Fluidized bed boiler gas production prediction model generation method and device, equipment and storage medium

Abstract

The invention discloses a fluidized bed boiler gas production prediction model generation method, device and equipment and a storage medium, and the method comprises the steps: dividing working condition data into process control type data, raw material type data, raw material physicochemical property type data and product gas information type data; and preprocessing the working condition data in the preset data data updating the process control type data; various data of the working condition data are used as feature data, and a prediction model with the purpose of predicting product gas information is established; the method comprises the following steps: performing multi-dimensional analysis on process control type data, raw material type data and product gas information type data of raw material processing, and preprocessing working condition data; according to the method, the relevance between various data of the raw material data and the product gas information is determined, and the high-accuracy prediction model is established, so that the optimization effect on the gas production performance of the fluidized bed boiler is improved, and the guidance on gas performance optimization is higher.

Description

Technical Field

[0001] The invention relates to the technical field of fluidized bed boiler gas production, and in particular to a method, device, equipment and storage medium for generating a fluidized bed boiler gas production prediction model. Background Art

[0002] Fluidized bed boilers include bubbling fluidized bed boilers, circulating fluidized bed boilers and double fluidized bed boilers; among them, fluidized bed boilers use fluidized combustion and have many advantages such as wide fuel adaptability, high combustion efficiency, high-efficiency desulfurization, low NOX emissions, simple structure and easy operation; double fluidized bed boilers are composed of a gasifier and a combustion furnace. When biomass is added to the gasifier, the raw materials will undergo a pyrolysis reaction and release heat. The generated synthesis gas is separated from the gas and solid by a high-temperature cyclone separator and then passed into the boiler for combustion after a series of purification processes by a purification device. The tar and fluidized medium generated in this process are circulated into the combustion furnace to burn and release heat, thereby realizing the recycling of heat. Therefore, fluidized bed boilers and double fluidized bed boilers are widely used.

[0003] When using fluidized bed boilers or double fluidized bed boilers for combustion or gasification of biomass, coal, or solid waste, the raw materials come from a wide range of sources and their types and compositions are relatively complex. In order to obtain higher gas production performance, it is necessary not only to coordinate the process controls of bottom air, primary air, secondary air, return air, and gasification air, but also to accurately control the correlation between the bed material circulation volume and the process conditions.

[0004] The inventors have found through research that the prior art method for optimizing the gas production performance of fluidized bed boilers still has at least the following defects:

[0005] At present, the process control based on different raw materials and current gas production results is carried out through manual experience, but the optimization effect on gas production performance is poor.

[0006] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0007] The purpose of the present invention is to be able to perform multi-dimensional analysis on the raw materials of product gas and improve the optimization effect of the gas production performance of the fluidized bed boiler.

[0008] The present invention provides a method for generating a gas production prediction model for a fluidized bed boiler, comprising the steps of:

[0009] S11, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0010] S12, preprocessing the working condition data in the preset data database, including:

[0011] The process control type data is updated, including:

[0012] respectively calculating the difference between each item of data in the first preset data and the corresponding standard data, and updating each item of data in the first preset data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the first preset data based on the set gas performance;

[0013] S13. Using various data of the operating condition data as characteristic data, a prediction model is established for the purpose of predicting gas performance.

[0014] Preferably, in an embodiment of the present invention, the preprocessing further includes:

[0015] The raw material type data is updated, including:

[0016] By re-encoding, each item of data in the raw material type data is converted into binary data.

[0017] Preferably, in an embodiment of the present invention, the preprocessing further includes:

[0018] A first dimensionality reduction process is performed on the raw material physical and chemical property type data and the process control type data.

[0019] Preferably, in an embodiment of the present invention, the preprocessing further includes:

[0020] The product gas information type data is updated, including:

[0021] According to the preset level classification rules, the levels corresponding to various information data of the product gas are determined respectively.

[0022] Preferably, in an embodiment of the present invention, the preprocessing further includes:

[0023] The raw material type data, as well as the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction processing are subjected to a second dimensionality reduction processing, and each data is updated to be a new feature data.

[0024] In another aspect of the present invention, there is also provided a device for generating a gas production prediction model for a fluidized bed boiler, comprising:

[0025] A data classification unit is used to classify the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0026] The preprocessing unit is used to preprocess the working condition data in the preset data database, including:

[0027] The process control type data is updated, including:

[0028] respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information;

[0029] The modeling unit is used to establish a prediction model for the purpose of predicting product gas information by taking various data of the operating condition data as feature data.

[0030] Preferably, in the embodiment of the present invention, the pre-processing unit further includes:

[0031] The second updating module is used to update the raw material type data, including:

[0032] By re-encoding, each item of data in the raw material type data is converted into binary data.

[0033] Preferably, in the embodiment of the present invention, the pre-processing unit further includes:

[0034] The third updating module is used to perform a first dimensionality reduction process on the raw material physical and chemical property type data and the process control type data.

[0035] Preferably, in the embodiment of the present invention, the pre-processing unit further includes:

[0036] The fourth updating module is used to update the product gas information type data, including:

[0037] According to the preset level classification rules, the levels corresponding to various information data of the product gas are determined respectively.

[0038] Preferably, in the embodiment of the present invention, the pre-processing unit further includes:

[0039] The fifth updating module is used to perform a second dimensionality reduction process on the raw material type data, the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction process, and update each data into new feature data.

[0040] On the other hand, an embodiment of the present invention further provides a fluidized bed boiler gas production prediction model generation device, which includes a computer program stored on a medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the methods described in the above aspects and achieves the same technical effects.

[0041] On the other hand, a storage medium is provided on which a computer program is stored. When the computer program is executed by a processor, each step of the method for generating a gas production prediction model for a fluidized bed boiler as described in any one of the above items is implemented.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The present invention provides a method for generating a gas production prediction model for a fluidized bed boiler. First, operating condition data are divided into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data include one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage; then the operating condition data are preprocessed and the process control type data are updated, including: respectively calculating the difference between each data in the process control type data and the corresponding standard data, and updating each data in the process control type data with the difference; determining the standard data includes, based on the set product gas information, determining the standard data corresponding to each data in the process control type data; finally, taking each data of the raw material data as feature data, a prediction model for the purpose of predicting product gas information is established.

[0044] Since the difference between each data in the process control type data and the corresponding standard data and the correspondence with the product gas information type data are clearer in the present invention, the prediction accuracy of the prediction model can be effectively improved by preprocessing the process control type data. Since the prediction model in the present invention is more accurate, the obtained control strategy is more effective and reasonable, thereby being able to obtain a better effect of optimizing gas performance.

[0045] The present invention conducts multi-dimensional analysis on process control type data of raw material processing, raw material type data and product gas information type data, and pre-processes the operating condition data; determines the correlation between various data of raw material data and product gas information, and establishes a high-accuracy prediction model, thereby improving the optimization effect on the gas production performance of the fluidized bed boiler, and thus being able to provide stronger guidance for optimizing gas performance.

[0046] In addition, the present invention is mainly based on feature data identification and prediction of product gas information, and establishes an optimal process control strategy that can quickly identify feature data and extract raw material processing, forming an intelligent decision-making method that integrates raw material property analysis, optimal process control strategy, product gas information and economic cost accounting.

[0047] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, and to make the above and other purposes, technical features and advantages of the present invention easier to understand, one or more preferred embodiments are listed below and described in detail with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 It is a step diagram of the method for generating a gas production prediction model for a fluidized bed boiler in the present invention;

[0050] Figure 2 It is a structural schematic diagram of a device for generating a gas production prediction model for a fluidized bed boiler in the present invention;

[0051] Figure 3 It is a structural schematic diagram of a fluidized bed boiler gas production prediction model generating device in the present invention. DETAILED DESCRIPTION

[0052] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings, but it should be understood that the protection scope of the present invention is not limited by the specific implementation modes.

[0053] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.

[0054] In this document, the terms "first", "second", etc. are used to distinguish two different elements or parts, and are not used to limit a specific position or relative relationship. In other words, in some embodiments, the terms "first", "second", etc. can also be interchangeable.

[0055] Embodiment 1

[0056] In order to improve the optimization effect of gas production performance of fluidized bed boiler, such as Figure 1 As shown, in an embodiment of the present invention, a method for generating a gas production prediction model for a fluidized bed boiler is provided, comprising the steps of:

[0057] S11, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0058] The inventive concept of the present invention includes: when using a fluidized bed boiler or a double fluidized bed reactor for biomass, coal, or solid waste combustion or gasification, the raw materials are widely sourced, and the types and components are complex. In order to obtain higher gas production performance, not only the bottom air, primary air, secondary air, return air and gasification air need to cooperate with each other, but also the correlation between the bed material circulation amount and the process conditions needs to be accurately controlled. Most of the existing prediction methods use single factor analysis methods, which are not comprehensive enough for the analysis of raw materials, resulting in poor optimization effect on gas performance, thereby bringing higher trial and error costs.

[0059] In order to solve the above problems, the present invention conducts multi-dimensional analysis on raw material type data, process control type data of raw material processing and product gas information type data, and pre-processes the operating condition data; determines the correlation between various data of raw material data and product gas information, and establishes a high-accuracy prediction model.

[0060] In an embodiment of the present invention, the operating condition data are classified, and the process control type data for raw material processing is set as the first preset data; wherein, the process control type data for raw material processing may specifically include the bottom air volume, primary air volume, secondary air volume, return air volume, furnace air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage corresponding to each process route.

[0061] S12, preprocessing the working condition data in the preset data database, including:

[0062] The process control type data is updated, including:

[0063] respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information;

[0064] The raw materials for combustion or gasification of biomass, coal, or solid waste using fluidized bed boilers or dual fluidized bed reactors include a large number of data items. If the original values ​​of these data items are used as feature data for model training and verification during modeling, the resulting prediction model will have poor prediction accuracy.

[0065] After research, the inventors found that some data items in the operating condition data, such as: process control type data for raw material processing, the numerical value of this data item for the raw material output gas can be determined to its optimal value based on the product gas information; for example, under a certain product gas information standard, the optimal value of each data item in the corresponding process control type data.

[0066] Next, in the embodiment of the present invention, for the set product gas information, the optimal value corresponding to the process control type data item is determined in advance, and the optimal value of the process control type data item is used as the standard data.

[0067] After the above settings, the specific process of preprocessing the process control type data may include:

[0068] For a certain data item, the difference between the data of the data item and the standard data is calculated according to the corresponding standard data, and the data of the corresponding data item is updated with the difference result, thereby replacing the original data; for example: the standard data for determining the temperature value in the process control type data is A, and the original temperature value obtained from the raw material data is B, and the difference between B and A is updated to the original temperature value.

[0069] Since the difference between the process control type data and the standard data has a clearer impact on the product gas information type data, the above preprocessing can effectively improve the prediction accuracy of the prediction model.

[0070] Furthermore, in an embodiment of the present invention, updating the process control type data may also include: normalizing the covariance of the process control type data, eliminating data with high correlation, performing a first dimensionality reduction process, performing a correlation coefficient analysis on each data in the newly obtained process control type data, the correlation coefficient analysis is the normalized covariance between each data, eliminating data with high correlation, and performing preliminary dimensionality reduction processing.

[0071] Furthermore, in order to further improve the accuracy of the prediction model, the preprocessing process in the embodiment of the present invention may also include: updating the raw material type data, including:

[0072] By re-encoding, each item of data in the raw material type data is converted into binary data.

[0073] In the implementation of the present invention, the raw material type data may include data items such as coal from different regions, different types of biomass such as wood and straw, urban solid waste sludge and mixed raw materials.

[0074] Since the data conditions of raw material types are relatively fixed and are all non-numerical data items, they are re-encoded according to one-hot encoding and each data item is converted into binary data.

[0075] Furthermore, in order to further improve the accuracy of the prediction model, in an embodiment of the present invention, the preprocessing process may also include: performing a first dimensionality reduction processing on the raw material physical and chemical property type data and the process control type data; the raw material physical and chemical property type data may include data items such as density, carbon-hydrogen ratio, elemental composition, volatile matter, porosity, water content and fixed carbon.

[0076] Furthermore, in an embodiment of the present invention, the preprocessing may also include: updating the product gas information type data, including: determining the levels corresponding to each item of product gas information data respectively according to a preset level classification rule.

[0077] Since whether the product gas information meets the standard is determined based on whether each indicator is within a set range, in order to reduce the difficulty of correspondence between the independent variable data and the dependent variable data in the prediction model and reduce the difficulty of modeling the prediction model, in the embodiment of the present invention, a method of characterizing the data of each indicator (i.e., each data) of the product gas by setting a grade score is also adopted. Specifically:

[0078] For each type of product gas data item, a corresponding classification interval is set. For example, for the gas calorific value data item, it can be divided into multiple classifications. In this way, the corresponding level can be determined according to the specific value of the product gas type data; similarly, other data items such as gas flow rate and gas nitrogen oxide concentration can also be divided into multiple classifications.

[0079] Among them, each item of product gas information in the product gas information type data item may specifically include: gas temperature, gas flow, gas composition, gas calorific value, gas tar content and gas nitrogen oxide concentration, etc.

[0080] It should be noted that, in the embodiment of the present invention, the number of classifications of the data items of the product gas information can be determined by those skilled in the art according to needs and experience, and is not specifically limited here.

[0081] Furthermore, in an embodiment of the present invention, the preprocessing also includes: performing a second dimensionality reduction processing on the raw material source type data, as well as the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction processing, and updating each data into new feature data; in actual applications, the principal component analysis PCA method can be used to further reduce the dimensionality of the above data to obtain new feature data for each item after reduction.

[0082] S13. Using various data of the operating condition data as characteristic data, a prediction model is established for the purpose of predicting product gas information.

[0083] In the embodiment of the present invention, the prediction model may select a neural network model or an integrated learning model according to the actual application; the neural network model or the integrated learning model is established according to the preprocessed data.

[0084] The prediction model is trained by the constructed training set, and the scores of each data item of the product gas information are output; the accuracy of the prediction model is verified by the constructed test set; in the embodiment of the present invention, the data set is composed of the new feature data and the scores of the product gas information data items; and the data set is reasonably divided into a training set and a test set.

[0085] The prediction model is machine-learned through the training set, and the prediction model outputs the predicted values ​​of various information items of the gas produced by the raw materials under various process control data. The accuracy of the prediction model is then continuously cross-validated through the test set to obtain a prediction model with high accuracy.

[0086] In the embodiment of the present invention, the prediction accuracy of the prediction model is improved to accurately predict the product gas information, and the product gas information can characterize the gas production performance; thus, in practical applications, according to the type of raw materials and the physical and chemical properties of the current raw materials, the prediction model can be used to determine the corresponding control strategy with the goal of obtaining the expected product gas performance; it should be noted that in the embodiment of the present invention, the control strategy value is a coordinated setting scheme of the control quantities of various control processes (such as bottom air volume, primary air volume, secondary air volume and return air volume, etc.).

[0087] Since the prediction model in the embodiment of the present invention is more accurate, the resulting control strategy is more effective and reasonable, such as obtaining the best process control type data (such as bottom air volume, primary air volume, secondary air volume and return air volume, etc.).

[0088] The following is a specific example to illustrate the effectiveness of the prediction model in the embodiment of the present invention in optimizing the gas production performance of the fluidized bed boiler. For example, the forest waste produced by a forest farm in Heilongjiang Province, after analysis and testing by the experimenters, the data of the type of raw materials and the type of physical and chemical properties of the raw materials and other data are submitted to the prediction system established according to the fluidized bed boiler in Heilongjiang Province (the prediction system includes the above prediction model). The prediction system can query that such raw materials are existing raw materials that have been processed in the operation history of the fluidized bed boiler, and output the process control information corresponding to the different gas calorific values ​​or hydrogen-carbon ratios of the raw material output gas. The decision maker finally selected the process control type data corresponding to the gas calorific value of 7 based on the process control type data and economic cost. The prediction system used the process control type data as the target process condition and sent adjustment instructions for the bottom air, primary air and secondary air according to the preset air volume and the corresponding adjustment speed. After all the adjustment instructions were reviewed by mature process engineers, the fluidized bed boiler accepted the instructions and operated according to the target process conditions. Compared with manually adjusting the process control type data, the calorific value of the gas generated was increased by 8.4%, thereby verifying the effectiveness of the prediction model in the embodiment of the present invention in optimizing the gas production performance of the fluidized bed boiler.

[0089] In addition, the prediction system implemented in the double fluidized bed boiler also illustrates the effectiveness of the prediction model in the embodiment of the present invention in optimizing the gas production performance of the fluidized bed boiler. For example, a mixture of solid waste and forest waste generated in a certain place in Liaoning Province, after analysis and testing by the experimenters, the raw material type data and the raw material physical and chemical property type data and other data are submitted to the prediction system established according to the double fluidized bed boiler in a certain place in Liaoning Province (the prediction system includes the above-mentioned prediction model). The prediction system can query that such raw materials are existing raw materials that have been processed in the operation history of the double fluidized bed boiler. Decision makers search according to the required gas information type data and hydrogen concentration, and the prediction system affects the physical and chemical property type data of the raw material. The raw material data items with greater hydrogen concentration are matched, and the prediction system outputs the process control type data and economic cost corresponding to the raw material data with the highest matching degree. The decision maker finally chooses to adjust according to the process control type data. The prediction system uses the process control type data as the target process condition and sends adjustment instructions for the bottom air, primary air and secondary air at the adjustment speed corresponding to the preset air volume. After all the adjustment instructions are reviewed by mature process engineers, the dual fluidized bed boiler accepts the instructions and operates according to the target process conditions. Compared with manually adjusting the process control type data, the hydrogen concentration is increased by 10%, which also verifies the effectiveness of the prediction model in the embodiment of the present invention for optimizing the gas production performance of the fluidized bed boiler.

[0090] In summary, the present invention provides a method for generating a fluidized bed boiler gas production prediction model, firstly, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage; then the operating condition data is preprocessed and the process control type data is updated, including: respectively calculating the difference between each data in the process control type data and the corresponding standard data, and updating each data in the process control type data with the difference; determining the standard data includes, based on the set product gas information, determining the standard data corresponding to each data in the process control type data; finally, taking each data of the raw material data as feature data, a prediction model for the purpose of predicting product gas information is established.

[0091] Since the correspondence between the difference between each data in the process control type data and the corresponding standard data and the gas performance is clearer in the present invention, the prediction accuracy of the prediction model can be effectively improved by preprocessing the process control type data. Since the prediction model in the embodiment of the present invention is more accurate, the obtained control strategy is more effective and reasonable, thereby being able to obtain a better effect of optimizing gas performance.

[0092] The present invention conducts multi-dimensional analysis on process control type data of raw material processing, raw material type data and product gas information type data, and pre-processes the operating condition data; determines the correlation between various data of raw material data and product gas information, and establishes a high-accuracy prediction model, thereby improving the optimization effect on the gas production performance of the fluidized bed boiler, and thus being able to provide stronger guidance for optimizing gas performance.

[0093] In addition, the present invention is mainly based on feature data identification and prediction of product gas information, and establishes an optimal process control strategy that can quickly identify feature data and extract raw material processing, forming an intelligent decision-making method that integrates raw material property analysis, optimal process control strategy, product gas information and economic cost accounting.

[0094] Embodiment 2

[0095] Corresponding to the method embodiment, in another aspect of the embodiment of the present invention, a device for generating a gas production prediction model for a fluidized bed boiler is provided. Figure 2 The schematic diagram of the structure of the fluidized bed boiler gas production prediction model generation device provided by the embodiment of the present invention is shown. The fluidized bed boiler gas production prediction model generation device is Figure 1 The device corresponding to the method for generating a gas production prediction model for a fluidized bed boiler described in the corresponding embodiment, that is, the method is implemented by a virtual device Figure 1In the corresponding embodiment, the method for generating a gas production prediction model for a fluidized bed boiler, each virtual module constituting the device for generating a gas production prediction model for a fluidized bed boiler can be executed by an electronic device, such as a network device, a terminal device or a server. Specifically, the device for generating a gas production prediction model for a fluidized bed boiler in the embodiment of the present invention includes:

[0096] The data classification unit 01 is used to classify the working condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0097] The preprocessing unit 02 is used to preprocess the working condition data in the preset data database, including:

[0098] The process control type data is updated, including:

[0099] respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information;

[0100] In the embodiment of the present invention, the optimal value corresponding to the process control type data item is determined in advance, and the optimal value of the process control type data item is used as standard data.

[0101] After the above settings, the specific process of preprocessing the process control type data may include:

[0102] For a certain data item, the difference between the data of the data item and the standard data is calculated according to the corresponding standard data, and the data of the corresponding data item is updated with the difference result, thereby replacing the original data; for example: the standard data for determining the temperature value in the process control type data is A, and the original temperature value obtained from the raw material data is B, and the difference between B and A is updated to the original temperature value.

[0103] Since the difference between the process control type data and the standard data has a clearer impact on the product gas information, the above preprocessing can effectively improve the prediction accuracy of the prediction model.

[0104] Furthermore, in the embodiment of the present invention, the preprocessing unit 02 may also include: a second updating module, used to update the raw material type data, including: converting each item of data in the raw material type data into binary data by re-encoding.

[0105] Raw material type data may include: coal from different regions, different types of biomass such as wood and straw, urban solid waste sludge and mixed raw materials, etc.

[0106] Since the data conditions of raw material types are relatively fixed and are all non-numerical data items, they are re-encoded according to one-hot encoding and each data item is converted into binary data.

[0107] Furthermore, in an embodiment of the present invention, the preprocessing unit 02 may also include: a third updating module, used to perform a first dimensionality reduction processing on the raw material physical and chemical property type data and the process control type data; wherein the raw material physical and chemical property type data may include data items such as density, carbon-hydrogen ratio, elemental composition, volatile matter, porosity, water content and fixed carbon.

[0108] Furthermore, in the embodiment of the present invention, the preprocessing unit 02 may further include: a fourth updating module, configured to update the fourth preset data, including:

[0109] Used to update the product gas information type data, including:

[0110] According to the preset level classification rules, the levels corresponding to various information data of the product gas are determined respectively.

[0111] Since whether the product gas information meets the standard is determined based on whether each indicator is within a set range, in order to reduce the difficulty of correspondence between the independent variable data and the dependent variable data in the prediction model and reduce the difficulty of modeling the prediction model, in the embodiment of the present invention, a method of characterizing the data of each indicator (i.e., each data) of the product gas by setting a grade score is also adopted. Specifically:

[0112] For each type of product gas data item, a corresponding classification interval is set. For example, for the gas calorific value data item, it can be divided into multiple classifications. In this way, the corresponding level can be determined according to the specific value of the product gas type data; similarly, other data items such as gas flow rate and gas nitrogen oxide concentration can also be divided into multiple classifications.

[0113] Furthermore, in the embodiment of the present invention, the preprocessing unit 02 may further include:

[0114] The fifth updating module is used to perform a second dimensionality reduction process on the raw material type data, the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction process, and update each data into new feature data.

[0115] The modeling unit 03 is used to establish a prediction model for predicting product gas information by taking various data of the operating condition data as feature data.

[0116] In an embodiment of the present invention, the prediction model is trained by using a constructed training set to output the scores of each data item of the product gas information; the accuracy of the prediction model is verified by using a constructed test set; in an embodiment of the present invention, a data set is formed by using new feature data and the scores of the product gas information data items; and the data set is reasonably divided into a training set and a test set.

[0117] The prediction model is machine-learned through the training set, and the prediction model outputs the scores of various information items of the gas produced by the raw materials under various process control data. The accuracy of the prediction model is then continuously cross-validated through the test set to obtain a prediction model with high accuracy.

[0118] It should be noted that the specific implementation method and technical effects of the fluidized bed boiler gas production prediction model generation device in the embodiment of the present invention can be referred to Figure 1 The corresponding method for generating the fluidized bed boiler gas production prediction model will not be described in detail here.

[0119] Embodiment 3

[0120] Corresponding to the method embodiment, in the embodiment of the present invention, a fluidized bed boiler gas production prediction model generation device is also provided, such as a terminal, a server, etc. Among them, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto.

[0121] An example diagram of a hardware structure block diagram of a fluidized bed boiler gas production prediction model generation device provided in an embodiment of the present invention is shown in FIG. Figure 3 As shown, it may include:

[0122] Processor 1, communication interface 2, memory 3 and communication bus 4;

[0123] The processor 1, the communication interface 2, and the memory 3 communicate with each other via the communication bus 4;

[0124] Optionally, the communication interface 2 may be an interface of a communication module, such as an interface of a GSM module;

[0125] The processor 1 may be a central processing unit CPU, or an application-specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0126] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0127] The processor 1 is specifically used to execute the computer program stored in the memory 3 to perform the following steps:

[0128] S11, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0129] S12, preprocessing the working condition data in the preset data database, including:

[0130] The process control type data is updated, including:

[0131] respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information;

[0132] S13. Using various data of the operating condition data as characteristic data, a prediction model is established for the purpose of predicting product gas information.

[0133] The above product can execute the method provided in the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method for generating a gas production prediction model for a fluidized bed boiler provided in the embodiment of the present invention.

[0134] Embodiment 4

[0135] In an embodiment of the present invention, a storage medium is further provided, which may store a program suitable for execution by a processor, wherein the program is used to:

[0136] S11, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage;

[0137] S12, preprocessing the working condition data in the preset data database, including:

[0138] The process control type data is updated, including:

[0139] respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information;

[0140] S13. Using various data of the operating condition data as characteristic data, a prediction model is established for the purpose of predicting product gas information.

[0141] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0142] The above product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the methods provided by other embodiments of the present invention.

[0143] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0145] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0146] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0147] It should be understood that in the embodiments of the present application, the various embodiments and features can be combined with each other to solve the aforementioned technical problems.

[0148] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0149] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a gas production prediction model for a fluidized bed boiler, characterized in that: Includes steps: S11, dividing the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; the process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage; S12, preprocessing the working condition data in the preset data database, including: The process control type data is updated, including: respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information; S13. Using various data of the operating condition data as characteristic data, a prediction model is established for the purpose of predicting product gas information.

2. The method for generating a gas production prediction model for a fluidized bed boiler according to claim 1, characterized in that: The preprocessing further comprises: The raw material type data is updated, including: By re-encoding, each item of data in the raw material type data is converted into binary data.

3. The method for generating a gas production prediction model for a fluidized bed boiler according to claim 2, characterized in that: The preprocessing further comprises: A first dimensionality reduction process is performed on the raw material physical and chemical property type data and the process control type data.

4. The method for generating a gas production prediction model for a fluidized bed boiler according to claim 3, characterized in that: The preprocessing further comprises: The product gas information type data is updated, including: According to the preset level classification rules, the levels corresponding to various information data of the product gas are determined respectively.

5. The method for generating a gas production prediction model for a fluidized bed boiler according to claim 4, characterized in that: The preprocessing further comprises: The raw material type data, as well as the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction processing are subjected to a second dimensionality reduction processing, and each data is updated to be a new feature data.

6. A fluidized bed boiler gas production prediction model generation device, characterized in that: include: A data classification unit, used to classify the operating condition data into process control type data, raw material type data, raw material physical and chemical property type data and product gas information type data; The process control type data includes one or more data of bottom air volume, primary air volume, secondary air volume, return air volume, furnace temperature, furnace pressure, bed material circulation volume and gasification medium dosage; The preprocessing unit is used to preprocess the working condition data in the preset data database, including: The process control type data is updated, including: respectively calculating the difference between each item of data in the process control type data and the corresponding standard data, and updating each item of data in the process control type data with the difference; determining the standard data includes determining the standard data corresponding to each item of data in the process control type data based on the set product gas information; The modeling unit is used to establish a prediction model for the purpose of predicting product gas information by taking various data of the operating condition data as feature data.

7. The device for generating a gas production prediction model for a fluidized bed boiler according to claim 6, characterized in that: The pre-processing unit further includes: The second updating module is used to update the raw material type data, including: By re-encoding, each item of data in the raw material type data is converted into binary data.

8. The device for generating a gas production prediction model for a fluidized bed boiler according to claim 7, characterized in that: The pre-processing unit further includes: The third updating module is used to perform a first dimensionality reduction process on the raw material physical and chemical property type data and the process control type data.

9. The device for generating a gas production prediction model for a fluidized bed boiler according to claim 8, characterized in that: The pre-processing unit further includes: The fourth updating module is used to update the product gas information type data, including: According to the preset level classification rules, the levels corresponding to various information data of the product gas are determined respectively.

10. The device for generating a gas production prediction model for a fluidized bed boiler according to claim 9, characterized in that: The pre-processing unit further includes: The fifth updating module is used to perform a second dimensionality reduction process on the raw material type data, the process control type data and the raw material physical and chemical property type data after the first dimensionality reduction process, and update each data into new feature data.

11. A fluidized bed boiler gas production prediction model generation device, characterized in that: include: Memory for storing computer programs; A processor is used to call and execute the computer program to implement the steps of the method for generating a fluidized bed boiler gas production prediction model as described in any one of claims 1 to 5.

12. A storage medium, characterized in that: The method comprises a software program, wherein the software program is suitable for a processor to execute the steps of the method for generating a gas production prediction model for a fluidized bed boiler as claimed in any one of claims 1 to 5.

Images