Variable-working-condition coal-fired unit high-temperature superheater outlet steam temperature prediction method and device

Abstract

The invention relates to the technical field of thermal power generating unit safety monitoring, in particular to a variable-working-condition coal-fired unit high-temperature superheater outlet steam temperature prediction method and device.The method comprises the steps that related variables of the coal-fired unit high-temperature superheater outlet steam temperature are collected, and a maximum information coefficient algorithm is used for screening the related variables; carrying out model training on data acquired on site by utilizing a plurality of network models; and weighting output results of the plurality of network models by utilizing an attention mechanism, establishing an integrated high-temperature superheater outlet steam temperature prediction model, training the model, and verifying the performance of the model by utilizing data acquired on site. According to the method, the high-temperature superheater outlet steam temperature prediction model adapting to the dynamic characteristics of different working conditions can be constructed, high-temperature superheater outlet steam temperature prediction under the complex working conditions can be achieved, an automatic superheated steam temperature control system can be optimized, a unit can always work in a normal range, and safe and stable operation of the unit is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of safety monitoring of thermal power units, and in particular to a method and device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit. Background Art

[0002] Long-term creep operation of coal-fired turbine components can easily cause mechanical damage, posing a significant threat to their safe and stable operation. This is especially true with the widespread integration of renewable energy, where frequent peak and frequency regulation of thermal power plants increases the risk of damage. Real-time acquisition of temperature field data from heat exchangers in coal-fired units is crucial for optimizing boiler operating parameters and preventing degradation of turbine components.

[0003] Existing steam temperature prediction models include mechanistic modeling and data-driven modeling. Mechanistic modeling typically calculates steam temperature by building thermodynamic and heat transfer models. However, due to the complex operating environment of coal-fired power plants and the diverse and highly nonlinear factors affecting steam temperature, these models struggle to fully and accurately describe actual operating conditions. Furthermore, physical models typically employ numerical simulation methods for prediction, but these methods are computationally complex, dependent on parameter selection and initial condition settings, and lack robustness, making them difficult to implement in engineering applications. Data-driven modeling is more efficient than numerical simulation and enables online prediction. Existing shallow and deep neural networks, such as LSTM (Long Short-Term Memory) networks and TCN (Temporal Convolutional Network), offer some accuracy in prediction. However, single models often fail to reflect the dynamic characteristics of diverse operating conditions, resulting in poor generalization and inability to address the complex operating conditions of coal-fired power plants during peak and frequency regulation.

[0004] In summary, existing technologies cannot fully and accurately describe actual working conditions, and are overly dependent on parameter selection and initial condition setting, making it difficult to reflect the dynamic characteristics of different working conditions. They also have poor generalization and robustness, which urgently need to be addressed. Summary of the Invention

[0005] The present application provides a method and device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, in order to solve the problems that the existing technology cannot fully and accurately describe the actual operating conditions, is overly dependent on parameter selection and initial condition setting, is difficult to reflect the dynamic characteristics of different operating conditions, and has poor generalization ability and robustness.

[0006] The first embodiment of the present application provides a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, which is applied to an offline training stage and includes the following steps: obtaining operating data corresponding to a target coal-fired unit, extracting multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain corresponding multidimensional standard features; performing nonlinear correlation analysis on each dimensional standard feature of the multidimensional standard feature and the outlet steam temperature data in the operating data to obtain corresponding analysis results, and screening out a plurality of key features from the multidimensional standard feature according to the analysis results, and based on a preset long short-term memory network, a time convolution network and an inverse propagation network, establish a corresponding outlet steam temperature prediction model; input the multiple key features into the outlet steam temperature prediction model respectively to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and based on the preset attention mechanism, empower the output features to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and train the integrated high-temperature superheater outlet steam temperature prediction model to use the trained integrated high-temperature superheater outlet steam temperature prediction model to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions in the online prediction stage.

[0007] According to the above technical means, the embodiment of the present application measures the average flow velocity of the grid area by arranging more measuring points in the grid closest to the wall, and at the same time measures the flow velocity in the grid of the non-near-wall area. According to the cross-sectional area of the flue, the average flow velocity of the flue section considering the near-wall effect is obtained, and the ratio of the flow velocity to the average flow velocity of the section measured using the grid method without considering the near-wall effect is calculated, thereby obtaining the wall effect adjustment coefficient, which provides reliable data support for improving the accuracy of manual measurement of flue gas flow velocity.

[0008] Optionally, in one embodiment of the present application, the obtaining of operating data corresponding to the target coal-fired unit, extracting multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain corresponding multidimensional standard features includes: collecting multidimensional data generated by the target coal-fired unit during operation, wherein the multidimensional data includes external feature data and target high-temperature superheater outlet steam temperature data; performing data cleaning on the multidimensional data, and performing smoothing filtering on the cleaned multidimensional data to generate corresponding denoised data, and normalizing the denoised data to obtain the multidimensional standard features.

[0009] According to the above technical means, the embodiment of the present application divides the flue measurement section to arrange multiple wall effect measurement points, thereby providing a reliable data basis for the subsequent measurement of the flue gas flow velocity in the circular flue.

[0010] Optionally, in one embodiment of the present application, a nonlinear correlation analysis is performed on each dimensional standard feature of the multidimensional standard feature and the outlet steam temperature data in the operating data to obtain a corresponding analysis result, and a plurality of key features are screened out from the multidimensional standard feature based on the analysis result, including: based on a preset maximum information coefficient algorithm, a correlation analysis is performed on each dimensional standard feature and the outlet steam temperature data to obtain a corresponding correlation analysis result; based on the correlation analysis result, a plurality of initial key features are screened out from the multidimensional standard feature, and a correlation analysis is performed on each of the plurality of initial key features and the outlet steam temperature data to obtain a corresponding correlation analysis result; based on the correlation analysis result, the plurality of key features that meet the preset correlation requirements are screened out from the plurality of initial key features.

[0011] According to the above technical means, the embodiment of the present application divides the wall effect measurement points into local wall effect measurement type and complete wall effect measurement type, thereby obtaining and recording the corresponding flow velocity measurement operations, effectively ensuring the reliability of the subsequent calculation of the wall effect adjustment coefficient.

[0012] Optionally, in one embodiment of the present application, the corresponding outlet steam temperature prediction model is established based on the preset long short-term memory network, temporal convolution network and back propagation network, including: connecting the long short-term memory network, the temporal convolution network and the back propagation network in parallel to obtain a corresponding comprehensive model; training the comprehensive model through the multiple key features to construct the outlet steam temperature prediction model.

[0013] According to the above technical means, the embodiment of the present application calculates the attenuation rate and sub-sector area corresponding to each sub-sector, thereby providing reliable data guidance and basis for the calculation of the wall adjustment coefficient by calculating the flue gas volume flow rate in the sub-sector between the inner wall of the external equal-area sector and the wall effect measuring point.

[0014] The second aspect of the present application provides a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, which is applied to the online prediction stage and includes the following steps: collecting actual operating data corresponding to the target coal-fired unit, and preprocessing the actual operating data to obtain corresponding actual standard data; inputting the actual standard data into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

[0015] According to the above technical means, the embodiment of the present application introduces the measurement and determination of the wall effect adjustment coefficient taking into account the wall effect, thereby correcting the average flue gas flow rate measured by the conventional grid method, effectively improving the accuracy of manual measurement of the flue gas flow rate in the circular flue.

[0016] Optionally, in one embodiment of the present application, after outputting the outlet steam temperature prediction result corresponding to the target coal-fired unit, it also includes: performing a prediction and evaluation operation on the outlet steam temperature prediction result based on at least one preset prediction and evaluation strategy to obtain corresponding actual prediction and evaluation data; fine-tuning the integrated high-temperature superheater outlet steam temperature prediction model according to the actual prediction and evaluation data, so as to re-execute the outlet steam temperature prediction operation of the target coal-fired unit using the fine-tuned integrated high-temperature superheater outlet steam temperature prediction model.

[0017] The third embodiment of the present application provides a device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, which is applied to an offline training stage and includes: an extraction module for obtaining operating data corresponding to a target coal-fired unit, extracting multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain corresponding multidimensional standard features; a modeling module for performing nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain corresponding analysis results, and screening out a plurality of key features from the multidimensional standard features according to the analysis results, and based on a preset long short-term memory network, a time convolution network and a A back propagation network is used to establish a corresponding outlet steam temperature prediction model; a training module is used to input the multiple key features into the outlet steam temperature prediction model respectively to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and based on a preset attention mechanism, the output features are weighted to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and the integrated high-temperature superheater outlet steam temperature prediction model is trained to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions.

[0018] Optionally, in one embodiment of the present application, the extraction module includes: an acquisition unit for acquiring multidimensional data generated by the target coal-fired unit during operation, wherein the multidimensional data includes external characteristic data and target high-temperature superheater outlet steam temperature data; a denoising unit for performing data cleaning on the multidimensional data, and performing smoothing filtering on the cleaned multidimensional data to generate corresponding denoised data, and normalizing the denoised data to obtain the multidimensional standard features.

[0019] Optionally, in one embodiment of the present application, the modeling module includes: a correlation analysis unit, which is used to perform a correlation analysis on the standard features of each dimension and the outlet steam temperature data based on a preset maximum information coefficient algorithm to obtain corresponding correlation analysis results; a correlation analysis unit, which is used to screen out multiple initial key features from the multidimensional standard features according to the correlation analysis results, and perform a correlation analysis on each of the multiple initial key features and the outlet steam temperature data to obtain corresponding correlation analysis results; a screening unit, which is used to screen out the multiple key features that meet the preset correlation requirements from the multiple initial key features based on the correlation analysis results.

[0020] Optionally, in one embodiment of the present application, the modeling module further includes: a parallel unit for connecting the long short-term memory network, the temporal convolution network and the back propagation network in parallel to obtain a corresponding comprehensive model; and a construction unit for training the comprehensive model through the multiple key features to construct the outlet steam temperature prediction model.

[0021] The fourth embodiment of the present application provides a device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, which is applied to the online prediction stage and includes: a preprocessing module for collecting actual operating data corresponding to the target coal-fired unit and preprocessing the actual operating data to obtain corresponding actual standard data; a prediction module for inputting the actual standard data into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

[0022] Optionally, in one embodiment of the present application, it also includes: an evaluation module, which is used to perform a prediction and evaluation operation on the outlet steam temperature prediction result corresponding to the target coal-fired unit based on at least one preset prediction and evaluation strategy after outputting the outlet steam temperature prediction result, so as to obtain corresponding actual prediction and evaluation data; a fine-tuning module, which is used to fine-tune the integrated high-temperature superheater outlet steam temperature prediction model according to the actual prediction and evaluation data, so as to re-execute the outlet steam temperature prediction operation of the target coal-fired unit using the fine-tuned integrated high-temperature superheater outlet steam temperature prediction model.

[0023] The fifth aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit as described in the above embodiment.

[0024] The sixth aspect of the present application provides a computer-readable storage medium, which stores a computer program. When the program is executed by a processor, it implements the above-mentioned method for predicting the outlet steam temperature of the high-temperature superheater of a variable-operating-condition coal-fired unit.

[0025] Therefore, the embodiments of the present application have the following beneficial effects: The embodiments of the present application can obtain the operating data corresponding to the target coal-fired unit, extract the multidimensional features corresponding to the operating data, and pre-process the multidimensional features to obtain the corresponding multidimensional standard features; perform nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain the corresponding analysis results, and screen out multiple key features from the multidimensional standard features based on the analysis results, and establish a corresponding outlet steam temperature prediction model based on a preset long short-term memory network, a temporal convolution network, and a back propagation network; input the multiple key features into the outlet steam temperature prediction model respectively to obtain the output features corresponding to the long short-term memory network, the temporal convolution network, and the back propagation network in the outlet steam temperature prediction model, and weight the output features based on a preset attention mechanism to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and train the integrated high-temperature superheater outlet steam temperature prediction model, so as to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions. This application can construct a high-temperature superheater outlet steam temperature prediction model that adapts to the dynamic characteristics of different operating conditions. It can realize high-temperature superheater outlet steam temperature prediction under complex operating conditions, which helps to optimize the superheated steam temperature automatic control system, so that the unit always operates within the normal range and ensures safe and stable operation of the unit. This solves the problems of existing technologies that cannot fully and accurately describe actual operating conditions, rely too much on parameter selection and initial condition setting, have difficulty reflecting the dynamic characteristics of different operating conditions, and have poor generalization and robustness.

[0026] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a flow chart of a method for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions in an offline training phase according to an embodiment of the present application; Figure 2A schematic diagram of the correlation of variables screened by the Maximum Information Coefficient (MIC) provided in an embodiment of the present application; Figure 3 A schematic diagram of the execution logic of a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit provided in an embodiment of the present application; Figure 4 This is a flow chart of a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit applied in an online prediction stage according to an embodiment of the present application; Figure 5 A visualization diagram of regression analysis of a backpropagation algorithm on a test set provided in an embodiment of the present application; Figure 6 A visualization diagram of regression analysis of a long short-term memory network on a test set provided in an embodiment of the present application; Figure 7 A visualization diagram of regression analysis of a temporal convolutional network on a test set provided in an embodiment of the present application; Figure 8 A schematic diagram comparing the prediction effects of an integrated model and a single model provided in an embodiment of the present application; Figure 8 (a) is a comparative schematic diagram of LSTM steam temperature prediction provided in an embodiment of the present application; Figure 8 (b) is a TCN steam temperature prediction comparison diagram provided in an embodiment of the present application; Figure 8 (c) is a comparative schematic diagram of steam temperature prediction using a BP (Back Propagation) network provided in an embodiment of the present application; Figure 8 (d) is a comparative schematic diagram of steam temperature prediction using an integrated model provided in an embodiment of the present application; Figure 9 This is an example diagram of a device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions, applied in an offline training phase, according to an embodiment of the present application; Figure 10 This is an example diagram of a device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit applied in an online prediction stage according to an embodiment of the present application; Figure 11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0028] Among them, 10-a device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions applied in the offline training stage, 20-a device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions applied in the online prediction stage; 101-an extraction module, 102-a modeling module, 103-a training module; 201-a preprocessing module, 202-a prediction module; 1101-a memory, 1102-a processor, 1103-a communication interface. DETAILED DESCRIPTION

[0029] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0030] The following describes the method and device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit with reference to the accompanying drawings according to an embodiment of the present application. In response to the problems mentioned in the above background technology, the present application provides a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit. In this method, the operating data corresponding to the target coal-fired unit is obtained, and the multidimensional features corresponding to the operating data are extracted, and the multidimensional features are preprocessed to obtain the corresponding multidimensional standard features; the nonlinear correlation analysis is performed on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain the corresponding analysis results, and a plurality of key features are screened out from the multidimensional standard features according to the analysis results, and based on the preset long short-term memory network and the time convolution network, a multidimensional standard feature prediction method is provided. and back propagation network to establish a corresponding outlet steam temperature prediction model; multiple key features are respectively input into the outlet steam temperature prediction model to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and based on the preset attention mechanism, the output features are weighted to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and the integrated high-temperature superheater outlet steam temperature prediction model is trained to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions. The present application can construct a high-temperature superheater outlet steam temperature prediction model with dynamic characteristics that adapts to different operating conditions, and can realize high-temperature superheater outlet steam temperature prediction under complex operating conditions, which is conducive to helping optimize the superheated steam temperature automatic control system, so that the unit always works within the normal range and ensures safe and stable operation of the unit. Thus, the existing technology solves the problems of being unable to fully and accurately describe the actual operating conditions, overly relying on parameter selection and initial condition setting, being difficult to reflect the dynamic characteristics of different operating conditions, and having poor generalization ability and robustness.

[0031] Specifically, Figure 1This is a flow chart of a method for predicting the high-temperature superheater outlet steam temperature of a variable-operating-condition coal-fired unit applied in an offline training phase, provided in an embodiment of the present application.

[0032] like Figure 1 As shown, the method for predicting the outlet steam temperature of the high-temperature superheater of a variable-operating-condition coal-fired unit includes the following steps: In step S101, operating data corresponding to a target coal-fired unit is acquired, and multidimensional features corresponding to the operating data are extracted, and the multidimensional features are preprocessed to obtain corresponding multidimensional standard features.

[0033] The embodiments of the present application can first obtain multidimensional features from the operating data of the coal-fired unit, including key variables such as main steam temperature, main steam pressure, unit load, and high-temperature superheater outlet steam temperature, and perform data cleaning, denoising and normalization on the multidimensional features to obtain corresponding multidimensional standard features.

[0034] Optionally, in one embodiment of the present application, operating data corresponding to the target coal-fired unit is obtained, and multidimensional features corresponding to the operating data are extracted, and the multidimensional features are preprocessed to obtain corresponding multidimensional standard features, including: collecting multidimensional data generated by the target coal-fired unit during the operation process, wherein the multidimensional data includes external feature data and target high-temperature superheater outlet steam temperature data; performing data cleaning on the multidimensional data, and performing smoothing filtering on the cleaned multidimensional data to generate corresponding denoised data, and normalizing the denoised data to obtain multidimensional standard features.

[0035] It should be noted that the embodiments of the present application can first collect multidimensional data generated during the operation of the coal-fired unit (i.e., variables related to the high-temperature superheater outlet steam temperature), including but not limited to external characteristics such as main steam flow, main steam temperature, main steam pressure, unit load, fuel average, primary air volume, air preheater outlet primary air temperature, air preheater outlet primary air pressure, secondary air volume, furnace inlet secondary air pressure, secondary air temperature, and high-temperature superheater outlet steam temperature data, as well as first-stage superheater desuperheating water flow and second-stage superheater desuperheating water flow. Specifically, the variables related to the high-temperature superheater outlet steam temperature are shown in the table:

[0036] In the actual implementation process, the outlet steam temperature of the high-temperature superheater of the coal-fired unit is affected by the coupling of multiple variables such as steam flow and combustion parameters. The original multi-dimensional feature data can be collected through the DCS system, which contains 40-dimensional variables and a sampling frequency of 6s.

[0037] Secondly, in order to deal with the possible noise problem in the thermal power unit data collected by the DCS, the embodiment of the present application can delete the missing values and abnormal values of the collected data, and use the smoothing filtering method to eliminate the noise in the data; normalize the data so that all eigenvalues fall into a unified interval to obtain multi-dimensional standard features, thereby improving the training stability of the subsequent model.

[0038] Therefore, the embodiment of the present application obtains the multidimensional features of the coal-fired unit operation data and performs pre-processing such as data cleaning, denoising and normalization, thereby effectively improving the data quality and providing reliable data support for the stability of subsequent model training.

[0039] In step S102, a nonlinear correlation analysis is performed on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain corresponding analysis results, and multiple key features are screened out from the multidimensional standard features based on the analysis results, and a corresponding outlet steam temperature prediction model is established based on the preset long short-term memory network, time convolution network and back propagation network.

[0040] Furthermore, the embodiments of the present application also need to use the maximum information coefficient algorithm to analyze the nonlinear correlation between each feature and the outlet steam temperature data to screen out the key features that have a significant impact on the outlet steam temperature change, thereby reducing the input variable dimension, reducing the model complexity, and improving the model training speed and accuracy; then, the embodiments of the present application can be based on the key features extracted by MIC, and combined with LSTM, TCN, and BP networks to establish outlet steam temperature prediction models respectively.

[0041] Optionally, in one embodiment of the present application, a nonlinear correlation analysis is performed on each dimensional standard feature of the multidimensional standard feature and the outlet steam temperature data in the operating data to obtain corresponding analysis results, and multiple key features are screened out from the multidimensional standard features based on the analysis results, including: based on a preset maximum information coefficient algorithm, a correlation analysis is performed on each dimensional standard feature and the outlet steam temperature data to obtain corresponding correlation analysis results; based on the correlation analysis results, multiple initial key features are screened out from the multidimensional standard features, and each initial key feature of the multiple initial key features is correlated with the outlet steam temperature data to obtain corresponding correlation analysis results; based on the correlation analysis results, multiple key features that meet the preset correlation requirements are screened out from the multiple initial key features.

[0042] As an achievable method, the embodiment of the present application can use the maximum information coefficient MIC to perform correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain the corresponding correlation analysis results; based on the correlation analysis results, highly correlated variables (i.e., multiple initial key features) are preliminarily screened out, and the correlation between the screened multiple initial key features and the outlet steam temperature data is then analyzed, so as to screen out multiple key features that meet the preset correlation requirements from the multiple initial key features based on the correlation analysis results. The correlation of the MIC screening variables is as follows: Figure 2 shown.

[0043] The embodiment of the present application can set variables , , the mathematical expression of the maximum information coefficient MIC is:

[0044]

[0045] in, and For Grid G The size of the partitioned interval; is the joint probability density function; is the marginal probability density function; For Grid The maximum number of messages under Mic is the normalized maximum mutual information number, where is the constraint on the total number of grid divisions; B The value is usually taken as 0.6 of the total number of data.

[0046] Based on the collected data related to the outlet steam temperature of the high-temperature superheater of the actual coal-fired unit, a calculation example test was performed. In the embodiment of the present application, the MIC algorithm can be used to screen the model input characteristic variables. The input characteristic variables are preliminarily screened using the MIC method with a threshold of 0.6, as shown in Table 2: Table 2

[0047] Therefore, the embodiments of the present application adopt the maximum information coefficient algorithm to perform feature selection and optimization to identify key features that are closely related to the changes in the superheater outlet steam temperature, thereby improving the accuracy and generalization ability of the steam temperature prediction model, screening out the most influential key features, reducing the input dimension, and avoiding the impact of redundant information on model training.

[0048] Optionally, in one embodiment of the present application, a corresponding outlet steam temperature prediction model is established based on a preset long short-term memory network, a time convolution network, and a back propagation network, including: connecting the long short-term memory network, the time convolution network, and the back propagation network in parallel to obtain a corresponding comprehensive model; training the comprehensive model through multiple key features to construct an outlet steam temperature prediction model Specifically, the embodiments of the present application can connect LSTM, TCN, and BP networks in parallel, such as Figure 3 As shown in the figure, a TCN-LSTM-BP network (i.e., a comprehensive network) is constructed, and based on the TCN-LSTM-BP network, outlet steam temperature prediction models are established using LSTM, TCN, and BP networks respectively.

[0049] Among them, the specific calculation expression of LSTM is as follows: Input Gate:

[0050] Forget Gate:

[0051] Output gate:

[0052] Memory Update:

[0053] Hidden state output:

[0054] Tangent hyperbolic function:

[0055] Sigmoid activation function

[0056] in, is the input at the current moment; 、 The state of memory cells at the previous moment and the current moment; 、 is the output of the LSTM layer at the previous moment and the current moment; is the sigmoid function; tanh(·) represents the tangent hyperbolic function; 、 、 The corresponding gates are input gate, forget gate and output gate; 、 、 The weights of the sigmoid function corresponding to the input gate, forget gate, and output gate; 、 、 are the corresponding biases respectively; 、 are the weights and biases in the tanh activation function.

[0057] The calculation expression of the TCN network is as follows:

[0058] in, is the convolution kernel weight of the lth layer; is the expansion rate; is the output of the last layer at time t; is the output of the previous layer; the input of layer 0 is the input feature .

[0059] The calculation expression of BP network is as follows: Forward propagation process: For a L ) layers of neural network, the ( l The output of the ) layer can be expressed as:

[0060]

[0061] in, Indicates the l The activation output of the layer; represents the result of linear combination; represents the weight matrix; represents the bias vector; is the activation function, and the GIA activation function includes the sigmoid function, the tanh function, and the ReLU function.

[0062] Back propagation process: loss function for the output layer E , which is about the value before the output layer activation The gradient of is:

[0063] in, Represents the derivative of the activation function. For the sigmoid function, its derivative is:

[0064] Gradient update:

[0065]

[0066] in, η is the learning rate.

[0067] During actual implementation, the embodiments of the present application can train a TCN-LSTM-BP network based on the key variables (i.e., key features) extracted through MIC screening to establish an outlet steam temperature prediction model. Specifically, the embodiments of the present application can use an optimization algorithm to adjust the model parameters of the TCN-LSTM-BP network and train it based on historical operating data to ensure the model's prediction performance under different operating conditions. Secondly, the embodiments of the present application can reduce prediction errors by adjusting the neural network parameters using a gradient descent method based on the loss function through iterative training, terminating iterative training when the loss function value stabilizes or before the loss function value diverges. In addition, by actively selecting a TCN-LSTM-BP network for a specific number of training rounds, it is possible to determine a model with better performance and also serves as a means to prevent model overfitting.

[0068] Therefore, the embodiments of the present application can adjust the model parameters based on the time and accuracy spent on each model training, and modify the number of key variables screened by MIC to achieve the optimal value of prediction accuracy and model prediction time.

[0069] It can be understood that the embodiments of the present application can use the TCN-LSTM-BP network to establish an outlet steam temperature prediction model based on the key features extracted by MIC, thereby combining the advantages of the three networks: LSTM neural network, TCN neural network and BP neural network, and integrating the three models through a parallel structure to enhance the model's time feature extraction capability.

[0070] In step S103, multiple key features are respectively input into the outlet steam temperature prediction model to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and the output features are weighted based on the preset attention mechanism to establish the corresponding integrated high-temperature superheater outlet steam temperature prediction model, and the integrated high-temperature superheater outlet steam temperature prediction model is trained to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions.

[0071] Afterwards, the embodiment of the present application uses the attention mechanism to weight the output results of the three prediction models of LSTM neural network, TCN neural network and BP neural network (i.e., TCN-LSTM-BP network) to establish an integrated high-temperature superheater outlet steam temperature prediction model and train it.

[0072] Specifically, embodiments of the present application extract the output features of different trained single models (such as LSTM neural networks, TCN neural networks, and BP neural networks), analyze the correlations between these features and the high-temperature superheater outlet steam temperature characteristics in different dimensions through the attention mechanism, and weight the output features of different models to establish and train an integrated high-temperature superheater outlet steam temperature prediction model. The prediction performance of the LSTM, TCN, and BP networks is then compared with the integrated high-temperature superheater outlet steam temperature prediction model. The calculation expression for the attention mechanism is:

[0073]

[0074] in, is the characteristic variable; is the load characteristic; 、 、 is the linear mapping weight matrix; tanh(·) represents the tangent hyperbolic function; softmax(·) represents the nonlinear activation function; is the output weight of different feature variables; Output result for attention.

[0075] Therefore, the embodiment of the present application empowers the output results of the TCN-LSTM-BP network through the attention mechanism to establish an integrated high-temperature superheater outlet steam temperature prediction model, thereby providing solid technical support for the prediction of the superheater outlet steam temperature of coal-fired units under complex operating conditions.

[0076] According to the variable operating condition coal-fired unit high temperature superheater outlet steam temperature prediction method applied to the offline training stage proposed in the embodiment of the present application, the operating data corresponding to the target coal-fired unit is obtained, and the multidimensional features corresponding to the operating data are extracted, and the multidimensional features are preprocessed to obtain the corresponding multidimensional standard features; the nonlinear correlation analysis is performed on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain the corresponding analysis results, and a plurality of key features are screened out from the multidimensional standard features according to the analysis results, and based on the preset long short-term memory network, time convolution network and back propagation network, the prediction method is used to predict the outlet steam temperature of the coal-fired unit in the variable operating condition coal-fired unit in the offline training stage. The invention relates to a method for predicting the outlet steam temperature of a coal-fired unit by using a back propagation network and a long short-term memory network to obtain the output features of the long short-term memory network, the time convolution network and the back propagation network in the outlet steam temperature prediction model; the invention also ... to obtain the output features of the long short-term memory network, the time convolution network and the back prop

[0077] Figure 4 This is a flow chart of a method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit applied in the online prediction stage, provided in an embodiment of the present application.

[0078] like Figure 4 As shown, the method for predicting the outlet steam temperature of the high-temperature superheater of a variable-operating-condition coal-fired unit includes the following steps: In step S401, actual operation data corresponding to the target coal-fired unit is collected and pre-processed to obtain corresponding actual standard data.

[0079] In step S402, the actual standard data is input into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

[0080] It should be noted that the embodiment of the present application is based on the trained integrated high-temperature superheater outlet steam temperature prediction model, and uses the actual operation data collected on-site to verify the model performance of the pre-trained integrated high-temperature superheater outlet steam temperature prediction model, thereby realizing the prediction of the superheater outlet steam temperature of the coal-fired unit under complex operating conditions.

[0081] Optionally, in one embodiment of the present application, after outputting the outlet steam temperature prediction result corresponding to the target coal-fired unit, it also includes: performing a prediction and evaluation operation on the outlet steam temperature prediction result based on at least one preset prediction and evaluation strategy to obtain corresponding actual prediction and evaluation data; fine-tuning the integrated high-temperature superheater outlet steam temperature prediction model according to the actual prediction and evaluation data, so as to re-execute the outlet steam temperature prediction operation of the target coal-fired unit using the fine-tuned integrated high-temperature superheater outlet steam temperature prediction model.

[0082] As an achievable method, the embodiment of the present application may adopt the root mean square error RMSE, mean absolute error MAE and determination coefficient R 2 To evaluate the prediction effect of the integrated model, the calculation expression of the evaluation index is:

[0083]

[0084]

[0085] in, is the actual value; is the predicted value; is the average of the actual values; is the sample size.

[0086] Specifically, the embodiment of the present application can use the mean square error (MSE), root mean square error (RMSE) and coefficient of determination (R 2 ) To evaluate the simulation and prediction capabilities of the integrated high-temperature superheater outlet steam temperature prediction model to determine whether the model meets expectations or needs further adjustment. Among them, the regression analysis of the back propagation algorithm, long short-term memory network and time convolution network on the test set are as follows: Figure 5 、 Figure 6 and Figure 7 shown.

[0087] Afterwards, based on the trained integrated high-temperature superheater outlet steam temperature prediction model, the model performance was verified using field collected data, and the overall prediction effect of the integrated high-temperature superheater outlet steam temperature prediction model (i.e., the integrated model) was compared with that of a single model. At the same time, the prediction effects of the integrated model and the single model under different operating conditions were also compared.

[0088] Figure 8 This is a schematic diagram comparing the prediction effects of the integrated model and the single model, where: Figure 8 (a) is a schematic diagram of LSTM steam temperature prediction comparison; Figure 8 (b) is a schematic diagram of TCN steam temperature prediction comparison; Figure 8 (c) is a comparison diagram of steam temperature prediction using BP network; Figure 8 (d) is a schematic diagram of the steam temperature prediction comparison of the integrated model. The comparison results are shown in Table 3: Table 3

[0089] Depend on Figure 8 As can be seen from Table 3, the mean square error, root mean square error and determination coefficient of the integrated model in the embodiment of the present application are better than those of other single models, which effectively ensures the reliability and robustness of the integrated model prediction.

[0090] In summary, the embodiments of the present application can perform feature screening through the maximum information coefficient, integrate long short-term memory networks, temporal convolutional networks, and back-propagation algorithms for time series modeling, and integrate them through the attention mechanism to construct a dynamic prediction model that adapts to complex working conditions. The embodiments of the present application can collect relevant parameters of the high-temperature superheater outlet steam temperature of coal-fired units based on the high-temperature superheater outlet steam temperature prediction model, and use the MIC algorithm to screen the model input feature variables; use the TCN-LSTM-BP network to train the model on the data collected on site, and use the attention mechanism to weight the output results of the single prediction model to establish an integrated high-temperature superheater outlet steam temperature prediction model, and use the field collected data to verify its performance. Therefore, the embodiments of the present application can construct a high-temperature superheater outlet steam temperature prediction model with dynamic characteristics that adapt to different operating conditions, and can realize the outlet steam temperature prediction under complex operating conditions, which is helpful to help optimize the superheated steam temperature automatic control system, provide a basis for advance action for the control module, and provide reference guidance for operating personnel, so that the unit always operates within the normal range, preventing the occurrence of accidents such as pipe burst caused by overheating of the high-temperature superheater pipeline, which is of great significance to improving the stability and safety of the operation of the coal-fired unit.

[0091] Secondly, the device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit with variable operating conditions proposed in accordance with an embodiment of the present application will be described with reference to the accompanying drawings.

[0092] Figure 9 It is a block diagram of a device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions, which is applied in an offline training phase, according to an embodiment of the present application.

[0093] like Figure 9 As shown, the device 10 for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions, which is applied in the offline training stage, includes: an extraction module 101 , a modeling module 102 and a training module 103 .

[0094] The extraction module 101 is used to obtain the operating data corresponding to the target coal-fired unit, extract the multidimensional features corresponding to the operating data, and pre-process the multidimensional features to obtain corresponding multidimensional standard features.

[0095] The modeling module 102 is used to perform nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain corresponding analysis results, and to screen out multiple key features from the multidimensional standard features based on the analysis results, and to establish a corresponding outlet steam temperature prediction model based on the preset long short-term memory network, time convolution network and back propagation network.

[0096] The training module 103 is used to input multiple key features into the outlet steam temperature prediction model respectively to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and to weight the output features based on the preset attention mechanism to establish the corresponding integrated high-temperature superheater outlet steam temperature prediction model, and to train the integrated high-temperature superheater outlet steam temperature prediction model so as to use the trained integrated high-temperature superheater outlet steam temperature prediction model to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions in the online prediction stage.

[0097] Optionally, in one embodiment of the present application, the extraction module 101 includes: a collection unit and a denoising unit.

[0098] The acquisition unit is used to acquire multi-dimensional data generated by the target coal-fired unit during operation, wherein the multi-dimensional data includes external characteristic data and target high-temperature superheater outlet steam temperature data.

[0099] The denoising unit is used to clean the multidimensional data, perform smoothing filtering on the cleaned multidimensional data to generate corresponding denoised data, and normalize the denoised data to obtain multidimensional standard features.

[0100] Optionally, in one embodiment of the present application, the modeling module 102 includes: a correlation analysis unit, a relevance analysis unit, and a screening unit.

[0101] Among them, the correlation analysis unit is used to perform correlation analysis on the standard features of each dimension and the outlet steam temperature data based on a preset maximum information coefficient algorithm to obtain corresponding correlation analysis results.

[0102] The correlation analysis unit is used to screen out multiple initial key features from the multidimensional standard features according to the correlation analysis results, and perform correlation analysis on each of the multiple initial key features and the outlet steam temperature data to obtain corresponding correlation analysis results.

[0103] The screening unit is used to screen out multiple key features that meet preset correlation requirements from the multiple initial key features based on the correlation analysis result.

[0104] Optionally, in one embodiment of the present application, the modeling module 102 further includes: a parallel unit and a construction unit.

[0105] Among them, the parallel unit is used to connect the long short-term memory network, the temporal convolutional network and the back propagation network in parallel to obtain the corresponding comprehensive model.

[0106] A construction unit is used to train a comprehensive model through multiple key features to build an outlet steam temperature prediction model.

[0107] It should be noted that the above explanation of the embodiment of the method for predicting the high-temperature superheater outlet steam temperature of a variable-operating-condition coal-fired unit applied in the offline training stage is also applicable to the device for predicting the high-temperature superheater outlet steam temperature of a variable-operating-condition coal-fired unit applied in the offline training stage of this embodiment, and will not be repeated here.

[0108] According to the embodiment of the present application, the device for predicting the outlet steam temperature of the high-temperature superheater of a coal-fired unit under variable operating conditions and applied to the offline training stage includes an extraction module 101 for obtaining the operating data corresponding to the target coal-fired unit, extracting the multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain the corresponding multidimensional standard features; a modeling module 102 for performing nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain the corresponding analysis results, and screening out multiple key features from the multidimensional standard features according to the analysis results, and based on the preset long short-term memory network and time convolution network, The invention relates to a method for predicting the outlet steam temperature of a coal-fired unit by using a high-temperature superheater outlet steam temperature prediction model and a back-propagation network. The method comprises the following steps: a) inputting a plurality of key features into the outlet steam temperature prediction model, respectively, to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back-propagation network in the outlet steam temperature prediction model, and weighting the output features based on the preset attention mechanism to establish the corresponding integrated high-temperature superheater outlet steam temperature prediction model, and training the integrated high-temperature superheater outlet steam temperature prediction model to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions. The present application can construct a high-temperature superheater outlet steam temperature prediction model with dynamic characteristics that adapt to different operating conditions, and can realize the prediction of high-temperature superheater outlet steam temperature under complex operating conditions, which is conducive to helping optimize the superheated steam temperature automatic control system, so that the unit always works within the normal range and ensures the safe and stable operation of the unit.

[0109] Figure 10 It is a block diagram of a device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit in an online pre-stage according to an embodiment of the present application.

[0110] like Figure 10 As shown, the device 20 for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions, which is applied in the online prediction stage, includes a preprocessing module 201 and a prediction module 202 .

[0111] The pre-processing module 201 is used to collect actual operation data corresponding to the target coal-fired unit and pre-process the actual operation data to obtain corresponding actual standard data.

[0112] The prediction module 202 is used to input actual standard data into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

[0113] Optionally, in one embodiment of the present application, the device 20 for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit applied in the online prediction stage of the embodiment of the present application further includes: an evaluation module and a fine-tuning module.

[0114] Among them, the evaluation module is used to perform a prediction and evaluation operation on the outlet steam temperature prediction result corresponding to the target coal-fired unit based on at least one preset prediction and evaluation strategy after outputting the outlet steam temperature prediction result to obtain corresponding actual prediction and evaluation data.

[0115] The fine-tuning module is used to fine-tune the integrated high-temperature superheater outlet steam temperature prediction model according to actual prediction evaluation data, so as to re-execute the outlet steam temperature prediction operation of the target coal-fired unit using the fine-tuned integrated high-temperature superheater outlet steam temperature prediction model.

[0116] It should be noted that the above explanation of the embodiment of the method for predicting the high-temperature superheater outlet steam temperature of a variable-operating-condition coal-fired unit applied in the online prediction stage is also applicable to the device for predicting the high-temperature superheater outlet steam temperature of a variable-operating-condition coal-fired unit applied in the online prediction stage of this embodiment, and will not be repeated here.

[0117] According to the embodiment of the present application, a device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions, which is applied to the online prediction stage, includes a preprocessing module 201 for collecting actual operating data corresponding to the target coal-fired unit and preprocessing the actual operating data to obtain corresponding actual standard data; and a prediction module 202 for inputting the actual standard data into an integrated high-temperature superheater outlet steam temperature prediction model to output a prediction result of the outlet steam temperature corresponding to the target coal-fired unit. The present application can construct a high-temperature superheater outlet steam temperature prediction model with dynamic characteristics that adapt to different operating conditions, and can realize the prediction of the high-temperature superheater outlet steam temperature under complex operating conditions, which is conducive to helping optimize the superheated steam temperature automatic control system, so that the unit always operates within the normal range and ensures the safe and stable operation of the unit.

[0118] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: A memory 1101 , a processor 1102 , and a computer program stored in the memory 1101 and executable on the processor 1102 .

[0119] When the processor 1102 executes the program, the method for predicting the outlet steam temperature of the high-temperature superheater of a variable-operating-condition coal-fired unit provided in the above-mentioned embodiment is implemented.

[0120] Furthermore, the electronic device further includes: The communication interface 1103 is used for communication between the memory 1101 and the processor 1102 .

[0121] The memory 1101 is used to store computer programs that can be run on the processor 1102 .

[0122] The memory 1101 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.

[0123] If the memory 1101, processor 1102, and communication interface 1103 are implemented independently, the communication interface 1103, memory 1101, and processor 1102 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0124] Optionally, in a specific implementation, if the memory 1101, the processor 1102 and the communication interface 1103 are integrated on a chip, the memory 1101, the processor 1102 and the communication interface 1103 can communicate with each other through an internal interface.

[0125] The processor 1102 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0126] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit with variable operating conditions.

[0127] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0128] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of this application, "N" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0129] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing a custom logical function or process step, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed in a different order than shown or discussed, including performing functions in a substantially simultaneous manner or in a reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application pertain.

[0130] The logic and / or steps represented in a flowchart or otherwise described herein, for example, can be considered a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" is any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (not exhaustive) of computer-readable media include: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program can be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing it in other suitable ways as necessary, and then storing it in a computer memory.

[0131] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented using hardware, as in another embodiment, any of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logical functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0132] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0133] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0134] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, applied in the offline training stage, characterized in that: The following steps are involved: Acquiring operating data corresponding to a target coal-fired unit, extracting multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain corresponding multidimensional standard features; performing a nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain corresponding analysis results, screening a plurality of key features from the multidimensional standard features based on the analysis results, and establishing a corresponding outlet steam temperature prediction model based on a preset long short-term memory network, a temporal convolutional network, and a back propagation network; The multiple key features are respectively input into the outlet steam temperature prediction model to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and the output features are weighted based on the preset attention mechanism to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and the integrated high-temperature superheater outlet steam temperature prediction model is trained to use the trained integrated high-temperature superheater outlet steam temperature prediction model to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions in the online prediction stage.

2. The method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit according to claim 1, characterized in that: The step of obtaining the operating data corresponding to the target coal-fired unit, extracting the multidimensional features corresponding to the operating data, and preprocessing the multidimensional features to obtain corresponding multidimensional standard features includes: Collecting multidimensional data generated during the operation of the target coal-fired unit, wherein the multidimensional data includes external characteristic data and target high-temperature superheater outlet steam temperature data; The multidimensional data is cleaned, and smoothing filtering is performed on the cleaned multidimensional data to generate corresponding denoised data, and the denoised data is normalized to obtain the multidimensional standard features.

3. The method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit according to claim 2, characterized in that: The nonlinear correlation analysis is performed on each dimensional standard feature of the multidimensional standard feature and the outlet steam temperature data in the operation data to obtain corresponding analysis results, and a plurality of key features are screened from the multidimensional standard feature according to the analysis results, including: Based on a preset maximum information coefficient algorithm, correlation analysis is performed on the standard features of each dimension and the outlet steam temperature data to obtain corresponding correlation analysis results; screening a plurality of initial key features from the multidimensional standard features according to the correlation analysis result, and performing a correlation analysis on each of the plurality of initial key features and the outlet steam temperature data to obtain a corresponding correlation analysis result; Based on the correlation analysis result, the multiple key features that meet preset correlation requirements are screened out from the multiple initial key features.

4. The method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit according to claim 1, characterized in that: The corresponding outlet steam temperature prediction model is established based on the preset long short-term memory network, time convolution network and back propagation network, including: Connecting the long short-term memory network, the temporal convolutional network, and the back propagation network in parallel to obtain a corresponding comprehensive model; The comprehensive model is trained using the multiple key features to construct the outlet steam temperature prediction model.

5. A method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, applied in the online prediction stage, characterized in that: The method for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired power plant under variable operating conditions applied to an offline training phase according to any one of claims 1 to 4 is adopted, wherein the method comprises the following steps: Collecting actual operating data corresponding to the target coal-fired unit and preprocessing the actual operating data to obtain corresponding actual standard data; The actual standard data is input into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

6. The method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit according to claim 5, characterized in that: After outputting the outlet steam temperature prediction result corresponding to the target coal-fired unit, the method further includes: Based on at least one preset prediction and evaluation strategy, performing a prediction and evaluation operation on the outlet steam temperature prediction result to obtain corresponding actual prediction and evaluation data; The integrated high-temperature superheater outlet steam temperature prediction model is fine-tuned according to the actual prediction evaluation data, so as to re-execute the outlet steam temperature prediction operation of the target coal-fired unit using the fine-tuned integrated high-temperature superheater outlet steam temperature prediction model.

7. A device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, used in the offline training phase, characterized in that: include: An extraction module is used to obtain operating data corresponding to the target coal-fired unit, extract multidimensional features corresponding to the operating data, and preprocess the multidimensional features to obtain corresponding multidimensional standard features; a modeling module, configured to perform nonlinear correlation analysis on the standard features of each dimension of the multidimensional standard features and the outlet steam temperature data in the operating data to obtain corresponding analysis results, screen a plurality of key features from the multidimensional standard features based on the analysis results, and establish a corresponding outlet steam temperature prediction model based on a preset long short-term memory network, a temporal convolutional network, and a back propagation network; A training module is used to input the multiple key features into the outlet steam temperature prediction model respectively to obtain the output features corresponding to the long short-term memory network, the temporal convolution network and the back propagation network in the outlet steam temperature prediction model, and based on a preset attention mechanism, weight the output features to establish a corresponding integrated high-temperature superheater outlet steam temperature prediction model, and train the integrated high-temperature superheater outlet steam temperature prediction model to use the trained integrated high-temperature superheater outlet steam temperature prediction model in the online prediction stage to predict the actual high-temperature superheater outlet steam temperature of the target coal-fired unit under different operating conditions.

8. A device for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit, used in the online prediction stage, characterized in that: The device for predicting the outlet steam temperature of a high-temperature superheater of a coal-fired unit under variable operating conditions for offline training according to claim 7 is used, wherein the device comprises: a preprocessing module, configured to collect actual operating data corresponding to the target coal-fired unit and preprocess the actual operating data to obtain corresponding actual standard data; The prediction module is used to input the actual standard data into the integrated high-temperature superheater outlet steam temperature prediction model to output the outlet steam temperature prediction result corresponding to the target coal-fired unit.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit as claimed in any one of claims 1 to 5 or claim 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for predicting the outlet steam temperature of a high-temperature superheater of a variable-operating-condition coal-fired unit according to any one of claims 1 to 5 or claim 6.

Images