Method, device, equipment, medium and product for determining optimal oxygen concentration at inlet of high-temperature economizer of circulating fluidized bed boiler

Abstract

This application discloses a method, apparatus, equipment, medium, and product for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler, relating to the field of CFB boiler parameter optimization. The method includes: acquiring historical data; preprocessing the historical data; constructing a machine learning model; training the machine learning model based on the preprocessed historical data; inputting the historical data into the trained machine learning model to obtain a predicted value of thermal efficiency; constructing an objective function based on the predicted thermal efficiency; and solving the objective function to obtain the optimal oxygen concentration. This application can determine the optimal oxygen concentration at the inlet of the high-temperature economizer in a CFB boiler, ensuring complete fuel combustion and reducing the emission of harmful substances.

Description

Technical Field

[0001] This application relates to the field of CFB boiler parameter optimization, and in particular to a method, apparatus, equipment, medium and product for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler. Background Technology

[0002] Circulating fluidized bed (CFB) boilers are an advanced boiler technology that can burn coal, oil, or other solid fuels. Compared with traditional combustion technologies, CFB boilers use a fluidized bed as the combustion medium, offering advantages such as high combustion efficiency, low emissions, and high flexibility, making them valuable for energy utilization and environmental protection. Optimizing CFB boiler parameters to achieve energy conservation, emission reduction, and improved efficiency is an important research direction in the current engineering field.

[0003] During the operation of CFB boilers, the oxygen concentration at the inlet of the high-temperature economizer has a significant impact on combustion efficiency and environmental emissions. Excessive oxygen concentration leads to overcombustion, fuel waste, and increased harmful emissions, while insufficient oxygen concentration results in incomplete combustion and the generation of toxic gases, thereby reducing boiler performance, increasing environmental pollution, and posing safety risks. Traditionally, operators set target oxygen levels based on their experience and understanding of boiler operation. This method typically determines the oxygen setpoint based on current boiler operating conditions, such as load, coal quantity, primary air volume, and secondary air volume, making it difficult to achieve efficient combustion. Therefore, finding a technical solution that utilizes data acquisition and machine learning methods to determine the optimal oxygen setpoint is crucial for improving boiler performance and environmental benefits. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment, medium, and product for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler, which can determine the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler, ensuring complete combustion of fuel and reducing the emission of harmful substances.

[0005] To achieve the above objectives, this application provides the following solution:

[0006] Firstly, this application provides a method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler, including:

[0007] Obtain historical data;

[0008] The historical data is preprocessed;

[0009] Build machine learning models;

[0010] The machine learning model is trained based on preprocessed historical data;

[0011] Historical data is input into a trained machine learning model to obtain predicted values ​​for thermal efficiency.

[0012] Construct an objective function based on the predicted value of the thermal efficiency;

[0013] The objective function is solved to obtain the optimal oxygen concentration.

[0014] Optionally, the historical data includes: coal feed flow rate, main steam flow rate, feed water pressure, feed water temperature, main steam temperature, main steam pressure, O2 concentration at the inlet of the high-temperature economizer, total primary air volume, total secondary air volume, secondary air fan outlet air pressure, pressure at the top of the combustion chamber, and pressure at the middle of the combustion chamber.

[0015] Optionally, preprocessing the historical data specifically includes: data filtering, missing value handling, outlier detection and handling, and smoothing.

[0016] Optionally, the machine learning model is one of CatBoost, LSTM and Transformer, wherein the structure of the LSTM includes: an input gate, a forget gate and an output gate, as well as cell state updates;

[0017] Specifically, the following steps are included:

[0018] Calculate the forget gate and select the information to be forgotten;

[0019] Calculate the memory gate and select the information to be memorized;

[0020] Calculate the cell state at the current moment;

[0021] Calculate the output gate and the hidden state at the current time.

[0022] Optionally, the calculation of the forget gate selects the information to be forgotten, specifically using the following formula:

[0023] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0024] Among them, f t The output of the forget gate represents the percentage of the cell state at the current time that needs to be forgotten. σ represents the sigmoid activation function, and W... f h represents the weight matrix of the forget gate. t-1 Let x represent the hidden state at the previous time step. t b represents the input at the current moment. f Indicates the bias of the forget gate;

[0025] Calculate the memory gate, select the information to be memorized, and use the following formula.

[0026] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0027]

[0028] Among them, i t The output of the memory gate indicates how much new information will be written to the cell state at the current moment; W i The weight matrix represents the memory gates, used to weight the hidden state from the previous time step and the current input; b i This indicates the bias of the memory gate, used to adjust the output of the memory gate; W represents the candidate cell state, which is a new candidate state generated based on the current input and past information. c The weight matrix representing the candidate cell state is used to weight the hidden state from the previous time step and the current input; b c This represents the bias of the candidate cell state, used to adjust the output of the candidate cell state;

[0029] The cell state at the current moment is calculated using the following formula.

[0030]

[0031] Among them, C t This represents the current cell state, used for long-term information storage. (C) t-1 This represents the cell state at the previous moment, and the information stored in the LSTM unit at the previous moment.

[0032] The output gate and the current hidden state are calculated using the following formula:

[0033] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0034] h t =o t ·tanh(C t )

[0035] Among them, O t The output of the output gate determines which part of the current cell state will use the output as the hidden state h. t W ob represents the weight matrix of the output gate, used to weight the hidden state from the previous time step with the current input; o The output gate's bias is indicated by h, which is used to adjust the output of the gate. t This indicates the hidden state at the current moment.

[0036] Optionally, the expression for the objective function is as follows:

[0037] j = w p (1-y p (k)) 2 +w k k 2

[0038] Where j represents the value of the objective function, w p For prediction weighting, w k To control the weighting, y p For the thermal efficiency prediction result, k is the oxygen content setpoint. By traversing the range of k values, j is minimized.

[0039] Secondly, this application provides a device for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer, the device comprising:

[0040] The historical data acquisition module is used to acquire historical data;

[0041] The preprocessing module is used to preprocess the historical data;

[0042] The machine learning model building module is used to build machine learning models;

[0043] The training module is used to train the machine learning model based on preprocessed historical data;

[0044] The thermal efficiency prediction module is used to input historical data into a trained machine learning model to obtain predicted values ​​of thermal efficiency.

[0045] The objective function construction module is used to construct an objective function based on the predicted value of the thermal efficiency;

[0046] The objective function solving module is used to solve the objective function to obtain the optimal oxygen concentration.

[0047] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for determining the optimal oxygen concentration at the inlet of the CFB boiler high-temperature economizer as described in any of the above.

[0048] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler as described above.

[0049] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler as described above.

[0050] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0051] This application provides a method, apparatus, equipment, medium, and product for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler. The method involves: acquiring historical data; preprocessing the historical data; constructing a machine learning model; training the machine learning model based on the preprocessed historical data; inputting the historical data into the trained machine learning model to obtain a predicted value of thermal efficiency; constructing an objective function based on the predicted thermal efficiency; and solving the objective function to obtain the optimal oxygen concentration. It is evident that this application, through machine learning algorithms, can respond in real-time to changes in the operating status of the CFB boiler, thereby maximizing combustion efficiency and reducing environmental emissions. It exhibits higher adaptability and optimization effects, and compared to traditional experience-based methods, it can more effectively improve boiler performance and environmental benefits. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is an application environment diagram of a method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler, as described in one embodiment of this application.

[0054] Figure 2 A flowchart illustrating a method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler, provided as an embodiment of this application;

[0055] Figure 3 This is a schematic diagram of the hidden layer structure of an LSTM provided in an embodiment of this application;

[0056] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0057] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] The method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer in a CFB boiler, provided in this application embodiment, can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be set up independently, integrated into server 104, or placed in the cloud or on another server. Terminal 102 can send the data to be processed to server 104. After receiving the data, server 104 performs the following steps: acquires historical data; preprocesses the historical data; constructs a machine learning model; trains the machine learning model based on the preprocessed historical data; inputs the historical data into the trained machine learning model to obtain a predicted value of thermal efficiency; constructs an objective function based on the predicted thermal efficiency; and solves the objective function to obtain the optimal oxygen concentration. Server 104 can then feed back the obtained optimal oxygen concentration to terminal 102. In addition, in some embodiments, the method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of the CFB boiler can also be implemented by the server 104 or the terminal 102 separately. For example, the terminal 102 can directly process the data to be processed, or the server 104 can obtain the data to be processed from the data storage system and process the data to be processed.

[0060] The terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0061] In one exemplary embodiment, such as Figure 2As shown, a method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 207. Wherein:

[0062] Step 201: Obtain historical data.

[0063] The data collection involves acquiring massive amounts of historical data from the MIS systems of various thermal power plants. This historical data includes: coal feed flow rate, main steam flow rate, feedwater pressure, feedwater temperature, main steam temperature, main steam pressure, O2 concentration at the inlet of the high-temperature economizer, total primary air volume, total secondary air volume, secondary air fan outlet air pressure, pressure at the top of the combustion chamber, and pressure at the middle of the combustion chamber.

[0064] The data collection interval should be less than 1 minute to avoid losing important information during modeling due to large data intervals. The data collection period should be at least 1 month and should be the most recent time.

[0065] Step 202: Preprocess the historical data.

[0066] Specifically, it includes:

[0067] Data processing includes data filtering, missing value handling, outlier detection and handling, smoothing, and normalization.

[0068] The "thermal efficiency" column is also calculated as follows:

[0069]

[0070] In addition, a single boiler typically has multiple coal feeders, each with a coal feeder flow measurement point. The total coal feed is the sum of the coal feed flows of all coal feeders in a single boiler. The boiler evaporation is the boiler main steam flow. The steam enthalpy and feedwater enthalpy are calculated and obtained from tables using the main steam pressure, main steam temperature, feedwater pressure, and feedwater temperature. The lower heating value of the fuel is obtained from on-site sampling.

[0071] Step 203: Build a machine learning model.

[0072] The specific models used include CatBoost, LSTM, and Transformer.

[0073] Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs) are two important variants of Recurrent Neural Networks (RNNs). They address the long-range dependency problem of traditional RNNs by introducing gating mechanisms and cell states. LSTMs can transmit long-range information within the network, thus capturing temporal dependencies in sequences. The core of LSTM is the cell state, which runs throughout the entire sequence and transmits long-range information within the network. The structure of an LSTM includes an input gate, a forget gate, an output gate, and cell state updates.

[0074] The hidden layer structure of LSTM is as follows Figure 3 As shown, it includes: input gate, forget gate, and output gate, as well as cell state updates.

[0075] Step 204: Train the machine learning model based on the preprocessed historical data.

[0076] Specifically, the following steps are included:

[0077] By using historical data columns as input and "thermal efficiency" column as output, a machine learning model can be built, which can be achieved through data preprocessing, feature engineering, and model parameter optimization.

[0078] ① Data preprocessing:

[0079] Data cleaning: Handling missing and outlier values, using interpolation to fill in missing data, or replacing missing values ​​with the mean or median.

[0080] Feature selection: Select the features most relevant to the target variable, and use the SelectKBest method to select the top K best features.

[0081] ② Feature engineering:

[0082] Feature enhancement: Create new features or transform existing features to enhance model performance, including logarithmic transformation, normalization, and multinomial feature generation.

[0083] Dimensionality reduction: Reduce the number of features to remove redundant or noisy features. Principal component analysis (PCA) is used to reduce features to a lower dimension.

[0084] ③ Model parameter optimization:

[0085] Hyperparameter tuning: Adjust the model's hyperparameters to find the optimal configuration, using grid search or random search to optimize the hyperparameters.

[0086] Adjusting the model architecture: Modifying the model's structure to improve performance, such as increasing or decreasing the number of hidden layers in a neural network.

[0087] Optimization algorithm selection: Choose a suitable optimization algorithm to speed up convergence. In deep learning, choose the Adam optimizer instead of SGD.

[0088] Regularization: Use regularization techniques to prevent overfitting, such as L2 regularization in regression models or Dropout layers in neural networks.

[0089] Finally, keep the pkl or hdf5 file after the model training is completed.

[0090] Step 205: Input historical data into the trained machine learning model to obtain the predicted value of thermal efficiency.

[0091] Specifically, the following steps are included:

[0092] Calculate the forget gate and select the information to be forgotten;

[0093] Input the hidden state h from the previous time step t-1 The input word at the current moment is x t Output the value f of the forget gate. t This can be seen from the lower right corner of the structure diagram.

[0094] Calculate the memory gate and select the information to be memorized;

[0095] Input the hidden state h from the previous time step t-1 The input word at the current moment is x t Output the value of the memory gate i t and temporary cell state

[0096] Calculate the cell state at the current moment;

[0097] Input the value of the memory gate i t The value f of the forget gate t Temporary cell state Previous cell state C t-1 Output the current cell state C. t .

[0098] Calculate the output gate and the hidden state at the current time.

[0099] Input the hidden state h from the previous time step t-1 The input word at the current moment is x t The current cell state C t Output the value of the output gate. t and hidden state h t .

[0100] The forgetting gate is calculated by selecting the information to be forgotten, using the following formula:

[0101] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0102] Among them, f t The output of the forget gate indicates how much of the cell's current state needs to be forgotten; its value ranges from 0 to 1, with values ​​closer to 0 indicating more forgetting. σ is the sigmoid activation function, which compresses the result of the linear combination to the range of 0 to 1, determining information retention and forgetting. W f h is the weight matrix for the forget gate, used to weight the hidden state from the previous time step and the current input; t-1 The hidden state from the previous time step is the output of the LSTM unit from the previous time step, a compressed representation containing past temporal information; x t b represents the input at the current time, indicating the external input data at the current time. f This is the bias of the forget gate, used to adjust the output of the forget gate.

[0103] The memory gate is used to select the information to be memorized, specifically using the following formula:

[0104] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0105]

[0106] Among them, i t The output of the memory gate indicates how much new information will be written to the cell state at the current moment. Its value ranges from 0 to 1; W i b is the weight matrix for the memory gates, used to weight the hidden state from the previous time step and the current input; i This is the bias setting for the memory gate, used to adjust its output. denoted as candidate cell state, representing a new candidate state generated based on the current input and past information, with values ​​ranging from -1 to 1; tanh is the hyperbolic tangent activation function, used to compress the linear combination result to the range of -1 to 1, enabling LSTM to better handle nonlinear relationships; W c b is the weight matrix for the candidate cell states, used to weight the hidden state from the previous time step and the current input; cThis is a bias for the candidate cell state, used to adjust the output of the candidate cell state.

[0107] The current cell state is calculated using the following formula:

[0108]

[0109] Among them, C t Representing the current cell state, this is the most important part of LSTM, used for long-term information storage. It combines past cell states C... t-1 and the candidate state at the current moment C t-1 This represents the cell state at the previous moment, and the information stored in the LSTM unit at the previous moment.

[0110] The output gate and the current hidden state are calculated using the following formula:

[0111] o t =σ(W o ·[h t-1 ,x t ]+b o )

[0112] h t =o t ·tanh(C t )

[0113] Among them, O t The output of the output gate determines which part of the current cell state will use the output as the hidden state ht; W o b is the weight matrix for the output gate, used to weight the hidden state from the previous time step and the current input; o The output gate's bias is used to adjust its output; h t The hidden state at the current time step is the output of the LSTM unit and can also be used as the input for the next time step. It is the cell state C. t The output is a result after filtering a portion of the information.

[0114] Step 206: Construct an objective function based on the predicted value of the thermal efficiency.

[0115] Based on the constructed model, historical data is used as fixed input, and oxygen content is used as a variable. All predicted results are then output, representing the predicted thermal efficiency values ​​for different oxygen contents under the current operating conditions. The objective function for optimization is expressed as follows:

[0116] j = w p (1-y p (k)) 2 +w k k2

[0117] Where j represents the value of the objective function, w p For prediction weighting, w k To control the weighting, y p For the thermal efficiency prediction result, k is the oxygen content setpoint. By traversing the range of k values, j is minimized.

[0118] When the predicted output j is at its minimum, the input oxygen setting value is the optimal value and can be used as the oxygen setting value under the current operating conditions.

[0119] Step 207: Solve the objective function to obtain the optimal oxygen concentration.

[0120] Based on the same inventive concept, this application also provides a device for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a boiler, used to implement the aforementioned method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a boiler. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the device for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a boiler provided below can be found in the limitations of the method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a boiler described above, and will not be repeated here.

[0121] In one exemplary embodiment, a device for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer is provided, comprising:

[0122] The historical data acquisition module is used to acquire historical data;

[0123] The preprocessing module is used to preprocess the historical data;

[0124] The machine learning model building module is used to build machine learning models;

[0125] The training module is used to train the machine learning model based on preprocessed historical data;

[0126] The thermal efficiency prediction module is used to input historical data into a trained machine learning model to obtain predicted values ​​of thermal efficiency.

[0127] The objective function construction module is used to construct an objective function based on the predicted value of the thermal efficiency;

[0128] The objective function solving module is used to solve the objective function to obtain the optimal oxygen concentration.

[0129] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data for determining the optimal oxygen concentration at the inlet of the high-temperature economizer in a CFB boiler. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer in a CFB boiler.

[0130] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0131] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0132] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0133] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0134] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0135] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0136] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0137] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for determining the optimal oxygen concentration at the inlet of a high-temperature economizer in a CFB boiler, characterized in that, The methods for determining the optimal oxygen concentration at the inlet of the high-temperature economizer in a CFB boiler include: Obtain historical data; The historical data is preprocessed; Build machine learning models; The machine learning model is trained based on preprocessed historical data; Historical data is input into a trained machine learning model to obtain predicted values ​​for thermal efficiency. Construct an objective function based on the predicted value of the thermal efficiency; The objective function is solved to obtain the optimal oxygen concentration; The expression for the objective function is as follows: ； in, This represents the value of the objective function. For prediction weighting, To control the weighting, For the thermal efficiency prediction results, Set a value for oxygen content by iterating through... Value range, to minimize .

2. The method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler according to claim 1, characterized in that, The historical data includes: coal feed flow rate, main steam flow rate, feed water pressure, feed water temperature, main steam temperature, main steam pressure, O2 concentration at the inlet of the high-temperature economizer, total primary air volume, total secondary air volume, secondary air fan outlet air pressure, pressure at the top of the combustion chamber, and pressure at the middle of the combustion chamber.

3. The method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler according to claim 1, characterized in that, The preprocessing of the historical data specifically includes: data filtering, missing value handling, outlier detection and handling, and smoothing.

4. The method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler according to claim 1, characterized in that, The machine learning model is one of CatBoost, LSTM and Transformer. The structure of the LSTM includes: an input gate, a forget gate and an output gate, as well as cell state updates. Specifically, the following steps are included: Calculate the forget gate and select the information to be forgotten; Calculate the memory gate and select the information to be memorized; Calculate the cell state at the current moment; Calculate the output gate and the hidden state at the current time.

5. The method for determining the optimal oxygen concentration at the inlet of the high-temperature economizer of a CFB boiler according to claim 4, characterized in that, The calculation of the forgetting gate selects the information to be forgotten, specifically using the following formula: ； in, The output of the forget gate indicates how much of the cell's current state needs to be forgotten. This represents the Sigmoid activation function. The weight matrix represents the forget gate. This represents the hidden state at the previous time step. This represents the input at the current moment. Indicates the bias of the forget gate; Calculate the memory gate, select the information to be memorized, and use the following formula. ； in, The output of the memory gate indicates how much new information will be written to the cell state at the current moment; The weight matrix represents the memory gates, used to weight the hidden state from the previous time step and the current input. This indicates the bias of the memory gate, used to adjust the output of the memory gate; Represents the candidate cell state, indicating a new candidate state generated based on the current input and past information. The weight matrix represents the candidate cell state and is used to weight the hidden state of the previous time step and the current input. This represents the bias of the candidate cell state, used to adjust the output of the candidate cell state; The cell state at the current moment is calculated using the following formula. ； in, This indicates the current cell state and is used for long-term information storage. This represents the cell state at the previous moment, and the information stored in the LSTM unit at the previous moment. The output gate and the current hidden state are calculated using the following formula: in, The output of the output gate determines which part of the current cell state will use the output as a hidden state. ; The weight matrix represents the output gate, used to weight the hidden state from the previous time step and the current input. This indicates the output gate's bias, used to adjust the output of the output gate. This indicates the hidden state at the current moment.

6. A device for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer, characterized in that, The device for determining the optimal oxygen concentration at the inlet of the boiler high-temperature economizer includes: The historical data acquisition module is used to acquire historical data; The preprocessing module is used to preprocess the historical data; The machine learning model building module is used to build machine learning models; The training module is used to train the machine learning model based on preprocessed historical data; The thermal efficiency prediction module is used to input historical data into a trained machine learning model to obtain predicted values ​​of thermal efficiency. The objective function construction module is used to construct an objective function based on the predicted value of the thermal efficiency; The objective function solving module is used to solve the objective function to obtain the optimal oxygen concentration; The expression for the objective function is as follows: ； in, This represents the value of the objective function. For prediction weighting, To control the weighting, For the thermal efficiency prediction results, Set a value for oxygen content by iterating through... Value range, to minimize .

7. A computer device, comprising: The memory and processor contain a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer as described in any one of claims 1-5.

9. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the method for determining the optimal oxygen concentration at the inlet of a boiler high-temperature economizer as described in any one of claims 1-5.

Images