WO2021157240A1 - Prediction device, plant, prediction method, program, and configuration program - Google Patents

Abstract

A deterioration prediction device of the present invention comprises: a model generation unit that generates, on the basis of learning data including first data related to a catalyst in a past operation of a first plant and second data related to the status of the past operation, a first prediction model for predicting a degree of deterioration of the catalyst in a second plant which is different from the first plant; and a deterioration degree prediction unit that predicts, on the basis of the first prediction model generated by the model generation unit, the degree of deterioration in the second plant.

Description

Predictor, plant, forecast method, program, and configuration program

The present disclosure relates to predictors, plants, prediction methods, programs, and configuration programs.
The present application claims priority over 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 are used here.

In plants equipped with boilers and the like, catalysts are used. In plants where catalysts are used, the catalysts deteriorate as the plant is operated. Therefore, in a plant where a catalyst is used, it is desirable to grasp the degree of deterioration of the catalyst.
Patent Document 1 discloses a technique for predicting the degree of deterioration of a catalyst as a related technique.

Japanese Patent No. 6278296

By the way, in order to predict the degree of deterioration of the catalyst in the plant, it is generally necessary to perform a complicated process such as analyzing the properties of the exhaust gas every time the user of the plant makes a prediction.
Therefore, in a plant, there is a demand for a technique that allows a user to know the degree of deterioration of a catalyst without performing complicated processing.

The present disclosure aims to provide a prediction device, a plant, a prediction method, a program, and a configuration program capable of solving the above problems.

In order to solve the above problems, the deterioration prediction device according to the present disclosure includes learning data including first data related to a catalyst in the past operation of the first plant and second data related to the past operation state. Based on the model generation unit that generates the first prediction model that predicts the deterioration degree of the catalyst in the second plant different from the first plant, and the first prediction model generated by the model generation unit, the first It is provided with a deterioration degree prediction unit for predicting the deterioration degree in the two plants.

The differential pressure prediction device according to the present disclosure includes the deterioration prediction device, a differential pressure estimation unit that estimates the differential pressure between the input and output of the air preheater based on the deterioration degree predicted by the deterioration prediction device, and the differential pressure estimation unit. To be equipped.

The plant according to the present disclosure is a plant in which a catalyst is used, and includes an apparatus for causing deterioration of the catalyst and the above-mentioned deterioration prediction apparatus for predicting the degree of deterioration of the catalyst.

The deterioration prediction method according to the present disclosure is based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state, and the first plant. It includes generating a first prediction model for predicting the degree of deterioration of a catalyst in a different second plant, and predicting the degree of deterioration in the second plant based on the generated first prediction model.

The differential pressure prediction method according to the present disclosure is based on learning data including first data relating to a catalyst in the past operation of the first plant and second data relating to the past operating state of the first plant. It was predicted that the first prediction model for predicting the deterioration degree of the catalyst in the second plant different from the above was generated, and that the deterioration degree in the second plant was predicted based on the generated first prediction model. It includes estimating the differential pressure between the input and output of the air preheater based on the degree of deterioration.

The program according to the present disclosure is based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state in the computer, and the first plant. To generate a first prediction model for predicting the degree of deterioration of the catalyst in the second plant different from the above, and to predict the degree of deterioration in the second plant based on the generated first prediction model. Let me.

The program for causing the computer to execute the configuration process according to the present disclosure is training data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. Based on the model generation unit that generates the first prediction model that predicts the deterioration degree of the catalyst in the second plant different from the first plant, and the first prediction model generated by the model generation unit. Each of the deterioration degree prediction units for predicting the deterioration degree in the two plants is configured as hardware.

According to the prediction device, the plant, the prediction method, the program, and the configuration program according to the embodiment of the present disclosure, the user of the plant can know the degree of deterioration of the catalyst without performing complicated processing.

It is a figure which shows an example of the structure of the plant by 1st Embodiment of this disclosure. It is a figure which shows an example of the structure of the sensor device by 1st Embodiment of this disclosure. It is a figure which shows an example of the structure of the deterioration prediction apparatus by 1st Embodiment of this disclosure. It is a figure which shows an example of the past performance data in various plants in 1st Embodiment of this disclosure. It is a figure for demonstrating the generation of the 1st model and the 2nd model in 1st Embodiment of this disclosure. It is a figure which shows an example of the input of the data from the 1st model and the 2nd model in the 1st Embodiment of this disclosure. It is the first figure which shows the processing flow of the deterioration prediction apparatus by 1st Embodiment of this disclosure. It is the 2nd figure which shows the processing flow of the deterioration prediction apparatus by 1st Embodiment of this disclosure. It is a figure which shows an example of the comparative result of the Example in 1st Embodiment of this disclosure. It is a figure which shows an example of the structure of the differential pressure prediction apparatus by 2nd Embodiment of this disclosure. It is a figure which shows an example of the structure of the plant by 2nd Embodiment of this disclosure. It is a figure which shows the processing flow of the differential pressure prediction apparatus by 2nd Embodiment of this disclosure. It is a figure for demonstrating the processing of the differential pressure prediction apparatus by the 2nd Embodiment of this disclosure. It is a figure which shows an example of the comparative result of the Example in the 2nd Embodiment of this disclosure. It is a schematic block diagram which shows the structure of the computer which concerns on at least one Embodiment.

<First Embodiment>
Hereinafter, embodiments will be described in detail with reference to the drawings.
The deterioration prediction system according to the first embodiment of the present disclosure predicts the deterioration state of the denitration device 20 included in the plant 1 based on the operation data of the plant 1 to be predicted. The deterioration prediction system includes a data server device 50 and a deterioration prediction device 40. The data server device 50 is a device that stores the data of the plant 1. The data of the plant 1 is data including, for example, fuel properties, operation data, and the like. The deterioration prediction device 40 predicts the deterioration state of the denitration device 20 included in the plant 1 based on the data stored in the data server device 50.

(Plant configuration)
The configuration of the plant 1 (an example of the second plant, an example of the plant) according to the first embodiment will be described.
The plant 1 according to the first embodiment is a plant that predicts the degree of deterioration of the catalyst performance with respect to the total operating time of the plant 1.
As shown in FIG. 1, the plant 1 includes a boiler 10, a denitration device 20 (an example of a device in which catalyst deterioration occurs), and a sensor device 30.

Boiler 10 is a boiler that uses coal or the like as fuel. The boiler 10 discharges the combustion gas when the fuel is burned to the denitration device 20.

The denitration device 20 is a device that reduces the NOx concentration in the combustion gas by decomposing NOx 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.
For example, the denitration device 20 keeps the denitration rate, which tends to decrease due to deterioration of the catalyst performance due to the poisonous component _{, constant by injecting ammonia (NH 3) at the inlet of the denitration device 20.} The poisonous component is a component that comes in from the outside, which is different from the main component of the catalyst.
Toxic components enter the denitration device 20 from the outside. Therefore, in the denitration device 20, the active sites decrease as the poisoning component increases. As a result, the value of the reaction rate constant K, which indicates the high performance (goodness) of the catalyst, decreases. That is, the performance of the catalyst deteriorates.
The plant 1 according to the first embodiment is a target for predicting the degree of deterioration of the performance of this catalyst.

As shown in FIG. 2, the sensor device 30 includes a first sensor 301, a second sensor 302, a third sensor 303, and a fourth sensor 304. The sensor device 30 is provided at the entrance of the denitration device 20.
The first sensor 301 detects the dust concentration at the inlet of the denitration device 20.
The second sensor 302 detects the NOx concentration at the inlet of the denitration device 20.
The third sensor 303 detects the SOx (socks) concentration at the inlet of the denitration device 20.
The fourth sensor 304 detects the O2 (oxygen) concentration at the inlet of the denitration device 20.
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.

(Configuration of deterioration prediction device)
The deterioration prediction device 40 is a device that predicts the degree of deterioration of the catalyst performance with respect to the total operating time of the plant 1.
As shown in FIG. 3, the deterioration prediction device 40 includes a storage unit 401, a planned value acquisition unit 402, a plant data acquisition unit 403, a first model generation unit 404 (an example of a model generation unit), and a second model generation unit 405 (an example of the model generation unit). A model generation unit), a post-poisoning data acquisition unit 406, and a catalyst deterioration degree acquisition unit 407 (an example of a deterioration degree prediction unit) are provided.

The storage unit 401 stores various information necessary for the processing of the deterioration prediction device 40.
For example, the storage unit 401 stores past actual data D1 in various plants (an example of the first plant).
Specifically, the storage unit 401 stores, for example, the planned value shown in FIG. 4, the plant data, and the catalyst data after poisoning.
The storage unit 401 stores the data that changes with the passage of time in association with the time.

As shown in FIG. 4, the planned values include catalyst performance, catalyst specifications, and device specifications.
Examples of catalyst performance are the initial value K0 of the reaction rate constant, the ratio of the reaction rate constant K to the initial value K0 of the reaction rate constant after a certain operating time has elapsed (K / K0) (degree of deterioration of the reaction rate constant). That is, the degree of deterioration of the catalyst) and the like. The initial value K0 of the reaction rate constant is a value given in advance.

Examples of catalyst specifications include the initial specific surface area of the catalyst, the initial pore volume of the catalyst, and the initial composition of the catalyst (for example, TIO2 (titania), WO3 (tungsten trioxide), V2O5 (vanadium pentoxide)). , SiO2 (silica), etc.) and the like. The catalyst specifications can be obtained, for example, by sampling and analyzing an actual catalyst.

Examples of device specifications include the flow velocity, the dust concentration at the inlet of the denitration device 20, the NOx (knox) concentration at the inlet of the denitration device 20, the SOx (socks) concentration at the inlet of the denitration device 20, and the inlet of the denitration device 20. Examples include O2 (oxygen) concentration.
The flow velocity 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 O2 concentration at the inlet of the denitration device 20 are sensors provided at the inlet of the denitration device 20. It is a value measured by the device 30. 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 O2 concentration at the inlet of the denitration device 20 are calculated from the fuel properties. May be good. The fuel properties here are data indicating the amount of each component in the ash flowing into the catalyst, and are data possessed by analyzing the fuel for each plant.

Further, as shown in FIG. 4, the plant data includes fuel properties and operation data.
Examples of fuel properties include the amount of inflow of components in fly ash such as SiO2 and Al2O3 into the catalyst.
Further, as an example of the operation data, the operation time of the plant can be mentioned.

Further, as shown in FIG. 4, the catalyst data after poisoning includes the catalyst physical characteristics after poisoning and the catalyst composition after poisoning.
Examples of the catalyst physical characteristics after poisoning include the specific surface area of the catalyst after poisoning, the pore volume of the catalyst after poisoning, and the like.
As an example of the catalyst composition after poisoning, the ratio of TIO2 in the catalyst after poisoning is normalized by the ratio of WO3 (TiO2 / WO3), and the ratio of SiO2 in the catalyst after poisoning is the ratio of WO3. Examples thereof include values normalized by (SiO2 / WO3).
The catalyst data after poisoning is the data of the catalyst inside the denitration device 20, and is difficult to acquire during the operation of the plant. Therefore, the catalyst data after poisoning will be acquired at the time of maintenance and inspection of the plant 1.

Specifically, the storage unit 401 stores the required degree of catalyst deterioration (that is, ratio (K / K0)) for past plant operations.

Further, for example, the storage unit 401 stores data used when predicting the degree of deterioration of the catalyst in the plant 1 using the model.
Specifically, the storage unit 401 includes a first model (an example of a second prediction model) generated using the past actual data D1 and a second model (first) generated using the past actual data D1. An example of a prediction model), the catalyst performance of the plant 1, the catalyst specifications of the plant 1, the equipment specifications of the plant 1, and the fuel properties of the plant 1 are stored. The first model is a model for predicting catalyst data after poisoning in plant 1. The second model is a model that predicts the degree of deterioration of the catalyst from the catalyst data after poisoning predicted by the first model. Each of the first model and the second model is a trained model trained using past performance data in various plants.
An example of the catalytic performance of the plant 1 is the initial value K0 of the reaction rate constant of the plant 1.
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.
Further, as an example of the device specifications of the plant 1, the flow velocity in the plant 1 and the like can be mentioned.
Further, as an example of the fuel property of the plant 1, the inflow amount of the component in the ash into the catalyst in the plant 1 (for example, the inflow amount into the catalyst such as TiO2 and SiO2) can be mentioned.
The generation of the first model and the generation of the second model will be described later.

The planned value acquisition unit 402 acquires the planned value of the plant 1.
For example, the planned value acquisition unit 402 reads out 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.
Specifically, the planned value acquisition unit 402 stores 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 velocity, and the like. Read from 401.

Further, for example, the planned value acquisition unit 402 includes the dust concentration detected by the sensor device 30, the NOx (knox) concentration at the inlet of the denitration device 20, the SOx (sox) concentration at the inlet of the denitration device 20, and the inlet of the denitration device 20. The detected values of each O2 (oxygen) concentration are acquired from the sensor device 30.
The detected value acquired by the planned value acquisition unit 402 may be stored in the storage unit 401, and the stored detected value may be used as actual data when determining the degree of deterioration of the catalyst in another plant. ..

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.

(Generation of the first model)
The first model generation unit 404 generates the 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 will be described using a neural network as an example.

The first model generation unit 404 has already learned the modeled neural network by machine learning the past actual data (planned value, plant data, catalyst data after poisoning) in various plants as learning data. Generate the first model.
The neural network is, for example, a folding neural network having an input layer, an intermediate layer, and an output layer. Further, the learning data here is data in which the planned value, the plant data, and the catalyst data after poisoning are associated with each other on a one-to-one basis.

Specifically, the first model generation unit 404 divides a plurality of learning data (planned value, plant data, catalyst data after poisoning) into training data, evaluation data, and test data. As shown in FIG. 5, the first model generation unit 404 inputs the planned value of the training data and the plant data to the neural network. The neural network (first model before learning in FIG. 5) outputs catalyst data after poisoning. The first model generation unit 404 performs backpropagation according to the output of the planned value of the training data and the plant data every time the planned value and the plant data are input to the neural network and the catalyst data after poisoning is output from the neural network. Changes the weighting of data joins between nodes (ie, changes the model of the neural network). Next, the first model generation unit 404 inputs the planned value of the evaluation data and the plant data into the neural network of the model changed by the planned value of the training data and the plant data. The neural network outputs the catalyst data after poisoning according to the input planned value and plant data. The first model generation unit 404 changes the weighting 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 generation unit 404 in this way is the trained first model. Next, as a final confirmation, the first model generation unit 404 inputs the planned value of the test data and the plant data into the trained neural network of the first model. The trained neural network of the first model outputs the catalyst data after poisoning according to the planned value of the input test data and the plant data. For all test data, the post-poisoning catalyst data output by the trained first model neural network is the post-poisoning catalyst data associated with the input test data planned values and plant data. If it is within a predetermined error range, it is determined that the trained neural network of the first model is the desired model. In addition, even in one of the test data, the post-poisoning catalyst data output by the trained neural network of the first model is associated with the input test data planned value and plant data after poisoning. If it is not within a predetermined error range with respect to the catalyst data of the above, a trained first model is generated using the new training data.
The generation of the trained model by the first model generation unit 404 is repeated until a desired trained first model is obtained.
The first model generation unit 404 writes the generated learned first model to the storage unit 401.

(Generation of the second model)
The second model generation unit 405 generates the second model.
For example, the second model generation unit 405 machine-learns the modeled neural network using past actual data (planned value, plant data, catalyst data after poisoning, catalyst deterioration data) in various plants as learning data. By doing so, a trained second model is generated.
The neural network is, for example, a folding neural network having an input layer, an intermediate layer, and an output layer. Further, the learning data here is data in which the planned value, the plant data, the catalyst data after poisoning, and the catalyst deterioration data are associated with each other on a one-to-one basis.

Specifically, the second model generation unit 405 divides a plurality of training data (planned value, plant data, catalyst data after poisoning, catalyst deterioration data) into training data, evaluation data, and test data. .. As shown in FIG. 5, the second model generation unit 405 inputs the planned value of the training data, the plant data, and the catalyst data after poisoning to the neural network. The neural network outputs catalyst deterioration data. Then, as in the case where the first model generation unit 404 generates the first model, the second model generation unit 405 uses the catalyst deterioration output by the trained neural network of the second model for all the test data. A trained second model neural network if the data is within a predetermined error with respect to the input planned values, plant data, and catalyst degradation data associated with post-poisoning catalyst data. Is determined to be the desired model. In addition, even in one of the test data, the catalyst deterioration data output by the trained neural network of the second model is associated with the planned value of the input test data, the plant data, and the catalyst data after poisoning. If the catalyst deterioration data is not within a predetermined error range, a trained second model is generated using the new training data.
The generation of the trained model by the second model generation unit 405 is repeated until a desired trained second model is obtained.
The second model generation unit 405 writes the generated and learned second model to the storage unit 401.

The post-poisoning data acquisition unit 406 acquires the post-poisoning catalyst data based on the planned value, the plant data, and the trained first model.

For example, the post-poisoning data acquisition unit 406 uses the catalyst performance of the plant 1 (for example, the initial value K0 of the reaction rate constant of the plant 1), the catalyst specifications of the plant 1 (for example, the initial specific surface area of the catalyst in the plant 1, the plant). Initial pore volume of catalyst in 1), equipment specifications of plant 1 (for example, flow velocity in plant 1), fuel properties of plant 1 (for example, amount of inflow of components in ash into catalyst in plant 1), learning The completed first model is read from the storage unit 401.
Further, the post-poisoning data acquisition unit 406 may use 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 inlet of the denitration device 20 during the operation of the plant 1. Information indicating each of the O2 concentrations in the above is acquired from the sensor device 30.
As shown in FIG. 6, the post-poisoning data acquisition unit 406 reads from 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, and the sensor device 30 read from the storage unit 401. Information indicating each of the acquired dust concentration at the inlet of the denitration device 20, NOx concentration at the inlet of the denitration device 20, SOx concentration at the inlet of the denitration device 20, O2 concentration at the inlet of the denitration device 20, and information indicating the operation time. Is input to the trained first model.
Then, the post-poisoning data acquisition unit 406 acquires the post-poisoning catalyst data (for example, the catalyst physical characteristics after poisoning and the catalyst composition after poisoning) output by the trained first model.
This post-poisoning catalyst data acquired by the post-poisoning data acquisition unit 406 is the poisoning obtained by the prediction when 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. Later catalyst data.

The catalyst deterioration degree acquisition unit 407 acquires catalyst deterioration data based on the planned value, plant data, catalyst data after poisoning, and the learned second model.

For example, the catalyst deterioration degree acquisition unit 407 includes catalyst specifications of plant 1 (for example, initial specific surface area of catalyst in plant 1, initial pore volume of catalyst in plant 1, etc.), equipment specifications of plant 1 (for example, plant). The second model that has been learned (such as the flow velocity in 1) is read out from the storage unit 401.
Further, during the operation of the plant 1, the catalyst deterioration degree acquisition unit 407 includes a dust concentration at the inlet of the denitration device 20, a NOx concentration at the inlet of the denitration device 20, an SOx concentration at the inlet of the denitration device 20, and a denitration device 20 at the inlet. Information indicating each of the O2 concentrations is acquired from the sensor device 30.
Further, the catalyst deterioration degree acquisition unit 407 acquires the post-poisoning catalyst data predicted by the post-poisoning data acquisition unit 406 from the post-poisoning data acquisition unit 406.
As shown in FIG. 6, the catalyst deterioration degree acquisition unit 407 includes the catalyst specifications of the plant 1, the device specifications of the plant 1, the fuel properties of the plant 1, and the inlet of the denitration device 20 acquired from the sensor device 30 as shown in FIG. Information indicating each of the dust concentration in 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, the O2 concentration at the inlet of the denitration device 20, and the post-poisoning data acquisition unit 406 predicted after detoxification. The catalyst data and the information indicating the operation time are input to the trained second model.
Then, the catalyst deterioration degree acquisition unit 407 acquires the catalyst deterioration data output by the trained second model.
The catalyst deterioration data acquired by the catalyst deterioration degree acquisition unit 407 is the catalyst deterioration data obtained by the prediction when 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.

(Model learning process)
Hereinafter, the operation of the deterioration prediction device 40 will be described. First, the learning process of the machine learning model used by the deterioration prediction device 40 to predict the deterioration of the catalyst will be described.
The manager or the like of the deterioration prediction device 40 acquires the catalyst deterioration data of the catalyst of the denitration device 20 included in the plant at the time of maintenance and inspection of a plant different from the plant 1 to be predicted in advance. The catalyst deterioration data is obtained by measuring the NOx concentration of gas at the input end and the NOx concentration of gas at the output end of the denitration device 20. In addition, the manager of the deterioration prediction device 40 acquires post-poisoning catalyst data including catalyst physical characteristics and catalyst composition by destructive inspection at the time of maintenance and inspection of a plant different from the prediction target plant 1 in advance. Keep it. The catalyst data after poisoning is the data of the catalyst inside the denitration device 20, and is difficult to obtain during the operation of the plant.

The first model generation unit 404 causes the first model, which is a machine learning model, to perform machine learning using past actual data (planned value, plant data, catalyst data after poisoning) in various plants as learning data. The planned value and the plant data are acquired from the data server device 50.
Specifically, the first model generation unit 404 generates training data using the planned value and the plant data as input samples and the catalyst data after poisoning as the output sample among the actual data. The first model generation unit 404 trains the first model so as to output the predicted value of the state of the catalyst after poisoning by inputting the planned value and the plant data based on the training data.
The first model generation unit 404 writes the learned first model to the storage unit 401.

The second model generation unit 405 uses the past actual data (planned value, plant data, catalyst data after poisoning, catalyst deterioration data) in various plants as learning data for the second model, which is a machine learning model. Let them learn.
Specifically, the second model generation unit 405 generates training data in which the planned value, the plant data, and the catalyst data after poisoning are input samples, and the catalyst deterioration data is an output sample, among the actual data. The second model generation unit 405 learns the second model so as to output the predicted value of the catalyst deterioration data by inputting the planned value, the plant data and the catalyst data after poisoning based on the training data. Let me.
The second model generation unit 405 writes the learned second model to the storage unit 401.

(Catalyst deterioration prediction processing)
Next, a process in which the deterioration prediction device 40 predicts the deterioration of the catalyst will be described using the processing flow of the deterioration prediction device 40 shown in FIGS. 7 and 8. The deterioration prediction device 40 predicts the catalyst data after poisoning based on the planned value and the plant data, and predicts the catalyst deterioration data using the predicted catalyst data after poisoning.

(Processing to predict catalyst data after poisoning)
First, a process in which the deterioration prediction device 40 predicts the catalyst data after poisoning based on the planned value, the plant data, and the learned first model will be described with reference to FIG. 7.

The post-poisoning data acquisition unit 406 uses the catalyst performance of plant 1 (for example, the initial value K0 of the reaction rate constant of plant 1), the catalyst specifications of plant 1 (for example, the initial specific surface area of the catalyst in plant 1, and the initial specific surface area of the catalyst in plant 1). The initial pore volume of the catalyst, etc.), the equipment specifications of the plant 1 (for example, the flow velocity in the plant 1), the fuel properties of the plant 1 (for example, the inflow amount of the components in the ash into the catalyst in the plant 1), and learned. The first model is read from the storage unit 401 (step S1).

During the operation of the plant 1, the post-poisoning data acquisition unit 406 determines 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 O2 at the inlet of the denitration device 20. Information indicating each of the concentrations is acquired from the sensor device 30 (step S2).

The post-poisoning data acquisition unit 406 uses the catalyst performance of the plant 1 read from the storage unit 401, the catalyst specifications of the plant 1, the device specifications of the plant 1, the fuel properties of the plant 1, and the inlet of the denitration device 20 acquired from the sensor device 30. Information indicating each of the dust concentration in 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, the O2 concentration at the inlet of the denitration device 20, and the information indicating the operation time have been learned. Input to the model (step S3).

The post-poisoning data acquisition unit 406 acquires the post-poisoning catalyst data (for example, the catalyst physical characteristics after poisoning and the catalyst composition after poisoning) output by the trained first model (step S4).

(Processing for predicting catalyst deterioration data)
Next, with reference to FIG. 8, the deterioration prediction device 40 predicts the catalyst deterioration data based on the planned value, the plant data, the catalyst data after poisoning, and the learned second model. Will be explained.

The catalyst deterioration degree acquisition unit 407 includes catalyst specifications of plant 1 (for example, initial specific surface area of catalyst in plant 1, initial pore volume of catalyst in plant 1, etc.), equipment specifications of plant 1 (for example, in plant 1). (Flow velocity, etc.), the learned second model is read out from the storage unit 401 (step S11).

During the operation of the plant 1, the catalyst deterioration degree acquisition unit 407 includes dust concentration at the inlet of the denitration device 20, NOx concentration at the inlet of the denitration device 20, SOx concentration at the inlet of the denitration device 20, and O2 concentration at the inlet of the denitration device 20. Information indicating each of the above is acquired from the sensor device 30 (step S12).

The catalyst deterioration degree acquisition unit 407 acquires the post-poisoning catalyst data predicted by the post-poisoning data acquisition unit 406 from the post-poisoning data acquisition unit 406 (step S13).

As shown in FIG. 6, the catalyst deterioration degree acquisition unit 407 includes the catalyst specifications of the plant 1, the device specifications of the plant 1, the fuel properties of the plant 1, and the inlet of the denitration device 20 acquired from the sensor device 30 as shown in FIG. Information indicating each of the dust concentration in 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, the O2 concentration at the inlet of the denitration device 20, and the post-poisoning data acquisition unit 406 predicted after detoxification. The catalyst data and the information indicating the operation time are input to the trained second model (step S14).

The catalyst deterioration degree acquisition unit 407 acquires the catalyst deterioration data output by the trained second model (step S15).

(Example)
For a certain plant, the deterioration degree of the catalyst was predicted by the deterioration prediction device 40 according to the first embodiment of the present disclosure described above, and compared with the actual measurement of the deterioration degree of the catalyst in the plant.
127 data were prepared for 11 plants with different manufacturing years and operating hours, 70% of which was used as learning data and 30% was used as data for accuracy verification.
FIG. 9 shows a comparison result between the prediction of the degree of deterioration of the catalyst and the actual measurement of the degree of deterioration of the catalyst.
In FIG. 9, the horizontal axis is the operating time. The vertical axis is the degree of deterioration of the catalyst. The degree of deterioration of the catalyst is a value obtained by dividing the reaction rate constant K at each operating time by the initial reaction rate constant K0.
The accuracy of predicting the degree of deterioration of the catalyst was RMSE (Root Mean Squared Error) 0.05.

The plant 1 according to the first embodiment of the present disclosure has been described above.
The deterioration prediction device (40) includes a model generation unit (404, 405) and a deterioration degree prediction unit (407).
The model generation unit is a second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. A first prediction model for predicting the degree of deterioration of the catalyst in (1) is generated.
The deterioration degree prediction unit (407) predicts the deterioration degree in the second plant (1) based on the first prediction model generated by the model generation unit (404, 405).

With this deterioration prediction device (40), the degree of deterioration of the catalyst in the second plant (1) can be predicted only by inputting the 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 a complicated analysis on the catalyst.

<Second Embodiment>
The deterioration prediction system according to the second embodiment of the present disclosure predicts the deterioration state of the denitration device 20 included in the plant 1 based on the operation data of the plant 1 to be predicted, and air preheating described later from the predicted deterioration state. Predict the differential pressure between the input and output of the device 60. The deterioration prediction system includes a data server device 50 and a differential pressure prediction device 70.

(Configuration of differential pressure prediction device)
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 deterioration of the catalyst performance.
As shown in FIG. 10, the differential pressure prediction device 70 includes a scale-up factor estimation unit 701, a leak estimation unit 702, a blockage degree estimation unit 703, and a differential pressure estimation unit 704 in addition to the deterioration prediction device 40.

The scale-up factor estimation unit 701 obtains a correction coefficient 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. This correction coefficient is the scale-up factor.
For example, we have acquired the performance characteristics of catalysts in the laboratory for various combinations of gas temperature, gas concentration, and oxygen concentration. Then, the performance characteristics of the acquired catalyst are expressed by an equation including the gas temperature, the gas concentration, and the oxygen concentration. However, in the actual plant 1, the performance of the catalyst deteriorates with respect to the performance characteristics of the catalyst acquired in the laboratory. The scale-up factor estimation unit 701 determines a scale-up factor that corrects the equation that indicates the performance characteristics of the catalyst acquired in the laboratory so that the equation can be applied to the actual plant 1. The scale-up factor is determined by an empirical rule obtained from the performance characteristics of the catalyst acquired in the laboratory and the performance characteristics of the catalyst in the actual plant 1.

The leak estimation unit 702 uses an empirical formula obtained by applying a scale-up factor to an equation showing the performance characteristics of the catalyst acquired in the laboratory, a NOx concentration at the inlet of the denitration device 20, a NOx concentration at the outlet of the denitration device 20, and catalyst deterioration data. Based on this, the amount of leaked ammonia is estimated. In this method, the performance characteristics of the catalyst are obtained from an empirical formula, and the amount of unreacted ammonia is estimated from the performance characteristics and the NOx concentration at the input / output of the denitration device 20.

The blockage estimation unit 703 uses an empirical formula obtained based on the differential pressure data when the differential pressure of the air preheater 60 in the actual plant 1 is rising and the operation data of the plant 1, and the air. The degree of obstruction in the preheater 60 is estimated.
For example, when the differential pressure of the air preheater 60 in the actual plant 1 is increasing, the differential pressure between the input and output in the element of the air preheater 60 is monitored and the differential pressure is calculated. The closing speed is calculated from the differential pressure, and an empirical formula is obtained using the operation data of the plant 1 at that time.
This can be estimated from the operation data, and if this leaked ammonia is deposited on the air preheater 60 as acidic ammonium sulfate (NH4HSO4), the amount of acidic ammonium sulfate and the blockage of the air preheater 60 It is based on the idea that the degree of blockage in the element of the air preheater 60 can be estimated because there is a correlation with the degree.

The differential pressure estimation unit 704 is based on the amount of leak ammonia estimated by the leak estimation unit 702 and the blockage degree in the element of the air preheater 60 estimated by the blockage degree estimation unit 703, between the input and output of the air preheater 60. Estimate the differential pressure of.

(Plant configuration)
The configuration of the plant 1 according to the second embodiment will be described.
As shown in FIG. 11, the plant 1 according to the second embodiment includes a boiler 10, a denitration device 20, a sensor device 30, and an air preheater 60.
The air preheater 60 is a device for preheating the combustion air to improve the combustion efficiency of the boiler. When leaked ammonia is generated in the denitration device 20, the leaked ammonia reacts with sulfur trioxide (SO3) in the combustion gas in the air preheater 60 to generate acidic ammonium sulfate. Precipitation of acidic ammonium sulfate causes clogging. In order to remove this acidic ammonium sulfate, it is necessary to stop the plant 1 and wash the air preheater 60 with water.

(Process for predicting the differential pressure in the air preheater 60)
Next, 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 with reference to FIGS. 12 and 13.
The deterioration prediction device 40 executes the processes shown in steps S11 to S15 to acquire catalyst deterioration data.

The scale-up factor estimation unit 701 obtains a scale-up factor that corrects the difference between the catalyst performance characteristics obtained in the laboratory and the catalyst performance characteristics in the actual plant 1 (step S21).

The leak estimation unit 702 uses an empirical formula obtained by applying a scale-up factor to an equation showing the performance characteristics of the catalyst acquired in the laboratory, a NOx concentration at the inlet of the denitration device 20, a NOx concentration at the outlet of the denitration device 20, and catalyst deterioration data. Based on this, the amount of leaked ammonia is estimated (step S22).

The blockage estimation unit 703 uses an empirical formula obtained based on the differential pressure data when the differential pressure of the air preheater 60 in the actual plant 1 is rising and the operation data of the plant 1, and the air. The degree of occlusion in the preheater 60 is estimated (step S23).

The differential pressure estimation unit 704 is based on the amount of leak ammonia estimated by the leak estimation unit 702 and the blockage degree in the element of the air preheater 60 estimated by the blockage degree estimation unit 703, between the input and output of the air preheater 60. Estimate the differential pressure of (step S24).
The differential pressure estimation unit 704 may notify the person involved in the catalyst work or the person in charge of the plant 1 of the estimated differential pressure. Based on this notification, catalyst replacement work planning and plant 1 operation support may be performed.

(Example)
For a plant, the differential pressure between the input and output of the air preheater 60 was estimated by the differential pressure predictor 70 according to the second embodiment of the present disclosure described above.
FIG. 14 shows a comparison result between the prediction of the differential pressure in the plant and the actual measurement.
In FIG. 14, the horizontal axis is the operating time. The vertical axis is the differential pressure between the input and output of the air preheater 60. In FIG. 14, the air preheater 60 is shown as an AH (Air Heater). Then, the air preheater 60 is washed once with water.
As can be seen from FIG. 14, the differential pressure prediction device 70 predicted the differential pressure between the input and output of the air preheater 60, and obtained a result close to the measured value.

The plant 1 according to the second embodiment of the present disclosure has been described above.
The differential pressure prediction device (70) includes a deterioration prediction device (40), a scale-up factor estimation unit (701), a leak estimation unit (702), a blockage degree estimation unit (703), and a differential pressure estimation unit (704).
The scale-up factor estimation unit (701) obtains a scale-up factor that corrects 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 leak estimation unit (702) is an empirical formula obtained by applying a scale-up factor to an equation showing the performance characteristics of the catalyst acquired in the laboratory, the NOx concentration at the inlet of the denitration device (20), and the NOx at the outlet of the denitration device (20). The amount of leaked ammonia is estimated based on the concentration and catalyst deterioration data. The blockage estimation unit (703) is obtained based on the differential pressure data when the differential pressure of the air preheater (60) in the actual plant (1) is increasing and the operation data of the plant (1). The degree of blockage in the air preheater 60 is estimated using the empirical formula. The differential pressure estimation unit (704) is based on the amount of leaked ammonia estimated by the leak estimation unit (702) and the blockage degree in the element of the air preheater (60) estimated by the blockage degree estimation unit (703). The differential pressure between the input and output of the air preheater (60) is estimated.

This differential pressure predictor (70) estimates the differential pressure between the input and output of the air preheater (60) when leaked ammonia occurs in the denitration device (20). As a result, the catalyst replacement work can be planned and the operation support of the plant 1 can be performed at an appropriate timing.

Note that the processing in the embodiment of the present disclosure may change the order of processing within the range in which appropriate processing is performed.

The storage unit 401, other storage devices (including registers and latches) in each embodiment of the present invention may be provided anywhere within a range in which appropriate information is transmitted and received. Further, there may be a plurality of storage units 401, other storage devices, and the like within a range in which appropriate information is transmitted and received, and the data may be distributed and stored.

In each embodiment of the present disclosure, the first model generation unit 404 has been described as generating the trained first model as software and storing it in the storage unit 401. Further, in each embodiment of the present disclosure, the second model generation unit 405 has been described as generating the learned second model as software and storing it in the storage unit 401.
However, in another embodiment of the present disclosure, each of the trained first model and the trained second model may be realized as hardware.
For example, the first model generation unit 404 writes a configuration program for realizing the processing performed by the learned first model stored in the storage unit 401 to programmable hardware such as FPGA (Field-Programmable Gate Array). It may be a thing.
Further, for example, a configuration program in which the second model generation unit 405 realizes the processing performed by the learned second model stored in the storage unit 401 is programmable hardware such as FPGA (Field-Programmable Gate Array). It may be written in.

Although the embodiment of the present disclosure has been described, the deterioration prediction device 40, the differential pressure prediction device 70, and other control devices described above may have a computer device inside. The process of the above-mentioned processing is stored in a computer-readable recording medium in the form of a program, and the above-mentioned processing is performed by the computer reading and executing this program. A specific example of a computer is shown below.
FIG. 10 is a schematic block diagram showing the configuration of a computer according to at least one embodiment.
As shown in FIG. 10, the computer 5 includes a CPU 6, a main memory 7, a storage 8, and an interface 9.
For example, each of the above-mentioned deterioration prediction device 40, differential pressure prediction device 70, and other control devices is mounted on the computer 5. The operation of each processing unit described above is stored in the storage 8 in the form of a program. The CPU 6 reads the program from the storage 8, expands it into the main memory 7, and executes the above processing according to the program. Further, the CPU 6 secures a storage area corresponding to each of the above-mentioned storage units in the main memory 7 according to the program.

Examples of the storage 8 include HDD (Hard Disk Drive), SSD (Solid State Drive), magnetic disk, optical magnetic disk, CD-ROM (Compact Disk Read Only Memory), DVD-ROM (Digital Versatile Disk) , Semiconductor memory and the like. The storage 8 may be internal media directly connected to the bus of computer 5, or external media connected to computer 5 via an interface 9 or a communication line. When this program is distributed to the computer 5 via a communication line, the distributed computer 5 may expand the program in the main memory 7 and execute the above processing. In at least one embodiment, the storage 8 is a non-temporary tangible storage medium.

Further, the above program may realize a part of the above-mentioned functions. Further, the program may be a file that can realize the above-mentioned functions in combination with a program already recorded in the computer device, that is, a so-called difference file (difference program).

Although some embodiments of the present disclosure have been described, these embodiments are examples and do not limit the scope of disclosure. These embodiments may be subject to various additions, various omissions, various replacements, and various modifications without departing from the gist of the disclosure.

<Additional notes>
The deterioration prediction device 40, the plant 1, the deterioration prediction method, the program, and the program for causing the computer to execute the processing of the configuration described in each embodiment of the present disclosure are grasped as follows, for example.

(1) The deterioration prediction device (40) according to the first aspect is
A catalyst in a second plant (1) different from the first plant based on learning data including first data relating to a catalyst in the past operation of the first plant and second data relating to the past operating state. Model generators (404, 405) that generate the first prediction model that predicts the degree of deterioration of
A deterioration degree prediction unit (407) that predicts the deterioration degree in the second plant (1) based on the first prediction model generated by the model generation units (404, 405).
To be equipped.

With this deterioration prediction device (40), the degree of deterioration of the catalyst in the second plant (1) can be predicted only by inputting the 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 a complicated analysis on the catalyst.

(2) The deterioration prediction device (40) according to the second aspect is the deterioration prediction device (40) of (1).
The training data includes data relating to the catalyst after poisoning in the past operation of the first plant and data relating to the degree of deterioration in the past operation of the first plant.
The model generators (404, 405)
A first model generation unit (404) that generates a second prediction model that predicts data related to the catalyst after poisoning in the second plant (1) based on the training data.
Based on the data relating to the catalyst after poisoning in the second plant (1) predicted by the second prediction model, the first prediction model for predicting the degree of deterioration in the second plant (1) is generated. Second model generator (405)
With
The deterioration degree prediction unit (407)
Based on the first prediction model and the second prediction model, the degree of deterioration in the second plant (1) is predicted.

With this deterioration prediction device (40), it is possible to predict data related to the catalyst in the device, which is difficult to predict for the second plant (1). As a result, the user of the second plant (1) can easily determine which data affects the degree of deterioration of the catalyst in the second plant (1).

(3) The differential pressure prediction device (70) according to the third aspect is
Deterioration prediction device (40) according to the first aspect or the second aspect, and
A differential pressure estimation unit (704) that estimates the differential pressure between the input and output of the air preheater (60) based on the deterioration degree predicted by the deterioration prediction device (40).
To be equipped.

This differential pressure predictor (70) estimates the differential pressure between the input and output of the air preheater (60) when leaked ammonia occurs in the denitration device (20). As a result, it becomes possible to plan the catalyst replacement work and support the operation of the plant 1 at an appropriate timing.

(4) The differential pressure prediction device (70) according to the fourth aspect is the differential pressure prediction device (70) of (3).
The blockage degree estimation unit (703), which estimates the blockage degree in the air preheater (60) based on the deterioration degree predicted by the deterioration prediction device (40),
With
The differential pressure estimation unit (704)
Based on the degree of blockage estimated by the degree of blockage estimation unit (703), the differential pressure between the input and output of the air preheater (60) is estimated.

The differential pressure prediction device (70) makes the configuration of the differential pressure prediction device (70) clearer, and when leak ammonia occurs in the denitration device (20), it is between the input and output of the air preheater (60). Estimate the differential pressure. As a result, it becomes possible to plan the catalyst replacement work and support the operation of the plant 1 at an appropriate timing.

(5) The differential pressure prediction device (70) according to the fifth aspect is the differential pressure prediction device (70) of (3) or (4).
The leak estimation unit (702), which estimates the amount of leak ammonia based on the degree of deterioration predicted by the deterioration prediction device (40),
With
The differential pressure estimation unit (704)
The differential pressure between the input and output of the air preheater (60) is estimated based on the amount of the leaked ammonia estimated by the leak estimation unit (702).

The differential pressure prediction device (70) makes the configuration of the differential pressure prediction device (70) clearer, and when leak ammonia occurs in the denitration device (20), it is between the input and output of the air preheater (60). Estimate the differential pressure. As a result, it becomes possible to plan the catalyst replacement work and support the operation of the plant 1 at an appropriate timing.

(6) The differential pressure prediction device (70) according to the fifth aspect is the differential pressure prediction device (70) according to any one of (3) to (5).
A scale-up factor estimation unit (701) that obtains a scale-up factor based on the degree of deterioration predicted by the deterioration prediction device (40).
With
The differential pressure estimation unit (704)
Based on the scale-up factor obtained by the scale-up factor estimation unit (701), the differential pressure between the input and output of the air preheater (60) is estimated.

The differential pressure prediction device (70) makes the configuration of the differential pressure prediction device (70) clearer, and when leak ammonia occurs in the denitration device (20), it is between the input and output of the air preheater (60). Estimate the differential pressure. As a result, it becomes possible to plan the catalyst replacement work and support the operation of the plant 1 at an appropriate timing.

(7) The plant (1) according to the seventh aspect is
A plant that uses catalysts
The device (20) for deteriorating the catalyst and
The deterioration predictor (40) of (1) or (2) for predicting the degree of deterioration of the catalyst, and
To be equipped.

With this plant (1), it is possible to predict the degree of deterioration of the catalyst in the plant (1) simply by inputting the 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 a complicated analysis on the catalyst.

(8) The deterioration prediction method according to the eighth aspect is
Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. To generate a first prediction model that predicts
Predicting the degree of deterioration in the second plant based on the generated first prediction model, and
including.

With this deterioration prediction method, it is possible to predict the degree of deterioration of the catalyst in the second plant (1) simply by inputting the 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 a complicated analysis on the catalyst.

(9) The program according to the tenth aspect is
On the computer
Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. To generate a first prediction model that predicts
Predicting the degree of deterioration in the second plant based on the generated first prediction model, and
To execute.

With this program, it is possible to predict the degree of deterioration of the catalyst in the second plant (1) simply by inputting the 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 a complicated analysis on the catalyst.

(10) The program for causing the computer to execute the configuration process according to the eleventh aspect is
A catalyst in a second plant (1) different from the first plant based on learning data including first data relating to a catalyst in the past operation of the first plant and second data relating to the past operating state. Model generators (404, 405), which generate a first prediction model for predicting the degree of deterioration of
A deterioration degree prediction unit (407) that predicts the deterioration degree in the second plant (1) based on the first prediction model generated by the model generation units (404, 405).
Configure each of these as hardware.

A program for causing a computer to perform this configuration process predicts the degree of catalyst deterioration in the second plant (1) simply by inputting data about the second plant (1) into the first prediction model. Can be done. As a result, the user of the second plant (1) can know the degree of deterioration of the catalyst without performing a complicated analysis on the catalyst.

According to the prediction device, the plant, the prediction method, the program, and the configuration program according to the embodiment of the present disclosure, the user of the plant can know the degree of deterioration of the catalyst without performing complicated processing.

1 ... Plant 5 ... Computer 6 ... CPU
7 ... Main memory 8 ... Storage 9 ... Interface 10 ... Boiler 20 ... Denitration device 30 ... Sensor device 40 ... Deterioration prediction device 50 ... Data server device 60 ...・ Air preheater 70 ・ ・ ・ Differential pressure prediction device 301 ・ ・ ・ 1st sensor 302 ・ ・ ・ 2nd sensor 303 ・ ・ ・ 3rd sensor 304 ・ ・ ・ 4th sensor 401 ・ ・ ・ Storage unit 402 ・ ・ ・Planned value acquisition unit 403 ... Plant data acquisition unit 404 ... First model generation unit 405 ... Second model generation unit 406 ... Post-poisoning data acquisition unit 407 ... Catalyst deterioration degree acquisition unit 701 ... Scale-up factor estimation unit 702 ... Leak estimation unit 703 ... Blockage degree estimation unit 704 ... Differential pressure estimation unit

Claims (11)

1. Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. A model generator that generates a first prediction model that predicts
   Based on the first prediction model generated by the model generation unit, the deterioration degree prediction unit that predicts the deterioration degree in the second plant, and the deterioration degree prediction unit.
   Deterioration prediction device including.
2. The training data includes data relating to the catalyst after poisoning in the past operation of the first plant and data relating to the degree of deterioration in the past operation of the first plant.
   The model generator
   A first model generation unit that generates a second prediction model that predicts data related to the catalyst after poisoning in the second plant based on the training data.
   A second model generator that generates the first prediction model that predicts the degree of deterioration in the second plant based on the data related to the catalyst after poisoning in the second plant predicted by the second prediction model. When,
   With
   The deterioration degree prediction unit
   Based on the first prediction model and the second prediction model, the degree of deterioration in the second plant is predicted.
   The deterioration prediction device according to claim 1.
3. The deterioration prediction device according to claim 1 or 2,
   A differential pressure estimation unit that estimates the differential pressure between the input and output of the air preheater based on the degree of deterioration predicted by the deterioration prediction device.
   A differential pressure predictor.
4. An obstruction degree estimation unit that estimates the obstruction degree in the air preheater based on the deterioration degree predicted by the deterioration predictor.
   With
   The differential pressure estimation unit
   Based on the degree of occlusion estimated by the degree of occlusion estimation unit, the differential pressure between the input and output of the air preheater is estimated.
   The differential pressure prediction device according to claim 3.
5. A leak estimation unit that estimates the amount of leaked ammonia based on the degree of deterioration predicted by the deterioration predictor.
   With
   The differential pressure estimation unit
   Based on the amount of leaked ammonia estimated by the leak estimation unit, the differential pressure between the input and output of the air preheater is estimated.
   The differential pressure predictor according to claim 3 or 4.
6. A scale-up factor estimation unit that obtains a scale-up factor based on the degree of deterioration predicted by the deterioration prediction device.
   With
   The differential pressure estimation unit
   Based on the scale-up factor obtained by the scale-up factor estimation unit, the differential pressure between the input and output of the air preheater is estimated.
   The differential pressure prediction device according to any one of claims 3 to 5.
7. A plant that uses catalysts
   The device in which the catalyst deteriorates and
   The deterioration prediction device according to claim 1 or 2, which predicts the degree of deterioration of the catalyst.
   Plant equipped with.
8. Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. To generate a first prediction model that predicts
   Predicting the degree of deterioration in the second plant based on the generated first prediction model, and
   Deterioration prediction method including.
9. Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. To generate a first prediction model that predicts
   Predicting the degree of deterioration in the second plant based on the generated first prediction model, and
   Estimating the differential pressure between the input and output of the air preheater based on the predicted degree of deterioration,
   Differential pressure prediction method including.
10. On the computer
    Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. To generate a first prediction model that predicts
    Predicting the degree of deterioration in the second plant based on the generated first prediction model, and
    A program that executes.
11. Deterioration of the catalyst in the second plant different from the first plant based on the learning data including the first data related to the catalyst in the past operation of the first plant and the second data related to the past operation state. Each of the model generation unit that generates the first prediction model that predicts the degree of deterioration and the degree of deterioration prediction unit that predicts the degree of deterioration in the second plant based on the first prediction model generated by the model generation unit are hardware. Configure as,
    A program that lets a computer perform configuration processing.

Images