40 becomes equal to or less than a threshold, the additional learning data selection unit 42 selects additional learning data to be added to the learning data for reconstructing the second learning model, and the second learning model is reconstructed using new learning data including the additional learning data. At this time, the additional learning data to be added to the learning data is appropriately selected based on the degree of deviation from the initial learning data conventionally included in the learning data, thereby effectively reducing the prediction error of the second learning model. As a result, even if the prediction error of the second learning model is reduced due to some factor, good prediction accuracy can be obtained by reconstructing the second learning model.

[0077] The control device 15 can accurately set a control target value for the absorbent slurry supply amount based on the predicted value of the second learning model by reconstructing the second learning model with improved prediction accuracy in the second learning model construction unit 39. As a result, the absorbent slurry supply control unit 37 can suitably control the supply amount of absorbent slurry by controlling the absorbent slurry supply amount control valve 23 based on the control target value.

[0078] In the above embodiment, the SO 2 As an absorbent, CaCO 3 However, CaCO 3 However, this is not limited to SO 2 As an absorbent for 2 ) etc. can also be used.

[0079] 12 as a device for executing each process in the control device 15, may be configured to be electrically connected to the edge server 42 via a cloud environment or VPN so as to be able to communicate with the edge server 42. In this case, the information processing device 52 may include an operation data receiving unit 30, a first relationship table creating unit 31, a circulation flow rate determining unit 32, a second relationship table creating unit 35, an absorbent slurry supply amount determining unit 36, a first learning model constructing unit 38, a second learning model constructing unit 39, a prediction error calculating unit 40, and an additional learning data selecting unit 42, and may communicate the control target values ​​determined by the circulation flow rate determining unit 32 and the absorbent slurry supply amount determining unit 36 ​​to the circulation pump adjusting unit 33 and the absorbent slurry supply control unit 37 in the control device 15 to control the circulation pump and the supply amount of the absorbent. Furthermore, the driving data receiving unit 30 may receive various driving data via the driving data relay unit 43 of the control device 15, or may receive various driving data from the driving data acquiring unit 20 as described above.

[0080] In particular, when performing calculations on a cloud environment, from the viewpoint of security, there are cases where the control target values ​​of the circulation pump and the absorbent are not directly controlled but are only displayed. For example, the operation index diagram generated on the cloud environment may be sent and illustrated on a customer's device (terminal 54) via a dedicated app, and the customer may manually update the local operation index diagram. On the other hand, the information processing device 52 may also include a circulation pump adjustment unit 33 and an absorbent slurry supply control unit 37, and may remotely control the circulation pump and the supply amount of absorbent. Furthermore, the information processing device 52 may be configured to execute each process in the information processing device 52 in response to a request from the terminal 54.

[0081] In addition, within the scope of the present disclosure, the components in the above-described embodiments may be replaced with well-known components as appropriate, and the above-described embodiments may be combined as appropriate.

[0082] The contents described in each of the above embodiments can be understood, for example, as follows.

[0083] (1) An apparatus according to one embodiment includes: An apparatus for executing a process related to plant control based on a prediction result using a learning model, an additional learning data selection unit for selecting, when a prediction result using the learning model satisfies a predetermined condition, data from the driving data that has a large deviation from the learning data used to construct the learning model as additional learning data; a learning model construction unit for reconstructing the learning model by using new learning data including the learning data and the additional learning data; Equipped with.

[0084] According to the above aspect (1), new learning data is created by selecting and adding additional learning data to the learning data for constructing the learning model, and the learning model is reconstructed using the learning data. At this time, the additional learning data to be added to the new learning data is selected so as to include data having a large deviation from the learning data, so that the learning model can be appropriately reconstructed. Then, a process related to the control of the plant is executed based on the prediction result of the learning model appropriately reconstructed in this way, and good control accuracy is obtained.

[0085] (2) In another embodiment, in the above embodiment (1), When a prediction result using the reconstructed new learning model satisfies a predetermined condition, the additional learning data selection unit further selects, from the driving data not selected as the additional learning data, data including data with a large deviation as the additional learning data, and the learning data construction unit reconstructs the learning model using the new learning data including the additional learning data further selected by the additional learning data selection unit.

[0086] According to the above aspect (2), when a prediction result using the new reconstructed learning model satisfies a predetermined condition, new additional learning data is further added to the learning data to create new learning data, and the learning model is reconstructed again using the new learning data. By repeatedly selecting such additional learning data and reconstructing the learning model, the prediction error of the learning model can be sufficiently reduced.

[0087] (3) In another embodiment, in the above embodiment (1) or (2), The additional learning data selection unit selects, as the additional learning data, an average value of a parameter included in the driving data for a predetermined period.

[0088] According to the above aspect (3), by using an average value over a predetermined period of time as an additional learning parameter, it is possible to effectively reduce the amount of calculation when reconstructing a learning model while ensuring learning accuracy.

[0089] (4) In another embodiment, in any one of the above (1) to (3), The learning model construction unit reconstructs the learning model when the prediction result continuously satisfies the specified condition for a specified period of time or more.

[0090] According to the above aspect (4), the judgment of whether the prediction result satisfies the predetermined condition is made based on whether the prediction result satisfies the predetermined condition continuously for a predetermined time. The prediction result may vary depending on the operating state of the plant, and if a short-term judgment is made, the learning model will be reconstructed frequently, which may increase the burden of model management. Therefore, by making a continuous judgment over a predetermined time as in this aspect, the learning model can be reconstructed appropriately, enabling efficient model management.

[0091] (5) In another embodiment, in any one of the above (1) to (4), The learning data is data used before the learning model is constructed or data used in the previous construction.

[0092] According to the above aspect (5), the learning model is reconstructed using new learning data created by adding additional learning data to the data before the learning model is constructed or the learning data used in the previous construction of the learning model.

[0093] (6) In another embodiment, in any one of the above (1) to (5), The additional learning data selection unit selects the additional learning data from the operation data acquired during steady operation of the plant.

[0094] According to the above aspect (6), the selection of additional learning data is performed from the operation data acquired during steady operation of the plant. For example, operation data acquired during non-steady operation such as when an abnormality occurs in the plant, when the plant starts up, or when the plant stops operating is excluded from the selection of additional learning data, so that the prediction result of the learning model can be appropriately obtained. Furthermore, when the operation data includes data acquired during such non-steady operation, the data may be excluded by performing preprocessing on the operation data.

[0095] (7) In another embodiment, in any one of the above (1) to (6), When the prediction result using the learning model satisfies a predetermined condition, it indicates that the prediction error of the predicted value obtained using the learning model satisfies a threshold value.

[0096] According to the above aspect (7), the determination of whether the prediction result satisfies the predetermined condition is performed based on whether the prediction error of the prediction value obtained using the learning model satisfies a threshold. As a result, even if the prediction error of the learning model decreases due to some factor, the learning model is reconstructed using new learning data including efficiently selected additional learning data, and good prediction accuracy is obtained. Then, good control accuracy is obtained by executing processing related to the control of the plant based on the prediction value of the learning model with improved prediction accuracy.

[0097] (8) In another embodiment, in the above embodiment (7), The additional learning data selection unit selects parameters to be included in the additional learning data from the driving data based on the degree of contribution to the predicted value.

[0098] According to the above aspect (8), by including some parameters selected from the driving data in the additional learning data, it is possible to effectively reduce the amount of calculation when reconstructing the learning model while ensuring the learning accuracy.

[0099] (9) In another embodiment, in the above embodiment (7) or (8), The plant is a wet flue gas desulfurization plant that performs desulfurization by bringing exhaust gas generated in a combustion device into gas-liquid contact with an absorbing liquid circulated in an absorption tower, The predicted value is the sulfur dioxide concentration of the flue gas at the outlet of the absorption tower.

[0100] According to the above aspect (9), the prediction accuracy of the learning model for predicting the sulfur dioxide concentration of the exhaust gas at the outlet of the absorption tower of the wet flue gas desulfurization system can be preferably ensured by reconstructing the learning model when the prediction error becomes larger than a predetermined value.

[0101] (10) In another embodiment, in the above embodiment (9), A control target value for the circulation flow rate of the absorption liquid is determined based on the predicted value calculated by the learning model.

[0102] According to the above aspect (10), a predicted value is calculated using a learning model in which the prediction error has been reduced by reconstruction, and a control target value for the circulation amount of the absorption liquid is determined based on the predicted value, thereby obtaining good control accuracy.

[0103] (11) In another embodiment, in the above embodiment (9) or (10), A control target value for the amount of absorbent supplied to the absorption tower is determined based on the predicted value calculated by the learning model.

[0104] According to the above aspect (11), a predicted value is calculated using a learning model in which the prediction error has been reduced by reconstruction, and a control target value for the supply amount of the supply agent is determined based on the predicted value, thereby obtaining good control accuracy.

[0105] (12) A remote monitoring system according to one aspect includes: A remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, The apparatus comprises: an additional learning data selection unit for selecting, when a prediction result using the learning model satisfies a predetermined condition, from the driving data, data having a large deviation from the learning data used to construct the learning model, as additional learning data, in response to a request from the terminal; a learning model construction unit for reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with.

[0106] According to the above aspect (12), new learning data is created by selecting and adding additional learning data to the learning data for constructing the learning model, and the learning model is reconstructed using the learning data. At this time, the additional learning data to be added to the new learning data is selected so as to include data having a large deviation from the learning data, so that the learning model can be appropriately reconstructed. Then, a process related to the control of the plant is executed based on the prediction result of the learning model appropriately reconstructed in this way, and good control accuracy is obtained.

[0107] (13) A method for controlling an apparatus according to one aspect includes the steps of: A method for controlling an apparatus for executing a process related to plant control based on a prediction result using a learning model, comprising: an additional learning data selection step of selecting, from the driving data, data having a large deviation from the learning data used to construct the learning model as additional learning data when a prediction result using the learning model satisfies a predetermined condition; a learning model construction step of reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with.

[0108] According to the above aspect (13), new learning data is created by selecting and adding additional learning data to the learning data for constructing the learning model, and the learning model is reconstructed using the learning data. At this time, the additional learning data to be added to the new learning data is selected so as to include data having a large deviation from the learning data, so that the learning model can be appropriately reconstructed. Then, a process related to the control of the plant is executed based on the prediction result of the learning model appropriately reconstructed in this way, and good control accuracy is obtained.

[0109] (14) A method for controlling a remote monitoring system according to one embodiment includes: A control method for a remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, an additional learning data selection step of selecting, when a prediction result using the learning model satisfies a predetermined condition, data having a large deviation from the learning data used to construct the learning model from the driving data, as additional learning data, in response to a request from the terminal; a learning model construction step of reconstructing the learning model using learning data including the learning data and the additional learning data; Equipped with.

[0110] According to the above aspect (14), new learning data is created by selecting and adding additional learning data to the learning data for constructing the learning model, and the learning model is reconstructed using the learning data. At this time, the additional learning data to be added to the new learning data is selected so as to include data having a large deviation from the learning data, so that the learning model can be appropriately reconstructed. Then, a process related to the control of the plant is executed based on the prediction result of the learning model appropriately reconstructed in this way, and good control accuracy is obtained. [Explanation of symbols]

[0111] 1 Combustion equipment 2 Piping 3 Circulation piping 5. Generator 10 Wet flue gas desulfurization equipment 11 Absorption Tower 12 Circulation pump 13 Absorbent slurry supply section 14 Gypsum Recovery Section 15 Control device 16 Outlet pipe 17 Gas analyzer 20 Operation data acquisition section 21 Absorbent slurry production facility 22 Absorbent slurry supply pipe 23 Absorbent slurry supply amount control valve 25 Gypsum separator 26 Gypsum slurry extraction pipe 27 Gypsum slurry extraction pump 30 Operation data receiver 31 First relation table creation unit 32 Circulation flow rate determination section 33 Circulation pump adjustment section 35 Second relation table creation unit 36 Absorbent slurry supply amount determination unit 37 Absorbent slurry supply control unit 38 First Learning Model Construction Unit 39 Second Learning Model Construction Section 40 Prediction error calculation unit 42 Additional learning data selection unit 43 Operation Data Relay Unit 44 Remote Monitoring System 48 Edge Servers 50 Remote monitoring device 52 Information processing equipment 54 Terminals 55 Information Processing Systems

Claims

1. An apparatus for executing a process related to plant control based on a prediction result using a learning model, an additional learning data selection unit for, when a prediction result using the learning model satisfies a predetermined condition, calculating a distance to a closest initial learning data among a plurality of initial learning data used in the previous construction of the learning model for a plurality of additional learning data candidates that are new operating data collected from the plant after the previous construction of the learning model in a space defined by a plurality of variables included in explanatory variables of the learning model, and selecting the data with the maximum distance as additional learning data for each of the plurality of initial learning data used in the previous construction of the learning model; a learning model construction unit for reconstructing the learning model by using new learning data including the initial learning data and the additional learning data; An apparatus comprising:

2. 2. The device according to claim 1, wherein, when a prediction result using the reconstructed new learning model satisfies a predetermined condition, the additional learning data selection unit further selects, from the driving data not selected as the additional learning data, data that has a large deviation from the initial learning data and the additional learning data selected at the time of reconstruction as the additional learning data, and the learning model construction unit reconstructs the learning model using the new learning data including the additional learning data further selected by the additional learning data selection unit.

3. The device according to claim 1 or 2, wherein the additional learning data selection unit selects, as the additional learning data, an average value of a parameter included in the driving data for a predetermined period.

4. The device according to claim 1 , wherein the learning model construction unit reconstructs the learning model when the prediction result continuously satisfies the predetermined condition for a predetermined period of time or more.

5. The device according to claim 1 , wherein the initial learning data is data used before the learning model is constructed or data used in a previous construction of the learning model.

6. The apparatus according to claim 1 , wherein the additional learning data selection unit selects the additional learning data from the operation data acquired during steady operation of the plant.

7. The device according to claim 1 , further comprising: a prediction error determining section for determining whether a prediction result obtained using the learning model satisfies a threshold value when the prediction result obtained using the learning model satisfies a predetermined condition;

8. The device according to claim 7 , wherein the additional learning data selection unit selects parameters to be included in the additional learning data from the driving data based on a contribution to the predicted value.

9. The plant is a wet flue gas desulfurization plant that performs desulfurization by bringing exhaust gas generated in a combustion device into gas-liquid contact with an absorbing liquid circulated in an absorption tower, The apparatus according to claim 7 or 8, wherein the predicted value is a sulfur dioxide concentration of the exhaust gas at an outlet of the absorption tower.

10. The apparatus according to claim 9 , further comprising: determining a control target value of the circulation flow rate of the absorption liquid based on the predicted value calculated by the learning model.

11. The apparatus according to claim 9 or 10, further comprising: determining a control target value of an absorbent supply amount to the absorption tower based on the predicted value calculated by the learning model.

12. A remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, The apparatus comprises: an additional learning data selection unit for, when a prediction result using the learning model satisfies a predetermined condition in response to a request from the terminal, calculating a distance to the closest initial learning data among the initial learning data used in the previous construction of the learning model for a plurality of additional learning data candidates that are new operating data collected from the plant after the previous construction of the learning model in a space defined by a plurality of variables included in explanatory variables of the learning model, and selecting the data with the maximum distance as additional learning data for each of the initial learning data used in the previous construction of the learning model; a learning model construction unit for reconstructing the learning model using learning data including the initial learning data and the additional learning data; A remote monitoring system comprising:

13. A method for controlling an apparatus for executing a process related to plant control based on a prediction result using a learning model, comprising: an additional learning data selection step of calculating, when a prediction result using the learning model satisfies a predetermined condition, a distance to the closest initial learning data among the initial learning data used in the previous construction of the learning model for a plurality of additional learning data candidates that are new operating data collected from the plant after the previous construction of the learning model in a space defined by a plurality of variables included in explanatory variables of the learning model, and selecting the data with the maximum distance as additional learning data for each of the initial learning data used in the previous construction of the learning model; a learning model construction step of reconstructing the learning model using learning data including the initial learning data and the additional learning data; A method for controlling an apparatus comprising:

14. A control method for a remote monitoring system including a device for executing a process related to plant control based on a prediction result using a learning model and a terminal capable of communicating therewith, an additional learning data selection step of, when a prediction result using the learning model satisfies a predetermined condition upon request from the terminal, calculating a distance to the closest initial learning data among the initial learning data used in the previous construction of the learning model for a plurality of additional learning data candidates that are new operating data collected from the plant after the previous construction of the learning model in a space defined by a plurality of variables included in explanatory variables of the learning model, and selecting the data with the maximum distance as additional learning data for each of the initial learning data used in the previous construction of the learning model; a learning model construction step of reconstructing the learning model using learning data including the initial learning data and the additional learning data; A method for controlling a remote monitoring system comprising:

Images