Air preheater intelligent control system and method based on thermal imaging and artificial intelligence

Abstract

The invention discloses an air preheater intelligent control system and method based on thermal imaging and artificial intelligence, and relates to the technical field of air preheater monitoring control, and the method comprises the following steps: collecting thermal imaging image data of the outer surface of an air preheater in real time by using a thermal imager; thermal imaging image data are preprocessed, an artificial intelligence algorithm is used for extracting temperature distribution characteristics of the outer surface of the air preheater, whether leakage or temperature abnormity occurs in the air preheater is judged by comparing the temperature distribution characteristics with a temperature distribution model in a normal state, and if yes, a worker is reminded to handle in time through a sound-light alarm signal; the temperature parameter of the outer surface of the air preheater is calculated, and when the detected temperature parameter exceeds the normal temperature range, the air preheater is automatically flushed, and the opening degree of a flue baffle is adjusted to control the temperature. According to the invention, rapid detection and intelligent control of leakage and temperature abnormity of the preheater can be realized, and the operation reliability and stability of the air preheater are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air preheater monitoring control, and more particularly to an air preheater intelligent control system and method based on thermal imaging and artificial intelligence. BACKGROUND

[0002] The air preheater is one of the core components of heat energy conversion equipment such as industrial boilers, metallurgical furnaces, and chemical reaction furnaces, and is widely used in power, steel, chemical, and building material industries. Its main function is to preheat the cold air entering the combustion system to the required temperature by efficiently recovering the waste heat of flue gas, thereby optimizing the efficiency of heat energy utilization and ensuring the stability of the production process, and achieving the cascade utilization of energy.

[0003] However, in actual operation, the air preheater faces many complex working condition challenges. High-temperature flue gas and cold air cross-flow leakage can cause energy efficiency to decline, equipment operating load to increase, environmental protection system to fail, equipment to be damaged, and even major safety hazards such as fire and explosion. Temperature anomalies cannot be ignored either. Local over-temperature can accelerate the fatigue aging of metal materials and shorten the service life of equipment. Low-temperature corrosion can cause heat exchange elements to perforate and fail, causing equipment failure and seriously affecting the continuity and stability of industrial production.

[0004] Currently, the monitoring and control of air preheaters mainly rely on traditional temperature sensors and manual inspection. Traditional temperature sensors can only obtain temperature data at limited positions and are difficult to fully reflect the temperature distribution of the preheater surface, which can easily lead to monitoring blind spots. Manual inspection has low efficiency, strong subjectivity, and cannot be monitored in real time, which cannot timely detect potential faults of the preheater.

[0005] With the development of industrial automation and intelligentization, how to realize rapid detection and intelligent control of preheater leakage and temperature anomalies and improve the reliability and stability of air preheater operation has become a technical problem that needs to be solved by those skilled in the art. SUMMARY

[0006] Therefore, the present application provides an air preheater intelligent control system and method based on thermal imaging and artificial intelligence, which solves the problems in the background art.

[0007] In order to achieve the above purpose, the present application provides the following technical solutions:

[0008] An air preheater intelligent control method based on thermal imaging and artificial intelligence, comprising the following steps:

[0009] Using a thermal imager, real-time acquisition of thermal imaging image data of the outer surface of the air preheater;

[0010] The thermal imaging image data is preprocessed, and the temperature distribution features of the outer surface of the air preheater are extracted by using an artificial intelligence algorithm. By comparing with the temperature distribution model in the normal state, it is judged whether the air preheater leaks or has temperature abnormality. If it does, an audible and visual alarm signal is used to remind the staff to handle it in time.

[0011] The temperature parameters of the outer surface of the air preheater are calculated. When the detected temperature parameters exceed the normal temperature range, the air preheater is automatically flushed and the opening of the flue damper is adjusted to control the temperature.

[0012] Optionally, the thermal imaging instrument is used to collect thermal imaging image data of the outer surface of the air preheater in real time, specifically:

[0013] The thermal imaging instrument is installed at a circumferential position at a preset length from the outer surface of the air preheater, and the overlapping area of the viewing angles of adjacent thermal imaging instruments is set to meet the full coverage of the target.

[0014] The thermal imaging instrument is parameter-configured to set the acquisition frame rate and temperature resolution;

[0015] The thermal imaging instrument is temperature-calibrated by the built-in blackbody radiation source. The calibrated thermal imaging instrument is used to collect thermal imaging image data and perform encryption processing, and then the data is transmitted in real time to the data processing equipment through Ethernet.

[0016] Optionally, the thermal imaging image data is preprocessed, specifically:

[0017] Non-local mean filtering algorithm and adaptive median filtering algorithm are used to perform multi-scale denoising on the thermal imaging image data;

[0018] Based on the structural feature points of the air preheater, the denoised image is geometrically corrected to obtain a corrected image;

[0019] A mask template is generated according to the CAD model of the air preheater, and the air preheater region is extracted by a template matching algorithm to obtain a preprocessed image.

[0020] Optionally, the temperature distribution features of the outer surface of the air preheater are extracted by using an artificial intelligence algorithm, specifically:

[0021] An improved U-Net architecture is used to generate a multi-scale feature map of the preprocessed image;

[0022] Based on the multi-scale feature map, the statistical features and spatial gradient features of the temperature field are calculated to form a basic feature vector of the temperature distribution;

[0023] The basic feature vector is input into a bidirectional LSTM network in time sequence to obtain a spatio-temporal fusion feature matrix;

[0024] The variational autoencoder is trained by using the space-time fusion feature matrix, and the reconstruction probability of the test sample is calculated; when the reconstruction probability is lower than a preset threshold, the DBSCAN clustering algorithm is used to segment the abnormal area, and the geometric features and temperature features of the abnormal area are extracted.

[0025] Optionally, the improved U-Net architecture includes an encoder and a decoder;

[0026] The encoder uses a ResNet-50 pre-trained model to extract image features, and an attention mechanism module is added after each residual block;

[0027] The decoder uses deconvolution layers to gradually restore the spatial resolution, and introduces channel attention and spatial attention fusion mechanisms in the skip connection.

[0028] Optionally, the specific process of reminding the staff to handle it in time is:

[0029] The spatial features, temporal features and statistical features of the abnormal area are extracted, and a Gaussian mixture model is used for fault type classification;

[0030] According to the area proportion of the abnormal area, the temperature deviation degree and the duration, three alarm levels are divided, and specific alarm threshold combinations are matched for different fault types;

[0031] The alarm level is indicated by the flashing frequency and combination mode of the yellow / red warning light, and the interval time and tone change are used to distinguish different fault types.

[0032] Optionally, the temperature parameters of the outer surface of the air preheater are calculated, specifically:

[0033] Based on the temperature distribution characteristics, the average temperature, the maximum temperature and the temperature gradient standard deviation of the outer surface of the air preheater are calculated; combined with the equipment operation load and the environment temperature, the temperature-load correlation model constructed by historical data is used to dynamically adjust the normal reference range of the temperature parameters;

[0034] When the average temperature exceeds 105% of the upper limit of the normal reference range or the maximum temperature exceeds 180℃, the preheater flushing device is started; the PID control algorithm is used to adjust the flushing water flow of the preheater flushing device until the average temperature is less than or equal to 95% of the upper limit of the normal reference range and the maximum temperature is less than or equal to 160℃, then the flushing is stopped;

[0035] Based on the temperature gradient standard deviation, the adjustment mode of the flue damper is selected, and the opening of the flue damper is adjusted according to the preset constraint condition.

[0036] A system for implementing the above-mentioned air preheater intelligent control method based on thermal imaging and artificial intelligence, comprising:

[0037] The collection module is used for collecting thermal imaging image data of the outer surface of the air preheater in real time through the thermal imager;

[0038] The data processing and analysis module is used for pre-processing the thermal imaging image data, extracting temperature distribution features of the outer surface of the air preheater by using an artificial intelligence algorithm, comparing with a temperature distribution model in a normal state, judging whether the air preheater appears a leakage or a temperature abnormality, and calculating temperature parameters of the outer surface of the air preheater and comparing with a normal temperature range;

[0039] The intelligent control module is used for automatically flushing the air preheater and adjusting the opening of the flue damper to control the temperature when the detected temperature parameters exceed the normal temperature range.

[0040] The alarm module is used for reminding the staff to handle in time through an audible and visual alarm signal when the leakage or the temperature abnormality appears.

[0041] According to the above technical solution, compared with the prior art, the present application provides an air preheater intelligent control system and method based on thermal imaging and artificial intelligence, which has the following beneficial effects:

[0042] (1) The present application uses thermal imaging technology to replace the traditional single-point temperature sensor, can obtain the temperature distribution information of the outer surface of the air preheater, eliminates the monitoring blind area, realizes the high-precision collection of temperature data, and improves the accuracy of fault detection;

[0043] (2) The present application uses artificial intelligence algorithm to analyze the temperature data from space, time, statistics and other dimensions, which can not only quickly identify temperature abnormalities, but also accurately distinguish different fault types such as leakage and local overheating; the hierarchical audible and visual alarm can shorten the response time of the staff and reduce the accident risk caused by manual inspection;

[0044] (3) The present application can automatically match the flushing water volume and the flue damper adjustment scheme for different fault scenarios, improve the energy utilization efficiency and the equipment operation stability;

[0045] (4) Through the deep mining of temperature distribution features, the potential faults of the equipment can be predicted, the passive maintenance can be changed to active maintenance, the service life of the air preheater can be prolonged, and the operation and maintenance cost can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or the prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creating any creative labor.

[0047] Figure 1 A flowchart of the intelligent control method for air preheaters based on thermal imaging and artificial intelligence provided by the present invention;

[0048] Figure 2 The structural diagram of the intelligent control system for an air preheater based on thermal imaging and artificial intelligence provided by the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0050] This invention discloses an intelligent control method for air preheaters based on thermal imaging and artificial intelligence, such as... Figure 1 As shown, it includes the following steps:

[0051] Thermal imaging data of the outer surface of the air preheater are acquired in real time using a thermal imager.

[0052] The thermal imaging data is preprocessed, and the temperature distribution characteristics of the outer surface of the air preheater are extracted using artificial intelligence algorithms. By comparing it with the temperature distribution model under normal conditions, it is determined whether there is a leak or abnormal temperature in the air preheater. If so, an audible and visual alarm signal is used to remind the staff to deal with it in time.

[0053] The system calculates the temperature parameters of the outer surface of the air preheater. When the detected temperature parameters exceed the normal temperature range, it automatically flushes the air preheater and adjusts the opening of the flue damper to control the temperature.

[0054] based on Figure 1 As shown in the process, this embodiment can obtain temperature distribution information on the surface of the preheater through thermal imaging technology, and combine it with artificial intelligence algorithms to analyze and process the data, so as to realize rapid detection and intelligent control of air preheater leakage and temperature anomaly, thereby improving the reliability and stability of air preheater operation.

[0055] In the prior art, a single-point temperature sensor can only obtain temperature data of the installation position, and it is difficult to cover the entire surface of the air preheater which is a large-area equipment; the single-point temperature sensor has low sensitivity to the non-contact fault of surface thermal radiation anomaly and the response is delayed; the single-point sensor cannot provide spatial information of temperature distribution and is difficult to identify complex fault modes, and misjudgment is prone to occur. Therefore, the application proposes to use a thermal imager to replace the single-point temperature sensor, which can simultaneously obtain tens of thousands of pixels of temperature data and cover the entire surface of the preheater. In the embodiment, the thermal imager is used to collect thermal imaging image data of the outer surface of the air preheater in real time, specifically:

[0056] The thermal imager is installed at a circumferential position at a preset length from the outer surface of the air preheater, and the overlapping area of the viewing angles of adjacent thermal imagers meets the target of full-coverage collection; specifically, the preset length can be 1-3 meters, and the overlapping area of the viewing angles is not less than 20% of the effective field of view, so as to realize full-coverage collection of the temperature distribution of the outer surface of the air preheater.

[0057] The thermal imager is configured with parameters, and the frame rate and temperature resolution are set to ensure that high-precision thermal imaging image data is collected; specifically, the frame rate can be 5-10 frames per second, balancing real-time performance and data volume, and the temperature resolution is 0.1℃, facilitating detection of early faults.

[0058] The thermal imager is calibrated by a built-in blackbody radiation source to ensure the accuracy of the collected data, and the calibrated thermal imager is used to collect thermal imaging image data and perform encryption processing, and then the data is transmitted in real time to a data processing device through Ethernet to prevent data loss or tampering during transmission.

[0059] In the embodiment, based on the deployment and parameter configuration of the above-mentioned thermal imager, the limitations of the traditional single-point sensor can be fundamentally solved, and a high-precision and high-reliability technical foundation is provided for intelligent monitoring of the air preheater.

[0060] In actual application, affected by factors such as environmental temperature fluctuations and electromagnetic interference, thermal imaging images may have Gaussian noise and salt and pepper noise, resulting in a decrease in image signal-to-noise ratio; thermal imager lens distortion, installation position deviation or air preheater vibration may cause image geometric distortion. To solve this problem, the embodiment proposes a technical solution for preprocessing thermal imaging image data, specifically:

[0061] A non-local mean filtering algorithm and an adaptive median filtering algorithm are used to perform multi-scale denoising on the thermal imaging image data.

[0062] The denoised image is geometrically corrected based on the structural feature points of the air preheater to obtain a corrected image; specifically, the SIFT algorithm can be adopted for feature point extraction, and the FLANN fast nearest neighbor search is adopted for matching, so that the re-projection error of the corrected image is controlled within 0.5 pixels;

[0063] A mask template is generated according to the CAD model of the air preheater, and the air preheater region is extracted through a template matching algorithm to obtain a preprocessed image.

[0064] In addition, before the region of interest is extracted, a multi-scale adaptive image enhancement algorithm based on the Retinex theory can be adopted to perform image enhancement processing on the corrected image, so as to improve the distinguishability of temperature detail features in the image.

[0065] In the prior art, the commonly used feature extraction methods have limitations: the manually designed features cannot capture complex heat distribution patterns; the single-point temperature data or simple statistical features ignore the spatiotemporal correlation of the temperature field; the fixed threshold method is difficult to adapt to changes in working conditions; and only shallow features are extracted, which cannot mine deep information of the data. To solve this problem, the present application proposes to replace the traditional technology with an artificial intelligence algorithm to perform deep analysis on the temperature data. In this embodiment, an artificial intelligence algorithm is used to extract the temperature distribution features of the outer surface of the air preheater, specifically:

[0066] An improved U-Net architecture is adopted to generate a multi-scale feature map of the preprocessed image;

[0067] Based on the multi-scale feature map, statistical features and spatial gradient features of the temperature field are calculated to form a basic feature vector of the temperature distribution; specifically, the statistical features include temperature mean, standard deviation, skewness and kurtosis

[0068] The basic feature vector is input into a bidirectional LSTM network in time sequence, and the attention mechanism in the LSTM can automatically adjust the weights of the features at different times according to the temperature change trend to obtain a spatiotemporal fusion feature matrix, which effectively captures the time sequence variation law of the temperature distribution; wherein, the bidirectional LSTM can consider both past and future time sequence information, and compared with the unidirectional LSTM, it can improve the prediction accuracy for temperature mutations;

[0069] The spatiotemporal fusion feature matrix is used to train a variational autoencoder to calculate the reconstruction probability of the test sample; when the reconstruction probability is lower than a preset threshold, a DBSCAN clustering algorithm is used to segment the abnormal region, and geometric features and temperature features of the abnormal region are extracted.

[0070] Further, the improved U-Net architecture includes an encoder and a decoder; the encoder uses a ResNet-50 pre-training model to extract image features, and adds an attention mechanism module after each residual block to automatically extract multi-scale information from low-level edge features to high-level semantic features; the decoder uses deconvolution layers to gradually restore the spatial resolution, and introduces channel attention and spatial attention fusion mechanisms in the skip connection.

[0071] In the present embodiment, based on the above artificial intelligence scheme, the transition from "passive detection" to "active early warning" and from "manual experience" to "data-driven" can be realized, significantly improving the accuracy and timeliness of air preheater fault detection.

[0072] Further, according to the processing and analysis results of the temperature data, the present embodiment reminds the staff to handle the emergency situation in time through an audible and visual alarm signal, and the specific process is as follows:

[0073] Extract the spatial features (such as area, temperature gradient), time features (such as change rate), and statistical features (such as mean, standard deviation) of the abnormal area, and use a Gaussian mixture model to classify the fault type;

[0074] According to the area proportion (0.5%-3%), temperature deviation (5-20℃), and duration (30-60 seconds) of the abnormal area, three alarm levels are divided, and specific alarm threshold combinations are matched for different fault types (leakage / overheating); specifically, the leakage detection focuses on low temperature gradient, and the overheating detection focuses on the number of high temperature points;

[0075] The alarm level is indicated by the flashing frequency and combination of yellow / red warning lights, and the interval time and tone change are used to distinguish different fault types.

[0076] On this basis, the present embodiment can further dynamically adjust the judgment threshold based on the equipment running time and load state, and establish a linear correction function. Normal operation data is automatically collected every week, and the parameters of the Gaussian mixture model are updated through incremental learning to maintain detection accuracy. Based on the above scheme, the present embodiment can achieve accurate detection through multi-dimensional comparison, improve response efficiency through a hierarchical alarm mechanism, and enhance system robustness through an adaptive threshold adjustment method.

[0077] Further, the temperature parameters of the outer surface of the air preheater are calculated, specifically:

[0078] Based on the temperature distribution characteristics, the average temperature, maximum temperature, and temperature gradient standard deviation of the outer surface of the air preheater are calculated; combined with the equipment running load and environmental temperature, a temperature-load correlation model is constructed based on historical data to dynamically adjust the normal reference range of the temperature parameters;

[0079] When the average temperature exceeds 105% of the upper limit of the normal reference range or the maximum temperature exceeds 180℃, the preheater flushing device is started; the flushing water flow of the preheater flushing device is adjusted by using a PID control algorithm until the average temperature is less than or equal to 95% of the upper limit of the normal reference range and the maximum temperature is less than or equal to 160℃, and then the flushing is stopped.

[0080] The adjustment mode of the flue damper is selected based on the standard deviation of the temperature gradient, and the opening of the flue damper is adjusted according to the preset constraint condition.

[0081] In the embodiment, the maximum temperature can be used to identify a local overheating fault, and the threshold triggering priority of the maximum temperature is higher than that of the average temperature; the standard deviation of the temperature gradient can quantify the uniformity of the temperature distribution and reflect the leakage or ash deposition state. The technical scheme sets up an independent emergency triggering logic for the maximum temperature, and forms a hierarchical control with the average temperature; for different fault scenarios, the flushing water flow and the adjustment scheme of the flue damper are automatically matched, adaptive intelligent control is realized, and the stability of system operation is enhanced.

[0082] With Figure 1 Corresponding to the method, the embodiment of the present application also provides an air preheater intelligent control system based on thermal imaging and artificial intelligence, which is used for Figure 1 The air preheater intelligent control system based on thermal imaging and artificial intelligence provided by the embodiment of the present application can be applied to a computer terminal or various mobile devices, such as Figure 2 As shown in the figure, specifically includes:

[0083] The acquisition module is used for collecting thermal imaging image data of the outer surface of the air preheater in real time through the thermal imager;

[0084] The data processing and analysis module is used for pre-processing the thermal imaging image data, and extracting the temperature distribution characteristics of the outer surface of the air preheater by using an artificial intelligence algorithm; by comparing with a temperature distribution model in a normal state, it is judged whether the air preheater appears a leakage or temperature abnormality; the temperature parameters of the outer surface of the air preheater are calculated and compared with a normal temperature range;

[0085] The intelligent control module is used for automatically flushing the air preheater and adjusting the opening of the flue damper to control the temperature when the detected temperature parameters exceed the normal temperature range;

[0086] The alarm module is used for reminding the staff to handle in time through an audible and visual alarm signal when a leakage or temperature abnormality occurs.

[0087] The various embodiments described in this specification are implemented in a progressive manner, each embodiment focusing on the differences from other embodiments, and the same or similar parts between embodiments referring to each other. For the system disclosed by the embodiments, since it corresponds to the method disclosed by the embodiments, the description is relatively simple, and the relevant parts refer to the method part description.

[0088] The above description of disclosed embodiments enables one of ordinary skill in the art to make or use the application. Various modifications to these embodiments will be readily apparent to those of ordinary skill in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent control method for air preheaters based on thermal imaging and artificial intelligence, characterized in that, The method comprises the following steps: Real-time acquisition of thermal imaging image data of the outer surface of the air preheater by using a thermal imager; Preprocessing of the thermal imaging image data and extraction of the temperature distribution characteristics of the outer surface of the air preheater by using an artificial intelligence algorithm, comparison with the temperature distribution model under the normal state to determine whether the air preheater has a leakage or temperature abnormality, and if so, reminding the staff to handle it in time through an audible and visual alarm signal; Calculation of the temperature parameters of the outer surface of the air preheater, automatic flushing of the air preheater and adjustment of the opening degree of the flue damper to control the temperature when the detected temperature parameters exceed the normal temperature range.

2. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 1, characterized in that, Real-time acquisition of thermal imaging image data of the outer surface of the air preheater by using a thermal imager, specifically: The thermal imager is installed at a circumferential position at a preset length from the outer surface of the air preheater, and the overlapping area of the visual angles of adjacent thermal imagers meets the target of full-coverage acquisition; Parameter configuration of the thermal imager, setting the acquisition frame rate and temperature resolution; Temperature calibration of the thermal imager by using the built-in blackbody radiation source, acquisition of thermal imaging image data by using the calibrated thermal imager and encryption processing, and real-time transmission to the data processing equipment through Ethernet.

3. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 1, characterized in that, Preprocessing of the thermal imaging image data, specifically: Non-local mean filtering algorithm and adaptive median filtering algorithm are used to perform multi-scale denoising on the thermal imaging image data; Based on the structural feature points of the air preheater, the denoised image is geometrically corrected to obtain a corrected image; A mask template is generated according to the CAD model of the air preheater, and the air preheater region is extracted by template matching algorithm to obtain a preprocessed image.

4. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 1, characterized in that, Extraction of the temperature distribution characteristics of the outer surface of the air preheater by using an artificial intelligence algorithm, specifically: An improved U-Net architecture is used to generate a multi-scale feature map of the preprocessed image; Based on the multi-scale feature map, the statistical features and spatial gradient features of the temperature field are calculated to form a basic feature vector of the temperature distribution; The basic feature vector is input into a bidirectional LSTM network in time sequence to obtain a spatio-temporal fusion feature matrix; A variational autoencoder is trained using the spatio-temporal fusion feature matrix to calculate the reconstruction probability of the test sample; when the reconstruction probability is lower than a preset threshold, a DBSCAN clustering algorithm is used to segment the abnormal area and extract the geometric features and temperature features of the abnormal area.

5. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 4, characterized in that, The improved U-Net architecture includes an encoder and a decoder; The encoder uses a ResNet-50 pre-trained model to extract image features, and adds an attention mechanism module after each residual block; The decoder uses deconvolution layers to gradually restore the spatial resolution, and introduces channel attention and spatial attention fusion mechanisms in the skip connection.

6. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 4, characterized in that, The specific process of reminding the staff to handle it in time is as follows: Extraction of the spatial features, temporal features and statistical features of the abnormal area, classification of fault types by using a Gaussian mixture model; According to the area proportion, temperature deviation degree and duration of the abnormal area, three alarm levels are divided, and specific alarm threshold combinations are matched for different fault types; The alarm level is indicated by the flashing frequency and combination mode of the yellow / red warning light, and different fault types are distinguished by the interval time and tone change.

7. The intelligent control method of air preheater based on thermal imaging and artificial intelligence according to claim 1, characterized in that, The temperature parameters of the outer surface of the air preheater are calculated, specifically: Based on the temperature distribution characteristics, the average temperature, the maximum temperature, and the standard deviation of the temperature gradient of the outer surface of the air preheater are calculated; combined with the equipment operating load and the environmental temperature, the temperature-load correlation model constructed through historical data is used to dynamically adjust the normal reference range of the temperature parameters; When the average temperature exceeds 105% of the upper limit of the normal reference range or the maximum temperature exceeds 180℃, the preheater flushing device is started; the PID control algorithm is used to adjust the flushing water flow of the preheater flushing device until the average temperature is less than or equal to 95% of the upper limit of the normal reference range and the maximum temperature is less than or equal to 160℃, then the flushing is stopped; Based on the standard deviation of the temperature gradient, the adjustment mode of the flue damper is selected, and the opening of the flue damper is adjusted according to the preset constraints.

8. A system for performing the intelligent control method of air preheaters based on thermography and artificial intelligence according to any one of claims 1-7, characterized in that, It includes: The acquisition module is used to collect the thermal imaging image data of the outer surface of the air preheater in real time through the thermal imager; The data processing and analysis module is used to preprocess the thermal imaging image data, extract the temperature distribution characteristics of the outer surface of the air preheater using artificial intelligence algorithms, compare with the temperature distribution model under normal state, and judge whether the air preheater has leakage or temperature abnormality; calculate the temperature parameters of the outer surface of the air preheater and compare with the normal temperature range; The intelligent control module is used to automatically flush the air preheater and adjust the opening of the flue damper to control the temperature when the detected temperature parameters exceed the normal temperature range; The alarm module is used to remind the staff to handle in time through the sound and light alarm signal when there is leakage or temperature abnormality.

Images