DE112020006667T5 - Prediction device, plant, prediction method, program and configuration program - Google Patents

Abstract

A deterioration prediction device includes a model generation unit configured to generate, based on learning data, first data related to a catalyst in a past operation of a first plant and second data related to a state of the past operation , included, generates a first prediction model that predicts a degree of deterioration of a catalyst in a second facility different from the first facility, and a deterioration degree prediction unit that is configured to predict the degree of deterioration in the second facility based on the from the first prediction model generated by the model generating unit.

Description

[Technical Field]

The present disclosure relates to a prediction device, a system, a prediction method, a program and a configuration program. Priority is given to Japanese Patent Application No. JP 2020-017156 A , filed February 4, 2020 and Japanese Patent Application No. JP 2020-106496 A , filed June 19, 2020, the contents of which are incorporated herein by reference.

[State of the art]

A catalyst is used in equipment having a boiler and the like. In the equipment using the catalyst, the catalyst deteriorates as the equipment is operated. Therefore, in the facility using the catalyst, it is desirable to detect a degree of deterioration of the catalyst.

Patent Document 1 discloses a method for predicting the degree of deterioration of a catalyst as a related art.

[quote list]

[patent document]

[Patent Document 1] Japanese Patent No. JP 6278296

[Summary of the Invention]

[Technical problem]

Incidentally, predicting the degree of deterioration of the catalyst in the system usually requires a complicated process such as analyzing an exhaust gas property every time a prediction is made by a user of the system.

Therefore, there is a need in the plant for a technique that allows the user to know the degree of deterioration of the catalyst without having to perform a complicated procedure.

An aim of the present disclosure is to provide a prediction device, a facility, a prediction method, a program and a configuration program for solving the above problem.

[The solution of the problem]

In order to solve the above problems, a deterioration prediction device according to the present disclosure includes a model generation unit configured to generate, based on learning data, first data related to a catalyst in a past operation of a first plant and second containing data relating to a state of past operation, generating a first prediction model that predicts the degree of deterioration of a catalytic converter in a second system, which is different from the first system, and a degree of deterioration prediction unit configured so that it predicts the degree of deterioration in the second plant based on the first prediction model generated by the model generation unit.

A differential pressure prediction device according to the present disclosure includes the deterioration prediction device described above, and a differential pressure estimation unit configured to estimate a differential pressure between the inlet and outlet of an air preheater based on the degree of deterioration predicted by the deterioration prediction device.

A system according to the present disclosure in which a catalyst is used includes a device in which deterioration of the catalyst occurs and the deterioration predicting device described above configured to predict the degree of deterioration of the catalyst.

A deterioration prediction method according to the present disclosure includes creating a first prediction model that predicts a degree of deterioration of a catalyst in a second plant, which is different from the first plant, based on learning data that is first data relating to a catalyst in relating to a past operation of a first plant, and comprising second data relating to a state of the past operation, and predicting the degree of deterioration in the second plant based on the generated first prediction model.

A differential pressure prediction method according to the present disclosure includes creating a first prediction model that predicts a degree of deterioration of a catalyst in a second facility, which is different from the first facility, based on learning data, which is first data relating to a Catalyst in a past operation of a first plant and second data relating to a condition of past operation comprises predicting the degree of deterioration in the second plant based on the generated first predictive model, and estimating the differential pressure between the inlet and outlet of an air preheater based on the predicted one degree of deterioration.

A program according to the present disclosure causes a computer, based on learning data including first data relating to a catalyst in a past operation of a first plant and second data relating to a condition of the past operation, to first instruct a computer generate a prediction model that predicts a degree of deterioration of a catalyst in a second plant, which is different from the first plant, and to predict the degree of deterioration in the second plant based on the generated first prediction model.

[Advantageous Effects of the Invention]

With the prediction device, the system, the prediction method, the program, and the configuration program according to the embodiment of the present disclosure, it is possible for the system user to grasp the degree of deterioration of the catalyst without performing a complicated process.

character list

• 1 12 is a diagram showing an example of a configuration of a plant according to a first embodiment of the present disclosure.
• 2 12 is a diagram showing an example of a configuration of a sensor device according to the first embodiment of the present disclosure.
• 3 12 is a diagram showing an example of a configuration of a deterioration prediction device according to the first embodiment of the present disclosure.
• 4 12 is a diagram showing an example of past result data in various plants according to the first embodiment of the present disclosure.
• 5 14 is a diagram for describing generation of a first model and a second model according to the first embodiment of the present disclosure.
• 6 12 is a diagram showing an example of data input from the first model and the second model according to the first embodiment of the present disclosure.
• 7 12 is a first diagram showing a process flow of the deterioration prediction device according to the first embodiment of the present disclosure.
• 8th FIG. 2 is a second diagram showing the process flow of the deterioration prediction device according to the first embodiment of the present disclosure.
• 9 14 is a graph showing an example of a comparison result of examples according to the first embodiment of the present disclosure.
• 10 14 is a diagram showing an example of a configuration of a differential pressure prediction device according to a second embodiment of the present disclosure.
• 11 14 is a diagram showing an example of a configuration of a plant according to the second embodiment of the present disclosure.
• 12 12 is a diagram showing the process flow of the differential pressure prediction device according to the second embodiment of the present disclosure.
• 13 14 is a diagram for describing a method of the differential pressure prediction device according to the second embodiment of the present disclosure.
• 14 14 is a graph showing an example of a comparison result of examples according to the second embodiment of the present disclosure.
• 15 12 is a schematic block diagram showing a configuration of a computer according to at least one embodiment.

[Description of the Embodiments]

<First Embodiment>

In the following, an embodiment will be described in detail with reference to the drawings.

A deterioration prediction system according to a first embodiment of the present disclosure predicts a deterioration state of a denitration apparatus 20 provided in a facility 1 based on operation data of the facility 1 which is a prediction target. The deterioration prediction system comprises a data server device 50 and a deterioration prediction device 40. The Data server device 50 is a device that stores data of the facility 1 . The data of the system 1 is data that includes, for example, a fuel property, the operating data and the like. The deterioration prediction device 40 predicts the deterioration state of the denitration device 20 provided in the facility 1 based on the data stored in the data server device 50 .

(system configuration)

A configuration of the plant 1 (an example of a second plant and an example of a plant) according to the first embodiment will be described.

The system 1 according to the first embodiment is a system that predicts the degree of performance deterioration of a catalyst in terms of the total running hours of the system 1 .

As in 1 As shown, the plant 1 comprises a boiler 10, the denitration device 20 (an example of a device in which deterioration of the catalyst occurs), and a sensor device 30.

The boiler 10 is a boiler using coal or the like as fuel. The boiler 10 forwards the combustion gas produced when the fuel is burned to the denitration device 20 .

The denitration device 20 is a device that reduces the concentration of nitrogen oxides (NOx) in the combustion gas by decomposing NOx contained in the combustion gas. For a stable operation of the plant 1, it is advantageous to keep the denitration rate of the denitration device 20 at a substantially fixed rate.

For example, by injecting ammonia (NH ₃ ) at an inlet of the denitration device 20, the denitration device 20 keeps the denitration rate, which tends to decrease due to deterioration of the catalyst performance by a poisoning component, at a fixed rate. The poisoning component is a component that is supplied from the outside and is different from a main component of the catalyst.

The poisoning component enters the denitration device 20 from the outside. Therefore, in the denitration device 20, an active site decreases as the poisoning component increases. As a result, the value of the reaction rate constant K, which is a measure of the performance of the catalyst, decreases. That is, the performance of the catalyst deteriorates.

The system 1 according to the first embodiment is a prediction target for the degree of deterioration of the performance of the catalyst.

As in 2 1, 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 inlet of the denitration device 20. FIG.

The first sensor 301 detects a dust concentration at the inlet of the denitration device 20.

The second sensor 302 detects the concentration of nitrogen oxides (NOx) at the inlet of the denitration device 20.

The third sensor 303 detects a concentration of sulfur oxides (Sox) at the inlet of the denitration device 20.

The fourth sensor 304 detects an oxygen concentration (02) 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. Thereby, the data of the facility 1 is accumulated in the data server device 50 .

(Configuration of Deterioration Predicting Device)

The deterioration prediction device 40 is a device that predicts the degree of performance degradation of the catalyst in relation to the total running hours of the facility 1 .

As in 3 As shown, the deterioration prediction device 40 includes a storage unit 401, a schedule value acquisition unit 402, a facility 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 deterioration degree acquisition unit 407 (an example of a deterioration degree prediction unit).

The storage unit 401 stores various information required for a Ver deterioration predictor 40 are required.

The storage unit 401 stores, for example, previous result data D1 in different plants (example of a first plant).

Specifically, the storage unit 401 stores, for example, a plan value, plant data, and the in 4 shown poisoned catalyst data (poisoning catalyst data) .

Note that the storage unit 401 stores the data that changes with the passage of time in association with time.

As in 4 As shown, the target value includes the performance of the catalyst, a specification of the catalyst, and a specification of the device.

Examples of the performance of the catalyst include an initial value K0 of the reaction rate constant and a ratio (K/K0) between the reaction rate constant K and the initial value K0 of the reaction rate constant after a certain operation hour has elapsed (degree of deterioration of the reaction rate constant, i.e. the degree of deterioration of the catalyst). Note that the initial value K0 of the reaction rate constant is a predetermined value.

Other examples of catalyst specification include an initial catalyst specific surface area, an initial catalyst pore volume, and an initial catalyst composition (e.g., titania (TiO2), tungsten trioxide (WO3), vanadium pentoxide (V2O5), and silica (SiO2)). It should be noted that the specification of the catalytic converter can be determined, for example, by sampling and analyzing the actual catalytic converter.

Other examples of the specification of the device include the flow rate, the dust concentration at the inlet of the denitrator 20, the concentration of nitrogen oxides (NOx) at the inlet of the denitrator 20, the concentration of sulfur oxides (Sox) at the inlet of the denitrator 20, and the concentration of oxygen (O2) at the inlet of the denitrator 20

It should be noted that the flow rate is a specific value for each device. Each of 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 is a value derived from that at the inlet of the Denitration device 20 provided sensor device 30 is measured. The dust concentration at the inlet of the denitrator 20, the NOx concentration at the inlet of the denitrator 20, the Sox concentration at the inlet of the denitrator 20, and the O2 concentration at the inlet of the denitrator 20 can each be obtained by calculation from the fuel property. Here, the fuel property is a value indicating the amount of each component in the ash that flows into the catalyst and is a value obtained by analyzing the fuel for each plant.

In addition, the system data, as in 4 shown, the fuel properties and the operational data.

Exemplary examples of the fuel property include the amount of components such as SiO 2 and Al 2 O 3 contained in the fly ash that flow into the catalyst.

The operating data also includes, for example, the operating hours of the system.

Additionally, as in 4 For example, the poisoned catalyst data includes a physical property of the poisoned catalyst and a composition of the poisoned catalyst.

Exemplary examples of the physical property of the poisoned catalyst are the poisoned catalyst specific surface area and the poisoned catalyst pore volume.

Moreover, exemplary examples of the composition of the poisoned catalyst include a value obtained by normalizing a ratio of TiO2 in the poisoned catalyst to a ratio of WO3 (T1O2/WO3) and a value obtained by normalizing the ratio of SiO2 in the poisoned Catalyst to a ratio of WO3 (SiO2/WO3) is obtained.

It should be noted that the poisoned catalyst data is the data of the catalyst in the denitration apparatus 20, which is difficult to acquire during the operation of the plant. Therefore, the data on the poisoned catalyst is collected during the maintenance and inspection of the plant 1.

In addition, the storage unit 401 specifically stores the degree of deterioration of the catalyst (ie, the ratio (K/K0)) obtained for the past operation of the plant.

In addition, the storage unit 401 stores, for example, data used in the case of predicting the degree of deterioration of the catalyst in the plant 1 using the model.

Specifically, the storage unit 401 stores a first model (an example of a second prediction model) generated using the past result data D1, a second model (an example of a first prediction model) generated using the past result data D1 that Plant 1 catalyst performance, Plant 1 catalyst specification, Plant 1 device specification, and Plant 1 fuel property. The first model is a model that predicts the plant 1 poisoned catalyst data. The second model is a model that predicts the degree of deterioration of the catalyst from the poisoned catalyst data predicted by the first model. Both the first model and the second model is a trained model that has been trained using the previous result data in different plants.

Exemplary examples of the performance of the plant 1 catalyst include the initial value K0 of the reaction rate constant of plant 1.

Other examples of the specification of the plant 1 catalyst are the initial specific surface area of the catalyst in plant 1 and the initial pore volume of the catalyst in plant 1.

Exemplary examples of the specification of the device of Plant 1 also include the flow rate in Plant 1.

Moreover, exemplary examples of the fuel property of the plant 1 include the amount of the component contained in the ash that flows into the catalyst of the plant 1 (e.g., the amount of TiO2, SiO2 and the like that flows into the catalyst).

It should be noted that the creation of the first model and the creation of the second model are described below.

Plan value acquisition unit 402 acquires the plan value of facility 1.

For example, the schedule value acquisition unit 402 reads out the performance of the plant 1 catalyst, the specification of the plant 1 catalyst, and the specification of the device 1 from the storage unit 401 .

Specifically, the map value acquiring 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.

In addition, the plan value acquisition unit 402 acquires, for example, from the sensor device 30 a respective acquisition value of the dust concentration detected by the sensor device 30, the nitrogen oxide (NOx) concentration at the inlet of the denitration device 20, the sulfur oxide (SOx) concentration at the inlet of the denitration device 20 and the oxygen (O2) concentration at the inlet of the denitrator 20.

It should be noted that the detection value detected by the schedule value detection unit 402 can be stored in the storage unit 401, and the stored detection value can be used as result data in a case where the degree of deterioration of the catalyst is detected in another facility.

The facility data acquiring unit 403 acquires the facility data of the facility 1 from the data server device 50 to store the acquired facility data in the storage unit 401 .

(creation of the first model)

The first model creation unit 404 creates the first model. Examples of the first model are a machine learning model such as a neural network, a "random forest" and a "support vector machine". In the following description, the generation of the first model is described using the example of a neural network.

The first model generation unit 404 generates a trained first model by subjecting a modeled neural network to machine learning using the past result data (the plan value, the plant data, and the poisoning catalyst data) in various plants as learning data.

The neural network is, for example, a neural convolutional network with an input layer, an intermediate layer and an output layer. In addition, the learning data is data in which the plan value, the facility data, and the post-poisoning catalyst data are linked one-to-one.

Specifically, the first model generation unit 404 divides a plurality of learning data (the plan value, the facility data, and the poisoning catalyst data) into training data, evaluation data, and test data. As in 5 1, the first model generation unit 404 inputs the plan value of the training data and the plant data into the neural network. The neural network (first model before training in 5 ) outputs the poisoning catalyst data. Every time the plan value and the plant data of the training data are input to the neural network and the poisoning catalyst data are output from the neural network, the first model generating unit 404 performs back propagation (“back propagation”) in accordance with the output to adjust the weight change the data coupling between the nodes (ie change the model of the neural network). Next, the first model generating unit 404 inputs the plan value and the plant data of the evaluation data into the neural network of the model changed by the plan value and the plant data of the training data. The neural network outputs the poisoning catalyst data in accordance with the input plan value and the facility data. The first model generating unit 404 changes the weight of the data coupling between the nodes based on the output of the neural network as needed. The neural network generated by the first model generation unit 404 as described above is the trained first model. Next, the first model generation unit 404 inputs the plan value and the plant data of the test data into the neural network of the trained first model as a final confirmation. The trained neural network of the first model outputs the poisoning catalyst data in accordance with the input plan value and the plant data of the test data. For each test data, in a case where the poisoning catalyst data output from the trained neural network of the first model is within a range of a predetermined error with respect to the poisoning catalyst data associated with the input plan value and the plant data of the test data, determines that the neural network of the trained first model is a desired model. In addition, for some of the test data, in a case where the poisoning catalyst data output from the trained neural network of the first model is not within the range of the predetermined error with respect to the post-poisoning catalyst data associated with the input plan value and the plant data of the test data, generates the trained first model using new learning data.

The generation of the trained model by the first model generation unit 404 is repeated until the desired trained first model is reached.

The first model generation unit 404 writes the generated trained first model to the storage unit 401.

(creation of the second model)

The second model creation unit 405 creates the second model.

For example, the second model generation unit 405 generates a trained second model by subjecting a modeled neural network to machine learning using the past result data (the plan value, the plant data, the poisoning catalyst data, and the catalyst deterioration data) in various plants as learning data.

The neural network is, for example, a convolutional neural network having an input layer, an intermediate layer and an output layer. Also, the learning data here is data in which the schedule value, the facility data, the poisoning catalyst data, and the catalyst deterioration data are linked on a one-to-one basis.

Specifically, the second model generation unit 405 divides a plurality of learning data (the plan value, the facility data, the poisoning catalyst data, and the catalyst deterioration data) into training data, evaluation data, and test data. As in 5 1, the second model generation unit 405 inputs the plan value, the plant data, and the poisoning catalyst data of the training data into the neural network. The neural network outputs the catalyst deterioration data. In addition, as in a case where the first model generating unit 404 generates the first model, each test data, in a case where the catalyst deterioration data output by the neural network of the trained second model is within a range of a predetermined error in With respect to the catalyst deterioration data associated with the input plan value, the plant data and the poisoning catalyst data, the second model generation unit 405 that the neural network of the trained second model a desired th model is. In addition, for one of the test data, in a case where the catalyst deterioration data output from the trained neural network of the second model is not within the range of the predetermined error with respect to the catalyst deterioration data compared with the input plan value, the plant data and the Poisoning catalyst data of the test data are connected, the trained second model generated using new training data.

The generation of the trained model by the second model generation unit 405 is repeated until the desired trained second model is reached.

The second model generation unit 405 writes the generated trained second model to the storage unit 401.

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

The post-poisoning data acquisition unit 406 reads, for example, the performance of the plant 1 catalyst (e.g., the initial value K0 of the reaction rate constant of the plant 1), the specification of the plant 1 catalyst (e.g., 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), the specification of the device of the plant 1 (e.g. the flow rate in the plant 1), the fuel property of the plant 1 (e.g. the inflow amount of the component in the ash into the catalyst in the plant 1) and the trained first model from storage unit 401.

In addition, the post-poisoning data acquisition unit 406 acquires from the sensor device 30 information each of 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 specify during the operation of the system 1.

As in 6 shows, the post-poisoning data acquisition unit 406 inputs to the trained first model the performance of the plant 1 catalyst, the specification of the plant 1 catalyst, the specification of the plant 1 device, and the fuel property of the plant 1, which are read out from the storage unit 401, the Information each showing 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, which are detected by the sensor device 30 , and information indicating the operating hours.

In addition, the post-poisoning data acquisition unit 406 acquires the poisoning catalyst data (e.g., the physical property of the poisoning catalyst and the composition of the poisoning catalyst) output from the trained first model.

The poisoning catalyst data acquired by the post-poisoning data acquisition unit 406 is the poisoning catalyst data obtained through the prediction in a case where the plant 1 is operated for the time indicated by the plant data under a condition indicated by the schedule value and the plant data.

The catalyst deterioration degree acquiring unit 407 acquires the catalyst deterioration data based on the map value, the facility data, the poisoning catalyst data, and the trained second model.

For example, the catalyst deterioration degree detection unit 407 reads the specification of the catalyst of plant 1 (e.g., the initial specific surface area of the catalyst in plant 1 and the initial pore volume of the catalyst in plant 1), the specification of the device of plant 1 (e.g., the flow rate in of the system 1) and the trained second model from the storage unit 401.

In addition, the catalyst deterioration degree detection unit 407 detects from the sensor device 30 information each of the dust concentration at the inlet of the denitrator 20, the NOx concentration at the inlet of the denitrator 20, the Sox concentration at the inlet of the denitrator 20, and the O2 concentration at the inlet of the denitration device 20 during operation of the plant 1.

In addition, the catalyst deterioration degree acquisition unit 407 acquires the post-poisoning catalyst data predicted by the poisoning data acquisition unit 406 from the post-poisoning data acquisition unit 406.

As in 6 shown, the catalyst deterioration degree detection unit 407 outputs the Spe Specification of the plant 1 catalyst, the specification of the plant 1 device and the fuel property of the plant 1 read out from the storage unit 401, the NOx concentration at the inlet of the denitrator 20, the Sox concentration at the inlet of the denitrator 20 and the O2 concentration at the inlet of the denitration device 20 detected by the sensor device 30, the poisoning catalyst data predicted by the post-poisoning data acquisition unit 406, and information indicating the operation hours into the trained second model.

In addition, the catalyst deterioration degree acquisition unit 407 acquires the catalyst deterioration data output from the trained second model.

The catalyst deterioration data acquired by the catalyst deterioration degree detection unit 407 is the catalyst deterioration data obtained through the prediction in a case where the plant 1 is operated for the time indicated by the plant data under a condition indicated by the schedule value and the plant data.

(learning process of the model)

The operation of the deterioration prediction device 40 will be described below. First, a learning process of the machine learning model used by the deterioration prediction device 40 to predict the deterioration of the catalyst will be described.

An administrator or the like of the deterioration prediction device 40 acquires in advance the catalyst deterioration data of the catalyst of the denitration device 20 provided in the facility during maintenance and inspection of the facility different from the facility 1 and which is the prediction target. The catalyst deterioration data is obtained by measuring the NOx concentration of the gas at an entrance and the NOx concentration of the gas at an exit of the denitration device 20 . In addition, the manager of the deterioration predicting device 40 obtains in advance the poisoning catalyst data 12 including the physical properties of the catalyst and the composition of the catalyst through the destructive inspection during the maintenance and inspection of the facility different from the facility 1 which is the prediction target is. The poisoning catalyst data is the data of the catalyst in the denitration device 20 and is the data which is difficult to grasp during the operation of the plant.

The first model generating unit 404 subjects the first model, which is the machine learning model, to machine learning using the past result data (the plan value, the plant data, and the poisoning catalyst data) in various plants as the learning data. The plan value and the plant data are acquired from the data server device 50 .

Specifically, the first model generation unit 404 generates the training data using the plan value and the plant data as the input sample and the poisoning catalyst data as the output sample in the result data. The first model generation unit 404 trains, based on the training data, the first model to output a predicted value of a state of the poisoning catalyst by inputting the plan value and the plant data.

The first model generation unit 404 writes the trained first model to the storage unit 401.

The second model generation unit 405 subjects the second model, which is a machine learning model, to a machine learning process using the past result data (the plan value, the plant data, the poisoning catalyst data, and the catalyst deterioration data) in various plants as learning data .

Specifically, the second model generation unit 405 generates the training data using the target value, the facility data, and the poisoning catalyst data as the input sample and the catalyst deterioration data as the output sample in the result data. The second model generation unit 405 trains, based on the training data, the second model to output a predicted value of the catalyst deterioration data by inputting the target value, the facility data, and the poisoning catalyst data.

The second model generation unit 405 writes the trained second model to the storage unit 401.

(Method of Predicting Catalyst Deterioration)

Next, a method for predicting the deterioration of the catalyst is performed the deterioration prediction device 40 using the data shown in FIGS 7 and 8th illustrated process flows of the deterioration prediction device 40 will be described. The deterioration predicting device 40 predicts the catalyst poisoning data based on the map value and the plant data, and predicts the catalyst deterioration data using the predicted catalyst poisoning data.

(Process of Predicting Poisoning Catalyst Data)

First, the process of predicting the poisoning catalyst data by the deterioration predicting device 40 based on the map value, the plant data, and the trained first model will be explained with reference to FIG 7 described.

The post-poisoning data acquisition unit 406 reads the performance of the catalyst of plant 1 (e.g., the initial value K0 of the reaction rate constant of plant 1), the specification of the catalyst of plant 1 (e.g., the initial specific surface area of the catalyst in plant 1, the initial specific surface area of the catalyst in of the plant 1, the initial pore volume of the catalyst in the plant 1 and the like), the specification of the device of the plant 1 (e.g. the flow rate in the plant 1), the fuel property of the plant 1 (e.g. the inflow amount of the component in the Ash in the catalyst in the plant 1) and the trained first model from the storage unit 401 (step S1).

The post-poisoning data acquisition unit 406 acquires from the sensor device 30 the information respectively 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 indicate during the operation of the plant 1 (step S2).

The post-poisoning data acquisition unit 406 inputs into the trained first model the plant 1 catalyst performance, the plant 1 catalyst specification, the plant 1 device specification, and the plant 1 fuel property read out from the storage unit 401, the information , each 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 O2 concentration at the inlet of the denitration device 20, which are detected by the sensor device 30, and information indicating the operating hour (step S3).

The post-poisoning data acquisition unit 406 acquires the poisoning catalyst data (e.g., the physical property of the poisoning catalyst and the composition of the poisoning catalyst) output from the trained first model (step S4).

(Process of Predicting Catalyst Deterioration Data)

Next, the process of predicting the catalyst deterioration data by the deterioration predicting device 40 based on the map value, the plant data, the poisoning catalyst data and the trained second model will be explained with reference to FIG 8th described.

The catalyst deterioration degree detection unit 407 reads the specification of the catalyst of the plant 1 (e.g. the initial specific surface area of the catalyst in the plant 1 and the initial pore volume of the catalyst in the plant 1), the specification of the device of the plant 1 (e.g. the flow rate in the Appendix 1) and the trained second model from the memory unit 401 (step S11).

The catalyst deterioration degree detection unit 407 detects, from the sensor device 30, the information each of 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 Specify denitration device 20 during operation of equipment 1 (step S12).

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

As in 6 1, the catalyst deterioration degree detecting unit 407 gives the specification of the catalyst of plant 1, the specification of the device of plant 1 and the fuel property of plant 1 read out from the storage unit 401, 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, which of of the sensor device 30, the poisoning catalyst data predicted by the post-poisoning data acquisition unit 406 and information indicating the operation hour are incorporated into the trained second model (step S14).

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

(examples)

For a particular plant, the deterioration level of the catalyst is predicted by the deterioration prediction device 40 according to the first embodiment of the present disclosure described above and compared to an actual measurement of the deterioration level of the catalyst in the plant.

127 data are prepared for 11 plants with different years of construction and operating hours, of which 70% is used as learning data and 30% as data for checking accuracy.

9 FIG. 12 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 9 the horizontal axis shows the operating hours. The vertical axis is the degree of deterioration of the catalyst. Note that the degree of deterioration of the catalyst is a value obtained by dividing the reaction rate constant K in each hour of operation by the initial reaction rate constant K0.

For the accuracy of predicting the degree of deterioration of the catalyst, a result of 0.05 root mean square error (RMSE) is obtained.

The system 1 according to the first embodiment of the present disclosure has been described above.

The deterioration prediction device 40 includes the model generation unit 404, 405 and the deterioration degree prediction unit 407.

Based on the learning data including the first data related to the catalyst in the past operation of the first system and the second data related to the state of the past operation, the model generating unit generates the first prediction model representing the degree of deterioration of the catalyst in the second plant (1) which differs from the first plant.

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 the deterioration predicting device 40, the degree of deterioration of the catalyst in the second system 1 can be predicted only by inputting the data on the second system 1 into the first prediction model. As a result, the user of the second equipment 1 can grasp the degree of deterioration of the catalyst without performing a complicated analysis of the catalyst.

<Second embodiment>

A deterioration prediction system according to a second embodiment of the present disclosure predicts the deterioration state of the denitration device 20 provided in the plant 1 based on the operational data of the plant 1, which is the prediction target, and predicts a differential pressure between the inlet and outlet of an air preheater 60, which will be described later is predicted based on the predicted state of deterioration. The deterioration prediction system includes the data server device 50 and a differential pressure prediction device 70.

(Configuration of differential pressure prediction device)

The differential pressure predicting device 70 is a device that predicts the differential pressure of the air preheater 60 based on the predicted degree of deterioration in the performance of the catalyst.

As in 10 As shown, the differential pressure prediction device 70 includes, in addition to the deterioration prediction device 40, a scaling factor estimation unit 701, a leak estimation unit 702, clogging degree estimation unit 703, and a differential pressure estimation unit 704.

The scaling factor estimating unit 701 obtains a correction coefficient for correcting a difference between a performance characteristic of the catalyst found in the laboratory and a performance characteristic of the catalyst in the actual plant 1. The correction coefficient is a scaling factor. For example, the performance characteristic of the catalyst determined in the laboratory for various combinations of gas temperature, gas concentration and oxygen concentration. In addition, the obtained performance characteristics of the catalyst are expressed by a formula containing gas temperature, gas concentration and oxygen concentration. However, in the actual plant 1, the performance of the catalyst decreases compared to the laboratory-determined catalyst performance characteristics. The scale factor estimating unit 701 determines the scale-up factor for correcting the formula indicating the performance characteristics of the catalyst found in the laboratory so that the formula can be applied to the actual plant 1. It should be noted that the scale-up factor is determined by an empirical rule resulting from the performance characteristics of the catalyst detected in the laboratory and the performance characteristics of the catalyst in the actual plant 1. The leak estimation unit 702 estimates the amount of leaked ammonia based on an empirical formula in which the scale-up factor is based on the laboratory-determined performance characteristics of the catalyst, the NOx concentration at the inlet of the denitration device 20, the NOx concentration at the outlet of the Denitration device 20 and the catalyst deterioration data is applied. In the above, the performance of the catalyst is obtained from the empirical formula, and the amount of unreacted ammonia is estimated from the performance and the NOx concentration at the entrance and exit of the denitration device 20 .

The clogging degree estimating unit 703 estimates a clogging degree in the air preheater 60 using the 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.

If, for example, the differential pressure of the air preheater 60 in the actual system 1 increases, the differential pressure between the inlet and outlet in the element of the air preheater 60 is monitored and the differential pressure is calculated. The degree of clogging is calculated from the differential pressure, and the empirical formula in this case is found using the operating data of plant 1.

The above is based on the thought that in a case where a leaked ammonia concentration can be estimated from the operational data and the leaked ammonia precipitates as acidic ammonium sulfate (NH4HSO4) in the air preheater 60, there is a correlation between the amount of acidic ammonium sulfate and the degree of clogging of the air preheater 60, the degree of clogging in the element of the air preheater 60 can be estimated.

The differential pressure estimation unit 704 estimates the differential pressure between the inlet and outlet of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimation unit 702 and the clogging degree acidic in the element of the air preheater 60 estimated by the clogging degree estimation unit 703.

(system configuration)

A configuration of the plant 1 according to the second embodiment will be described. As in 11 1, the plant 1 according to the second embodiment includes the boiler 10, the denitration device 20, the sensor device 30, and the air preheater 60. The air preheater 60 is a device that preheats the combustion air to improve the combustion efficiency of the boiler. In a case where the 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 the acidic ammonium sulfate. The precipitation of the acidic ammonium sulfate causes the clogging. It should be noted that in order to remove the acidic ammonium sulfate it is necessary to shut down the plant 1 and wash the air preheater 60 with water.

(Method of Predicting Differential Pressure in Air Preheater 60)

Next, a method of predicting the differential pressure between the inlet and outlet of the air preheater 60 by the differential pressure predicting device 70 will be described with reference to FIG 12 and 13 described.

The deterioration predicting device 40 executes the processes shown in steps S11 to S15 to acquire the catalyst deterioration data.

The scale factor estimating unit 701 obtains the scale-up factor for correcting the difference between the laboratory catalyst performance and the catalyst performance in the actual plant 1 (step S21).

The leak estimation unit 702 estimates the amount of leaked ammonia based on the empirical formula in which the scale-up factor is based on the laboratory-determined performance characteristics of the catalyst, the Nox-Con concentration at the inlet of the denitrator 20, the NOx concentration at the outlet of the denitrator 20, and the catalyst deterioration data (step S22).

The clogging degree estimating unit 703 estimates the degree of occlusion in the air preheater 60 using the 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 (step S23) .

The differential pressure estimation unit 704 estimates the differential pressure between the inlet and outlet of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimation unit 702 and the degree of occlusion in the element of the air preheater 60 estimated by the clogging degree estimation unit 703 (step S24).

Note that the differential pressure estimating unit 704 may notify the estimated differential pressure to a person involved in catalyst works or a person in charge of the facility 1 . The planning of the catalyst replacement work or the support of the operation of plant 1 can take place on the basis of the report.

(examples)

For a specific plant, the differential pressure between the inlet and outlet of the air preheater 60 is estimated by the differential pressure predictor 70 according to the second embodiment of the present disclosure described above.

14 shows a comparison result between the prediction and the actual measurement of the differential pressure in the system.

In 14 a horizontal axis shows the operating hours. A vertical axis shows the differential pressure between the inlet and outlet of the air preheater 60. Note that the air preheater 60 in 14 is shown as an air heater (AH). In addition, the air preheater 60 is washed once with water.

How out 14 As can be seen, by predicting the differential pressure between the inlet and outlet of the air preheater 60 by the differential pressure predictor 70, a result close to the actual measured value is obtained.

The system 1 according to the second embodiment of the present disclosure has been described above.

The differential pressure prediction device 70 includes the deterioration prediction device 40, the scale factor estimation unit 701, the leakage estimation unit 702, the clogging degree estimation unit 703, and the differential pressure estimation unit 704.

The scaling factor estimation unit 701 obtains the scaling factor for correcting the difference between the catalyst performance obtained in the laboratory and the catalyst performance in the actual plant 1. The leakage estimation unit 702 estimates the amount of ammonia leakage based on the empirical formula 12 in which the scaling factor is applied to the laboratory-determined catalyst performance characteristics, the NOx concentration at the inlet of the denitrator 20, the NOx concentration at an outlet of the denitrator 20, and the catalyst deterioration data. The clogging degree estimating unit 703 estimates the clogging degree in the air preheater 60 using the 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. The differential pressure estimating unit 704 estimates the differential pressure between the inlet and outlet of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimation unit 702 and the clogging degree in the element of the air preheater (60) estimated by the clogging degree estimation unit 703.

The differential pressure predicting device 70 estimates the differential pressure between the inlet and outlet of the air preheater 60 when ammonia leaks in the denitration device 20 . As a result, the planning of the catalyst replacement work or the support of the operation of the plant 1 can take place at a suitable point in time.

It should be noted that in the method according to the present disclosure, the order of the process can be changed within a range in which an appropriate process is performed.

It should be noted that the storage unit 401 according to each embodiment of the present invention, other storage devices (including a register and a latch), and the like throughout of an area in which appropriate information is transmitted and received can be provided. In addition, a plurality of storage units 401, other storage devices, and the like can be present within the area where appropriate information is transmitted and received, and can distribute and store the data.

It should be noted that, as described, in each embodiment of the present disclosure, the first model creation unit 404 creates the trained first model as software and stores the created trained first model in the storage unit 401 . As also described, in each embodiment of the present disclosure, the second model generation unit 405 generates the trained second model as software and stores the generated trained second model 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 implemented in hardware.

For example, the first model generation unit 404 can write a configuration program that implements the process performed by the trained first model stored in the memory unit 401 in programmable hardware, such as a field programmable gate array (FPGA) .

Furthermore, the second model generation unit 405 can write, for example, a configuration program realizing the process performed by the trained second model stored in the storage unit 401 in programmable hardware such as a field programmable gate array (FPGA).

Although the embodiments of the present disclosure have been described, the deterioration prediction device 40, the differential pressure prediction device 70, and other control devices described above may include a computing device inside. In addition, a flow of the method described above is stored in a computer-readable recording medium in the form of a program, and the method described above is performed by a computer reading and executing the program. A specific example of a computer is described below.

10 12 is a schematic block diagram showing a configuration of the computer according to at least one embodiment.

As in 10 shown, a computer 5 includes a CPU 6, a main memory 7, a memory 8 and an interface 9.

For example, each of the deterioration prediction device 40, the differential pressure prediction device 70, and other control devices described above is mounted on the computer 5. FIG. In addition, the operation of each processing unit described above is stored in the memory 8 in the form of a program. The CPU 6 reads the program from the memory 8, develops the read program in the main memory 7, and executes the process described above in accordance with the program. In addition, the CPU 6 secures a storage area corresponding to each storage unit described above in the main storage 7 in accordance with the program.

Examples of the memory 8 are a hard disk drive (HDD), a solid state drive (SSD), a magnetic disk, a magnetic optical disk, a read-only memory for compact discs (CD-ROM), a read-only memory for Digital Versatile Discs (DVD-ROM) and a semiconductor memory. The memory 8 can be an internal medium that is directly connected to a bus of the computer 5, or an external medium that is connected to the computer 5 via the interface 9 or a communication line. When the program is distributed to the computer 5 through the communication line, the distributed computer 5 can develop the program in the main memory 7 and execute the process described above. In at least one embodiment, memory 8 is a non-portable tangible storage medium.

In addition, the program described above can implement some of the functions described above. In addition, the program can be a file that can realize the functions described above in combination with a program already recorded in the computer, i. H. a so-called difference file (difference program) .

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

<Supplementary Note>

• (1) Ein erster Aspekt betrifft eine Verschlechterungsvorhersagevorrichtung 40, die eine Modellerzeugungseinheit 404, 405 enthält, die so konfiguriert ist, dass sie auf Basis von Lerndaten, die erste Daten, die sich auf einen Katalysator in einem vergangenen Betrieb einer ersten Anlage beziehen, und zweite Daten, die sich auf einen Zustand des vergangenen Betriebs beziehen, enthalten, ein erstes Vorhersagemodell erzeugt, das einen Verschlechterungsgrad eines Katalysators in einer zweiten Anlage 1, die sich von der ersten Anlage unterscheidet, vorhersagt, und eine Verschlechterungsgrad-Vorhersageeinheit 407, die so konfiguriert ist, dass sie den Verschlechterungsgrad in der zweiten Anlage 1 auf der Grundlage des von der Modellerzeugungseinheit 404, 405 erzeugten ersten Vorhersagemodells vorhersagt.

The deterioration prediction device 40, the equipment 1, the deterioration prediction method, the program, and the program that makes the computer execute the configuration process described in each embodiment of the present disclosure are understood as follows, for example.

• (1) A first aspect relates to a deterioration prediction apparatus 40 including a model generation unit 404, 405 configured to generate, based on learning data, first data relating to a catalyst in a past operation of a first plant, and second data related to a state of past operation, generates a first prediction model that predicts a degree of deterioration of a catalyst in a second plant 1 different from the first plant, and a degree of deterioration prediction unit 407 that so is configured to predict the degree of deterioration in the second plant 1 based on the first prediction model generated by the model generation unit 404, 405.

• (2) Ein zweiter Aspekt bezieht sich auf die Verschlechterungsvorhersagevorrichtung 40 gemäß (1), wobei die Lerndaten Daten, die sich auf einen Vergiftungskatalysator im vergangenen Betrieb der ersten Anlage beziehen, und Daten, die sich auf den Verschlechterungsgrad im vergangenen Betrieb der ersten Anlage beziehen, enthalten, wobei die Modellerzeugungseinheit 404, 405 eine erste Modellerzeugungseinheit 404, die so konfiguriert ist, dass sie auf der Grundlage der Lerndaten ein zweites Vorhersagemodell erzeugt, das Daten vorhersagt, die sich auf den Vergiftungskatalysator in der zweiten Anlage 1 beziehen, und eine zweite Modellerzeugungseinheit 405, die konfiguriert ist, um das erste Vorhersagemodell zu erzeugen, das den Verschlechterungsgrad in der zweiten Anlage 1 basierend auf den Daten, die sich auf den Vergiftungs-Katalysator in der zweiten Anlage 1 beziehen, die durch das zweite Vorhersagemodell vorhergesagt werden, vorhersagt, beinhaltet, und die Verschlechterungsgradvorhersageeinheit 407 ist konfiguriert, um den Verschlechterungsgrad in der zweiten Anlage 1 basierend auf dem ersten Vorhersagemodell und dem zweiten Vorhersagemodell vorherzusagen.

With the deterioration predicting device 40, the degree of deterioration of the catalyst in the second system 1 can be predicted only by inputting the data on the second system 1 into the first prediction model. As a result, the user of the second equipment 1 can grasp the degree of deterioration of the catalyst without performing a complicated analysis of the catalyst.

• (2) A second aspect relates to the deterioration predicting device 40 according to (1), wherein the learning data is data related to a poisoning catalyst in the past operation of the first system and data related to the degree of deterioration in the past operation of the first system , included, wherein the model generation unit 404, 405 includes a first model generation unit 404 configured to generate, based on the learning data, a second prediction model that predicts data related to the poisoning catalyst in the second plant 1, and a second Model generation unit 405 configured to generate the first prediction model that predicts the degree of deterioration in the second plant 1 based on the data related to the poisoning catalyst in the second plant 1 predicted by the second prediction model , includes, and the deterioration degree predictionei Unit 407 is configured to predict the degree of deterioration in the second plant 1 based on the first prediction model and the second prediction model.

• (3) Ein dritter Aspekt bezieht sich auf eine Differenzdruckvorhersagevorrichtung 70, die die Verschlechterungsvorhersagevorrichtung 40 gemäß dem ersten oder zweiten Aspekt und eine Differenzdruckschätzeinheit 704 enthält, die so konfiguriert ist, dass sie einen Differenzdruck zwischen dem Eingang und dem Ausgang eines Luftvorwärmers 60 auf der Grundlage des von der Verschlechterungsvorhersagevorrichtung 40 vorhergesagten Verschlechterungsgrades schätzt.

With the deterioration predicting device 40, it is possible to predict the hard-to-predict data related to the catalyst in the device for the second plant 1. Thereby, the user of the second facility 1 can easily ascertain which data affects the degree of deterioration of the catalyst in the second facility 1 .

• (3) A third aspect relates to a differential pressure prediction device 70 that includes the deterioration prediction device 40 according to the first or second aspect and a differential pressure estimation unit 704 that is configured to calculate a differential pressure between the inlet and outlet of an air preheater 60 based of the degree of deterioration predicted by the deterioration predicting device 40 .

• (4) Ein vierter Aspekt bezieht sich auf die Differenzdruck-Vorhersagevorrichtung 70 gemäß (3), die ferner eine Verstopfungsgrad-Schätzeinheit 703 enthält, die so konfiguriert ist, dass sie einen Verstopfungsgrad in dem Luftvorwärmer 60 auf der Grundlage des von der Verschlechterungs-Vorhersagevorrichtung 40 vorhergesagten Verschlechterungsgrads schätzt, wobei die Differenzdruck-Schätzeinheit 704 so konfiguriert ist, dass sie den Differenzdruck zwischen Eingang und Ausgang des Luftvorwärmers 60 auf der Grundlage des von der Verstopfungsgrad-Schätzeinheit 703 geschätzten Verstopfungsgrads schätzt.

The differential pressure prediction device (70) estimates the differential pressure between the inlet and outlet of the air preheater 60 in the event that ammonia leaks in the denitration device (20). This makes it possible to carry out the planning of the catalyst change or the support of the operation of the plant 1 at a suitable point in time.

• (4) A fourth aspect relates to the differential pressure prediction device 70 according to (3), further including a degree of clogging estimating unit 703 configured to estimate a degree of clogging in the air preheater 60 based on that from the deterioration prediction device 40 predicted degree of deterioration, wherein the differential pressure estimation unit 704 is configured to estimate the differential pressure between the inlet and outlet of the air preheater 60 based on the degree of clogging estimated by the degree of clogging estimating unit 703 .

• (5) Ein fünfter Aspekt betrifft die Differenzdruckvorhersagevorrichtung 70 gemäß (3) oder (4), die ferner eine Leckabschätzungseinheit 702 enthält, die so konfiguriert ist, dass sie die Menge an ausgetretenem Ammoniak auf der Grundlage des von der Verschlechterungsvorhersagevorrichtung 40 vorhergesagten Verschlechterungsgrads abschätzt, wobei die Differenzdruckabschätzungseinheit 704 so konfiguriert ist, dass sie den Differenzdruck zwischen Eingang und Ausgang des Luftvorwärmers 60 auf der Grundlage der von der Leckabschätzungseinheit 702 abgeschätzten Menge an ausgetretenem Ammoniak abschätzt.

The differential pressure prediction device 70 clarifies the structure of the differential pressure prediction device 70 and estimates the differential pressure between the inlet and outlet of the air preheater 60 in the case that ammonia leakage is generated in the denitration device 20 . This makes it possible to carry out the planning of the catalyst change or the support of the operation of the plant 1 at a suitable point in time.

• (5) A fifth aspect relates to the differential pressure prediction device 70 according to (3) or (4), which further includes a leak estimation unit 702 configured to estimate the amount of leaked ammonia based on the degree of deterioration predicted by the deterioration prediction device 40, wherein the differential pressure estimation unit 704 is configured to estimate the differential pressure between the inlet and outlet of the air preheater 60 based on the amount of leaked ammonia estimated by the leakage estimation unit 702 .

• (6) Ein fünfter Aspekt bezieht sich auf die Differenzdruck-Vorhersagevorrichtung 70 gemäß einem der Punkte (3) bis (5), die ferner eine Skalierungsfaktor-Schätzeinheit 701 enthält, die so konfiguriert ist, dass sie einen Skalierungsfaktor auf der Grundlage des von der Verschlechterungs-Vorhersagevorrichtung 40 vorhergesagten Verschlechterungsgrades erhält, wobei die Differenzdruck-Schätzeinheit 704 so konfiguriert ist, dass sie den Differenzdruck zwischen Eingang und Ausgang des Luftvorwärmers 60 auf der Grundlage des von der Skalierungsfaktor-Schätzeinheit 701 erhaltenen Skalierungsfaktors schätzt.

The differential pressure prediction device 70 clarifies the structure of the differential pressure prediction device 70 and estimates the differential pressure between the inlet and outlet of the air preheater 60 in the case that ammonia leakage is generated in the denitration device 20 . This makes it possible to carry out the planning of the catalyst change or the support of the operation of the plant 1 at a suitable point in time.

• (6) A fifth aspect relates to the differential pressure prediction device 70 according to any one of (3) to (5), further including a scale factor estimating unit 701 configured to calculate a scale factor based on the Deterioration predicting device 40 obtains the predicted degree of deterioration, wherein the differential pressure estimating unit 704 is configured to estimate the differential pressure between the inlet and outlet of the air preheater 60 based on the scaling factor obtained from the scaling factor estimating unit 701 .

• (7) Ein siebter Aspekt betrifft eine Anlage 1, in der ein Katalysator verwendet wird, wobei die Anlage eine Vorrichtung 20 enthält, in der eine Verschlechterung des Katalysators auftritt, und die Verschlechterungsvorhersagevorrichtung 40 gemäß (1) oder (2) so konfiguriert ist, dass sie den Verschlechterungsgrad des Katalysators vorhersagt.

The differential pressure prediction device (70) clarifies the structure of the differential pressure prediction device (70) and estimates the differential pressure between the inlet and outlet of the air preheater (60) in the case that in the denitration device (20) leaked ammonia is generated. This makes it possible to carry out the planning of the catalyst change or the support of the operation of the plant 1 at a suitable point in time.

• (7) A seventh aspect relates to a system 1 using a catalyst, the system including a device 20 in which deterioration of the catalyst occurs, and the deterioration predicting device 40 is configured according to (1) or (2) so that that it predicts the degree of deterioration of the catalytic converter.

• (8) Ein achter Aspekt bezieht sich auf ein Verschlechterungsvorhersageverfahren, das das Erzeugen eines ersten Vorhersagemodells, das einen Verschlechterungsgrad eines Katalysators in einer zweiten Anlage, die sich von der ersten Anlage unterscheidet, vorhersagt, auf der Grundlage von Lerndaten, die erste Daten, die sich auf einen Katalysator in einem vergangenen Betrieb einer ersten Anlage beziehen, und zweite Daten, die sich auf einen Zustand des vergangenen Betriebs beziehen, umfassen, und das Vorhersagen des Verschlechterungsgrads in der zweiten Anlage auf der Grundlage des erzeugten ersten Vorhersagemodells beinhaltet.

With the plant 1, the degree of deterioration of the catalyst in the plant 1 can be predicted only by inputting the data regarding the plant 1 into the first prediction model. Consequently, the user of the system 1 can grasp the degree of deterioration of the catalyst without having to perform a complicated analysis of the catalyst.

• (8) An eighth aspect relates to a deterioration prediction method that includes creating a first prediction model that predicts a degree of deterioration of a catalyst in a second facility, which is different from the first facility, based on learning data that is first data that relate to a catalytic converter in a past operation of a first plant, and comprise second data relating to a state of the past service, and predicting the degree of deterioration in the second plant based on the generated first prediction model.

• (9) Ein zehnter Aspekt betrifft ein Programm, das einen Computer veranlasst, die Erzeugung eines ersten Vorhersagemodells, das einen Verschlechterungsgrad eines Katalysators in einer zweiten Anlage, die sich von der ersten Anlage unterscheidet, vorhersagt, auf der Grundlage von Lerndaten, die erste Daten, die sich auf einen Katalysator in einem vergangenen Betrieb einer ersten Anlage beziehen, und zweite Daten, die sich auf einen Zustand des vergangenen Betriebs beziehen, umfassen, und die Vorhersage des Verschlechterungsgrads in der zweiten Anlage auf der Grundlage des erzeugten ersten Vorhersagemodells auszuführen.

With the deterioration prediction method, the degree of deterioration of the catalyst in the second plant 1 can be predicted only by inputting the data of the second plant 1 into the first prediction model. Consequently, the user of the second equipment 1 can grasp the degree of deterioration of the catalyst without having to perform complicated analysis of the catalyst.

• (9) A tenth aspect relates to a program that causes a computer to generate a first prediction model that predicts a degree of deterioration of a catalyst in a second facility different from the first facility, based on learning data that is first data relating to a catalytic converter in a past operation of a first plant and second data relating to a state of the past operating, and to carry out the prediction of the degree of deterioration in the second plant based on the generated first prediction model.

• (10) Ein elfter Aspekt betrifft ein Programm, das einen Computer veranlasst, einen Konfigurationsprozess des Konfigurierens, als Hardware, sowohl einer Modellerzeugungseinheit 404, 405, die konfiguriert ist, auf Basis von Lerndaten, die sich auf einen Katalysator beziehenden erste Daten in einer vergangenen Operation einer ersten Anlage und sich auf einen Zustand der vergangenen Operation beziehende zweite Daten beinhalten, , ein erstes Vorhersagemodell zu erzeugen, das einen Verschlechterungsgrad eines Katalysators in einer zweiten Anlage 1, die sich von der ersten Anlage unterscheidet, vorhersagt, als auch einer Verschlechterungsgrad-Vorhersageeinheit 407, die so konfiguriert ist, dass sie den Verschlechterungsgrad in der zweiten Anlage (1) auf der Grundlage des von der Modellerzeugungseinheit (404, 405) erzeugten ersten Vorhersagemodells vorhersagt, auszuführen.

The program can predict the degree of deterioration of the catalyst in the second plant (1) only by inputting the data of the second plant (1) into the first prediction model. Consequently, the user of the second equipment (1) can grasp the degree of deterioration of the catalyst without having to perform complicated analysis of the catalyst.

• (10) An eleventh aspect relates to a program that causes a computer to perform a configuration process of configuring, as hardware, both a model generation unit 404, 405 configured to generate a first predictive model based on learning data including first data relating to a catalyst in a past operation of a first plant and second data relating to a state of the past operation that predicts a degree of deterioration of a catalyst in a second facility 1 different from the first facility, and a deterioration degree prediction unit 407 configured to predict the degree of deterioration in the second facility (1) based on the of the first prediction model generated by the model generation unit (404, 405) predicts to execute.

With the program causing the computer to execute the configuration process, the degree of deterioration of the catalyst in the second plant 1 can be predicted only by inputting the data regarding the second plant 1 into the first prediction model. As a result, the user of the second equipment 1 can grasp the degree of deterioration of the catalyst without performing a complicated analysis of the catalyst.

[Industrial Applicability]

With the prediction device, the system, the prediction method, the program, and the configuration program according to the embodiment of the present disclosure, it is possible for the system user to grasp the degree of deterioration of the catalyst without performing a complicated process.

Reference List

1

Attachment

5

computer

6

CPU

7

main memory

8th

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

First sensor

302

Second sensor

303

Third sensor

304

Fourth sensor

401

storage unit

402

planned value entry unit

403

Production data acquisition device

404

first model generation unit

405

second model generation unit

406

Post-poisoning data collection unit

407

Catalyst deterioration degree detection unit

701

scale factor estimation unit

702

leak assessment unit

703

congestion degree estimating unit

704

differential pressure estimation unit

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• JP 2020017156 A [0001]
• JP 2020106496 A [0001]
• JP 6278296 [0004]

Claims (11)

Deterioration prediction device comprising: a model generation unit configured to generate, based on learning data, first data related to a catalytic converter in a past operation of a first plant and second data related to a state of the past operation, generates a first prediction model that predicts a degree of deterioration of a catalyst in a second plant different from the first plant; and a deterioration degree prediction unit configured to predict the deterioration degree in the second plant based on the first prediction model generated by the model generation unit. deterioration prediction device claim 1 , wherein the learning data includes data relating to a poisoning catalyst in past operation of the first facility and data relating to the degree of deterioration in past operation of the first facility, the model generation unit includes a first model generation unit configured so that it generates, based on the learning data, a second prediction model that predicts data related to the poisoning catalyst in the second facility, and a second model generation unit that is configured to generate the first prediction model that predicts the degree of deterioration in the second plant based on the data related to the post-poisoning catalyst in the second plant predicted by the second prediction model, and the deterioration degree prediction unit is configured to predict the deterioration degree in the second plant based on the basis of the er th prediction model and the second prediction model. A differential pressure prediction device comprising: the deterioration prediction device according to FIG claim 1 or 2 ; and a differential pressure estimating unit configured to estimate a differential pressure between an inlet and an outlet of an air preheater based on the degree of deterioration predicted by the deterioration prediction device. Differential pressure prediction device claim 3 , further comprising: a clogging degree estimation unit configured to estimate a clogging degree in the air preheater based on the deterioration degree predicted by the deterioration prediction device, wherein the differential pressure estimation unit is configured to calculate the differential pressure between the inlet and outlet of the air preheater based on the degree of congestion estimated by the degree of congestion estimating unit. Differential pressure prediction device claim 3 or 4 , further comprising: a leakage estimation unit configured to estimate an amount of leaked ammonia based on the degree of deterioration predicted by the deterioration prediction device, wherein the differential pressure estimation unit is configured to estimate the differential pressure between the inlet and outlet of the air preheater based on the amount of leaked ammonia estimated by the leak estimation unit. Differential pressure prediction device according to one of claims 3 until 5 , further comprising: a scaling factor estimation unit configured to obtain a scaling factor based on the degree of deterioration predicted by the deterioration prediction device, wherein the differential pressure estimation unit is configured to estimate the differential pressure between the inlet and the outlet of the air preheater based on the scale factor obtained from the scale factor estimation unit. A system using a catalyst, the system comprising: a device in which deterioration of the catalyst occurs; and the deterioration prediction device claim 1 or 2 , which is configured to predict the degree of deterioration of the catalyst. A method for predicting deterioration, comprising: generating a first predictive model that predicts a degree of deterioration of a catalyst in a second plant different from the first plant, based on learning data, the first data relating to a catalyst in a previous operation of a first plant and second data, relating to a state of prior operation; and Prediction of the degree of deterioration in the second plant based on the first prediction model created. A differential pressure prediction method, comprising: generating, based on learning data including first data related to a catalyst in a previous operation of a first plant and second data related to a condition of the previous operation, a first prediction model that predicts a degree of deterioration of a catalyst in a predicts a second facility that is different from the first facility; predicting the degree of degradation in the second plant based on the generated first prediction model; and estimating a differential pressure between an inlet and outlet of an air preheater based on the predicted degree of deterioration. Program that causes a computer to run: Generating, on the basis of learning data, which includes first data relating to a catalytic converter in a previous operation of a first plant and second data relating to a state of the previous operation, include a first prediction model which has a degree of deterioration of a catalyst in a second plant different from the first plant; and Predicting the degree of deterioration in the second plant based on the created first predictive model. Program that causes a computer to run a configuration process that configures each of the following items as hardware: a model generation unit configured to generate based on learning data including first data related to a catalyst in a past operation of a first plant and second data related to a state of the past operation generates a first prediction model that predicts a degree of deterioration of a catalyst in a second plant different from the first plant; and a deterioration degree prediction unit configured to predict the deterioration degree in the second plant based on the first prediction model generated by the model generation unit.

Images