Air preheater control system and method based on big data model analysis

Abstract

The invention discloses an air preheater control system and method based on big data model analysis, and belongs to the technical field of air preheater control. The system comprises a data acquisition module, a data preprocessing module, a big data analysis module, an optimization control module, a fault early warning module and a man-machine interaction module. The big data analysis module extracts a performance trend by using polynomial regression fitting, and determines a causal relationship through a Bayesian network; the optimization control module generates an optimization operation scheme by adopting a graph optimization algorithm and a dynamic adjustment strategy based on a causal relationship; and the fault early warning module constructs a prediction network according to the causal relationship and the performance trend to realize fault early warning. The method comprises the steps of collecting and preprocessing operation data, analyzing and determining a causal relationship, dynamically adjusting an operation scheme, predicting faults, performing early warning and performing visual display. Accurate monitoring, dynamic optimization and fault early warning of the operation state of the air preheater are achieved, and the operation efficiency and reliability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air preheater control, and more particularly to an air preheater control system and method based on big data model analysis. Background Art

[0002] Air preheater is an important heat exchange equipment in thermal power generation system. Its main function is to use the waste heat of flue gas to heat the air entering the boiler, thereby improving combustion efficiency, reducing fuel consumption and reducing pollutant emissions.

[0003] However, the current traditional air preheater control system has exposed many problems. From the perspective of energy consumption, once there is a load fluctuation or a change in operating conditions, the system cannot respond in real time and adjust the operating parameters. For example, when the factory production load decreases, the air preheater still maintains the original fan speed and valve opening, resulting in energy being consumed in vain, causing great waste. In terms of equipment loss, long-term operation in non-optimal conditions has increased the wear of various equipment components. For example, the fan has been running at an unmatched speed for a long time, which has accelerated the wear of key parts such as bearings, causing a significant increase in the equipment failure rate, which not only shortens the service life of the equipment, but also increases the high maintenance and replacement costs. In terms of control accuracy, facing complex and changeable working conditions, such as changes in fuel types and large fluctuations in ambient temperature, it is difficult for traditional control systems to achieve precise control. This makes it impossible for the operating parameters of the air preheater to accurately adapt to changes, seriously affecting the control effect, and thus interfering with the stability and efficiency of the entire production process.

[0004] In recent years, with the development of big data technology, its application in the industrial field has gradually increased. Big data model analysis can realize real-time monitoring, fault prediction and optimization control of equipment operation status through mining and analysis of massive operation data. For example, models based on BP neural network and support vector regression (SVR) have been applied to the prediction of air preheater inlet and outlet pressure difference. By optimizing the model structure and algorithm, the prediction accuracy can be effectively improved. In addition, the rise of digital twin technology has also provided new ideas for the intelligent control of air preheaters. By constructing a digital twin model, combining multi-source heterogeneous data fusion and real-time data collection, dynamic monitoring of internal thermal parameters of air preheaters and dust accumulation prediction can be achieved.

[0005] Despite this, the existing technology still has some shortcomings. For example, some big data models have problems such as high computational complexity and insufficient model generalization ability in practical applications; and the application of digital twin technology in the field of air preheaters is still in its early stages and has not yet formed a mature control system. Therefore, how to provide an air preheater control system and method based on big data model analysis to achieve accurate monitoring of the operating status of the air preheater, fault prediction and optimization control is a problem that technicians in this field urgently need to solve. Summary of the invention

[0006] In view of this, the present invention provides an air preheater control system and method based on big data model analysis, which realizes intelligent operation management of the air preheater through big data model analysis, improves equipment operation efficiency, reduces energy consumption and failure rate, and enhances operational convenience with the help of a visual interface, thereby promoting the transformation of the air preheater from the traditional experience management mode to the data-driven intelligent management mode.

[0007] In order to achieve the above object, the present invention adopts the following technical solution: In one aspect, the present invention provides an air preheater control system based on big data model analysis, comprising: Data acquisition module: used to collect the operating data of the air preheater in real time; Data preprocessing module: used for preprocessing the operation data; Big data analysis module: used to analyze the preprocessed operation data, extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and the operation data to determine the cause-effect relationship between different operation data; Optimization control module: used for dynamically adjusting the operation plan of the air preheater according to the cause-effect relationship; Fault warning module: used to analyze the operation data according to the performance trend and the causal relationship, predict potential faults and issue warnings; Human-computer interaction module: used to provide a visual interface to facilitate operators to monitor the operating status and adjust parameters of the air preheater.

[0008] Preferably, the big data analysis module includes: a trend extraction unit, configured to extract a performance trend of the air preheater by fitting the preprocessed operation data using a polynomial regression; The causal analysis unit is used to perform causal analysis on the operation data using a Bayesian network to determine the causal relationship between different operation data.

[0009] Preferably, the causal analysis unit is configured as follows: Automatically learning the structure of the Bayesian network from the operation data, and constructing a causal relationship map between the operation data according to the results of the Bayesian network; Determine the conditional probability distribution of each node in the Bayesian network according to the performance trend, and quantify the causal relationship between the operating data; Causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data.

[0010] Preferably, the optimization control module includes: A graph structure building unit, which is used to reconfigure the process network structure of the air preheater operation process based on cause-effect relationships and using a graph optimization algorithm; A dynamic adjustment unit, for adjusting the operation state according to the real-time operation data and generating a dynamic operation plan based on the process network structure and adopting a dynamic adjustment strategy; A performance monitoring subunit is used to continuously monitor the operating status of the air preheater based on a dynamic adjustment scheme and a performance evaluation method, detect the performance of the air preheater in real time according to performance indicators, and generate performance monitoring results; The optimization unit is used to optimize the dynamic adjustment plan based on the performance monitoring result by using a mixed integer linear programming algorithm to generate an optimized operation plan.

[0011] Preferably, the fault warning module is configured as follows: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

[0012] Preferably, performing causal inference based on the Bayesian network to identify the causal relationship of the target operation data includes: Use Bayesian network for causal inference and calculate the target node V of interest t Conditional probability distribution, let the target node be V t , the variable of interest is X, and the causal inference formula is: ; Among them, Parents represents the target node V t All parent nodes of Through causal inference, we identify the variables that have the greatest impact on the target node.

[0013] On the other hand, the present invention provides an air preheater control method based on big data model analysis, comprising the following steps: Collecting the operating data of the air preheater in real time and preprocessing the operating data; Analyze the preprocessed operation data to extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and the operation data to determine the cause-effect relationship between different operation data; According to the cause-effect relationship, dynamically adjust the operation plan of the air preheater; Analyze the operation data according to the performance trend and the causal relationship, predict potential failures and issue early warnings; The operation scheme and operation status are displayed visually.

[0014] Preferably, the preprocessed operation data is analyzed to extract the performance trend of the air preheater, and a data analysis model is established based on the performance trend and the operation data to determine the causal relationship between different operation data, including: Fitting the preprocessed operating data using polynomial regression to extract performance trends of the air preheater; Automatically learning the structure of the Bayesian network from the operation data, and constructing a causal relationship map between the operation data according to the results of the Bayesian network; Determine the conditional probability distribution of each node in the Bayesian network according to the performance trend, and quantify the causal relationship between the operating data; Causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data.

[0015] Preferably, dynamically adjusting the operation scheme of the air preheater according to the cause-effect relationship includes: Based on the cause-effect relationship, the graph optimization algorithm is used to reconfigure the process network structure of the air preheater operation process; Based on the process network structure, a dynamic adjustment strategy is adopted to adjust the operation status according to real-time operation data to generate a dynamic operation plan; Based on the dynamic adjustment scheme, a performance evaluation method is adopted to continuously monitor the operating status of the air preheater, detect the performance of the air preheater in real time according to the performance index, and generate a performance monitoring result; Based on the performance monitoring results, a mixed integer linear programming algorithm is used to optimize the dynamic adjustment plan to generate an optimized operation plan.

[0016] Preferably, analyzing the operating data according to the performance trend and the causal relationship, predicting potential failures and issuing early warnings, includes: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

[0017] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses an air preheater control system and method based on big data model analysis, which realizes the intelligent operation and management of the air preheater through modules such as data acquisition, preprocessing, big data analysis, optimization control, fault warning and human-computer interaction. The present invention uses polynomial regression and Bayesian network to extract performance trends and quantify the causal relationship of operation data; optimizes the operation plan through graph optimization algorithm and dynamic adjustment strategy; combines performance monitoring with mixed integer linear programming algorithm to achieve continuous optimization of the operation plan; and fault warning function based on performance trend and causal relationship. The present invention can significantly improve the operating efficiency of the air preheater, reduce energy consumption and operation and maintenance costs. Its dynamic adjustment and fault warning functions can reduce equipment downtime, improve equipment reliability and service life. At the same time, the visual human-computer interaction interface improves the convenience of operation and monitoring efficiency. The present invention not only provides new technical means for the intelligent control of industrial equipment, but also promotes the development of industrial production towards high efficiency, energy saving and intelligence, and has important application value and promotion significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0019] Figure 1 A schematic diagram of the structure provided by the present invention.

[0020] Figure 2 A schematic diagram of the process provided by the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] The embodiment of the present invention discloses an air preheater control system based on big data model analysis, such as Figure 1 As shown, including: Data acquisition module: used to collect the operating data of the air preheater in real time; the operating data includes parameters such as flue gas temperature, flue gas oxygen content, flow, pressure, equipment vibration, etc. High-precision sensors can be used in combination with a distributed data acquisition system to achieve multi-channel, high-frequency data acquisition. For example, the temperature data acquisition module measures the temperature data in real time through sensors attached to each compartment of the air preheater, and the pressure acquisition module is used to collect the pressure difference information of each compartment.

[0023] Data preprocessing module: used to preprocess the operation data; preprocessing includes cleaning, normalizing, and denoising the collected operation data to improve data quality. Data transformation techniques, such as standardization and normalization, can be used to ensure uniform data formats. In addition, by introducing data filtering algorithms, outliers and noise data are removed to provide a reliable data basis for subsequent analysis. Furthermore, for time series data, it is also necessary to perform time series decomposition on the acquired time series data to obtain trend series data.

[0024] Big data analysis module: used to analyze the preprocessed operation data, extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and operation data to determine the cause-effect relationship between different operation data; Optimization control module: used to dynamically adjust the operation plan of the air preheater based on cause and effect relationships; Fault warning module: used to analyze operating data based on performance trends and causal relationships, predict potential faults and issue warnings; for example, based on short-term and medium-term evolution models of key parameters such as pressure difference and temperature, the blockage and corrosion risks of air preheaters are evaluated. When the probability of a fault exceeding the preset threshold, the system will automatically issue an alarm signal to remind the operator to take measures.

[0025] Human-computer interaction module: It is used to provide a visual interface to facilitate operators to monitor the operating status of the air preheater and adjust parameters. The interface can display the 3D model, real-time data, historical trends and warning information of the equipment, and support operators to adjust parameters and optimize strategies through the interface. In addition, through digital twin technology, the real-time visualization of the equipment operating status is realized, which enhances the operator's intuitive understanding of the equipment operation status.

[0026] Furthermore, the big data analysis module includes: The trend extraction unit is used to extract the performance trend of the air preheater by fitting the preprocessed operating data using polynomial regression, including: The preprocessed operating data are arranged in time series, and key performance indicators (such as temperature, pressure, flow rate, etc.) are selected as dependent variables and time as the independent variable.

[0027] Select the appropriate polynomial order (such as quadratic or cubic polynomial) according to the data characteristics, and fit the polynomial model through the least squares method or other optimization algorithms.

[0028] By fitting the polynomial model, the long-term trend and short-term fluctuation of equipment performance indicators are calculated, providing a basis for subsequent causal analysis and optimization control.

[0029] The causal analysis unit is used to perform causal analysis on the operation data using a Bayesian network to determine the causal relationship between different operation data.

[0030] Furthermore, the causal analysis unit is configured to: The structure of the Bayesian network is automatically learned from the operating data, and a causal relationship map between the operating data is constructed based on the results of the Bayesian network.

[0031] The Bayesian network consists of a directed acyclic graph (DAG) and a conditional probability table (CPT), where DAG is used to represent the causal relationship between variables. The goal of structural learning is to determine the dependency between variables. Commonly used algorithms include the PC algorithm, the K2 algorithm, and the Hill-Climb Search algorithm. For example, the PC algorithm can be used to gradually build a network structure through conditional independence tests. Specifically: Define each running data as a node, and set the node set to be V = {V1, V2, ..., V N}, where V i represents the i-th running data, the edges between nodes represent causal relationships, and the edge set is E = {E1, E2, ..., E M}, where E i represents the i-th causal relationship edge; Use the Bayesian network structure learning algorithm to learn the network structure from the data, iteratively update the nodes and edges until the scoring function converges or reaches the preset number of iterations, as shown below: Suppose the scoring function is S(G|D), where G is the candidate network structure and D is the data set. The optimal network structure G* is found by maximizing the scoring function S(G|D), which is expressed as follows: .

[0032] Determine the conditional probability distribution of each node in the Bayesian network based on the performance trend and quantify the causal relationship between the running data. Specifically: For each node V i , learn its conditional probability table P (V i |Parents(V i )), where Parents(V i ) represents node V i The parent node set of Vi The parent node set is {V i1 , V i2 , …, V ik}, then the conditional probability table P (V i |V i1 , V i2 , …, V ik ) is calculated as follows: ; Use the expectation maximization algorithm to estimate the parameters in the conditional probability table. Let the parameter be θ, and estimate the parameter by maximizing the log-likelihood function L(θ|D), which is expressed as follows: ; The specific form of the log-likelihood function L(θ|D) is: ; Among them, M represents the number of samples observed in the conditional probability table, N represents the number of nodes in the Bayesian network, and D ij represents the observed value of the i-th node in the j-th sample, and θ represents the parameter set in the conditional probability table.

[0033] Causal inference is performed based on Bayesian networks to identify the causal relationship of target operation data.

[0034] In another embodiment, the optimization control module includes: The graph structure building unit is used to reconfigure the process network structure of the air preheater operation process based on causal relationships and using a graph optimization algorithm.

[0035] According to the Bayesian network analysis results, a directed graph is constructed by taking the operating data as nodes in the graph and the causal relationships as directed edges. For example, temperature changes directly affect pressure changes, so a directed edge from the temperature node to the pressure node is added to the graph.

[0036] Use graph theory algorithms (such as Dijkstra algorithm, Floyd-Warshall algorithm, etc.) to optimize the process network structure. For example, the Dijkstra algorithm is used to find the shortest path from input parameters to target performance indicators, thereby optimizing the operation process and ensuring efficient transmission and adjustment of key parameters.

[0037] The dynamic adjustment unit is used to adjust the operation status according to the real-time operation data based on the process network structure and adopt the dynamic adjustment strategy to generate a dynamic operation plan.

[0038] The operation data of the air preheater is collected in real time through sensors, and combined with the optimized process network structure, the operation data of the air preheater is dynamically adjusted. For example, when a temperature anomaly is detected, the relevant parameters are automatically adjusted according to the cause-effect relationship in the graph structure to restore normal operation.

[0039] Drawing on the collaborative mechanism of multi-agent systems, the different components of the air preheater are regarded as agents, and information is transmitted and coordinated through the graph structure. For example, when a component fails, other components can dynamically compensate based on the information in the graph structure.

[0040] The performance monitoring subunit is used to continuously monitor the operating status of the air preheater based on the dynamic adjustment scheme and the performance evaluation method, detect the performance of the air preheater in real time according to the performance indicators, and generate performance monitoring results; The optimization unit is used to optimize the dynamic adjustment plan based on the performance monitoring results by using a mixed integer linear programming algorithm to generate an optimized operation plan.

[0041] The performance monitoring results are used as input, combined with the upper and lower limit constraints of the operating data, and the MILP algorithm is used to optimize the operating plan. For example, the MILP algorithm is used to determine the optimal temperature, pressure and flow combination to improve equipment efficiency.

[0042] Feedback the optimization results into the graph structure, dynamically adjust the graph structure and operation data, and achieve continuous optimization. For example, update the weights or paths in the graph based on the optimization results to further improve device performance.

[0043] In another embodiment, the fault warning module is configured to: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

[0044] Furthermore, causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data, including: Use Bayesian network for causal inference and calculate the target node V of interest t Conditional probability distribution, let the target node be V t , the variable of interest is X, and the causal inference formula is: ; Among them, Parents represents the target node V t All parent nodes of Through causal inference, we identify the variables that have the greatest impact on the target node.

[0045] On the other hand, the present invention provides an air preheater control method based on big data model analysis, such as Figure 2 As shown, the following steps are included: Collect the operating data of the air preheater in real time and pre-process the operating data; Analyze the preprocessed operation data to extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and operation data to determine the cause-effect relationship between different operation data; Dynamically adjust the operation plan of the air preheater based on cause and effect relationships; Analyze operating data based on performance trends and cause-and-effect relationships to predict potential failures and provide early warnings; Visualize the operation plan and operation status.

[0046] Furthermore, the preprocessed operation data is analyzed to extract the performance trend of the air preheater, and a data analysis model is established based on the performance trend and operation data to determine the causal relationship between different operation data, including: Use polynomial regression to fit preprocessed operating data to extract performance trends of air preheaters; Automatically learn the structure of the Bayesian network from the operating data, and construct a causal relationship map between the operating data based on the results of the Bayesian network; Determine the conditional probability distribution of each node in the Bayesian network based on performance trends and quantify the causal relationship between operating data; Causal inference is performed based on Bayesian networks to identify the causal relationship of target operation data.

[0047] In another embodiment, dynamically adjusting the operation scheme of the air preheater according to the cause-effect relationship includes: Based on the cause-effect relationship, the graph optimization algorithm is used to reconfigure the process network structure of the air preheater operation process; Based on the process network structure, a dynamic adjustment strategy is adopted to adjust the operation status according to real-time operation data and generate a dynamic operation plan; Based on the dynamic adjustment scheme, the performance evaluation method is adopted to continuously monitor the operating status of the air preheater, detect the air preheater performance in real time according to the performance indicators, and generate performance monitoring results; Based on the performance monitoring results, a mixed integer linear programming algorithm is used to optimize the dynamic adjustment plan and generate an optimized operation plan.

[0048] Furthermore, the operating data is analyzed based on performance trends and causal relationships to predict potential failures and provide early warnings, including: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

[0049] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0050] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An air preheater control system based on big data model analysis, characterized in that: include: Data acquisition module: used to collect the operating data of the air preheater in real time; Data preprocessing module: used for preprocessing the operation data; Big data analysis module: used to analyze the preprocessed operation data, extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and the operation data to determine the cause-effect relationship between different operation data; Optimization control module: used for dynamically adjusting the operation plan of the air preheater according to the cause-effect relationship; Fault warning module: used to analyze the operation data according to the performance trend and the causal relationship, predict potential faults and issue warnings; Human-computer interaction module: used to provide a visual interface to facilitate operators to monitor the operating status and adjust parameters of the air preheater.

2. The air preheater control system based on big data model analysis according to claim 1 is characterized in that: The big data analysis module includes: a trend extraction unit, configured to extract a performance trend of the air preheater by fitting the preprocessed operation data using a polynomial regression; The causal analysis unit is used to perform causal analysis on the operation data using a Bayesian network to determine the causal relationship between different operation data.

3. The air preheater control system based on big data model analysis according to claim 2 is characterized in that: The causal analysis unit is configured to: Automatically learning the structure of the Bayesian network from the operation data, and constructing a causal relationship map between the operation data according to the results of the Bayesian network; Determine the conditional probability distribution of each node in the Bayesian network according to the performance trend, and quantify the causal relationship between the operating data; Causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data.

4. The air preheater control system based on big data model analysis according to claim 1, characterized in that: The optimization control module comprises: A graph structure building unit, which is used to reconfigure the process network structure of the air preheater operation process based on cause-effect relationships and using a graph optimization algorithm; A dynamic adjustment unit, for adjusting the operation state according to the real-time operation data and generating a dynamic operation plan based on the process network structure and adopting a dynamic adjustment strategy; A performance monitoring subunit is used to continuously monitor the operating status of the air preheater based on a dynamic adjustment scheme and a performance evaluation method, detect the performance of the air preheater in real time according to performance indicators, and generate performance monitoring results; The optimization unit is used to optimize the dynamic adjustment plan based on the performance monitoring result by using a mixed integer linear programming algorithm to generate an optimized operation plan.

5. The air preheater control system based on big data model analysis according to claim 1, characterized in that: The fault warning module is configured as follows: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

6. The air preheater control system based on big data model analysis according to claim 3 is characterized in that: Causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data, including: Use Bayesian network for causal inference and calculate the target node V of interest t Conditional probability distribution, let the target node be V t , the variable of interest is X, and the causal inference formula is: ； Among them, Parents represents the target node V t All parent nodes of Through causal inference, we identify the variables that have the greatest impact on the target node.

7. An air preheater control method based on big data model analysis, characterized in that: The following steps are involved: Collecting the operating data of the air preheater in real time and preprocessing the operating data; Analyze the preprocessed operation data to extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and the operation data to determine the cause-effect relationship between different operation data; According to the cause-effect relationship, dynamically adjust the operation plan of the air preheater; Analyze the operation data according to the performance trend and the causal relationship, predict potential failures and issue early warnings; The operation scheme and operation status are displayed visually.

8. The air preheater control method based on big data model analysis according to claim 7 is characterized in that: Analyze the preprocessed operation data to extract the performance trend of the air preheater, and establish a data analysis model based on the performance trend and the operation data to determine the causal relationship between different operation data, including: Fitting the preprocessed operating data using polynomial regression to extract performance trends of the air preheater; Automatically learning the structure of the Bayesian network from the operation data, and constructing a causal relationship map between the operation data according to the results of the Bayesian network; Determine the conditional probability distribution of each node in the Bayesian network according to the performance trend, and quantify the causal relationship between the operating data; Causal inference is performed based on the Bayesian network to identify the causal relationship of the target operation data.

9. The air preheater control method based on big data model analysis according to claim 7, characterized in that: According to the cause-effect relationship, dynamically adjusting the operation plan of the air preheater includes: Based on the cause-effect relationship, the graph optimization algorithm is used to reconfigure the process network structure of the air preheater operation process; Based on the process network structure, a dynamic adjustment strategy is adopted to adjust the operation status according to real-time operation data to generate a dynamic operation plan; Based on the dynamic adjustment scheme, a performance evaluation method is adopted to continuously monitor the operating status of the air preheater, detect the performance of the air preheater in real time according to the performance index, and generate a performance monitoring result; Based on the performance monitoring results, a mixed integer linear programming algorithm is used to optimize the dynamic adjustment plan to generate an optimized operation plan.

10. The air preheater control method based on big data model analysis according to claim 7, characterized in that: Analyzing the operating data according to the performance trend and the causal relationship, predicting potential failures and issuing early warnings, including: Constructing a prediction network to predict a plurality of second operation data having a causal relationship with the target operation data and the performance trend according to the causal relationship of the target operation data and the performance trend; The probability of a fault occurring is determined according to the performance trends of the plurality of second operating data, and a fault alarm is issued when the probability is higher than a preset value.

Images