Deterioration prediction device and method, differential pressure prediction device and method, complete set of equipment and storage medium
By generating a prediction model, the evaluation of catalyst degradation in a complete set of equipment is simplified, the problem of complex analysis is solved, and the rapid and accurate evaluation of catalyst degradation and differential pressure prediction is achieved, which improves the operation efficiency of the equipment.
Patent Information
- Application Number
- CN202080092582.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-06-19
- Filing Date
- 2020-12-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-12-23
AI Technical Summary
In complete sets of equipment, predicting the degree of catalyst degradation usually requires complex exhaust gas trait analysis, making it difficult for users to quickly understand the degradation of the catalyst.
By generating a prediction model based on past operation data, predicting the degree of catalyst degradation in different complete sets of equipment, and combining differential pressure prediction, the differential pressure between the input and output of the air preheater is estimated, simplifying the evaluation process of catalyst degradation.
Users can learn the degree of deterioration of the catalyst without performing complex processing, which improves the operation efficiency of complete sets of equipment and the accuracy of maintenance plans.
Smart Images

Figure CN114929366B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a prediction device, a complete set of equipment, a prediction method, a program and a configuration program.
[0002] This application claims priority to Japanese Patent Application No. 2020-017156 filed in Japan on February 4, 2020 and Japanese Patent Application No. 2020-106496 filed in Japan on June 19, 2020, the contents of which are incorporated herein by reference. Background Art
[0003] Catalysts are used in plants equipped with boilers, etc. In plants using catalysts, the catalysts deteriorate as the plants are operated. Therefore, it is desirable to understand the degree of catalyst deterioration in plants using catalysts.
[0004] Patent Document 1 discloses a technique for predicting the degree of catalyst degradation as a related technique.
[0005] Prior art literature
[0006] Patent Literature
[0007] Patent Document 1: JP Patent No. 6278296 Summary of the invention
[0008] Problems to be solved by the invention
[0009] However, in order to predict the degree of catalyst degradation in a plant, a plant user generally needs to perform complicated processing such as analyzing the properties of exhaust gas each time when making a prediction.
[0010] For this reason, in a plant, a technology is sought that allows a user to know the degree of catalyst degradation without performing complicated processing.
[0011] The present disclosure aims to provide a prediction device, a plant, a prediction method, a program, and a configuration program that can solve the above-mentioned problems.
[0012] Means for solving problems
[0013] In order to solve the above-mentioned problems, the degradation prediction device involved in the present disclosure comprises: a model generating unit, which generates a first prediction model for predicting the degree of degradation of the catalyst in the second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and a degradation degree prediction unit, which predicts the degree of degradation in the second set of equipment based on the first prediction model generated by the model generating unit.
[0014] The differential pressure prediction device according to the present disclosure includes: the above-mentioned degradation prediction device; and a differential pressure estimation unit that estimates the differential pressure between the input and output of an air preheater based on the degradation degree predicted by the degradation prediction device.
[0015] The plant according to the present disclosure is a plant using a catalyst, and includes: a device for causing degradation of the catalyst; and the above-mentioned degradation prediction device for predicting the degree of degradation of the catalyst.
[0016] The degradation prediction method involved in the present disclosure includes: generating a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and predicting the degree of degradation in the second set of equipment based on the generated first prediction model.
[0017] The differential pressure prediction method involved in the present disclosure includes: generating a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; predicting the degree of degradation in the second set of equipment based on the generated first prediction model; and estimating the differential pressure between the input and output of an air preheater based on the predicted degree of degradation.
[0018] The program involved in the present disclosure enables a computer to execute: generating a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and predicting the degree of degradation in the second set of equipment based on the generated first prediction model.
[0019] The program for causing a computer to execute the processing of the configuration involved in the present disclosure respectively constitutes the following elements as hardware: a model generation unit, which generates a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and a degradation degree prediction unit, which predicts the degree of degradation in the second set of equipment based on the first prediction model generated by the model generation unit.
[0020] Effects of the Invention
[0021] According to the prediction device, the plant, the prediction method, the program, and the configuration program of the embodiment of the present disclosure, the user of the plant can obtain the degree of degradation of the catalyst without performing complicated processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 This is a diagram showing an example of the configuration of a plant according to the first embodiment of the present disclosure.
[0023] Figure 2 It is a diagram showing an example of the structure of the sensor device according to the first embodiment of the present disclosure.
[0024] Figure 3 It is a diagram showing an example of the configuration of the degradation prediction device according to the first embodiment of the present disclosure.
[0025] Figure 4 This is a diagram showing an example of past performance data in various plants in the first embodiment of the present disclosure.
[0026] Figure 5 It is a diagram for explaining the generation of the first model and the second model in the first embodiment of the present disclosure.
[0027] Figure 6 This is a diagram showing an example of input of data from the first model and the second model in the first embodiment of the present disclosure.
[0028] Figure 7 This is the first diagram showing the processing flow of the degradation prediction device according to the first embodiment of the present disclosure.
[0029] Figure 8 FIG. 2 is a second diagram showing a processing flow of the degradation prediction device according to the first embodiment of the present disclosure.
[0030] Fig. 9 This is a diagram showing an example of comparison results of Examples in the first embodiment of the present disclosure.
[0031] Fig.10 It is a diagram showing an example of the configuration of a differential pressure prediction device according to a second embodiment of the present disclosure.
[0032] Fig.11 It is a diagram showing an example of the configuration of a plant according to the second embodiment of the present disclosure.
[0033] Fig.12 It is a diagram showing a processing flow of the differential pressure prediction device according to the second embodiment of the present disclosure.
[0034] Fig.13 It is a diagram for explaining the processing of the differential pressure prediction device according to the second embodiment of the present disclosure.
[0035] Fig.14 This is a diagram showing an example of comparison results of Examples in the second embodiment of the present disclosure.
[0036] Fig.15 This is a schematic block diagram showing the structure of a computer according to at least one embodiment. DETAILED DESCRIPTION
[0037] <First Embodiment>
[0038] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.
[0039] The degradation prediction system according to the first embodiment of the present disclosure predicts the degradation state of the denitration device 20 provided in the plant 1 based on the operation data of the plant 1 as the prediction object. The degradation prediction system includes a data server device 50 and a degradation prediction device 40. The data server device 50 is a device for storing data of the plant 1. The data of the plant 1 includes, for example, fuel properties, operation data, etc. The degradation prediction device 40 predicts the degradation state of the denitration device 20 provided in the plant 1 based on the data stored in the data server device 50.
[0040] (Structure of complete equipment)
[0041] The structure of the plant 1 (an example of a second plant, an example of a plant) according to the first embodiment will be described.
[0042] The plant 1 of the first embodiment is a plant that predicts the degree of degradation of the performance of the catalyst with respect to the total operating time of the plant 1 .
[0043] Complete set of equipment 1 Figure 1 As shown, a boiler 10 , a denitration device 20 (an example of a device in which catalyst degradation occurs), and a sensor device 30 are provided.
[0044] The boiler 10 is a boiler that uses coal or the like as fuel. The boiler 10 discharges combustion gas generated when burning the fuel to the denitration device 20 .
[0045] The denitration device 20 is a device for reducing the NOx concentration in the combustion gas by decomposing NOx (nitrogen oxides) contained in the combustion gas. For stable operation of the plant 1, it is preferable to keep the denitration rate of the denitration device 20 substantially constant.
[0046] For example, the denitration device 20 injects ammonia (NH 3 ) to keep the denitration rate, which tends to decrease due to the degradation of the catalyst performance caused by the poisoning component, constant. The so-called poisoning component is a component that comes from the outside and is different from the main component of the catalyst.
[0047] In the denitration device 20, poisoning components enter from the outside. Therefore, in the denitration device 20, the active points decrease with the increase of poisoning components. As a result, the value of the reaction rate constant K indicating the degree (goodness) of the performance of the catalyst decreases. That is, the performance of the catalyst deteriorates.
[0048] The plant 1 of the first embodiment is an object for predicting the degree of degradation of the performance of the catalyst.
[0049] The sensor device 30 is as follows Figure 2 As shown in the figure, the first sensor 301, the second sensor 302, the third sensor 303, and the fourth sensor 304 are provided. The sensor device 30 is mounted at the inlet of the denitration device 20.
[0050] The first sensor 301 detects the dust concentration at the inlet of the denitration device 20 .
[0051] The second sensor 302 detects the NOx (nitrogen oxide) concentration at the inlet of the denitration device 20 .
[0052] The third sensor 303 detects the SOx (sulfur oxide) concentration at the inlet of the denitration device 20 .
[0053] The fourth sensor 304 detects the O at the inlet of the denitration device 20. 2 (Oxygen) concentration.
[0054] The data measured by the sensor device 30 is transmitted to the data server device 50 via a network such as the Internet. As a result, the data of the plant 1 is accumulated in the data server device 50.
[0055] (Structure of Degradation Prediction Device)
[0056] The degradation prediction device 40 is a device that predicts the degree of degradation of the performance of the catalyst relative to the total operating time of the plant 1 .
[0057] The degradation prediction device 40 is as follows Figure 3 As shown, it includes a storage unit 401, a planned value acquisition unit 402, a complete set of equipment data acquisition unit 403, a first model generation unit 404 (an example of a model generation unit), a second model generation unit 405 (an example of a model generation unit), a post-poisoning data acquisition unit 406, and a catalyst degradation degree acquisition unit 407 (an example of a degradation degree prediction unit).
[0058] The storage unit 401 stores various information necessary for the processing of the degradation prediction device 40 .
[0059] For example, the storage unit 401 stores past performance data D1 in various plants (an example of the first plant).
[0060] Specifically, the storage unit 401 stores, for example, Figure 4 The planned values, complete equipment data, and catalyst data after poisoning are shown.
[0061] Furthermore, the storage unit 401 stores data that changes with the passage of time in association with the time.
[0062] like Figure 4 As shown, the planned values include catalyst performance, catalyst specifications, and device specifications.
[0063] Examples of catalyst performance include the initial value K0 of the reaction rate constant, the ratio of the reaction rate constant K after a certain operation time to the initial value K0 of the reaction rate constant (K / K0) (the degree of degradation of the reaction rate constant, i.e., the degree of degradation of the catalyst), etc. In addition, the initial value K0 of the reaction rate constant is a value given in advance.
[0064] In addition, examples of catalyst specifications include the initial specific surface area of the catalyst, the initial pore volume of the catalyst, the initial composition of the catalyst (e.g., TiO 2 (TiO2), WO 3 (tungsten trioxide), V 2 O 5 (vanadium pentoxide), SiO 2 (silicon dioxide), etc. In addition, the specifications of the catalyst can be obtained by, for example, sampling and analyzing the actual catalyst.
[0065] In addition, examples of device specifications include flow rate, dust concentration at the inlet of the denitrification device 20, NOx (nitrogen oxide) concentration at the inlet of the denitrification device 20, SOx (sulfur oxide) concentration at the inlet of the denitrification device 20, and O 2 (Oxygen) concentration, etc.
[0066] The flow rate is a value determined for each device. The dust concentration at the inlet of the denitration device 20, the NOx concentration at the inlet of the denitration device 20, the SOx concentration at the inlet of the denitration device 20, and the O 2 The concentrations are values measured by the sensor device 30 provided at the inlet of the denitrification device 20. The dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the O 2 The concentrations can be calculated based on the fuel properties. The fuel properties here are data indicating the amount of each component in the ash flowing into the catalyst, and are data obtained by analyzing the fuel for each plant.
[0067] In addition, if Figure 4 As shown, the plant data includes fuel properties and operation data.
[0068] Examples of fuel properties include SiO 2 、Al 2 O 3 The amount of components in the fly ash flowing into the catalyst.
[0069] Furthermore, an example of the operation data is the operation time of the plant.
[0070] In addition, if Figure 4 As shown, the catalyst data after poisoning includes the catalyst properties after poisoning and the catalyst composition after poisoning.
[0071] Examples of the physical properties of the poisoned catalyst include the specific surface area of the poisoned catalyst and the pore volume of the poisoned catalyst.
[0072] In addition, as an example of the composition of the poisoned catalyst, TiO in the poisoned catalyst can be cited. 2 The ratio of WO 3 The value obtained by normalizing the ratio of TiO 2 / WO 3 ), SiO in the poisoned catalyst 2 The ratio of WO 3 The value obtained by normalizing the ratio of 2 / WO 3 )wait.
[0073] The poisoned catalyst data is data of the catalyst inside the denitration device 20, and is difficult to obtain during operation of the plant. Therefore, the poisoned catalyst data is obtained during maintenance of the plant 1, etc.
[0074] Specifically, the storage unit 401 stores the degree of catalyst degradation (ie, the ratio (K / K0)) obtained from past plant operations.
[0075] In addition, for example, the storage unit 401 stores data used when predicting the degree of degradation of the catalyst in the plant 1 using a model.
[0076] Specifically, the storage unit 401 stores the first model (an example of the second prediction model) generated using the past performance data D1, the second model (an example of the first prediction model) generated using the past performance data D1, the catalyst performance of the plant 1, the catalyst specifications of the plant 1, the device specifications of the plant 1, and the fuel properties of the plant 1. The first model is a model for predicting the catalyst data after poisoning in the plant 1. The second model is a model for predicting the degree of catalyst degradation based on the catalyst data after poisoning predicted by the first model. The first model and the second model are respectively learned models obtained by learning using the past performance data in various plants.
[0077] As an example of the catalytic performance of the plant 1 , the initial value K0 of the reaction rate constant of the plant 1 can be cited.
[0078] In addition, examples of the catalyst specifications of the plant 1 include the initial specific surface area of the catalyst in the plant 1 , the initial pore volume of the catalyst in the plant 1 , and the like.
[0079] Furthermore, examples of the equipment specifications of the plant 1 include the flow rate in the plant 1 and the like.
[0080] In addition, as an example of the fuel properties of the plant 1, the amount of the components in the ash in the plant 1 flowing into the catalyst (for example, TiO 2 、SiO 2 isotropic catalyst inflow).
[0081] In addition, the generation of the first model and the generation of the second model will be described later.
[0082] The planned value acquisition unit 402 acquires the planned value of the plant 1 .
[0083] For example, the planned value acquisition unit 402 reads the catalyst performance of the plant 1 , the catalyst specifications of the plant 1 , and the device specifications of the plant 1 from the storage unit 401 .
[0084] Specifically, the planned value acquisition unit 402 reads out the initial value K0 of the reaction rate constant of the plant 1 , the initial specific surface area of the catalyst, the initial pore volume of the catalyst, the initial composition of the catalyst, the flow rate, and the like from the storage unit 401 .
[0085] For example, the planned value acquisition unit 402 acquires from the sensor device 30 the dust concentration detected by the sensor device 30, the NOx (nitrogen oxide) concentration at the inlet of the denitrification device 20, the SOx (sulfur oxide) concentration at the inlet of the denitrification device 20, and the Ox concentration at the inlet of the denitrification device 20. 2 (Oxygen) concentration respective detection values.
[0086] Furthermore, the detection values acquired by the planned value acquisition unit 402 may be stored in the storage unit 401, and the stored detection values may be used as actual performance data when obtaining the degree of degradation of the catalyst in other plants.
[0087] The plant data acquisition unit 403 acquires the plant data of the plant 1 from the data server device 50 , and records the acquired plant data in the storage unit 401 .
[0088] (Generation of the first model)
[0089] The first model generation unit 404 generates a first model. Examples of the first model include machine learning models such as neural networks, random forests, and support vector machines. In the following description, the generation of the first model is described using a neural network as an example.
[0090] The first model generating unit 404 generates a learned first model by performing machine learning on the modeled neural network using past performance data (planned values, plant data, catalyst data after poisoning) in various plants as learning data.
[0091] The neural network is, for example, a convolutional neural network having an input layer, an intermediate layer, and an output layer. In addition, the learning data here is data in which the planned value, the plant data, and the catalyst data after poisoning are respectively established in a one-to-one correspondence.
[0092] Specifically, the first model generation unit 404 divides a plurality of learning data (planned values, plant data, catalyst data after poisoning) into training data, evaluation data, and test data. Figure 5 As shown in FIG. 1 , the planned values of the training data and the equipment data are input into the neural network. The neural network ( Figure 5The first model generating unit 404 outputs the poisoned catalyst data whenever the planned values of the training data and the equipment data are input to the neural network and the poisoned catalyst data is output from the neural network, and the weight of the data combination between the nodes is changed (i.e., the model of the neural network is changed) by back propagating in accordance with the output. Next, the first model generating unit 404 inputs the planned values of the evaluation data and the equipment data to the neural network of the model changed by the planned values of the training data and the equipment data. The neural network outputs the poisoned catalyst data corresponding to the input planned values and the equipment data. The first model generating unit 404 changes the weight of the data combination between the nodes based on the output of the neural network as needed. The neural network generated by the first model generating unit 404 in this way is the first model that has been learned. Next, the first model generating unit 404 inputs the planned values of the test data and the equipment data to the neural network of the first model that has been learned as a final confirmation. The neural network of the first model that has been learned outputs the poisoned catalyst data corresponding to the input planned values of the test data and the equipment data. When the poisoned catalyst data output by the neural network of the first model after learning for all test data are within a given error range relative to the planned value of the input test data and the poisoned catalyst data associated with the complete set of equipment data, the neural network of the first model after learning is determined to be the desired model. In addition, even if the poisoned catalyst data output by the neural network of the first model after learning for one of the test data are not within a given error range relative to the planned value of the input test data and the poisoned catalyst data associated with the complete set of equipment data, the first model after learning is generated using new learning data.
[0093] The generation of the learned model by the first model generating unit 404 described above is repeated until a desired learned first model is obtained.
[0094] The first model generation unit 404 writes the generated learned first model into the storage unit 401 .
[0095] (Creation of the second model)
[0096] The second model generating unit 405 generates a second model.
[0097] For example, the second model generation unit 405 generates a learned second model by performing machine learning on the modeled neural network using past performance data (planned values, plant data, catalyst data after poisoning, and catalyst degradation data) in various plants as learning data.
[0098] The neural network is, for example, a convolutional neural network having an input layer, an intermediate layer, and an output layer. In addition, the learning data here is data in which the planned value, the plant data, the catalyst data after poisoning, and the catalyst degradation data are respectively established one-to-one.
[0099] Specifically, the second model generation unit 405 divides a plurality of learning data (planned values, plant data, catalyst data after poisoning, catalyst degradation data) into training data, evaluation data, and test data. Figure 5 As shown, the planned values of the training data, the complete equipment data, and the catalyst data after poisoning are input into the neural network. The neural network outputs the catalyst degradation data. Then, similarly to the case where the first model generation unit 404 generates the first model, for all test data, when the catalyst degradation data output by the neural network of the second model after learning is within a given error range relative to the catalyst degradation data associated with the input planned values, complete equipment data, and the catalyst data after poisoning, the second model generation unit 405 determines that the neural network of the second model after learning is the desired model. In addition, even if there is one test data in which the catalyst degradation data output by the neural network of the second model after learning is not within a given error range relative to the catalyst degradation data associated with the planned values of the input test data, complete equipment data, and the catalyst data after poisoning, the new learning data is used to generate the learned second model.
[0100] The generation of the learned model by the second model generating unit 405 is repeated until a desired learned second model is obtained.
[0101] The second model generation unit 405 writes the generated learned second model into the storage unit 401 .
[0102] The post-poisoning data acquisition unit 406 acquires catalyst data after poisoning based on the planned value, the plant data, and the learned first model.
[0103] For example, the post-poisoning data acquisition unit 406 reads out the catalytic performance of the complete set of equipment 1 (for example, the initial value K0 of the reaction rate constant of the complete set of equipment 1), the catalyst specifications of the complete set of equipment 1 (for example, the initial specific surface area of the catalyst in the complete set of equipment 1, the initial pore volume of the catalyst in the complete set of equipment 1, etc.), the device specifications of the complete set of equipment 1 (for example, the flow rate in the complete set of equipment 1, etc.), the fuel properties of the complete set of equipment 1 (for example, the inflow amount of ash components in the complete set of equipment 1 into the catalyst), and the learned first model from the storage unit 401.
[0104] In addition, the post-poisoning data acquisition unit 406 acquires data indicating the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the O concentration at the inlet of the denitrification device 20 from the sensor device 30 during the operation of the plant 1. 2 Concentration information.
[0105] The post-poisoning data acquisition unit 406 is as follows Figure 6 As shown in FIG. 1 , the catalyst performance of the plant 1, the catalyst specification of the plant 1, the device specification of the plant 1, the fuel property of the plant 1, the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the Ox concentration at the inlet of the denitrification device 20 obtained from the sensor device 30 are respectively represented. 2 The concentration information and the information indicating the operation time are input to the learned first model.
[0106] Then, the post-poisoning data acquisition unit 406 acquires the post-poisoning catalyst data (eg, post-poisoning catalyst properties, post-poisoning catalyst composition) output by the first model after learning.
[0107] The catalyst data after poisoning acquired by the data acquisition unit 406 after poisoning is the catalyst data after poisoning obtained by predicting the situation in which the plant 1 is operated for the time indicated by the plant data under the conditions indicated by the planned value and the plant data.
[0108] The catalyst degradation degree acquisition unit 407 acquires catalyst degradation data based on the planned value, the plant data, the catalyst data after poisoning, and the learned second model.
[0109] For example, the catalyst degradation degree acquisition unit 407 reads out the catalyst specifications of the complete set of equipment 1 (for example, the initial specific surface area of the catalyst in the complete set of equipment 1, the initial pore volume of the catalyst in the complete set of equipment 1, etc.), the device specifications of the complete set of equipment 1 (for example, the flow rate in the complete set of equipment 1, etc.), and the learned second model from the storage unit 401.
[0110] In addition, the catalyst degradation degree acquisition unit 407 acquires from the sensor device 30 during the operation of the plant 1 the dust concentration at the inlet of the denitration device 20, the NOx concentration at the inlet of the denitration device 20, the SOx concentration at the inlet of the denitration device 20, and the O 2 Concentration information.
[0111] Furthermore, the catalyst degradation degree acquisition unit 407 acquires the catalyst data after poisoning predicted by the post-poisoning data acquisition unit 406 from the post-poisoning data acquisition unit 406 .
[0112] The catalyst degradation degree acquisition unit 407 is as follows: Figure 6 As shown in FIG. 1 , the catalyst specification of the plant 1, the device specification of the plant 1, the fuel property of the plant 1, the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the Ox concentration at the inlet of the denitrification device 20 obtained from the sensor device 30 are respectively represented. 2 The information on the concentration, the catalyst data after poisoning predicted by the post-poisoning data acquisition unit 406, and the information on the operation time are input to the learned second model.
[0113] Then, the catalyst degradation degree acquisition unit 407 acquires the catalyst degradation data output by the learned second model.
[0114] The catalyst degradation data acquired by the catalyst degradation degree acquisition unit 407 is catalyst degradation data obtained by predicting a situation in which the plant 1 is operated under the conditions indicated by the planned values and the plant data for a period of time indicated by the plant data.
[0115] (Model learning process)
[0116] The following describes the operation of the degradation prediction device 40. First, the learning process of the machine learning model used by the degradation prediction device 40 to predict the degradation of the catalyst will be described.
[0117] The manager of the degradation prediction device 40, etc., obtains the catalyst degradation data of the catalyst of the denitration device 20 provided in the plant 1 in advance when performing maintenance on the plant different from the prediction target plant 1. The catalyst degradation data is obtained by measuring the NOx concentration of the gas at the input and the NOx concentration of the gas at the output of the denitration device 20. In addition, the manager of the degradation prediction device 40, etc. obtains the poisoned catalyst data including the catalyst properties and the catalyst composition through a destructive inspection in advance when performing maintenance on the plant different from the prediction target plant 1. The poisoned catalyst data is the data of the catalyst inside the denitration device 20, which is difficult to obtain during the operation of the plant.
[0118] The first model generation unit 404 performs machine learning on the first model, which is a machine learning model, using past performance data (planned values, plant data, and catalyst data after poisoning) in various plants as learning data. The planned values and plant data are acquired from the data server device 50 .
[0119] Specifically, the first model generation unit 404 generates training data, which uses the planned value and the complete plant data in the performance data as input samples and the catalyst data after poisoning as output samples. The first model generation unit 404 learns the first model based on the training data so that the predicted value of the state of the catalyst after poisoning is output by inputting the planned value and the complete plant data.
[0120] The first model generation unit 404 writes the learned first model into the storage unit 401 .
[0121] The second model generating unit 405 causes the second model, which is a machine learning model, to perform machine learning using past performance data (planned values, plant data, catalyst data after poisoning, and catalyst degradation data) in various plants as learning data.
[0122] Specifically, the second model generation unit 405 generates training data, which uses the planned value, plant data, and catalyst data after poisoning in the performance data as input samples and the catalyst degradation data as output samples. The second model generation unit 405 learns the second model based on the training data so that the predicted value of the catalyst degradation data is output by inputting the planned value, plant data, and catalyst data after poisoning.
[0123] The second model generation unit 405 writes the learned second model into the storage unit 401 .
[0124] (Catalyst degradation prediction process)
[0125] Next, use Figure 7 as well as Figure 8 The process flow of the degradation prediction device 40 shown in the figure will be used to explain the process of predicting catalyst degradation by the degradation prediction device 40. The degradation prediction device 40 predicts catalyst data after poisoning based on the planned value and the plant data, and predicts catalyst degradation data using the predicted catalyst data after poisoning.
[0126] (Processing of catalyst data after prediction of poisoning)
[0127] First, the degradation prediction device 40 refers to Figure 7 The following describes the processing of catalyst data after poisoning predicted based on the planned value, plant data, and the learned first model.
[0128] The post-poisoning data acquisition unit 406 reads out the catalytic performance of the complete set of equipment 1 (for example, the initial value K0 of the reaction rate constant of the complete set of equipment 1), the catalyst specifications of the complete set of equipment 1 (for example, the initial specific surface area of the catalyst in the complete set of equipment 1, the initial pore volume of the catalyst in the complete set of equipment 1, etc.), the device specifications of the complete set of equipment 1 (for example, the flow rate in the complete set of equipment 1, etc.), the fuel properties of the complete set of equipment 1 (for example, the inflow amount of the ash components in the complete set of equipment 1 into the catalyst), and the learned first model (step S1) from the storage unit 401.
[0129] The post-poisoning data acquisition unit 406 acquires data indicating the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the O concentration at the inlet of the denitrification device 20 from the sensor device 30 during the operation of the plant 1. 2 concentration information (step S2).
[0130] The post-poisoning data acquisition unit 406 stores the catalyst performance of the plant 1, the catalyst specifications of the plant 1, the device specifications of the plant 1, the fuel properties of the plant 1, the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the Ox concentration at the inlet of the denitrification device 20 obtained from the sensor device 30. 2 The concentration information and the information indicating the operation time are input to the learned first model (step S3).
[0131] The post-poisoning data acquisition unit 406 acquires the post-poisoning catalyst data (eg, post-poisoning catalyst properties, post-poisoning catalyst composition) output by the learned first model (step S4).
[0132] (Processing of Catalyst Deterioration Prediction Data)
[0133] Next, refer to Figure 8 The following describes a process in which the degradation prediction device 40 predicts catalyst degradation data based on the planned value, plant data, catalyst data after poisoning, and the learned second model.
[0134] The catalyst degradation degree acquisition unit 407 reads out the catalyst specifications of the complete set of equipment 1 (for example, the initial specific surface area of the catalyst in the complete set of equipment 1, the initial pore volume of the catalyst in the complete set of equipment 1, etc.), the device specifications of the complete set of equipment 1 (for example, the flow rate in the complete set of equipment 1, etc.), and the learned second model from the storage unit 401 (step S11).
[0135] The catalyst degradation degree acquisition unit 407 acquires, during operation of the plant 1, data indicating the dust concentration at the inlet of the denitration device 20, the NOx concentration at the inlet of the denitration device 20, the SOx concentration at the inlet of the denitration device 20, and the O 2 concentration information (step S12).
[0136] The catalyst degradation degree acquisition unit 407 acquires the catalyst data after poisoning predicted by the post-poisoning data acquisition unit 406 from the post-poisoning data acquisition unit 406 (step S13).
[0137] The catalyst degradation degree acquisition unit 407 is as follows: Figure 6 As shown in FIG. 1 , the catalyst specification of the plant 1, the device specification of the plant 1, the fuel property of the plant 1, the dust concentration at the inlet of the denitrification device 20, the NOx concentration at the inlet of the denitrification device 20, the SOx concentration at the inlet of the denitrification device 20, and the Ox concentration at the inlet of the denitrification device 20 obtained from the sensor device 30 are respectively represented. 2 The concentration information, the catalyst data after poisoning predicted by the post-poisoning data acquisition unit 406, and the information indicating the operation time are input to the learned second model (step S14).
[0138] The catalyst degradation degree acquisition unit 407 acquires the catalyst degradation data output by the learned second model (step S15 ).
[0139] (Example)
[0140] For a certain plant, the prediction of the degree of degradation of the catalyst by the degradation prediction device 40 according to the first embodiment of the present disclosure is performed, and the prediction is compared with the actual measurement of the degree of degradation of the catalyst in the plant.
[0141] We prepared 127 data sets for 11 plants with different manufacturing years and operating hours, of which 70% were used as learning data and 30% as data for accuracy verification.
[0142] Fig. 9 This is the result of comparison between the predicted degree of catalyst degradation and the actual measurement of the degree of catalyst degradation.
[0143] exist Fig. 9 In the graph, the horizontal axis is the operation time. In addition, the vertical axis is the degree of catalyst degradation. In addition, the degree of catalyst degradation is a value obtained by dividing the reaction rate constant K in each operation time by the initial reaction rate constant K0.
[0144] Regarding the accuracy of prediction of the degree of catalyst degradation, a result of RMSE (Root Mean Squared Error) of 0.05 was obtained.
[0145] The plant 1 according to the first embodiment of the present disclosure has been described above.
[0146] A degradation prediction device (40) includes a model generation unit (404, 405) and a degradation degree prediction unit (407).
[0147] A model generating unit generates a first prediction model for predicting the degree of degradation of a catalyst in a second plant (1) different from the first plant based on learning data including first data related to the catalyst in past operation of the first plant and second data related to the state of the past operation.
[0148] A degradation degree prediction unit (407) predicts the degradation degree in the second plant (1) based on the first prediction model generated by the model generation unit (404, 405).
[0149] The degradation prediction device (40) can predict the degradation degree of the catalyst in the second plant (1) by simply inputting data about the second plant (1) into the first prediction model. As a result, the user of the second plant (1) can know the degradation degree of the catalyst without performing complex analysis on the catalyst.
[0150] <Second Embodiment>
[0151] The degradation prediction system according to the second embodiment of the present disclosure predicts the degradation state of the denitration device 20 provided in the plant 1 based on the operation data of the plant 1 as the prediction object, and predicts the differential pressure between the input and output of the air preheater 60 described later based on the predicted degradation state. The degradation prediction system includes a data server device 50 and a differential pressure prediction device 70.
[0152] (Structure of differential pressure prediction device)
[0153] The differential pressure prediction device 70 is a device that predicts the differential pressure of the air preheater 60 based on the predicted degree of degradation of the performance of the catalyst.
[0154] The differential pressure prediction device 70 is as follows Fig.10 As shown, in addition to the degradation prediction device 40 , a scaling factor estimation unit 701 , a leakage estimation unit 702 , an occlusion degree estimation unit 703 , and a differential pressure estimation unit 704 are provided.
[0155] The scaling factor estimation unit 701 obtains a correction factor for correcting the difference between the performance characteristics of the catalyst obtained in the laboratory and the performance characteristics of the catalyst in the actual plant 1. The correction factor is the scaling factor.
[0156] For example, for various combinations of gas temperature, gas concentration, and oxygen concentration, the performance characteristics of the catalyst are obtained in the laboratory. Then, the performance characteristics of the catalyst obtained are represented by a formula including gas temperature, gas concentration, and oxygen concentration. In the actual complete set of equipment 1, the performance of the catalyst is reduced relative to the performance characteristics of the catalyst obtained in the laboratory. In order to apply the formula representing the performance characteristics of the catalyst obtained in the laboratory to the actual complete set of equipment 1, the proportional amplification factor estimation unit 701 determines the proportional amplification factor to correct the formula. In addition, the proportional amplification factor is obtained by using the empirical rule obtained based on the performance characteristics of the catalyst obtained in the laboratory and the performance characteristics of the catalyst in the actual complete set of equipment 1.
[0157] The leakage estimation unit 702 estimates the amount of leaked ammonia based on an empirical formula that applies a scaling factor to a formula representing the performance characteristics of the catalyst obtained in the laboratory, the NOx concentration at the inlet of the denitrification device 20, the NOx concentration at the outlet of the denitrification device 20, and catalyst degradation data. The performance characteristics of the catalyst are obtained according to the empirical formula, and the amount of unreacted ammonia is estimated based on the performance characteristics and the NOx concentration in the input and output of the denitrification device 20.
[0158] The blockage degree estimating unit 703 estimates the blockage degree in the air preheater 60 using an empirical formula obtained based on differential pressure data when the differential pressure in the air preheater 60 in the actual plant 1 increases and the operation data of the plant 1 .
[0159] For example, when the differential pressure of the air preheater 60 in the actual plant 1 increases, the differential pressure between the input and output of the elements of the air preheater 60 is monitored and the differential pressure is calculated. The blockage degree is calculated based on the differential pressure, and the empirical formula is obtained using the operating data of the plant 1 at that time.
[0160] This is based on the following idea: the concentration of leaked ammonia can be estimated from the operating data. If the leaked ammonia is converted into acidic ammonium sulfate (NH 4 HSO 4 ) is precipitated, since there is a correlation between the amount of acidic ammonium sulfate and the occlusion degree of the air preheater 60, the occlusion degree of the elements of the air preheater 60 can be estimated.
[0161] The differential pressure estimating unit 704 estimates the differential pressure between the input and output of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimating unit 702 and the blockage degree among the elements of the air preheater 60 estimated by the blockage degree estimating unit 703 .
[0162] (Structure of complete equipment)
[0163] The structure of the plant 1 according to the second embodiment will be described.
[0164] The plant 1 of the second embodiment is as follows Fig.11 As shown, the boiler 10 , the denitration device 20 , the sensor device 30 , and the air preheater 60 are provided.
[0165] The air preheater 60 is a device for preheating the combustion air to improve the combustion efficiency of the boiler. If ammonia leaks from the denitrification device 20, the leaked ammonia will react with sulfur trioxide (SO 3 ) reaction to generate acidic ammonium sulfate. The precipitation of acidic ammonium sulfate becomes a cause of clogging. In addition, in order to remove the acidic ammonium sulfate, it is necessary to stop the plant 1 and clean the air preheater 60 with water.
[0166] (Processing of Predicting Differential Pressure in Air Preheater 60)
[0167] Next, refer to Fig.12 , Fig.13 The process of predicting the differential pressure between the input and output of the air preheater 60 by the differential pressure prediction device 70 will be described.
[0168] The degradation prediction device 40 executes the processing shown in step S11 to step S15 to obtain catalyst degradation data.
[0169] The scaling factor estimation unit 701 obtains a scaling factor for correcting the difference between the performance characteristics of the catalyst obtained in the laboratory and the performance characteristics of the catalyst in the actual plant 1 (step S21 ).
[0170] The leakage estimation unit 702 estimates the amount of leaked ammonia based on an empirical formula obtained by applying a proportionality magnification factor to a formula representing the performance characteristics of the catalyst obtained in the laboratory, the NOx concentration at the inlet of the denitrification device 20, the NOx concentration at the outlet of the denitrification device 20, and catalyst degradation data (step S22).
[0171] The blockage degree estimating unit 703 estimates the blockage degree in the air preheater 60 using an empirical formula obtained based on differential pressure data when the differential pressure in the air preheater 60 in the actual plant 1 increases and the operation data of the plant 1 (step S23).
[0172] The differential pressure estimating unit 704 estimates the differential pressure between the input and output of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimating unit 702 and the blockage degree among the elements of the air preheater 60 estimated by the blockage degree estimating unit 703 (step S24 ).
[0173] Furthermore, the differential pressure estimating unit 704 can notify the estimated differential pressure to the person involved in the catalyst work and the person in charge of the plant 1. Based on this notification, the catalyst replacement work can be planned and the operation support of the plant 1 can be performed.
[0174] (Example)
[0175] In a certain plant, the differential pressure between the input and the output of the air preheater 60 is estimated by the differential pressure prediction device 70 according to the second embodiment of the present disclosure described above.
[0176] Fig.14 It is the comparison result of the predicted and measured differential pressure in the equipment.
[0177] exist Fig.14 In FIG. 1 , the horizontal axis is the operation time. In addition, the vertical axis is the differential pressure between the input and output of the air preheater 60. Fig.14 In FIG. 1 , the air preheater 60 is indicated as AH (Air Heater). In addition, the air preheater 60 is cleaned once with water.
[0178] As from Fig.14 As can be seen, according to the differential pressure prediction device 70, the prediction of the differential pressure between the input and output of the air preheater 60 can obtain a result close to the actual measured value.
[0179] The plant 1 according to the second embodiment of the present disclosure has been described above.
[0180] The differential pressure prediction device (70) includes a degradation prediction device (40), a proportional amplification factor estimation unit (701), a leakage estimation unit (702), an occlusion degree estimation unit (703), and a differential pressure estimation unit (704).
[0181] The proportional amplification factor estimation unit (701) obtains a proportional amplification factor for correcting the difference between the performance characteristics of the catalyst obtained in the laboratory and the performance characteristics of the catalyst in the actual plant (1). The leakage estimation unit (702) estimates the amount of leaked ammonia based on an empirical formula obtained by applying the proportional amplification factor to a formula representing the performance characteristics of the catalyst obtained in the laboratory, the NOx concentration at the inlet of the denitrification device (20), the NOx concentration at the outlet of the denitrification device (20), and catalyst degradation data. The occlusion degree estimation unit (703) estimates the occlusion degree in the air preheater (60) using an empirical formula obtained based on differential pressure data when the differential pressure of the air preheater (60) in the actual plant (1) increases and the operation data of the plant (1). A differential pressure estimating unit (704) estimates a differential pressure between an input and an output of the air preheater (60) based on the amount of leaked ammonia estimated by the leakage estimating unit (702) and the degree of blockage among the elements of the air preheater (60) estimated by the blockage degree estimating unit (703).
[0182] According to the differential pressure prediction device (70), when ammonia leakage occurs in the denitration device (20), the differential pressure between the input and output of the air preheater (60) is estimated. As a result, the catalyst replacement work can be planned and the operation support of the plant 1 can be carried out at an appropriate timing.
[0183] In addition, the order of the processes in the embodiments of the present disclosure may be changed within the scope of performing appropriate processes.
[0184] In addition, the storage unit 401 and other storage devices (including registers and latches) in each embodiment of the present invention can be provided anywhere within the scope of sending and receiving appropriate information. In addition, the storage unit 401 and other storage devices can exist in multiple locations within the scope of sending and receiving appropriate information to store data in a distributed manner.
[0185] In addition, in each embodiment of the present disclosure, it is described that the first model generation unit 404 generates the learned first model as software and stores it in the storage unit 401. In addition, in each embodiment of the present disclosure, it is described that the second model generation unit 405 generates the learned second model as software and stores it in the storage unit 401.
[0186] However, in other embodiments of the present disclosure, the learned first model and the learned second model may be realized as hardware, respectively.
[0187] For example, the first model generation unit 404 may write a configuration program that implements the processing performed by the learned first model stored in the storage unit 401 into programmable hardware such as an FPGA (Field-Programmable Gate Array).
[0188] Furthermore, for example, the second model generation unit 405 may write a configuration program that implements the processing performed by the learned second model stored in the storage unit 401 into programmable hardware such as an FPGA (Field-Programmable Gate Array).
[0189] The embodiments of the present disclosure have been described, but the degradation prediction device 40, the differential pressure prediction device 70, and other control devices may also have a computer device inside. Furthermore, the above-mentioned processing process is stored in a computer-readable recording medium in the form of a program, and the program is read and executed by a computer to perform the above-mentioned processing. A specific example of a computer is shown below.
[0190] Fig.10 This is a schematic block diagram showing the structure of a computer according to at least one embodiment.
[0191] Computer 5 Fig.10 As shown, the CPU 6 , the main memory 7 , the storage 8 , and the interface 9 are provided.
[0192] For example, the above-mentioned degradation prediction device 40, differential pressure prediction device 70, and other control devices are installed in the computer 5, respectively. Then, the operations of the above-mentioned processing units are stored in the form of programs in the storage 8. The CPU 6 reads the program from the storage 8 and expands it in the main memory 7, and executes the above-mentioned processing according to the program. In addition, the CPU 6 ensures the storage area corresponding to the above-mentioned storage units in the main memory 7 according to the program.
[0193] Examples of the storage device 8 include a HDD (Hard Disk Drive), an SSD (Solid State Drive), a magnetic disk, an optical magnetic disk, a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), and a semiconductor memory. The storage device 8 may be an internal medium directly connected to the bus of the computer 5, or an external medium connected to the computer 5 via an interface 9 or a communication line. In addition, when the program is published to the computer 5 via a communication line, the computer 5 that receives the publication may expand the program in the main memory 7 and execute the above-mentioned processing. In at least one embodiment, the storage device 8 is a non-temporary tangible storage medium.
[0194] The program may realize part of the aforementioned functions. Furthermore, the program may be a file that can realize the aforementioned functions by combining with a program that has already been recorded in a computer device, that is, a so-called differential file (differential program).
[0195] Although several embodiments of the present disclosure have been described, these embodiments are examples and do not limit the scope of the disclosure. These embodiments may be variously added, omitted, replaced, and modified within the scope of the disclosure.
[0196] <Notes>
[0197] The degradation prediction device 40 , the plant 1 , the degradation prediction method, and the program for causing a computer to execute the program and the configured processing described in each embodiment of the present disclosure can be understood as follows, for example.
[0198] (1) The degradation prediction device (40) involved in the first embodiment comprises: a model generation unit (404, 405) which generates a first prediction model for predicting the degree of degradation of the catalyst in the second set of equipment (1) different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and a degradation degree prediction unit (407) which predicts the degree of degradation in the second set of equipment (1) based on the first prediction model generated by the model generation unit (404, 405).
[0199] According to the degradation prediction device (40), the degradation degree of the catalyst in the second plant (1) can be predicted by simply inputting data about the second plant (1) into the first prediction model. As a result, the user of the second plant (1) can know the degradation degree of the catalyst without performing complex analysis on the catalyst.
[0200] (2) The degradation prediction device (40) involved in the second method is based on the degradation prediction device (40) of (1), wherein the learning data includes data related to the poisoned catalyst in the past operation of the first set of equipment and data related to the degree of degradation in the past operation of the first set of equipment, and the model generating unit (404, 405) comprises: a first model generating unit (404), which generates a second prediction model for predicting the data related to the poisoned catalyst in the second set of equipment (1) based on the learning data; a second model generating unit (405), which generates the first prediction model for predicting the degree of degradation in the second set of equipment (1) based on the data related to the poisoned catalyst in the second set of equipment (1) predicted by the second prediction model, and the degradation degree prediction unit (407) predicts the degree of degradation in the second set of equipment (1) based on the first prediction model and the second prediction model.
[0201] According to the degradation prediction device (40), data related to the catalyst in the device which is difficult to predict can be predicted for the second plant (1). As a result, the user of the second plant (1) can easily determine which data affects the degradation degree of the catalyst in the second plant (1).
[0202] (3) The differential pressure prediction device (70) involved in the third embodiment comprises: the degradation prediction device (40) involved in the first embodiment or the second embodiment; and a differential pressure estimation unit (704) which estimates the differential pressure between the input and output of the air preheater (60) based on the degradation degree predicted by the degradation prediction device (40).
[0203] According to the differential pressure prediction device (70), when ammonia leakage occurs in the denitration device (20), the differential pressure between the input and output of the air preheater (60) is estimated. As a result, the catalyst replacement work can be planned and the operation support of the plant 1 can be carried out at an appropriate timing.
[0204] (4) The differential pressure prediction device (70) involved in the fourth method is based on the differential pressure prediction device (70) of (3), and includes: an occlusion degree estimating unit (703), which estimates the occlusion degree in the air preheater (60) based on the degradation degree predicted by the degradation prediction device (40); and the differential pressure estimating unit (704) estimates the differential pressure between the input and output of the air preheater (60) based on the occlusion degree estimated by the occlusion degree estimating unit (703).
[0205] According to the differential pressure prediction device (70), the structure of the differential pressure prediction device (70) becomes clearer, and when ammonia leakage occurs in the denitration device (20), the differential pressure between the input and output of the air preheater (60) is estimated. As a result, the catalyst replacement project can be planned and the operation of the plant 1 can be supported at an appropriate time.
[0206] (5) The differential pressure prediction device (70) involved in the fifth method is a differential pressure prediction device (70) based on (3) or (4), and comprises: a leakage estimation unit (702) which estimates the amount of leakage ammonia based on the degree of degradation predicted by the degradation prediction device (40); and the differential pressure estimation unit (704) estimates the differential pressure between the input and output of the air preheater (60) based on the amount of leakage ammonia estimated by the leakage estimation unit (702).
[0207] According to the differential pressure prediction device (70), the structure of the differential pressure prediction device (70) becomes clearer, and when ammonia leakage occurs in the denitration device (20), the differential pressure between the input and output of the air preheater (60) is estimated. As a result, the catalyst replacement project can be planned and the operation of the plant 1 can be supported at an appropriate time.
[0208] (6) The differential pressure prediction device (70) involved in the fifth method is based on any one of the differential pressure prediction devices (70) from (3) to (5), and comprises: a proportional amplification factor estimating unit (701), which obtains the proportional amplification factor based on the degradation degree predicted by the degradation prediction device (40), and the differential pressure estimation unit (704) drives the differential pressure between the input and output of the air preheater (60) based on the proportional amplification factor obtained by the proportional amplification factor estimating unit (701).
[0209] According to the differential pressure prediction device (70), the structure of the differential pressure prediction device (70) becomes clearer, and when ammonia leakage occurs in the denitration device (20), the differential pressure between the input and output of the air preheater (60) is estimated. As a result, the catalyst replacement project can be planned and the operation of the plant 1 can be supported at an appropriate time.
[0210] (7) The plant (1) according to the seventh aspect is a plant using a catalyst, comprising: a device (20) for degrading the catalyst; and a degradation prediction device (40) according to (1) or (2) for predicting the degree of degradation of the catalyst.
[0211] According to the plant (1), the degree of deterioration of the catalyst in the plant (1) can be predicted by simply inputting data about the plant (1) into the first prediction model. As a result, the user of the plant (1) can know the degree of deterioration of the catalyst without performing complex analysis on the catalyst.
[0212] (8) The degradation prediction method involved in the eighth embodiment includes: generating a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and predicting the degree of degradation in the second set of equipment based on the generated first prediction model.
[0213] According to the degradation prediction method, the degradation degree of the catalyst in the second plant (1) can be predicted by simply inputting data about the second plant (1) into the first prediction model. As a result, the user of the second plant (1) can know the degradation degree of the catalyst without performing complex analysis on the catalyst.
[0214] (9) The program involved in the tenth method causes a computer to execute: generating a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and predicting the degree of degradation in the second set of equipment based on the generated first prediction model.
[0215] According to the program, the degree of deterioration of the catalyst in the second plant (1) can be predicted by simply inputting data about the second plant (1) into the first prediction model. As a result, the user of the second plant (1) can know the degree of deterioration of the catalyst without performing complex analysis on the catalyst.
[0216] (10) A program for causing a computer to execute the processing of the configuration involved in the eleventh embodiment, which is respectively constituted as the following hardware: a model generation unit (404, 405) which generates a first prediction model for predicting the degree of degradation of the catalyst in a second set of equipment (1) different from the first set of equipment based on learning data including first data related to the catalyst in the past operation of the first set of equipment and second data related to the state of the past operation; and a degradation degree prediction unit (407) which predicts the degree of degradation in the second set of equipment (1) based on the first prediction model generated by the model generation unit (404, 405).
[0217] By using a program for causing a computer to execute the processing of the configuration, the degree of degradation of the catalyst in the second equipment (1) can be predicted by simply inputting data about the second equipment (1) into the first prediction model. As a result, the user of the second equipment (1) can know the degree of degradation of the catalyst without performing complex analysis on the catalyst.
[0218] Industrial Applicability
[0219] According to the prediction device, the plant, the prediction method, the program, and the configuration program of the embodiment of the present disclosure, the user of the plant can obtain the degree of degradation of the catalyst without performing complicated processing.
[0220] Description of Reference Numerals
[0221] 1…Complete set of equipment
[0222] 5…Computer
[0223] 6…CPU
[0224] 7…Main memory
[0225] 8… Storage
[0226] 9…Interface
[0227] 10…Boiler
[0228] 20…Denitrification device
[0229] 30…Sensor device
[0230] 40…Degradation prediction device
[0231] 50…Data server device
[0232] 60…Air preheater
[0233] 70…Differential pressure prediction device
[0234] 301…1st sensor
[0235] 302…Second sensor
[0236] 303…3rd sensor
[0237] 304…4th sensor
[0238] 401... Storage
[0239] 402…Plan value acquisition department
[0240] 403…Equipment Data Acquisition Department
[0241] 404…1st model generation unit
[0242] 405…2nd model generation unit
[0243] 406…Data acquisition after poisoning
[0244] 407 ... Catalyst degradation degree acquisition unit
[0245] 701…Proportional amplification factor estimation unit
[0246] 702…Leakage estimation department
[0247] 703…Occlusion degree estimation unit
[0248] 704 ... differential pressure estimation unit.
Claims
1. A degradation prediction device, characterized in that: have: a model generating unit for generating a second prediction model for predicting a degree of degradation of a catalyst in a second plant different from the first plant based on learning data including first data related to a catalyst in past operation of the first plant and second data related to the state of the past operation; and a degradation degree prediction unit that predicts the degradation degree in the second plant based on the second prediction model generated by the model generation unit, The model generation unit comprises: a first model generating unit for generating a first prediction model for predicting data related to the poisoned catalyst in the second plant based on the learning data that takes the planned value of the catalyst and the plant data as input and outputs the poisoned catalyst data; and a second model generating unit for generating the second prediction model for predicting the degree of degradation in the second plant based on the learning data which takes the planned value, the plant data and the catalyst data after poisoning as input and outputs catalyst degradation data, The catalyst degradation data is data related to the degree of degradation in past operation of the first plant, The degradation degree prediction unit uses the first prediction model to predict data related to the poisoned catalyst in the second set of equipment based on data of the second set of equipment, and uses the second prediction model to predict the degradation degree in the second set of equipment based on the data related to the poisoned catalyst predicted by the first prediction model.
2. A differential pressure prediction device, characterized in that: have: The degradation prediction device according to claim 1; and A differential pressure estimating unit estimates a differential pressure between an input and an output of the air preheater based on the degree of degradation predicted by the degradation predicting device.
3. The differential pressure prediction device according to claim 2, characterized in that: The differential pressure prediction device comprises: a blockage degree estimating unit for estimating a blockage degree in the air preheater based on the degradation degree predicted by the degradation predicting device, The differential pressure estimating unit estimates a differential pressure between an input and an output of the air preheater based on the blockage degree estimated by the blockage degree estimating unit.
4. The differential pressure prediction device according to claim 2 or 3, characterized in that: The differential pressure prediction device comprises: a leakage estimating unit that estimates the amount of leaked ammonia based on the degree of degradation predicted by the degradation predicting device, The differential pressure estimating unit estimates a differential pressure between an input and an output of the air preheater based on the amount of the leaked ammonia estimated by the leakage estimating unit.
5. The differential pressure prediction device according to claim 2 or 3, characterized in that: The differential pressure prediction device comprises: a scaling factor estimating unit for obtaining a scaling factor based on the degree of degradation predicted by the degradation predicting device, The differential pressure estimating unit estimates the differential pressure between the input and output of the air preheater based on the proportional amplification factor obtained by the proportional amplification factor estimating unit.
6. A complete set of equipment using a catalyst, characterized in that the complete set of equipment comprises: means for causing degradation of said catalyst; and The degradation prediction device according to claim 1 is configured to predict the degree of degradation of the catalyst.
7. A degradation prediction method, characterized in that: Include: generating a second prediction model for predicting the degree of degradation of the catalyst in a second plant different from the first plant based on learning data including first data related to the catalyst in past operation of the first plant and second data related to the state of the past operation; and predicting the degree of degradation in the second plant based on the generated second prediction model, The degradation prediction method further comprises: generating a first prediction model for predicting data related to the poisoned catalyst in the second complete set of equipment based on the planned value of the catalyst and the complete set of equipment data as input and the data of the catalyst after poisoning as output, and generating a second prediction model for predicting the degree of degradation in the second complete set of equipment based on the learning data that takes the planned value, the complete set of equipment data and the data of the catalyst after poisoning as input and the data of catalyst degradation as output, The catalyst degradation data is data related to the degree of degradation in past operation of the first plant, In the prediction of the degree of degradation, the first prediction model is used to predict the data related to the poisoned catalyst in the second set of equipment based on the data of the second set of equipment, and the data related to the poisoned catalyst predicted by the first prediction model is input into the second prediction model to predict the degree of degradation.
8. A differential pressure prediction method, characterized in that: Include: generating a second prediction model for predicting a degree of degradation of a catalyst in a second plant different from the first plant based on learning data including first data related to a catalyst in past operation of the first plant and second data related to the state of the past operation; predicting the degree of degradation in the second plant based on the generated second prediction model; and Based on the predicted degree of degradation, the differential pressure between the input and output of the air preheater is estimated. The differential pressure prediction method further comprises: generating a first prediction model for predicting data related to the poisoned catalyst in the second complete set of equipment based on the planned value of the catalyst and the complete set of equipment data as input and the data of the catalyst after poisoning as output, and generating a second prediction model for predicting the degree of degradation in the second complete set of equipment based on the learning data that takes the planned value, the complete set of equipment data and the data of the catalyst after poisoning as input and the data of catalyst degradation as output, The catalyst degradation data is data related to the degree of degradation in past operation of the first plant, In the prediction of the degree of degradation, the first prediction model is used to predict the data related to the poisoned catalyst in the second set of equipment based on the data of the second set of equipment, and the data related to the poisoned catalyst predicted by the first prediction model is input into the second prediction model to predict the degree of degradation.
9. A computer-readable storage medium storing a program, characterized in that: Programs cause computers to: generating a second prediction model for predicting the degree of degradation of the catalyst in a second plant different from the first plant based on learning data including first data related to the catalyst in past operation of the first plant and second data related to the state of the past operation; and predicting the degree of degradation in the second plant based on the generated second prediction model, The program also causes the computer to execute: generating a first prediction model for predicting data related to the poisoned catalyst in the second complete set of equipment based on the planned value of the catalyst and the complete set of equipment data as input and the data of the catalyst after poisoning as output, and generating a second prediction model for predicting the degree of degradation in the second complete set of equipment based on the learning data that takes the planned value, the complete set of equipment data and the data of the catalyst after poisoning as input and the data of catalyst degradation as output, The catalyst degradation data is data related to the degree of degradation in past operation of the first plant, In the prediction of the degree of degradation, the first prediction model is used to predict the data related to the poisoned catalyst in the second set of equipment based on the data of the second set of equipment, and the data related to the poisoned catalyst predicted by the first prediction model is input into the second prediction model to predict the degree of degradation.
10. A computer-readable storage medium storing a program, characterized in that: The program is used to cause the computer to execute the configured processing, and the following elements are respectively configured as hardware: a model generating unit for generating a second prediction model for predicting a degree of degradation of a catalyst in a second plant different from the first plant based on learning data including first data related to a catalyst in past operation of the first plant and second data related to the state of the past operation; and a degradation degree prediction unit that predicts the degradation degree in the second plant based on the second prediction model generated by the model generation unit, The model generation unit comprises: a first model generating unit for generating a first prediction model for predicting data related to the poisoned catalyst in the second plant based on the learning data that takes the planned value of the catalyst and the plant data as input and outputs the poisoned catalyst data; and a second model generating unit for generating the second prediction model for predicting the degree of degradation in the second plant based on the learning data which takes the planned value, the plant data and the catalyst data after poisoning as input and outputs catalyst degradation data, The catalyst degradation data is data related to the degree of degradation in past operation of the first plant, The degradation degree prediction unit uses the first prediction model to predict data related to the poisoned catalyst in the second set of equipment based on data of the second set of equipment, and uses the second prediction model to predict the degradation degree in the second set of equipment based on the data related to the poisoned catalyst predicted by the first prediction model.
