Boiler temperature monitoring system and method based on artificial intelligence

Abstract

The invention discloses a boiler temperature monitoring system and method based on artificial intelligence, and belongs to the technical field of boiler temperature monitoring, the boiler temperature monitoring system based on artificial intelligence comprises a data acquisition module, a database, a data processing module and an alarm processing module; the data acquisition module is used for acquiring environment information when the boiler works and operating parameters and process parameters when the boiler runs; the database is used for storing all the collected data; the data processing module is used for extracting features from the collected data and monitoring the temperature of the boiler by using an isolated forest model; according to the boiler temperature real-time detection system and method, through multi-parameter comprehensive analysis and artificial intelligence, real-time detection of the boiler temperature is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler temperature monitoring, and in particular to an artificial intelligence-based boiler temperature monitoring system and method. Background Art

[0002] During the operation of industrial boilers, the temperature of the inner and outer walls of the furnace is one of the important indicators to measure its operating status and efficiency. Due to long-term operation, incomplete fuel combustion, accumulation of combustion products and water quality problems, scaling or ash accumulation often forms on the inner and outer walls of the furnace. These scaling or ash accumulations will reduce the heat transfer efficiency and make the inner and outer wall temperatures of the boiler abnormal. Abnormal temperatures not only affect the operating efficiency of the boiler and increase energy consumption, but may also cause serious safety hazards. Scaling will cause the temperature of the inner wall of the boiler to rise, increase the pressure in the furnace, and may cause equipment damage; ash accumulation will hinder heat transfer, causing the boiler to overheat or even explode. Therefore, timely and accurate monitoring and diagnosis of abnormal boiler temperatures and identification of scaling or ash accumulation are crucial to ensuring safe and efficient operation of the boiler.

[0003] Traditional boiler temperature monitoring methods mainly rely on manual inspections and simple temperature sensors. These methods have the following limitations: manual inspections require operators to conduct regular on-site inspections, which is not only labor-intensive but also easily affected by subjective factors, resulting in temperature anomalies not being discovered in time. In addition, the frequency of manual inspections is limited, and real-time monitoring of boiler temperature cannot be achieved, which can easily miss early abnormal signals. Secondly, traditional temperature sensors can only provide local temperature data and it is difficult to fully reflect the operating status of the boiler, especially in complex industrial environments. The data from a single sensor may not be representative. The number of sensors is limited, and when the layout is unreasonable, it may not be able to capture temperature changes in key areas. Traditional methods lack intelligent data analysis methods, making it difficult to extract useful features from a large amount of temperature data for accurate anomaly detection and diagnosis. Temperature data is complex and changeable, and traditional methods have difficulty coping with the diversity of data and are prone to missing or misreporting anomalies. Summary of the invention

[0004] The purpose of the present invention is to provide a boiler temperature monitoring system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a boiler temperature monitoring system based on artificial intelligence, the system comprising: a data acquisition module, a database, a data processing module and an alarm processing module; the data acquisition module is used to collect environmental information when the boiler is working, operating parameters and process parameters when the boiler is running; the database is used to store all collected data; the data processing module is used to extract features from the collected data and monitor the boiler temperature using an isolation forest model; the alarm processing module immediately issues an alarm when an abnormal state is identified.

[0006] Furthermore, the data acquisition module includes an environmental parameter acquisition unit, an operating parameter acquisition unit and a process parameter acquisition unit. The environmental parameter acquisition unit is used to collect ambient temperature and ambient humidity; the operating parameter acquisition unit is used to collect damper opening and blower speed; the process parameter acquisition unit is used to collect flue gas flow rate and furnace inner and outer wall temperature data, and transmit all collected data to the database.

[0007] Furthermore, the data processing module includes a feature extraction unit and an anomaly detection unit, the feature extraction unit extracts features from ambient temperature, ambient humidity, damper opening, blower speed and flue gas flow rate data, and calculates the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate and flue gas flow rate change rate; the anomaly detection unit uses the extracted feature data to train an isolation forest model to monitor the boiler temperature.

[0008] Furthermore, the alarm processing module includes an alarm triggering unit and an alarm device. The alarm triggering unit immediately triggers an alarm when an abnormal boiler temperature is detected; the alarm device issues an alarm through an audible and visual alarm device and notifies relevant management personnel.

[0009] A boiler temperature monitoring method based on artificial intelligence comprises the following steps:

[0010] S1: Collect the ambient temperature, ambient humidity, air door opening, blower speed, flue gas flow rate data and furnace inner and outer wall temperature data when the boiler is working;

[0011] S2: Further data cleaning is performed on the obtained boiler data to remove abnormal points and noise;

[0012] S3: extract features from the cleaned data and calculate the rate of change of ambient temperature, ambient humidity, damper opening, blower speed and flue gas flow rate;

[0013] S4: Use the features extracted by the feature extraction unit to train the isolation forest model and establish a boiler temperature anomaly detection model.

[0014] Furthermore, in step S1: installing ambient temperature and humidity sensors at the air inlet and outlet of the boiler to obtain ambient temperature and humidity data, which can reflect the dynamic changes of external environmental conditions and monitor the impact of the boiler operating status in real time; installing a damper opening sensor on the boiler damper to monitor the damper opening in real time; installing a blower speed sensor on the rotating shaft of the blower to monitor the blower speed in real time; installing a flue gas flow velocity sensor on the straight pipe section in the boiler flue to monitor the flue gas flow velocity in real time; installing an infrared temperature sensor on the inner and outer walls of the boiler to obtain the inner and outer wall temperature data of the furnace.

[0015] Furthermore, in step S2: the obtained boiler data is further cleaned, and the Z-score method and the IQR method are used to remove outliers and remove noise.

[0016] Furthermore, in step S3: extract features from the cleaned data and assign different weights to each feature. The specific steps are as follows:

[0017] S3-1: Extract features from the cleaned data and calculate the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate and flue gas flow rate change rate. The feature calculation formula is as follows: env (t) = T env (t)-T env (t-1) Calculate the ambient temperature change rate, where ΔT env (t) represents the rate of change of ambient temperature, T env (t) represents the ambient temperature at time t, T env (t) represents the ambient temperature at time t-1; according to ΔH env (t) = H env (t)-H env (t-1) Calculate the rate of change of ambient humidity, where ΔH env (t) represents the rate of change of ambient humidity, H env (t) represents the ambient humidity at time t, H env(t-1) represents the ambient humidity at time t-1; the damper opening change rate is calculated according to ΔD(t)=D(t)-D(t-1), where ΔD(t) represents the damper opening change rate, D(t) represents the damper opening at time t, and D(t-1) represents the damper opening at time t-1; the blower speed change rate is calculated according to ΔF(t)=F(t)-F(t-1), where ΔF(t) represents the blower speed change rate, and F(t) represents the speed change rate at time t. The speed of the blower is F(t-1), and the speed of the blower at time t-1 is F(t-1). The flue gas velocity change rate is calculated according to ΔV(t)=V(t)-V(t-1), wherein ΔV(t) represents the flue gas velocity change rate, V(t) represents the flue gas velocity at time t, and V(t-1) represents the flue gas velocity at time t-1. The change rate calculation provides real-time parameter change information, so that the system can dynamically monitor the operating status of the boiler and respond quickly to abnormal situations.

[0018] S3-2: Different weights are assigned to each feature according to its impact on the boiler operation status. The specific weights are as follows: the weight of ambient temperature is w1, the weight of ambient humidity is w2, the weight of damper opening is w3, the weight of blower speed is w4, and the weight of flue gas flow rate is w5, where w1 represents the weight assigned to ambient temperature, w2 represents the weight assigned to ambient humidity, w3 represents the weight assigned to damper opening, w4 represents the weight assigned to blower speed, and w5 represents the weight assigned to flue gas flow rate;

[0019] S3-3: Perform weighted calculation on the feature change rate to obtain a comprehensive score, and preliminarily judge whether the data point is abnormal based on the preset feature change rate threshold. The comprehensive score calculation formula is as follows:

[0020] W all =ΔT env (t)*w1+ΔH env (t)*w2+ΔD(t)*w3+ΔF(t)*w4+ΔV(t)*w5

[0021] Among them, W all It means that the feature change rate is weightedly calculated to obtain a comprehensive score, which integrates multiple complex features into a single indicator, making it easier to intuitively understand the operating status of the boiler. By considering the changes in each feature and its weight, the comprehensive score can reflect the changes in the operating status of the boiler, thereby improving the accuracy and timeliness of anomaly detection.

[0022] Further, in step S4: the features extracted by the feature extraction unit and the historical data are used to train the isolation forest model to establish a boiler temperature anomaly detection model. The specific steps of training the isolation forest model are as follows:

[0023] S4-1: Standardize the extracted features through normalization;

[0024] S4-2: Use the standardized historical feature data to train the isolation forest model, establish a boiler temperature anomaly detection model, and input the standardized real-time feature data into the trained isolation forest model for real-time detection;

[0025] S4-3: The model performs an anomaly score on each data point and further determines whether the data point is abnormal based on a preset anomaly score threshold. If the anomaly score exceeds the threshold, the data point is judged to be abnormal.

[0026] S4-4: Analyze the comprehensive score and the weighted score of each feature to determine whether there is a problem with internal scaling or dust accumulation.

[0027] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention improves the recognition accuracy of abnormal boiler temperature by introducing ambient temperature, ambient humidity, damper opening, blower speed and flue gas flow rate parameters for comprehensive analysis; by combining the abnormality score of the isolation forest model, abnormal data points in boiler operation can be accurately detected, thereby improving the accuracy and efficiency of abnormality detection; sensors are installed at different positions of the boiler to collect parameter data in real time to ensure that the monitoring system can reflect the operating status of the boiler in real time; the model can process and analyze the collected data in real time, promptly discover boiler temperature abnormalities and trigger the alarm system, reducing delays and misjudgments of manual inspections; artificial intelligence technology is used to automatically process and analyze data, reducing the need for manual intervention, reducing operational complexity and labor intensity; the system can issue an early warning at the early stage of temperature abnormality to prevent further deterioration of boiler scaling or ash accumulation problems, thereby reducing the risk of equipment failure and safety accidents; through intelligent monitoring and early warning, the present invention reduces the emergency maintenance and repair costs caused by sudden failures and extends the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0029] Figure 1 It is a structural diagram of a boiler temperature monitoring system based on artificial intelligence of the present invention;

[0030] Figure 2 The present invention is a flow chart of a boiler temperature monitoring method based on artificial intelligence. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] See also Figure 1-Figure 2 The present invention provides a technical solution: a boiler temperature monitoring system based on artificial intelligence, the system comprising: a data acquisition module, a database, a data processing module and an alarm processing module; the data acquisition module is used to collect environmental information when the boiler is working, operating parameters and process parameters when the boiler is running; the database is used to store all collected data; the data processing module is used to extract features from the collected data and monitor the boiler temperature using an isolation forest model; the alarm processing module immediately issues an alarm when an abnormal state is identified.

[0033] The data acquisition module includes an environmental parameter acquisition unit, an operating parameter acquisition unit and a process parameter acquisition unit. The environmental parameter acquisition unit is used to collect ambient temperature and ambient humidity; the operating parameter acquisition unit is used to collect damper opening and blower speed; the process parameter acquisition unit is used to collect flue gas flow rate and furnace inner and outer wall temperature data, and transmit all collected data to the database.

[0034] The data processing module includes a feature extraction unit and an anomaly detection unit. The feature extraction unit extracts features from the ambient temperature, ambient humidity, damper opening, blower speed and flue gas flow rate data, and calculates the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate and flue gas flow rate change rate; the anomaly detection unit uses the extracted feature data to train an isolation forest model to monitor the boiler temperature.

[0035] The alarm processing module includes an alarm triggering unit and an alarm device. The alarm triggering unit immediately triggers an alarm when an abnormal condition of the boiler temperature is detected; the alarm device sends out an alarm through an audible and visual alarm device and notifies relevant management personnel.

[0036] A boiler temperature monitoring method based on artificial intelligence comprises the following steps:

[0037] S1: Collect the ambient temperature, ambient humidity, air door opening, blower speed, flue gas flow rate data and furnace inner and outer wall temperature data when the boiler is working;

[0038] S2: Further data cleaning is performed on the obtained boiler data to remove abnormal points and noise;

[0039] S3: Extract features from the cleaned data and assign different weights to each feature;

[0040] S4: Use the features extracted by the feature extraction unit to train the isolation forest model and establish a boiler temperature anomaly detection model.

[0041] In step S1: install ambient temperature and humidity sensors at the air inlet and outlet of the boiler to obtain ambient temperature and humidity data; install a damper opening sensor on the boiler damper to monitor the damper opening in real time; install a blower speed sensor on the rotating shaft of the blower to monitor the blower speed in real time; install a flue gas flow velocity sensor on the straight pipe section in the boiler flue to monitor the flue gas flow velocity in real time; install infrared temperature sensors on the inner and outer walls of the boiler to obtain the inner and outer wall temperature data of the furnace.

[0042] In step S2: the obtained boiler data is further cleaned, and the Z-score method and IQR method are used to remove outliers and remove noise.

[0043] In step S3: extract features from the cleaned data and assign different weights to each feature. The specific steps are as follows:

[0044] S3-1: Extract features from the cleaned data and calculate the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate and flue gas flow rate change rate. The feature calculation formula is as follows: env (t) = T env (t)-T env (t-1) Calculate the rate of change of ambient temperature, where ΔT env (t) represents the rate of change of ambient temperature, T env (t) represents the ambient temperature at time t, T env (t) represents the ambient temperature at time t-1; according to ΔH env (t) = H env (t)-H env (t-1) Calculate the rate of change of ambient humidity, where ΔH env (t) represents the rate of change of ambient humidity, H env (t) represents the ambient humidity at time t, H env(t-1) represents the ambient humidity at time t-1; the damper opening change rate is calculated according to ΔD(t)=D(t)-D(t-1), wherein ΔD(t) represents the damper opening change rate, D(t) represents the damper opening at time t, and D(t-1) represents the damper opening at time t-1; the blower speed change rate is calculated according to ΔF(t)=F(t)-F(t-1), wherein ΔF(t) represents the blower speed change rate, F(t) represents the blower speed at time t, and F(t-1) represents the blower speed at time t-1; the flue gas velocity change rate is calculated according to ΔV(t)=V(t)-V(t-1), wherein ΔV(t) represents the flue gas velocity change rate, V(t) represents the flue gas velocity at time t, and V(t-1) represents the flue gas velocity at time t-1;

[0045] S3-2: Different weights are assigned to each feature according to its impact on the boiler operation status. The specific weights are as follows: the weight of ambient temperature is w1, the weight of ambient humidity is w2, the weight of damper opening is w3, the weight of blower speed is w4, and the weight of flue gas flow rate is w5, where w1 represents the weight assigned to ambient temperature, w2 represents the weight assigned to ambient humidity, w3 represents the weight assigned to damper opening, w4 represents the weight assigned to blower speed, and w5 represents the weight assigned to flue gas flow rate;

[0046] S3-3: Perform weighted calculation on the feature change rate to obtain a comprehensive score, and preliminarily judge whether the data point is abnormal based on the preset feature change rate threshold. The comprehensive score calculation formula is as follows:

[0047] W all =ΔT env (t)*w1+ΔH env (t)*w2+ΔD(t)*w3+ΔF(t)*w4+ΔV(t)*w5

[0048] Among them, W all It means that the weighted calculation of the feature change rate is used to obtain a comprehensive score, which integrates multiple features into a single indicator, making it easier to intuitively understand the operating status of the boiler. By considering the changes in each feature and its weight, the comprehensive score can reflect the changes in the operating status of the boiler, thereby improving the accuracy and timeliness of anomaly detection.

[0049] In step S4: the features extracted by the feature extraction unit and the historical data are used to train the isolation forest model to establish a boiler temperature anomaly detection model. The specific steps of training the isolation forest model are as follows:

[0050] S4-1: Standardize the extracted features through normalization;

[0051] S4-2: Use the standardized historical feature data to train the isolation forest model, establish a boiler temperature anomaly detection model, and input the standardized real-time feature data into the trained isolation forest model for real-time detection;

[0052] S4-3: The model performs an anomaly score on each data point and determines whether the data point is abnormal based on a preset threshold. If the anomaly score exceeds the threshold, the data point is judged to be abnormal.

[0053] S4-4: Analyze the comprehensive score and the weighted score of each feature to determine whether there is a problem with internal scaling or dust accumulation.

[0054] In the embodiment: the ambient temperature, ambient humidity, air door opening, blower speed, flue gas flow rate data and furnace inner and outer wall temperature data are collected when the boiler is working, and the boiler data at time t1 [25, 60, 55, 1600, 6.5, 900, 150], the boiler data at time t2 [25, 60, 55, 1200, 4, 910, 155], the boiler data at time t3 [25, 60, 55, 1600, 6.5, 915, 155], the boiler data at time t4 [25, 60, 55, 1600, 6.5, 930, 160], and the boiler data at time tS [25, 60, 55, 1600, 6.5, 950, 165] are obtained; the extracted feature data are standardized and feature extracted using the Z-score standardization method, and according to ΔT env (t) = T env (t)-T env (t-1) Calculate the rate of change of ambient temperature; according to ΔH env (t) = H env (t)-H env (t-1) calculates the rate of change of ambient humidity; calculates the rate of change of damper opening according to ΔD(t)=D(t)-D(t-1); calculates the rate of change of blower speed according to ΔF(t)=F(t)-F(t-1); calculates the rate of change of flue gas flow velocity according to ΔV(t)=V(t)-V(t-1), and obtains the following characteristic data: the rate of change of ambient temperature, the rate of change of ambient humidity, the rate of change of damper opening, the rate of change of blower speed and the rate of change of flue gas flow velocity at time t2 are: 0, 0, 0, - 400, -2.5; the characteristic data at t3 are: 0, 0, 0, 400, 2.5; the characteristic data at t4 are: 0, 0, 0, 0, 0; the characteristic data at t5 are: 0, 0, 0, 0, 0. Different weights are assigned according to the impact of each feature on the boiler operation status. The specific distribution is as follows: ambient temperature weight 0.1, ambient humidity weight 0.1, damper opening weight 0.25, blower speed weight 0.25, flue gas flow rate weight 0.30; through W all =ΔT env(t)*w1+ΔH env (t)*w2+ΔD(t)*w3+ΔF(t)*w4+ΔV(t)*w5 calculates the comprehensive scores of each time point: -100.75, 102.75, 7.5, 10.5. The standardized historical feature data is used to train the isolation forest model, and the boiler temperature anomaly detection model is established. The standardized real-time feature data is input into the trained isolation forest model for real-time detection. The model calculates the anomaly score of each data point, and the anomaly scores at t2, t3, t4 and t5 are 0.8, 0.3, 0.5, and 0.7 respectively. The comprehensive score threshold is set to 10, and the anomaly score threshold is 0.72. According to the comprehensive score, it can be judged Abnormalities occurred at t2 and t3. By analyzing the specific data at t2, it was concluded that the fan speed and flue gas flow rate were abnormal due to short-term operation adjustments or equipment adjustment problems. The boiler continued to be monitored. By analyzing the specific data at t3, it can be determined that the fan speed and flue gas flow rate returned to normal, but the inner wall temperature began to rise, requiring further observation. At t4, the abnormality score was higher than the threshold of 0.7. At this time, there was no significant abnormality in the changes in ambient temperature, ambient humidity, damper opening, fan speed, and flue gas flow rate, indicating that the abnormalities were mainly concentrated in the inner and outer wall temperatures. The inner wall temperature increased significantly and the outer wall temperature also increased. It can be determined that internal scaling or ash accumulation caused the heat transfer efficiency to decrease.

[0055] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0056] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A boiler temperature monitoring system based on artificial intelligence, characterized in that: The system includes: a data acquisition module, a database, a data processing module and an alarm processing module; the data acquisition module is used to collect environmental information when the boiler is working, and the operating parameters and process parameters of the boiler during operation; the database is used to store all collected data; the data processing module is used to extract features from the collected data and monitor the boiler temperature using an isolation forest model; and the alarm processing module immediately issues an alarm when an abnormal state is identified.

2. The artificial intelligence-based boiler temperature monitoring system according to claim 1, characterized in that: The data acquisition module includes an environmental parameter acquisition unit, an operating parameter acquisition unit and a process parameter acquisition unit. The environmental parameter acquisition unit is used to collect ambient temperature and ambient humidity; the operating parameter acquisition unit is used to collect damper opening and blower speed; the process parameter acquisition unit is used to collect flue gas flow rate and furnace inner and outer wall temperature data, and all collected data are transmitted to the database.

3. The artificial intelligence-based boiler temperature monitoring system according to claim 1, characterized in that: The data processing module includes a feature extraction unit and an anomaly detection unit. The feature extraction unit extracts features from ambient temperature, ambient humidity, damper opening, blower speed and flue gas flow rate data, and calculates the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate and flue gas flow rate change rate; the anomaly detection unit uses the extracted feature data to train an isolation forest model to monitor the boiler temperature.

4. The artificial intelligence-based boiler temperature monitoring system according to claim 1, characterized in that: The alarm processing module includes an alarm triggering unit and an alarm device. The alarm triggering unit immediately triggers an alarm when an abnormality in the boiler temperature is detected; the alarm device issues an alarm through an audible and visual alarm device and notifies relevant management personnel.

5. A boiler temperature monitoring method based on artificial intelligence, comprising the following steps: S1: Collect the ambient temperature, ambient humidity, damper opening, blower speed, flue gas flow rate data and furnace inner and outer wall temperature data when the boiler is working; S2: Further clean the acquired boiler data to remove abnormal points and noise; S3: Extract features from the cleaned data and assign different weights to each feature; S4: Use the features extracted by the feature extraction unit to train the isolation forest model and establish a boiler temperature anomaly detection model.

6. The method for monitoring boiler temperature based on artificial intelligence according to claim 5, characterized in that: In step S1: install ambient temperature and humidity sensors at the air inlet and outlet of the boiler to obtain ambient temperature and humidity data; install a damper opening sensor on the boiler damper to monitor the damper opening in real time; install a blower speed sensor on the rotating shaft of the blower to monitor the blower speed in real time; install a flue gas flow rate sensor on the straight pipe section in the boiler flue to monitor the flue gas flow rate in real time; install infrared temperature sensors on the inner and outer walls of the boiler to obtain the inner and outer wall temperature data of the furnace.

7. The method for monitoring boiler temperature based on artificial intelligence according to claim 5, characterized in that: In step S3: In step S2: the boiler data obtained is further cleaned, and the Z-score method and IQR method are used to remove outliers and remove noise.

8. The method for monitoring boiler temperature based on artificial intelligence according to claim 5, characterized in that: Extract features from the cleaned data and assign different weights to each feature. The specific steps are as follows: S3-1: Extract features from the cleaned data and calculate the ambient temperature change rate, ambient humidity change rate, damper opening change rate, blower speed change rate, and flue gas flow rate change rate. The feature calculation formula is as follows: env (t) = T env (t)-T env (t-1) Calculate the ambient temperature change rate, where ΔT env (t) represents the rate of change of ambient temperature, T env (t) represents the ambient temperature at time t, T env (t) represents the ambient temperature at time t-1; according to ΔH env (t) = H env (t)-H env (t-1) Calculate the rate of change of ambient humidity, where ΔH env (t) represents the rate of change of ambient humidity, H env (t) represents the ambient humidity at time t, H env (t-1) represents the ambient humidity at time t-1; the damper opening change rate is calculated according to ΔD(t)=D(t)-D(t-1), wherein ΔD(t) represents the damper opening change rate, D(t) represents the damper opening at time t, and D(t-1) represents the damper opening at time t-1; the blower speed change rate is calculated according to ΔF(t)=F(t)-F(t-1), wherein ΔF(t) represents the blower speed change rate, F(t) represents the blower speed at time t, and F(t-1) represents the blower speed at time t-1; the flue gas flow rate change rate is calculated according to ΔV(t)=V(t)-V(t-1), wherein ΔV(t) represents the flue gas flow rate change rate, V(t) represents the flue gas flow rate at time t, and V(t-1) represents the flue gas flow rate at time t-1; S3-2: Assign different weights to each feature based on its impact on the boiler's operating status. Specific weights are as follows: ambient temperature is weighted by w1, ambient humidity is weighted by w2, damper opening is weighted by w3, blower speed is weighted by w4, and flue gas flow rate is weighted by w5. Here, w1 represents the weight assigned to ambient temperature, w2 represents the weight assigned to ambient humidity, w3 represents the weight assigned to damper opening, w4 represents the weight assigned to blower speed, and w5 represents the weight assigned to flue gas flow rate. S3-3: Perform weighted calculation on the feature change rate to obtain a comprehensive score, and preliminarily determine whether the data point is abnormal based on the preset feature change rate threshold. The comprehensive score calculation formula is as follows: W all =ΔT env (t)*w1+ΔH env (t)*w2+ΔD(t)*w3+ΔF(t)*w4+ΔV(t)*w5 Among them, W all Indicates that the comprehensive score is obtained by weighted calculation of the feature change rate.

9. The method for monitoring boiler temperature based on artificial intelligence according to claim 5, characterized in that: In step S4: the features extracted by the feature extraction unit and the historical data are used to train the isolation forest model to establish a boiler temperature anomaly detection model. The specific steps for training the isolation forest model are as follows: S4-1: Standardize the extracted features through normalization; S4-2: Use the standardized historical feature data to train the isolation forest model and establish a boiler temperature anomaly detection model. Then, input the standardized real-time feature data into the trained isolation forest model for real-time detection. S4-3: The model performs an anomaly score on each data point and further determines whether the data point is abnormal based on the preset anomaly score threshold. If the anomaly score exceeds the threshold, the data point is considered abnormal. S4-4: Analyze the overall score and the weighted score of each feature to determine whether there is a problem with internal scaling or dust accumulation.

Images