Early warning method and system for leakage of four pipes of boiler

Abstract

The invention relates to the technical field of thermal power generation, in particular to a boiler four-tube leakage early warning method and system, and the method comprises the following steps: S1, obtaining operation parameters of a power station and dissolved hydrogen data of boiler four tubes; s2, the power station operation parameters and the dissolved hydrogen data are stored in a time sequence database through data middleware; s3, fault characteristic parameters related to boiler four-tube leakage in the power station operation parameters are selected, and data preprocessing is carried out; s4, determining an optimal neural network prediction model based on the fault feature parameters; s5, boiler parameters are obtained, and a pipe wall temperature prediction model of four pipes of the boiler is established; s6, establishing an oxide film thickness calculation model, and establishing a corrosion monitoring model based on the oxide film thickness calculation model, the pipe wall temperature prediction model and the dissolved hydrogen data; and S7, a fusion model is constructed based on the optimal neural network prediction model and the corrosion monitoring model, boiler four-tube leakage early warning is carried out based on a fusion value, and the method has higher accuracy and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of thermal power generation, and in particular to a boiler four-tube leakage early warning method and system. Background Art

[0002] A thermal power plant is abbreviated as a thermal power plant. It is a factory that uses combustibles as fuel to produce electricity. A thermal power plant includes a fuel system, a combustion system, a steam-water system, an electrical system, and a control system. The most important equipment in the above systems is the boiler, steam turbine, and generator, which are installed in the main plant of the power plant.

[0003] Once a thermal power plant is turned on, it will be in a state of long-term operation. The economizer tubes, water-cooled wall tubes, superheater tubes and reheater tubes in the boiler are collectively referred to as the four tubes, which play an important role in heating, evaporation and transmission. When the boiler works in a high temperature and high pressure environment for a long time, pipeline leakage of the four tubes of the boiler will occur from time to time.

[0004] In the related art, the leakage warning of the four boiler tubes mainly adopts the acoustic detection method based on signal processing technology and the ray detection method based on ray detection. In the acoustic detection method based on signal processing technology, the detection result of the acoustic detection device is highly dependent on the quantity and quality of process data, the deployment position of the acoustic detection device, the interference of background noise and the attenuation of sound waves. In the ray detection method based on ray detection, the ray detection device can detect pores, cracks and oxide layers that cannot be detected by the naked eye. Therefore, the deployment position, deployment number, refraction angle of the ray, decay coefficient and imaging technology of the ray detection device will affect the detection result of the leakage. It can be seen that the above two methods cannot detect boiler pipeline leakage in advance, and can only detect the abnormal sound of "burst pipe" after the leakage occurs.

[0005] Since the four tubes in the boiler are interconnected, the operating parameters are highly coupled. This structure usually shows certain similarities in the failure phenomena when the pipeline leaks. In this context, the neural network technology based on big data shows a natural advantage in the detection of boiler four-tube leaks. However, relying solely on data processing may cause the results to deviate from the actual situation. In particular, considering the diversity of coal-fired boiler operating conditions, even if the same type of heating surface leaks under different operating conditions, the corresponding operating parameter changes may be different. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a boiler four-tube leakage early warning method and system. The above method and system use a dissolved hydrogen detection device and deep learning to warn of the leakage of the boiler four tubes, which has higher accuracy and reliability than the existing technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a boiler four-tube leakage early warning method, comprising the following steps:

[0009] S1: Obtain the power plant operation parameters and obtain the dissolved hydrogen data of the four boiler tubes based on the dissolved hydrogen detection device;

[0010] S2: Build database middleware and store power plant operation parameters and dissolved hydrogen data in a time series database through the data middleware;

[0011] S3: selecting fault characteristic parameters related to boiler four-tube leakage from power plant operation parameters, and performing data preprocessing on the fault characteristic parameters;

[0012] S4: Determine the optimal neural network prediction model based on the fault characteristic parameters after data preprocessing;

[0013] S5: Obtain boiler parameters, and establish a wall temperature prediction model for four boiler tubes based on the boiler parameters;

[0014] S6: Establish an oxide film thickness calculation model based on the tube wall temperature prediction model, and establish a corrosion monitoring model based on the oxide film thickness calculation model, the tube wall temperature prediction model and dissolved hydrogen data;

[0015] S7: A fusion model is constructed based on the optimal neural network prediction model and the corrosion monitoring model. A fusion value is obtained by weighted averaging of multi-source redundant information based on the fusion model. A boiler four-tube leakage warning is performed based on the fusion value.

[0016] By adopting the above technical solution, when the boiler four tubes are warned of leakage, the power plant operation parameters can be directly obtained, and the dissolved hydrogen data in the water vapor of the boiler four tubes can be obtained through the dissolved hydrogen detection device. After that, the power plant operation parameters and dissolved hydrogen data can be stored in the time series database through the constructed data middleware for calling. Then the power plant operation parameters are screened to obtain the fault characteristic parameters related to the leakage of the boiler four tubes. Then, after processing the fault characteristic parameters, the optimal neural network prediction model for determining the fault can be obtained. At the same time, by obtaining the boiler parameters, a tube wall temperature prediction model of the boiler four tubes can be established, and the tube wall temperature of the boiler four tubes can be obtained through the tube wall temperature prediction model. Based on the tube wall temperature, the oxide film thickness calculation model can be determined. The oxide film thickness calculation model can obtain the oxide film thickness in the boiler four tubes, and the corrosion monitoring model of the boiler four tubes can be established through the tube wall temperature, oxide film thickness and dissolved hydrogen data.

[0017] Finally, the optimal neural network prediction model is fused with the corrosion monitoring model to obtain a fusion model. This model can perform weighted average on the leakage warning information output by the optimal neural network prediction model and the leakage warning information output by the corrosion monitoring model to obtain a fusion value. By analyzing this fusion value, we can determine the leakage of the four boiler tubes and thus realize the leakage warning of the four boiler tubes.

[0018] Optionally, step S3 includes the following steps:

[0019] S31: selecting fault characteristic parameters related to boiler four-tube leakage from power plant operation parameters;

[0020] S32: performing data cleaning on the fault characteristic parameters to obtain a cleaned data set;

[0021] S33: Performing data dimensionality reduction on the cleaned data set to obtain low-dimensional feature parameters;

[0022] S34: Divide the low-dimensional feature parameters into training set, validation set and test set.

[0023] By adopting the above technical solution, the power plant operation parameters include the operation parameters of the boiler, steam turbine, generator and other equipment in the power plant. By selecting the power plant operation parameters, the fault characteristic parameters related to the boiler four-tube leakage can be obtained. Then, the fault characteristic parameters are preprocessed and the fault characteristic parameters are subjected to data dimension reduction to obtain low-dimensional feature parameters. Data dimension reduction can keep the original information structure unchanged or maximize the characteristics of specific aspects. Finally, the low-dimensional feature parameters are subjected to data normalization processing to be divided into training set, verification set and test set for use in subsequent steps.

[0024] Optionally, the fault characteristic parameters include: actual load, main steam pressure, feed water flow, induced draft impeller opening, left high-temperature superheater outlet steam temperature, right high-temperature superheater outlet steam temperature, left high-temperature superheater outlet steam pressure, right high-temperature superheater outlet steam pressure, left superheater secondary cooling water flow, right superheater secondary cooling water flow, left high-temperature reheater outlet steam temperature, right high-temperature reheater outlet steam temperature, left high-temperature reheater outlet steam pressure, right high-temperature reheater outlet steam pressure, left high-temperature reheater cooling water flow, right reheater cooling water flow, left and right economizer outlet flue gas temperature difference, economizer outlet feed water temperature, water-cooled wall outlet header temperature, steam-water separator outlet steam pressure.

[0025] By adopting the above technical solution and acquiring the above fault characteristic parameters, data related to the leakage of the four boiler tubes can be obtained.

[0026] Optionally, step S4 includes the following steps:

[0027] S41: Construct a CNN-LSTM neural network model based on the training set, validation set, convolutional neural network model and long short-term memory network model;

[0028] S42: Determine the structural parameters of the optimal neural network prediction model based on the validation set, test set and CNN-LSTM neural network model.

[0029] By adopting the above technical solution, a CNN-LSTM neural network model can be constructed through a convolutional neural network model and a long short-term memory network model, the CNN-LSTM neural network model can be trained through a training set, and the CNN-LSTM neural network model can be tested through a test set.

[0030] Optionally, the step S5 includes the following steps:

[0031] S51: Obtain boiler parameters, and establish a combustion numerical simulation model based on the boiler parameters;

[0032] The boiler parameters include: fuel characteristics, fuel flow, furnace size, furnace temperature, pressure and flue gas output. The boiler parameters are divided using unstructured grids. The boundary conditions of the combustion numerical simulation model include: primary air volume and temperature, air volume and temperature around the primary air nozzle, exhaust gas volume and temperature, air volume and temperature around the exhaust gas nozzle, distribution of secondary air volume and secondary air temperature, and burnout air volume and temperature.

[0033] S52: Establish a prediction model for the wall temperature of the four boiler tubes based on the combustion numerical simulation model.

[0034] By adopting the above technical solution, a combustion numerical simulation model can be established through boiler parameters, and then a tube wall temperature prediction model of the four tubes of the boiler can be established through the combustion numerical simulation model.

[0035] Optionally, the combustion numerical simulation model is verified by comparing the simulation values ​​and design values ​​of the average flue gas temperatures at the inlet and outlet of boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes under boiler maximum output conditions, boiler rated load conditions, heat rate acceptance conditions, 40% THA, 50% THA, and 75% THA conditions.

[0036] By adopting the above technical scheme, the combustion numerical simulation model can be verified by comparing the simulation values ​​and design values ​​of the average flue gas temperatures at the inlet and outlet of the boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes under the boiler maximum output condition, boiler rated load condition, heat rate acceptance condition, 40% THA, 50% THA, and 75% THA conditions.

[0037] Optionally, step S6 includes the following steps:

[0038] S61: determining the outer wall temperature of the four boiler tubes based on the tube wall temperature prediction model and the power station operation parameters;

[0039] S62: Establishing a calculation model for oxide film thickness based on the outer wall temperature of the four boiler tubes;

[0040] S63: A corrosion monitoring model is established based on the oxide film thickness calculation model, the pipe wall temperature prediction model and the dissolved hydrogen data.

[0041] By adopting the above technical solution, the outer wall temperature of the four boiler tubes can be determined by inputting the power plant operation parameters into the tube wall temperature prediction model, and the oxide film thickness calculation model can be established through the outer wall temperature. Finally, the corrosion monitoring model can be established through the tube wall temperature and dissolved hydrogen data after the oxide film.

[0042] Optionally, the boiler four-tube leakage early warning method further includes the following steps:

[0043] S8: Construct a cockpit for boiler four-tube leakage warning data and display the boiler four-tube leakage warning results.

[0044] By adopting the above technical solution, the cockpit can display the leakage warning results of the four boiler tubes.

[0045] In a second aspect, the present invention provides a boiler four-tube leakage warning system, comprising a dissolved hydrogen detection device and a processing device, wherein the dissolved hydrogen detection device is connected to the economizer tube, the water-cooled wall tube, the superheater tube and the reheater tube, and the processing device is communicatively connected to the dissolved hydrogen detection device. The processing device comprises a processor, a memory and a communication bus, wherein the communication bus is used to realize the communication connection between the processor and the memory, and the processor is used to execute a computer program stored in the memory to realize the boiler four-tube leakage warning method as described in any one of the first aspects.

[0046] By adopting the above technical solution, the dissolved hydrogen detection device can obtain the dissolved hydrogen data in the water vapor of the four boiler tubes and send it to the processing equipment. The processing equipment can obtain and store the dissolved hydrogen data, power plant operation parameters, and boiler parameters. After that, the processing equipment can screen and preprocess the fault characteristic parameters in the power plant operation data to determine the optimal neural network prediction model. And establish the corresponding tube wall temperature prediction model of the four boiler tubes through the boiler parameters, and then establish the oxide film thickness calculation model according to the tube wall temperature, and then establish the corrosion monitoring model according to the oxide film thickness, tube wall temperature and dissolved hydrogen data. Finally, a fusion model is constructed by the optimal neural network prediction model and the corrosion monitoring model to obtain the fusion value, and the leakage of the four boiler tubes is determined by the fusion value, so as to realize the leakage warning, which has higher accuracy and reliability than the existing technology.

[0047] Optionally, the boiler four-tube leakage warning system also includes a cockpit, which is communicatively connected to the processing equipment.

[0048] By adopting the above technical solution, the cockpit can display various data obtained by the processing equipment.

[0049] In summary, the present invention at least includes the following beneficial technical effects:

[0050] The chemical water samples collected by this boiler four-tube leakage warning method can fully represent the operation of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tubes. Since the metal wall is always accompanied by extremely small amounts of corrosion in high-temperature water and steam, and produces extremely small amounts of dissolved hydrogen, trace dissolved hydrogen is a corrosion product, most of which is dissolved in water vapor. Dissolved hydrogen can more truly reflect the corrosion situation. The corrosion monitoring model combined with the long-term monitoring data of trace dissolved hydrogen can effectively monitor and warn of boiler four-tube leakage. In addition, this boiler four-tube leakage warning method does not require the layout of the entire boiler. By detecting the trace dissolved hydrogen in the boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tube samples, the oxidation status of the boiler pipes can be effectively identified, which greatly improves the accuracy of leakage detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is the application architecture diagram of the boiler four-tube leakage early warning method;

[0052] Figure 2 It is the data flow diagram of the optimal neural network prediction model and corrosion monitoring model in the boiler four-tube leakage early warning method;

[0053] Figure 3 It is a structural schematic diagram of the optimal neural network prediction model in the boiler four-tube leakage early warning method;

[0054] Figure 4It is a structural schematic diagram of the tube wall temperature prediction model in the boiler four-tube leakage early warning method;

[0055] Figure 5 This is the structural diagram of the corrosion monitoring model in the boiler four-tube leakage early warning method;

[0056] Figure 6 This is a fusion model diagram of the boiler four-tube leakage early warning method based on the optimal neural network prediction model and the corrosion model;

[0057] Figure 7 It is a curve diagram of accuracy and loss change of different network structures in Example 1;

[0058] Figure 8 This is a diagram showing changes in the steam temperature of the water wall outlet header and the opening of the induced draft fan caused by leakage in Example 1;

[0059] Fig. 9 This is a comparison chart of the calculation results of the oxide film in Example 1;

[0060] Fig.10 This is a diagram showing the prediction results of key parameters in pipeline leakage in Example 1.

[0061] Explanation of the reference numerals: 1. dissolved hydrogen detection device; 2. processing equipment; 3. cockpit. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following Figure 1-10 It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0063] An embodiment of the present invention provides a boiler four-tube leakage early warning method.

[0064] refer to Figure 1 , the boiler four-tube leakage early warning method includes the following steps:

[0065] S1: Obtain the power plant operation parameters, and obtain the dissolved hydrogen data of the four boiler tubes based on the dissolved hydrogen detection device 1.

[0066] During the operation of the thermal power plant, corresponding power plant operation parameters will be generated, and the power plant operation parameters can be obtained through a detection device installed in the thermal power plant. This is a prior art and will not be described in detail here. Then the power plant operation parameters are obtained.

[0067] The dissolved hydrogen detection device 1 is connected to the four boiler tubes respectively. The dissolved hydrogen detection device 1 can detect the dissolved hydrogen in the water vapor of the four boiler tubes, thereby obtaining the corresponding dissolved hydrogen data, and then the dissolved hydrogen data is obtained through multiple dissolved hydrogen detection devices 1. Specifically, the dissolved hydrogen detection device 1 is used to detect the dissolved hydrogen data generated by the tube, water-cooled wall tube, superheater tube and reheater tube.

[0068] S2: Build Figure 2 The database middleware shown in the figure stores the power plant operation parameters and dissolved hydrogen data in the time series database through the data middleware. Specifically, step S2 includes the following steps S21 and S22:

[0069] S21: The time series database uses the Timescale time series database, which is developed through Python and installs the ODBC (Open Database Connectivity) library to access the PI (Plant Information System) database and time series database to read, store, create, and delete databases and data.

[0070] S22: Encapsulate the database access API (Application Programming Interface), and the processor builds the database middleware. The database middleware uses Python middleware, and the Python middleware contains the H5 (HTML5) model. The power plant operation parameters and dissolved hydrogen data are stored in the time series database through the data middleware, realizing the two-way real-time and historical data interaction between the simulation model and the real-time database of the physical equipment.

[0071] S3: Select fault characteristic parameters related to boiler four-tube leakage from the power plant operation parameters, and perform data preprocessing on the fault characteristic parameters. Specifically, step S3 includes the following steps S31 to S34:

[0072] S31: Select fault characteristic parameters related to boiler four-tube leakage from the power plant operation parameters.

[0073] The fault characteristic parameters include: actual load, main steam pressure, feed water flow, induced draft impeller opening, left high temperature superheater outlet steam temperature, right high temperature superheater outlet steam temperature, left high temperature superheater outlet steam pressure, right high temperature superheater outlet steam pressure, left superheater secondary cooling water flow, right superheater secondary cooling water flow, left high temperature reheater outlet steam temperature, right high temperature reheater outlet steam temperature, left high temperature reheater outlet steam pressure, right high temperature reheater outlet steam pressure, left high temperature reheater cooling water flow, right reheater cooling water flow, left and right economizer outlet flue gas temperature difference, economizer outlet feed water temperature, water wall outlet header temperature, and steam-water separator outlet steam pressure.

[0074] S32: Perform data cleaning on the fault characteristic parameters to obtain a cleaned data set.

[0075] The noisy, incomplete and inconsistent fault characteristic parameters are preprocessed to obtain a cleaned data set. Specifically, the preprocessing methods include missing value processing, error data processing, and redundant data processing.

[0076] S33: Perform data dimensionality reduction on the cleaned data set to obtain low-dimensional feature parameters.

[0077] The cleaned data set is processed by zero mean through principal component analysis (Principle Component Analysis, PCA), and then the covariance matrix is ​​calculated, and the eigenvectors and eigenvalues ​​of the covariance matrix are obtained. The eigenvectors and eigenvalues ​​form a new feature space and generate low-dimensional feature parameters. Through principal component analysis, the dimensions of the existing fault feature parameters are effectively reduced without losing important information of the fault feature parameters, thus achieving the purpose of data dimensionality reduction.

[0078] S34: Divide the low-dimensional feature parameters into training set, validation set and test set.

[0079] The low-dimensional feature parameters are normalized and divided into training set, validation set and test set after normalization. The training set, validation set and test set are stored in the Timescale time series database for easy access through the ODBC library.

[0080] S4: Determine the optimal neural network prediction model based on the fault characteristic parameters after data preprocessing. Specifically, step S4 includes the following steps S41 and S42:

[0081] S41: Construct a CNN-LSTM neural network model (Convolutional Neural Network-Long Short-Term Memory) based on the training set, validation set, convolutional neural network model (CNN, Convolutional Neural Networks) and long short-term memory network model (LSTM, Long Short-Term Memory).

[0082] refer to Figure 3 First, the hyperparameters are optimized through the training set to ensure the best performance of the CNN-LSTM neural network model. In this process, it is necessary to carefully select the appropriate optimizer (Adam), activation function (ReLU), and dropout rate to reduce overfitting and improve the generalization ability of the CNN-LSTM neural network model.

[0083] Secondly, the process of training the CNN-LSTM neural network model includes several key steps: first, forward propagation, which passes the input data through the network and generates predictions; then loss calculation, which evaluates the gap between the output of the CNN-LSTM neural network model and the true label through the loss function (using cross entropy loss); then backpropagation is performed, and the weights in the CNN-LSTM neural network model are updated using the gradient descent method; finally, iterative training is performed, and the above steps are repeated continuously to gradually improve the performance of the CNN-LSTM neural network model.

[0084] Finally, the training effect of the CNN-LSTM neural network model is monitored by observing the changes in the loss values ​​of the training set and the validation set. If the change range of the loss values ​​of the training set and the validation set is maintained within a small range and does not decrease significantly, it means that the CNN-LSTM neural network model is close to convergence. At the same time, if the accuracy on the validation set remains stable in multiple consecutive training rounds and no longer increases significantly, it further indicates that the model tends to converge. In addition, if the loss value on the validation set does not decrease after several iterations, training can be stopped to prevent overfitting.

[0085] S42: Determine the structural parameters of the optimal neural network prediction model based on the validation set, test set and CNN-LSTM neural network model.

[0086] Design multiple layers of 1D-CNN-LSTM neural network models, including 3C1L, 4C1L and 4C2L structures. Among them, 1D-CNN-LSTM (3C1L) consists of three convolutional layers and one LSTM layer; on this basis, 1D-CNN-LSTM (4C1L) adds one convolutional layer and one pooling layer, and the size of the convolution kernel remains unchanged. Furthermore, 1D-CNN-LSTM (4C2L) adds one LSTM layer based on the 4C1L structure.

[0087] Through multiple rounds of experiments, the training performance of different CNN-LSTM neural network models is compared, and the accuracy and loss value of each CNN-LSTM neural network model after hyperparameter adjustment are recorded. Finally, by evaluating the performance of the CNN-LSTM neural network model on the validation set and the test set, the generalization ability of the CNN-LSTM neural network model is monitored, and the hierarchical structure with stable performance, high accuracy and low loss value in the training set and the test set is selected, and finally the structural parameters of the CNN-LSTM neural network model with the best training effect are determined, that is, the optimal neural network prediction model is determined, and finally the h5 file of the optimal neural network prediction model is output.

[0088] S5: Obtain boiler parameters, and establish a prediction model for the wall temperature of the four boiler tubes based on the boiler parameters. Specifically, step S5 includes the following steps S51 and S52:

[0089] S51: Obtain boiler parameters, and establish a combustion numerical simulation model based on the boiler parameters.

[0090] Boiler parameters include: fuel characteristics, fuel flow, furnace size, furnace temperature, pressure and flue gas output, etc. The boiler parameters are divided by selecting appropriate grid division methods. Considering that if the screen superheater and high-temperature superheater are modeled according to the actual situation, the number of grids may increase sharply, so after comprehensive consideration, the unstructured grid is finally adopted. Then set the boundary conditions of the combustion numerical simulation model. The boundary conditions include: primary air volume and air temperature, air volume and air temperature around the primary air nozzle, exhaust gas air volume and air temperature, air volume and air temperature around the exhaust gas nozzle, secondary air volume distribution and secondary air temperature, and burnout air volume and air temperature. In addition, the turbulence intensity and hydraulic diameter need to be set for each inlet during the calculation process; for primary air and exhaust gas, the given combustion volume needs to be clearly specified. At the same time, different wall boundary conditions are set for each area, and the outlet boundary condition is set to the pressure outlet boundary, and the gauge pressure is set to -100Pa. Finally, in order to ensure the reliability of the combustion numerical simulation model, it is necessary to verify and compare the numerical simulation results.

[0091] Specifically, the combustion numerical simulation model is verified by comparing the simulation values ​​and design values ​​of the average flue gas temperatures at the inlet and outlet of boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes under various operating conditions such as BMCR (maximum boiler output condition), BRL (boiler rated load condition), THA (heat rate acceptance condition), 40% THA, 50% THA, and 75% THA.

[0092] S52: Based on the combustion numerical simulation model, the following is established: Figure 4 The wall temperature prediction model of the four boiler tubes is shown.

[0093] By using the design values ​​of the average inlet and outlet flue gas temperatures of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tubes under various working conditions to verify the simulation results, the wall temperature distribution of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tubes under different working conditions can be obtained. Then, a wall temperature prediction model for the four boiler tubes is established based on the neural network algorithm. The use of the neural network algorithm can avoid the need to consider the angular coefficient when calculating the wall temperature through the flue gas temperature, thereby reducing iterative operations and speeding up the calculation speed. At the same time, this method makes the wall temperature prediction model of the four heat exchangers unified.

[0094] Specifically, the temperature distribution data of boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tube walls obtained under different operating conditions in the combustion numerical simulation model are used as training samples, and a neural network tube wall temperature prediction model for boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer is established. The test sample temperature data is compared with the predicted value to verify the accuracy of the tube wall temperature prediction model.

[0095] S6: Based on the tube wall temperature prediction model, an oxide film thickness calculation model is established. Based on the oxide film thickness calculation model, the tube wall temperature prediction model and the dissolved hydrogen data, a calculation model is established as follows: Figure 5 The corrosion monitoring model shown. Specifically, step S6 includes the following steps S61 to S63:

[0096] S61: Determine the outer wall temperature of the four boiler tubes based on the tube wall temperature prediction model and the power plant operation parameters.

[0097] The tube wall temperature prediction model reads the on-site data of the thermal power plant operation through the data middleware, inputs the on-site data of the power plant operation such as boiler load, fuel consumption, primary air volume, primary air temperature, secondary air volume, secondary air temperature and the air volume percentage of each air layer, and calculates the outer wall temperature of the water-cooled wall tube, superheater tube, reheater tube and economizer tube.

[0098] S62: Establish an oxide film thickness calculation model based on the outer wall temperature of the four boiler tubes.

[0099] When the oxide film inside the four-tube boiler is evenly distributed, heat is transferred only in the radial direction, and is tightly combined with the tube wall with no gap in the middle, the oxide film thickness σ when the outer wall temperature is T is calculated by the tube wall temperature prediction model.

[0100] S63: Based on the oxide film thickness calculation model, tube wall temperature prediction model and dissolved hydrogen data, the following is established: Figure 5 The corrosion monitoring model shown.

[0101] A corrosion monitoring model can be established based on the relationship between the oxide film thickness and the temperature of the outer wall pipe and the content of online trace dissolved hydrogen. The oxide film thickness is obtained by the oxide film thickness calculation model, the outer wall pipe temperature is obtained by the pipe wall temperature prediction model, and the dissolved hydrogen data is obtained by the online trace dissolved hydrogen from the dissolved hydrogen detection device 1.

[0102] Since the oxide film thickness changes significantly only every 100 hours, the calculated results of the outer wall temperature need to be input into the database, and the oxide film thickness generated is calculated every 24 hours. The corrosion monitoring model regularly monitors whether there are over-temperature warnings for water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes, and determines whether the oxidation rate of the tube wall is accelerated based on the dissolved hydrogen content.

[0103] S7: Based on the optimal neural network prediction model and corrosion monitoring model construction Figure 6 The fusion model shown in the figure performs weighted averaging of redundant information from multiple sources based on the fusion model to obtain a fusion value, and performs a boiler four-tube leakage warning based on the fusion value.

[0104] The weighted data fusion technology is used to input the results of the optimal neural network prediction model and the corrosion monitoring model to build a fusion model. Based on the fusion model, the weighted average of multi-source redundant information is performed to obtain the fusion value, and the boiler four-tube leakage warning is performed based on the obtained fusion value.

[0105] In an optional implementation manner of the present application, the boiler four-tube leakage early warning method further includes:

[0106] S8: Construct the cockpit 3 of boiler four-tube leakage warning data, and display the boiler four-tube leakage warning results.

[0107] The cockpit 3 for boiler four-tube leakage warning data is constructed to output the leakage warning results of the optimal neural network prediction model, corrosion monitoring model and fusion model to display the alarm information in real time.

[0108] Specifically, the program of the cockpit 3 includes an offline training module, a data storage module, an online prediction module, a data display module and an early warning module. Each module runs independently of each other and is called through a function.

[0109] The function call specifically includes: in the data storage module, calling the handle and request set in the program to send data filtering conditions, and obtaining the return data of the Timescale time series database through the getData function. In the online prediction module, real-time data is used as input, and the prediction results are generated through the plotModel function. Finally, the online prediction module pushes the prediction results to the data display module for display on the front end. At the same time, the early warning module obtains the alarm information (alarmInfo) output by the fusion model through the getAlarm function and displays it on the interface so that users can understand the system status in time.

[0110] The online prediction module controls whether the online prediction model needs to be retrained offline through parameters. After receiving the retraining instruction output by the online prediction module, the offline training module retrains the CNN-LSTM neural network model and outputs the h5 algorithm file of the optimal neural network prediction model for use by the online prediction module.

[0111] Compared with the prior art, the present boiler four-tube leakage warning method adopts an electrochemical online dissolved hydrogen detection device 1, which can effectively avoid the problem of measuring point arrangement of the acoustic emission device and the ray device. In actual use, the dissolved hydrogen detection device 1 can be installed on the steam-water sampling rack of the water treatment workshop as needed, and then the feed water, boiler water or steam-water separator, main steam, and reheat steam samples corresponding to the boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes are connected to the dissolved hydrogen detection device 1. The water treatment workshop also has sufficient space to install the dissolved hydrogen detection device 1.

[0112] The chemical water samples collected by this boiler four-tube leakage early warning method can fully represent the operation conditions of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tubes. Since the metal wall surface is always accompanied by extremely trace corrosion in high-temperature water and steam, and extremely trace dissolved hydrogen is produced, trace dissolved hydrogen is a corrosion product, most of which is dissolved in water vapor. Dissolved hydrogen can more truly reflect the corrosion situation. Therefore, this boiler four-tube leakage early warning method can effectively monitor and warn the oxidation condition and leakage of the boiler four tubes by long-term monitoring of the power plant operation parameters through a fusion model established based on the changing relationship between the oxide film thickness and the tube wall temperature and the trace dissolved hydrogen content.

[0113] In addition, this boiler four-tube leakage warning method does not require the layout of the entire boiler, and can make full use of the dissolved hydrogen data of the water treatment workshop samples for fault prediction. The corrosion monitoring model is only related to dissolved hydrogen, and the combustion conditions during boiler operation do not directly affect leakage detection. At the same time, after the data is connected to the control room, the boiler four-tube leakage warning method can realize real-time and dynamic data warning of boiler leakage; this boiler four-tube leakage warning method does not require any changes to the heating surface, and the historical data connected from the water treatment workshop can be directly uploaded to the data center. In addition, the operating parameters of the optimal neural network prediction model can adapt to most supercritical boilers, making this boiler four-tube leakage warning method sufficiently adaptable and applicable to different power plant boilers.

[0114] It should be noted that GB / T10184 "Technical Requirements for Industrial Boilers" and GB / T10185 "Performance Test Methods for Industrial Boilers" contain all the operating parameters that this boiler four-tube leakage warning method relies on, and can be found in the supercritical units of all coal-fired power plants.

[0115] The present method is described below by way of Example 1:

[0116] In Example 1, the effectiveness of the boiler four-tube leakage early warning method is verified in combination with the operating data of a domestic 600MW supercritical thermal power unit. The specific implementation steps are as follows:

[0117] First, the power plant operating parameters and dissolved hydrogen data are obtained according to step S1, and then the data middleware is constructed according to step S2. The data is cleaned in combination with the power plant operating parameters in step S2, and noisy, incomplete and inconsistent power plant operating parameters are preprocessed, including missing value processing, erroneous data, redundant / related data, etc. Then, the preprocessed power plant operating parameters are zero-mean processed by PCA principal component analysis, and then the covariance matrix is ​​calculated, and then the eigenvectors and eigenvalues ​​of the covariance matrix are calculated. These eigenvectors constitute a new feature space. Reduce the dimension of the existing fault feature parameters without losing important information, thereby achieving data dimensionality reduction. Finally, prepare the historical data of relevant operations (with labels), determine the data normalization method, and divide the data into training set, verification set and test set.

[0118] These include the following:

[0119] Data collection: The high level of automation in coal-fired power plants collects and stores a large amount of production process data and equipment operation data. The power plant database data, including operation data and chemical water parameters, is directly read through the data interaction middleware (this project uses the PI database system, which supports ODBC access).

[0120] Data storage: ODBC technology associated with the time series database (Timescale) is used to realize the two-way interaction of real-time and historical data between the simulation model and the real-time database of the physical device.

[0121] Middleware service: After installing the ODBC library, Python can access the PI database and the timing library of this system, read, store, create, and delete database tables and data. Middleware services developed based on Python enable users to complete data interaction.

[0122] Secondly, according to step S4, a suitable CNN-LSTM neural network model (DNN, DCNN, LSTM) is selected. Different CNN-LSTM neural network models have different prediction effects, and hyperparameters are determined for different CNN-LSTM neural network models.

[0123] Finally, according to step S4, the batch size and training times of each CNN-LSTM neural network model are determined, and each CNN-LSTM neural network model is evaluated and the optimal neural network prediction model is generated through the mean absolute error (MAE), root mean square error (RMSE) and score to complete the offline training.

[0124] In the case of the same data set, the performance of three different levels of 1D-CNN-LSTM, as well as DNN, CNN and LSTM neural network models were compared. 76% of the continuous time series data in the data set was used as the training set, and the remaining data was used as the test set. To improve the training efficiency, the initial value of the learning rate was set to 0.005 during the training process and the Adam algorithm was used to adjust the learning rate after each iteration. This method significantly suppressed the volatility in the later stage, allowing the CNN-LSTM neural network model to converge to the optimal solution faster. Figure 7 As shown in Figure 1, the training process of each CNN-LSTM neural network model in the test set is demonstrated.

[0125] Compared with DNN, CNN and LSTM neural network models, 1D-CNN-LSTM performs better in terms of training accuracy and loss. Compared with other neural network models, DNN is often difficult to converge during training because it cannot effectively extract the temporal features of data, resulting in large loss values. In addition, 1D-CNN and LSTM neural network models are relatively sensitive to local spatial features and temporal features, but even with a sufficient number of training times, the accuracy and loss will still fluctuate to a certain extent. The 1D-CNN-LSTM neural network model combines the advantages of CNN and LSTM, and can effectively capture the global and temporal features of the data while ensuring the performance of the neural network model, thereby achieving better prediction results. The final experimental results show that the 1D-CNN-LSTM neural network model has significant advantages in this task, which is better than a single CNN or LSTM neural network model, proving the effectiveness and superiority of the neural network model, and thus confirming it as the optimal neural network prediction model.

[0126] According to the leakage records of the power plant, the operator discovered a leak one day and shut down the plant for maintenance. This leak caused the plant to be shut down for maintenance for 160 hours. The fault characteristic parameters during this operation period are as follows: Figure 8 As shown in Figure 2, when the normalized actual power remains at about 0.816, the normalized water wall outlet header steam temperature in the two groups of fault characteristic parameters remains at about 0.978 and 0.982, respectively, and the normalized induced draft fan opening remains at about 0.427 and 0.418, respectively. Figure 8 It can be seen that within the time period, the steam temperature of the water wall outlet header dropped to about 0.892 two days before the shutdown, and the opening of the induced draft fan blades increased to 0.482. At the same time, the fault characteristic parameters during this period are divided into three consecutive data sets, of which SetA and SetB are before the leakage, and SetC is after the leakage. At this time, the data set of the corresponding time period is imported into the optimal neural network prediction model. The leakage detection results output by the optimal neural network prediction model are as follows: Fig.10 shown. Fig.10 In c, we can see that the ratio of ms / mf also increases. This phenomenon shows that more feed water is needed when the steam flow rate is constant. These phenomena are consistent with the performance of water wall leakage.

[0127] In addition, due to the fluctuation of the operating load of the supercritical boiler, the false alarm rate of the leakage detection of the neural network is high. The present invention establishes an adaptive alarm threshold for the boiler four-tube leakage prediction method by normalizing the relationship between power and mass flow to ensure the accuracy of leakage detection and control the false alarm rate within the minimum range. Fig.10The changes of key parameters for leak detection after being processed by the adaptive alarm threshold method are shown. The left sub-figure shows the comparison between the predicted and measured values ​​of the key parameters after the adaptive alarm threshold is added. The right sub-figure counts the cumulative number of times the predicted and measured values ​​exceed the adaptive threshold. In the normal operation data set, the percentage of predicted values ​​crossing the fixed threshold is 18% and 14%, respectively, while the measured values ​​are 24% and 21%. In the leaking operation data, the predicted and measured values ​​exceed the fixed threshold at a higher frequency, 42% and 34%, respectively. Under normal operating conditions, the percentage of predicted values ​​passing the variable threshold is reduced to 4% and 3%, respectively, while the percentage of measured values ​​passing the variable threshold is 23% and 19%, respectively. Under fault conditions, the measured and predicted values ​​of the key parameters cross the variable threshold at a higher frequency, 37% and 31%, respectively. Therefore, the introduction of adaptive alarm thresholds greatly reduces the frequency of false alarms generated by key parameters when transitioning from one stable state to another. The variable threshold method provides a minimum false alarm rate under normal conditions while also maintaining the ability to detect leaks when faults occur.

[0128] After the water-cooled wall leak occurred, the site was shut down for maintenance and a pipeline leak was discovered during the maintenance process. The crack at the water-cooled wall leak was small, about 56mm long and about 7mm wide at the widest point. The fracture surface of the crack was rough and uneven, with blunt edges and a large number of longitudinal cracks around it. After analysis, it was found that the leak was caused by high-temperature creep of the material after the pipeline was overheated for a long time. Pipeline leakage accidents are usually sudden processes. The optimal neural network prediction model can detect abnormal signs in advance, thereby reducing the losses caused by leakage. Fig.10 As shown in Figure 3, this case illustrates that fault characteristic parameters can detect abnormal status of on-site operation data of coal-fired power plants in advance.

[0129] According to step S6, a corrosion monitoring model is established, and the combustion numerical simulation of the W-type flame boiler furnace is completed through Fluent. The simulation results are checked by the design values ​​of the average inlet and outlet flue gas temperatures of the boiler water-cooled wall, superheater, reheater, and economizer under various operating conditions, so as to obtain the wall temperature distribution of the boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes under various operating conditions, and then the neural network algorithm is used to establish a tube wall temperature prediction model for the four boiler tubes.

[0130] First, according to the boiler parameters, the CFD furnace combustion numerical simulation model is designed. Taking the high-temperature superheater as an example, it includes the exhaust steam tuyere, the main primary tuyere and the oil gun tuyere, the three-layer secondary air distribution on the front and rear walls under the arch, the burnout air nozzle arranged above the main combustion zone, the screen superheater arranged on the upper part of the furnace, and the high-temperature superheater arranged in the horizontal flue. After setting a reasonable number of grids, the inlet boundary conditions, avoidance boundary conditions, and outlet boundary conditions are determined, and the CFD calculation results are verified by comparing the average inlet and outlet temperatures of the high-temperature superheater with the design value.

[0131] Table 1 Comparison of errors in CFD calculation results for main working conditions

[0132]

[0133] Secondly, the temperature distribution data of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer tube walls obtained under different working conditions in the simulation model are stored in the system through the above-mentioned middleware service. The data such as boiler load, fuel consumption, primary air volume and air temperature, secondary air volume and air temperature and the percentage of air volume in each air layer are input into the deep neural network model and the wall temperature distribution prediction value is output. The neural network wall temperature distribution prediction model of the boiler water-cooled wall tubes, superheater tubes, reheater tubes and economizer is trained with the test data set to improve the prediction accuracy. Compared with the traditional wall temperature calculation method through the flue gas heat load calculation model and the working fluid measurement mechanism model, the wall temperature distribution prediction value obtained based on the neural network training method can reduce complex calculations such as iterative calculations and pipe section angle coefficient calculations, and improve the prediction operation speed. According to the pipe material and creep temperature data in the high-temperature superheater, the temperature of the two can be directly compared to realize the over-temperature warning.

[0134] Finally, the temperature and generation rate of the oxide film are added to the wall temperature calculation program to obtain the calculation results of the oxide film thickness. After coupling the wall temperature calculation results, the oxide film thickness calculation is coupled with the wall temperature calculation. The higher the inner wall temperature of the boiler water-cooled wall tube, superheater tube, reheater tube and economizer tube, the faster the oxide film is generated. As the thickness of the oxide film increases, the outer wall temperature also increases and even over-temperature creep occurs. In addition, the metal matrix on the steam side of the heating surface is corroded to form an oxide film, which will cause the thickness of the tube wall to continue to decrease. Finally, after the stress on the tube wall exceeds the critical stress, the oxide film will peel off. The peeled oxide scale is easy to block the pipe elbow, causing the elbow wall temperature to rise and an over-temperature pipe burst accident to occur.

[0135] Under the assumption that the oxide film in the tube is uniformly distributed, heat transfer is only carried out in the radial direction, axial heat transfer is not considered, and it is tightly combined with the metal tube wall without gaps in the middle, the oxidation temperature T when the oxide film thickness is σ is calculated by the wall temperature calculation program, and the oxide film thickness σ is calculated by the oxidation temperature T and the oxidation time Δτ, and the superheater operation time τ is updated, and finally the operation reaches the set time. The results of the oxide film calculation model are obtained when the working fluid temperature is 600℃. Compared with the literature data, it can be seen that Fig. 9 As shown in the figure, the calculated results are close to the experimental data in the literature. The oxide film calculation model can obtain more accurate data and be used in engineering projects.

[0136] The fusion model takes the results of the optimal neural network prediction model and the corrosion monitoring model and outputs the prediction results to form a "four-pipe" leakage and early warning system fusion model based on big data analysis, real-time monitoring and deep learning. The weighted average method includes weighted average, Kalman filtering, Bayesian estimation, fuzzy logic, neural network, etc. This project uses the weighted average method for information fusion fault prediction. Finally, the system's software program is divided into offline training module, data storage module, online prediction module, and display / drawing module. Each module runs independently and is called through functions. The offline training module saves the output evaluation parameters of the trained model as a file in h5 format and provides it to the online prediction module. The online prediction module controls whether the online prediction model needs to be retrained offline through parameters.

[0137] The embodiment of the present invention also provides a boiler four-tube leakage early warning system.

[0138] The boiler four-tube leakage warning system includes a dissolved hydrogen detection device 1 and a processing device 2. The dissolved hydrogen detection device 1 is connected to the economizer tube, the water-cooled wall tube, the superheater tube and the reheater tube, and the dissolved hydrogen detection device 1 is connected to the processing device 2 for communication. The dissolved hydrogen detection device 1 is used to detect the dissolved hydrogen generated by the economizer tube, the water-cooled wall tube, the superheater tube and the reheater tube. The processing device 2 is the execution subject of the boiler four-tube leakage warning method, and is used to execute all steps except the detection of dissolved hydrogen and the display of various data in the cockpit 3. The processing device 2 can be implemented in various forms, including mobile phones, tablet computers, PDAs, laptop computers, desktop computers and other devices.

[0139] The boiler four-tube leakage warning system further includes a cockpit 3, which is communicatively connected to the processing device 2. It should be understood that the cockpit 3 can be a display device such as a display screen.

[0140] The various variations and specific examples of the boiler four-tube leakage warning method provided in the above embodiment are applicable to the two boiler four-tube leakage warning systems in this embodiment. Through the above detailed description of the boiler four-tube leakage warning method, those skilled in the art can clearly know the implementation method of the boiler four-tube leakage warning system in this embodiment. For the sake of brevity of the specification, it will not be described in detail here.

[0141] As described above, the above embodiments are only used to introduce the technical solution of the present invention in detail, but the description of the above embodiments is only used to help understand the method and core idea of ​​the present invention and should not be understood as limiting the present invention. Changes or substitutions that can be easily thought of by technicians in this technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A boiler four-tube leakage early warning method, characterized in that: The following steps are involved: S1: obtaining power plant operation parameters and obtaining dissolved hydrogen data of four boiler tubes based on the dissolved hydrogen detection device (1); S2: Build database middleware and store power plant operation parameters and dissolved hydrogen data in a time series database through the data middleware; S3: selecting fault characteristic parameters related to boiler four-tube leakage from power plant operation parameters, and performing data preprocessing on the fault characteristic parameters; S4: Determine the optimal neural network prediction model based on the fault characteristic parameters after data preprocessing; S5: Obtain boiler parameters, and establish a wall temperature prediction model for four boiler tubes based on the boiler parameters; S6: establishing an oxide film thickness calculation model based on the tube wall temperature prediction model, and establishing a corrosion monitoring model based on the oxide film thickness calculation model, the tube wall temperature prediction model and dissolved hydrogen data; S7: A fusion model is constructed based on the optimal neural network prediction model and the corrosion monitoring model. A fusion value is obtained by weighted averaging of multi-source redundant information based on the fusion model. A boiler four-tube leakage warning is performed based on the fusion value.

2. The boiler four-tube leakage early warning method according to claim 1, characterized in that: The step S3 comprises the following steps: S31: selecting fault characteristic parameters related to boiler four-tube leakage from power plant operation parameters; S32: performing data cleaning on the fault characteristic parameters to obtain a cleaned data set; S33: Performing data dimensionality reduction on the cleaned data set to obtain low-dimensional feature parameters; S34: Divide the low-dimensional feature parameters into training set, validation set and test set.

3. The boiler four-tube leakage early warning method according to claim 2, characterized in that: The fault characteristic parameters include: actual load, main steam pressure, feed water flow, induced draft impeller opening, left high-temperature superheater outlet steam temperature, right high-temperature superheater outlet steam temperature, left high-temperature superheater outlet steam pressure, right high-temperature superheater outlet steam pressure, left superheater secondary cooling water flow, right superheater secondary cooling water flow, left high-temperature reheater outlet steam temperature, right high-temperature reheater outlet steam temperature, left high-temperature reheater outlet steam pressure, right high-temperature reheater outlet steam pressure, left high-temperature reheater cooling water flow, right reheater cooling water flow, left and right economizer outlet flue gas temperature difference, economizer outlet feed water temperature, water-cooled wall outlet header temperature, steam-water separator outlet steam pressure.

4. The boiler four-tube leakage early warning method according to claim 2, characterized in that: The step S4 comprises the following steps: S41: Construct a CNN-LSTM neural network model based on the training set, validation set, convolutional neural network model and long short-term memory network model; S42: Determine the optimal neural network prediction model based on the validation set, test set and CNN-LSTM neural network model.

5. The boiler four-tube leakage early warning method according to claim 1, characterized in that: The step S5 comprises the following steps: S51: Obtain boiler parameters, and establish a combustion numerical simulation model based on the boiler parameters; The boiler parameters include: fuel characteristics, fuel flow, furnace size, furnace temperature, pressure and flue gas output. The boiler parameters are divided using unstructured grids. The boundary conditions of the combustion numerical simulation model include: primary air volume and temperature, air volume and temperature around the primary air nozzle, exhaust gas volume and temperature, air volume and temperature around the exhaust gas nozzle, distribution of secondary air volume and secondary air temperature, and burnout air volume and temperature. S52: Establish a prediction model for the wall temperature of the four boiler tubes based on the combustion numerical simulation model.

6. The boiler four-tube leakage early warning method according to claim 5, characterized in that: The combustion numerical simulation model is verified by comparing the simulation values ​​and design values ​​of the average flue gas temperatures at the inlet and outlet of boiler water-cooled wall tubes, superheater tubes, reheater tubes, and economizer tubes under boiler maximum output conditions, boiler rated load conditions, heat rate acceptance conditions, 40% THA, 50% THA, and 75% THA conditions.

7. The boiler four-tube leakage early warning method according to claim 5, characterized in that: The step S6 comprises the following steps: S61: determining the outer wall temperature of the four boiler tubes based on the tube wall temperature prediction model and the power station operation parameters; S62: Establishing a calculation model for oxide film thickness based on the outer wall temperature of the four boiler tubes; S63: A corrosion monitoring model is established based on the oxide film thickness calculation model, the pipe wall temperature prediction model and the dissolved hydrogen data.

8. The boiler four-tube leakage early warning method according to any one of claims 1 to 8, characterized in that: The following steps are also included: S8: Construct a cockpit (3) for boiler four-tube leakage warning data, and display the boiler four-tube leakage warning results.

9. A boiler four-tube leakage early warning system, characterized in that: The invention comprises a dissolved hydrogen detection device (1) and a processing device (2), wherein the dissolved hydrogen detection device (1) is connected to an economizer tube, a water-cooled wall tube, a superheater tube and a reheater tube, and the processing device (2) is communicatively connected to the dissolved hydrogen detection device (1). The processing device (2) comprises a processor, a memory and a communication bus, wherein the communication bus is used to realize a communication connection between the processor and the memory, and the processor is used to execute a computer program stored in the memory to realize a boiler four-tube leakage early warning method according to any one of claims 1 to 8.

10. The boiler four-tube leakage early warning system according to claim 9, characterized in that: It also comprises a cockpit (3), wherein the cockpit (3) is communicatively connected with the processing device (2).

Images