Thermal control instrument fault self-diagnosis system based on multi-source data fusion

Abstract

The application discloses a thermal control instrument fault self-diagnosis system based on multi-source data fusion and relates to the technical field of prediction and health management. The thermal control instrument fault self-diagnosis system based on multi-source data fusion comprises the following steps: a data collection and arrangement module, which is used for collecting thermal control state data in real time and pre-processing the thermal control state data; a feature construction and extraction module, which is used for determining thermal bias abnormality of the pre-processed thermal control state data; a state intelligent evaluation module, which is used for comparing temperature trends of the thermal control state data after thermal bias abnormality determination; a diagnosis linkage control module, which is used for controlling and adjusting according to the results of thermal bias abnormality determination and temperature trend comparison; and a result display and archiving module, which is used for arranging and recording the diagnosis results and processing states of measuring points. The problems that false alarms are caused by wall temperature thermocouple signal drift and mutation, adjacent measuring points and operating states are not fused and analyzed, and real overheat and instrument faults are difficult to distinguish are solved.

Description

Technical Field

[0001] This invention relates to the field of prediction and health management technology, specifically to a self-diagnostic system for thermal control instruments based on multi-source data fusion. Background Technology

[0002] With the continuous advancement of predictive and health management technologies in the power and chemical industries, thermal control instruments, as key sensing components, play a crucial role in acquiring equipment operating status, monitoring anomalies, and ensuring operational safety. Especially in boiler systems, information from measuring points such as wall temperature and steam temperature is the fundamental data source for achieving precise control and risk prevention; their health status directly impacts the stability of system operation.

[0003] For example, the invention patent with announcement number CN115167366B discloses a board fault diagnosis method based on the weighted sum of squares of sensor signal residual sequences, relating to the field of electronic controller fault diagnosis technology. This method includes: acquiring raw temperature process signals from each temperature sensor in the gas turbine electronic controller system, performing data preprocessing to form a raw dataset; inputting the raw dataset into a residual sequence weighted sum of squares calculation model; selecting the weighted sum of squares of all sensors associated with the I / O board, inputting it into the I / O board health calculation model to obtain the board's health value; calculating the difference between the board's health value and a preset normal range boundary line, and determining whether the board has malfunctioned based on the difference. This method can quickly diagnose the fault status of thermocouple I / O boards in an engineering manner, providing an engineering-applicable means for timely identification of I / O board faults in electronic controllers, and providing important support for fault diagnosis of gas turbine control system loops.

[0004] For example, invention patent CN111459145B discloses a transformer temperature controller calibration device. This device includes a temperature conduction component, a temperature equalization block, a temperature sensor, and a temperature controller. The temperature equalization block is built into the temperature conduction component and is in close contact with it to maintain temperature uniformity. The temperature equalization block is used to place the thermocouple of the transformer temperature controller to be calibrated and is matched with the thermocouple. The temperature sensor is used to collect the temperature of the temperature equalization block and send the temperature to the temperature controller. The temperature controller is used to control the temperature of the temperature conduction component. Compared with existing technologies, this calibration device is small in size, portable, easy to operate, and can realize on-site calibration of transformer temperature controllers. The temperature equalization block can improve the heat conduction speed, shorten the calibration time, and improve work efficiency. The temperature equalization block can maintain the uniformity of the axial and radial temperature of the thermocouple, ensuring accurate and reliable calibration results and guaranteeing work quality.

[0005] Currently, the status judgment of thermal control instruments in the industry mostly relies on single-point measurement over-limit strategies, lacking integrated analysis of multi-point data trends and operating conditions. In complex thermal scenarios, problems such as measurement point drift, false alarms and misjudgments, and linkage failures are common, making it difficult to achieve early identification and graded intervention of potential anomalies, thus limiting the effective implementation of predictive maintenance and intelligent diagnostic technologies.

[0006] To address the above issues, there is an urgent need for a self-diagnostic system for thermal control instruments based on multi-source data fusion. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a self-diagnostic system for thermal control instruments based on multi-source data fusion. This system solves the problems of false alarms caused by drift and sudden changes in wall temperature thermocouple signals, as well as the lack of fusion analysis of adjacent measuring points and operating status, making it difficult to distinguish between actual overheating and instrument malfunctions.

[0009] Technical solution

[0010] To achieve the above objectives, this invention provides the following technical solution: a self-diagnostic system for thermal control instrument faults based on multi-source data fusion, comprising the following steps: a data collection and processing module for real-time acquisition of thermal control status data and preprocessing of the data; a feature construction and extraction module for determining thermal deviation anomalies in the preprocessed thermal control status data and extracting key feature variables for status deviation identification; a status intelligent evaluation module for comparing temperature trends in the thermal control status data after thermal deviation anomaly determination and analyzing the matching degree between the measurement point change trend and neighboring points; a diagnostic linkage control module for integrating the results of thermal deviation anomaly determination and temperature trend comparison to execute control participation and adjustment, adjusting the participation weight and response strategy of the measurement points in the control loop; and a result display and archiving module for organizing and recording the diagnostic results and processing status of the measurement points, supporting operators in status confirmation and manual operation, and generating complete log information and structured reports.

[0011] Further, the specific steps for real-time acquisition of thermal control status data are as follows: Thermal control status data includes: sliding window, current measuring point temperature, main steam temperature, load fitting temperature, and number of neighboring points; a sliding window is obtained by setting a local time series interval with the current time as the endpoint, a fixed time length backward, and the corresponding number of data points; the current measuring point temperature is obtained by acquiring data through thermocouple temperature sensors; the main steam temperature is obtained by acquiring data through thermal control measuring points in the main steam pipeline; the load fitting temperature is obtained by fitting a historical function between the boiler load and the current measuring point temperature; and the number of neighboring points is obtained through the configuration of the measuring point layout structure.

[0012] Furthermore, the specific steps for preprocessing the thermal control status data are as follows: The preprocessing steps include: interpolating data of different frequencies according to a set time step to correct the main steam temperature and the current measuring point temperature, as well as variables with response lag; performing sliding interpolation on short-term missing data, and retaining missing markers for long-term missing data; marking out-of-limit values ​​and abrupt change points as abnormal, not directly removing them, and providing them for subsequent module judgment; standardizing all continuous variables to unify their numerical scale, and compressing their numerical range through normalization.

[0013] Further, the specific steps for determining thermal deviation anomalies in the preprocessed thermal control status data are as follows: Obtain the sliding window, current measuring point temperature, main steam temperature, and load fitting temperature; calculate the temperature sliding mean by setting the current measuring point temperature sequence within the sliding window; calculate the temperature fluctuation intensity by setting the standard deviation function within the sliding window; subtract the main steam temperature from the current measuring point temperature, divide the result by the sum of the main steam temperature and the minimum correction term, and then square the quotient to obtain the main steam temperature relative deviation term; subtract the load fitting temperature from the current measuring point temperature, divide the result by the sum of the current measuring point temperature and the minimum correction term, and then square the quotient to obtain the load fitting temperature relative deviation term; divide the temperature fluctuation intensity by the sum of the temperature sliding mean and the minimum correction term to obtain the temperature fluctuation relative deviation term; finally, add the above three relative deviation terms to obtain the comprehensive temperature difference value.

[0014] Further, the specific steps for extracting key feature variables for state deviation identification are as follows: Real-time comparison of the comprehensive temperature difference value with the deviation assessment threshold, which includes a primary deviation threshold and a secondary deviation threshold; when the comprehensive temperature difference value is greater than or equal to the primary deviation threshold, the measuring point is removed from the control loop and alarm logic, a neighboring point substitution sampling mechanism is activated and interpolation estimation compensation is performed, a maintenance prompt is triggered and a fault report is generated and written to the anomaly log, and the measuring point is marked as failed in the status panel; when the comprehensive temperature difference value is greater than the secondary deviation threshold but less than the primary deviation threshold, an enhanced sampling mechanism is activated to shorten the sampling period and improve data resolution, temperature sequence smoothing is enabled to reduce occasional jump interference, the measuring point control weight is reduced to 50% but still participates in but does not dominate control, an early warning flag is set in the system and a trend tracking buffer period is entered; when the comprehensive temperature difference value is less than or equal to the secondary deviation threshold, the default sampling period is maintained, no encrypted sampling or additional processing is required, the measuring point data participates in full control and alarm logic, and the measuring point status is marked as normal and reliable.

[0015] Further, the specific steps for comparing the temperature trend of the thermal control status data after the thermal deviation anomaly determination are as follows: obtain the current measuring point temperature and the number of neighboring points; calculate the current temperature change by the difference between the current measuring point temperature value and the previous time point; calculate the temperature change of neighboring points by the difference between the current measuring point temperature value and the previous value; calculate the absolute value of the difference between the current temperature change and the temperature change of neighboring points as the numerator; then calculate the larger of the absolute values ​​of the current temperature change and the temperature change of neighboring points, and add a minimum value correction term as the denominator; divide the numerator of each group by the denominator to obtain the normalized difference degree; sum all the temperature changes of neighboring points after the above calculation, and divide by the number of neighboring points to finally obtain the temperature trend comparison value.

[0016] Further, the specific steps for analyzing the trend of the measuring point and the matching degree of neighboring points are as follows: Real-time comparison of the temperature trend comparison value with the trend deviation threshold, which includes a primary deviation threshold and a secondary deviation threshold; When the temperature trend comparison value is greater than or equal to the primary deviation threshold: the measuring point is automatically isolated and does not participate in real-time control calculations; spatial interpolation logic is activated to dynamically compensate for this point using the neighboring point trend; a trend anomaly signal is sent to the alarm system; manual inspection of wiring and sensors is recommended; When the temperature trend comparison value is greater than the secondary deviation threshold but less than the primary deviation threshold: the neighboring point trend monitoring program is activated; three or more adjacent measuring points are dynamically compared; the control weight is automatically reduced to 50%; the secondary adjustment role is retained; a trend fluctuation suspicion label is added to the system panel for on-duty personnel to follow up; the measuring point data is retained for subsequent analysis, but its participation scope is limited; When the temperature trend comparison value is less than or equal to the secondary deviation threshold: the current measuring point is kept in a consistent trend with neighboring points; the default trust level and control weight of the current measuring point are maintained; no additional comparison or intervention operations are performed.

[0017] Further, the specific steps for implementing control participation adjustment based on the fusion of thermal deviation anomaly judgment and temperature trend comparison results are as follows: Obtain the comprehensive temperature deviation value and the temperature trend comparison value; obtain a real-time scoring sequence by continuously calculating the comprehensive temperature difference value, temperature trend comparison value, and control weight adjustment value of each measuring point during operation, and recording their dynamic changes in chronological order; obtain the control adjustment threshold by fitting historical operating data of normal measuring points in the real-time scoring sequence; subtract the control adjustment threshold from the sum of the comprehensive temperature deviation value and the temperature trend comparison value, using this as the numerator; divide the two values ​​using the control smoothing coefficient as the denominator, and use the result as the input value of the exponential function; calculate the exponential value with the natural constant as the base, obtaining the output of the exponential function; add one to the output of the exponential function, using this as the denominator, and use one as the numerator to obtain the fractional result, ultimately obtaining the control weight adjustment value.

[0018] Furthermore, the specific steps for adjusting the participation weight and response strategy of the measurement point in the control loop are as follows: Real-time comparison of the control weight adjustment value and the weight evaluation interval, which includes a high confidence interval, a relatively high confidence interval, a low confidence interval, and an extremely low confidence interval: When the control weight adjustment value is greater than or equal to 0.9, the current measurement point is determined to be highly reliable, and the measurement point participates in all control logic and alarm judgments. The controller uses its original adjustment value without the need for enhanced sampling or replacement processing; when the control weight adjustment value is greater than or equal to 0.6 and less than 0.9, the current measurement point is determined to be relatively reliable, and the control system retains it. Its participation right will be moderately scaled up and down, and a trend tracking mechanism will be enabled to closely observe changes in the measuring points. The alarm system will slightly increase the trigger threshold. When the control weight adjustment value is greater than or equal to 0.3 and less than 0.6, the current measuring point will be judged as low confidence. The controller will reduce the control weight of the measuring point, start the neighbor point interpolation substitution mechanism, and enter the observation state. The alarm can only be responded to when it is triggered by other measuring points. When the control weight adjustment value is less than 0.3, the current measuring point will be judged as extremely low confidence. The measuring point will be excluded from all control and alarm logic. The substitution strategy will be used to completely shield its control output and set it to a red warning state in the interface.

[0019] Furthermore, the specific steps for organizing and recording the diagnostic results and processing status of measurement points, supporting operators in status confirmation and manual operations, and generating complete log information and structured reports are as follows: Organize the diagnostic tags, scores, and processing status of each measurement point to generate structured diagnostic result data; record the diagnostic results and processing status of measurement points at different times, constructing a status trend sequence to achieve operational traceability and comparative analysis; provide a manual operation interface to support operators in confirming the status of measurement points, modifying diagnostic tags, and adding remarks; automatically record the diagnostic process, alarm trigger responses, and manual operation behaviors to generate traceable log records; integrate and output the measurement point operating status, diagnostic results, and maintenance suggestions to generate a structured document for archiving management and operation and maintenance decision-making.

[0020] Beneficial effects

[0021] The present invention has the following beneficial effects:

[0022] (1) By constructing a multi-source data collection and processing module, this invention realizes the unified collection and standardized preprocessing of thermal control measurement point status data, which can solve the problems of difficulty in time alignment of original data and non-standard handling of missing anomalies, and improve the quality and timeliness consistency of input data for subsequent diagnostic analysis.

[0023] (2) This invention extracts the relative deviation index between the current measuring point and the main steam temperature and the load fitting temperature, and integrates the sliding mean and the fluctuation intensity to construct a comprehensive temperature difference value. This can effectively identify the thermal deviation anomaly of the measuring point, realize the transformation from single-point over-limit judgment to trend modeling judgment, and improve the accuracy and stability of anomaly identification.

[0024] (3) The present invention utilizes temperature trend comparison values ​​to achieve consistency analysis of the changing trends of the measuring point and its neighboring points, and combines deviation thresholds to distinguish trends, which can significantly enhance the sensitivity to thermal control measuring point drift and sudden changes, and avoid system false alarms and false linkage control caused by local anomalies.

[0025] (4) Based on the control weight adjustment value, the present invention dynamically adjusts the control strategy of the measuring point, and combined with the diagnostic linkage and result marking mechanism, realizes the automatic classification, response optimization and weight correction of the measuring point status, effectively improving the adaptive ability and robustness of the control system to the instrument fault status.

[0026] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0027] Figure 1 This is a structural diagram of the thermal control instrument fault self-diagnosis system based on multi-source data fusion of the present invention;

[0028] Figure 2 This is a bar chart showing the temperature trend comparison values ​​of the present invention.

[0029] Figure 3 This is a comparison chart of multiple indicators fused according to the present invention;

[0030] Figure 4 This is a dynamic change curve of the control weight adjustment value according to the present invention. Detailed Implementation

[0031] 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.

[0032] Please see Figures 1-4This invention provides a technical solution: a self-diagnostic system for thermal control instrument faults based on multi-source data fusion, comprising the following steps: a data collection and processing module for real-time acquisition of thermal control status data and preprocessing of the thermal control status data; a feature construction and extraction module for determining thermal deviation anomalies in the preprocessed thermal control status data and extracting key feature variables for status deviation identification; a status intelligent evaluation module for comparing temperature trends in the thermal control status data after thermal deviation anomaly determination and analyzing the matching degree between the measurement point change trend and neighboring points; a diagnostic linkage control module for integrating the results of thermal deviation anomaly determination and temperature trend comparison to execute control participation adjustment, adjusting the participation weight and response strategy of the measurement points in the control loop; and a result display and archiving module for organizing and recording the diagnostic results and processing status of the measurement points, supporting operators in status confirmation and manual operation, and generating complete log information and structured reports.

[0033] Specifically, the real-time acquisition of thermal control status data involves the following steps: Thermal control status data includes a sliding window, current measuring point temperature, main steam temperature, load fitting temperature, and the number of neighboring points, covering both temporal and spatial modeling elements of thermal state characteristics. A local time series interval is formed by setting a fixed time length and corresponding number of data points with the current time as the endpoint, thus obtaining a sliding window for constructing a dynamic feature sequence. The current measuring point temperature is acquired by thermocouple temperature sensors deployed in the high-temperature region, serving as the core input variable reflecting real-time changes in wall temperature. The main steam temperature is acquired by thermal control measuring points located in the main steam outlet pipeline, providing a global reference baseline. The load fitting temperature is obtained by fitting the historical function relationship between boiler load and the current measuring point temperature, thereby introducing the theoretical expected value of the operating condition. The number of neighboring points is obtained by analyzing the measuring point layout structure and physical adjacency configuration, providing a structural basis for subsequent trend consistency judgment.

[0034] In this implementation plan, this step, by comprehensively collecting thermal control status data and constructing structured time-series inputs, provides high-quality basic data support for subsequent feature extraction and status diagnosis. By unifying the time benchmark, clarifying the spatial correlation of measurement points, and introducing expected operating conditions and global references for the main steam, this acquisition process not only ensures the integrity and consistency of the data but also improves the background adaptability and trend discriminability of anomaly identification, laying the foundation for accurate assessment of thermal control status under multi-source fusion.

[0035] Specifically, the preprocessing steps for thermal control status data are as follows: Preprocessing includes: uniformly interpolating data of different frequencies according to a set time step to ensure alignment of various variables on the same time axis; correcting the temporal offset between the main steam and the current measuring point temperature, as well as variables with response lag, to improve time matching accuracy; performing sliding interpolation on short-term missing data to smoothly fill blank segments, while retaining missing markers for long-term missing data to avoid introducing spurious signals; marking out-of-limit values ​​and abrupt change points as anomalies, not directly removing them, but retaining them as the basis for subsequent anomaly identification and status reasoning analysis; standardizing all continuous variables to unify their numerical scale, eliminating the influence of different dimensions, and compressing their numerical range through normalization to adapt the data to subsequent input specifications.

[0036] In this implementation plan, this step improves the consistency, integrity, and usability of thermal control status data through systematic preprocessing, providing a stable and standardized input foundation for subsequent feature extraction and anomaly identification. Unifying the time series, filling in missing data, marking anomalies, and standardizing numerical ranges not only effectively reduce the interference risks caused by data noise and structural biases but also enhance the robustness of processing multi-class data fusion, ensuring the accuracy and continuity of diagnostic analysis.

[0037] Specifically, the steps for judging thermal deviation anomalies in the preprocessed thermal control status data are as follows: Obtain the sliding window, current measuring point temperature, main steam temperature, and load fitting temperature to construct a core data set for capturing thermal control status deviations; calculate the temperature sliding mean by setting the current measuring point temperature sequence within the sliding window to reflect the temperature stability level within a short period; calculate the temperature fluctuation intensity by setting the standard deviation function within the sliding window to identify local jump characteristics at the measuring points; subtract the main steam temperature from the current measuring point temperature, divide the result by the sum of the main steam temperature and the minimum correction term, and then calculate the average value. The following steps are performed: First, the relative deviation term of the main steam temperature is obtained, which measures the degree of deviation of the measuring point from the global thermal state. Second, the load-fitted temperature is subtracted from the current measuring point temperature, and the result is divided by the sum of the current measuring point temperature and the minimum correction term. The quotient is then squared to obtain the load-fitted temperature relative deviation term, which reflects the degree of deviation between the measuring point and the expected operating temperature. Third, the temperature fluctuation intensity is divided by the sum of the temperature sliding mean and the minimum correction term to obtain the temperature fluctuation relative deviation term, which measures the short-term instability of the measuring point. Finally, the above three relative deviation terms are added together to obtain the comprehensive temperature difference value, which serves as a key indicator for judging thermal deviation anomalies.

[0038] The specific calculation method for the comprehensive value of temperature difference is as follows:

[0039]

[0040] In the formula, This represents the overall value of the temperature difference. This indicates the current temperature at the measuring point. Indicates the main steam temperature. Indicates the load fitting temperature. Indicates the intensity of temperature fluctuations. This represents the moving average of temperature. This represents the minimum value correction term.

[0041] In this implementation plan, this step constructs a comprehensive index that measures the degree of deviation of the thermal state at the measuring point from multiple perspectives by building a temperature sliding mean, fluctuation intensity, and relative deviation from the fitted temperatures of the main steam and load. This effectively improves the ability to identify thermal anomalies. Compared with traditional single-point over-limit judgment methods, this method can combine the fluctuation characteristics of the measuring point itself with the background of operating conditions to achieve more refined and stable detection of thermal control anomalies, providing an accurate state basis for subsequent trend analysis and control strategies.

[0042] Specifically, the key feature variables extracted for state deviation identification are as follows: Real-time comparison of the comprehensive temperature difference value with the deviation assessment threshold. The deviation assessment threshold includes a primary deviation threshold and a secondary deviation threshold, used to distinguish different levels of thermal deviation anomalies. The deviation threshold is set based on statistical analysis of the distribution of measurement point deviation values ​​under normal operating conditions, combined with the temperature fluctuation range and relative deviation characteristics of measurement points in historical stable operating conditions, possessing engineering adaptability and dynamic adjustment capabilities. When the comprehensive temperature difference value is greater than or equal to the primary deviation threshold, the thermal state of the measurement point is judged to be severely deviated, and its observation results are considered unable to reflect the true thermal conditions. It is then removed from the control loop and alarm logic, and simultaneously, a neighboring point substitution sampling mechanism is activated, using trend information from adjacent normal measurement points for interpolation estimation compensation. This ensures the continuity and stability of the control link without relying on faulty measurement points. A maintenance prompt is triggered, and a fault report is generated and written to the anomaly log. The measurement point is marked as "failed" in the status panel, clearly indicating the alarm level, facilitating timely location by maintenance personnel. When the comprehensive temperature difference value is greater than the secondary deviation threshold but less than the primary deviation threshold, it is identified as a slight deviation. Such measuring points are not directly eliminated, but a suspicious trend has emerged. An enhanced sampling mechanism is activated to shorten the sampling period and improve data resolution, enhancing observation sensitivity over time. Temperature sequence smoothing is also enabled to filter local interference and stabilize the observation curve. Simultaneously, the measuring point's control weight is reduced to 50%, retaining only auxiliary control roles to prevent adverse interference with the overall control strategy. An early warning flag is set at this stage, and the measuring point is included in the trend tracking buffer period for continuous monitoring of its subsequent changes, supporting further judgment and dynamic strategy switching. Compared to existing technologies, this "trend tracking buffer mechanism" can achieve intelligent monitoring of suspicious data without forced elimination, significantly reducing the probability of misjudgment and miscontrol. When the comprehensive temperature difference value is less than or equal to the secondary deviation threshold, the measuring point is determined to be in a normal and reliable state. The default sampling period is maintained, without encrypted sampling or additional processing. Its data continues to participate in full control and alarm logic, and its status is marked as "normal and reliable," entering the regular closed-loop execution path. This processing flow constructs a linkage logic from judgment to control strategy based on a quantitative multi-level response mechanism, which effectively solves the problems of rough fault judgment and rigid adjustment response in traditional thermal control systems.

[0043] In this implementation plan, this step achieves graded judgment and response control of the thermal control measurement point status by comparing the comprehensive temperature difference value with the set deviation evaluation threshold in real time. Based on the degree of deviation, the system is divided into failure, suspected, and normal zones, and the sampling strategy, control weights, and alternative mechanisms are dynamically adjusted. This allows the system to improve diagnostic sensitivity and flexibility while ensuring operational safety, effectively reducing the risk of false alarms and misadjustments caused by single-point anomalies.

[0044] Specifically, the steps for comparing the temperature trend of thermal control status data after thermal deviation anomaly determination are as follows: Obtain the current measuring point temperature and the number of neighboring points to construct a local spatial measuring point set; calculate the current temperature change by the difference between the current measuring point temperature value and the previous time point, which is used to characterize the short-term dynamic fluctuations of the measuring point; calculate the temperature change of neighboring points by the difference between the current measuring point temperature value and the previous value, and extract their response behavior at the same time scale; calculate the absolute value of the difference between the current temperature change and the temperature change of neighboring points, as the numerator of the trend deviation, reflecting the local synchronicity between the current measuring point and neighboring measuring points; then calculate the larger of the absolute values ​​of the current temperature change and the temperature change of neighboring points, and add a minimum value correction term as the denominator to avoid numerical instability caused by a zero denominator; divide the numerator of each group by the denominator to obtain the normalized difference degree, achieving a unified scale for trend differences; sum all the temperature changes of neighboring points after the above calculations, and divide by the number of neighboring points to finally obtain the temperature trend comparison value, which is used to quantify the consistency of trend changes between the current measuring point and its neighboring measuring points. This method integrates local spatial dynamic comparison and normalized ratio calculation, which improves the sensitivity and anti-interference ability of abnormal trend identification.

[0045] The specific calculation method for the temperature trend comparison value is as follows:

[0046]

[0047] In the formula, This indicates a comparison value of temperature trends. This indicates the current temperature change. This represents the change in temperature at adjacent points. Indicates the number of neighboring nodes. This represents the minimum value correction term.

[0048] Table 1 shows the temperature trend comparison data provided in this embodiment. In this embodiment, the current temperature change of measuring point 1 is set to 0.5, the number of neighboring points is set to 3, the temperature change of neighboring point 1 is set to 0.52, the temperature change of neighboring point 2 is set to 0.48, and the temperature change of neighboring point 3 is set to 0.50; the current temperature change of measuring point 2 is set to 0.6, the number of neighboring points is set to 3, the temperature change of neighboring point 1 is set to 0.55, the temperature change of neighboring point 2 is set to 0.65, and the temperature change of neighboring point 3 is set to 0.60; the current temperature change of measuring point 3 is set to 0.9, and the temperature change of neighboring point 3 is set to 0.9. The quantity is set to 3, the temperature change of neighboring points 1 is set to 1.00, the temperature change of neighboring points 2 is set to 0.85, and the temperature change of neighboring points 3 is set to 0.92; the current temperature change of measuring point 4 is set to 1.2, the number of neighboring points is set to 3, the temperature change of neighboring points 1 is set to 1.40, the temperature change of neighboring points 2 is set to 1.20, and the temperature change of neighboring points 3 is set to 1.30; the current temperature change of measuring point 5 is set to 1.6, the number of neighboring points is set to 3, the temperature change of neighboring points 1 is set to 2.00, the temperature change of neighboring points 2 is set to 1.70, and the temperature change of neighboring points 3 is set to 1.80.

[0049] Table 1. Comparison of Temperature Trends

[0050]

[0051] like Figure 2 The image shows a bar chart of temperature trend comparison values ​​provided in an embodiment of the present invention. According to the data in the image and table, the set secondary deviation threshold is 0.05, and the primary deviation threshold is 0.10, used to identify the level of difference between the trend change of the measuring point and its neighboring points. As can be seen from the bar chart, the temperature trend comparison values ​​of measuring points 1 to 4 are all below the secondary deviation threshold, indicating that the temperature change trends of these measuring points are basically consistent with those of their neighboring points under continuous time sampling, without showing obvious deviation behavior, belonging to the trend consistency range. However, the trend comparison value of measuring point 5 is 0.0752, exceeding the secondary deviation threshold, and is in the suspicious deviation range, indicating that this measuring point has a certain difference in thermal change response compared with surrounding measuring points. Further judgment should be made based on other indicators to determine whether it is a local anomaly or an isolated point outside the group. This figure can intuitively reflect the degree of local consistency of each measuring point in the dimension of temperature change trend and help to identify potential risk measuring points. The threshold line in the figure serves as a judgment boundary, providing quantitative support for the system to dynamically identify abnormal trend behavior.

[0052] In this implementation scheme, this step calculates the difference in temperature change trends between the current measuring point and its neighboring points over a short timescale to form a temperature trend comparison value, thereby achieving a quantitative assessment of the spatial consistency of the measuring points. Compared with the traditional absolute temperature comparison method, this method, based on the change quantity, integrates normalization processing and a multi-point trend comparison mechanism. It can effectively identify hidden fault behaviors of measuring points when the fluctuation amplitude is normal but the trend is abnormal, improves the ability to identify atypical anomalies such as drift and jump, and helps to enhance the system's dynamic perception and analysis depth of complex thermal control states.

[0053] Specifically, the steps for analyzing the trend of a measuring point and its matching degree with neighboring points are as follows: Real-time comparison of the temperature trend comparison value with the trend deviation threshold. The trend deviation threshold includes a primary deviation threshold and a secondary deviation threshold, used to classify the consistency between the measuring point and its neighboring trends into three categories: abnormal, suspicious, and normal. When the temperature trend comparison value is greater than or equal to the primary deviation threshold, it is determined that the trend of the measuring point is significantly different from that of its neighboring points, and it is automatically isolated and not included in real-time control calculations to avoid interference from abnormal trends in control decisions. Simultaneously, spatial interpolation logic is activated to dynamically extrapolate the changes at this point based on the trend trajectories of the current adjacent measuring points, filling the gaps in its impact on the system response, and sending a trend abnormality signal to the alarm system to prompt operators to conduct manual verification, prioritizing checks for loose wiring and sensor failure. When the temperature trend comparison value is greater than the secondary deviation threshold but less than the primary deviation threshold, the measuring point is identified as being in a state of slight trend deviation. The neighboring point trend monitoring program is activated, continuously comparing the synchronicity of three or more adjacent measuring points. The control weight is automatically reduced to 50%, retaining it as a secondary adjustment reference. Simultaneously, a suspicious trend fluctuation label is marked on the system interface to guide on-duty personnel to focus their observation. The measuring point data is retained for subsequent trend evolution analysis but does not directly dominate control. When the temperature trend comparison value is less than or equal to the secondary deviation threshold, the measuring point is determined to have a stable trend and good matching with neighboring points. The original control weight and trust state are maintained, and no additional judgment or intervention measures are executed, ensuring system operating efficiency and resource utilization stability. This hierarchical discrimination and dynamic response mechanism can significantly improve the system's accuracy and fault tolerance in identifying trend anomalies. Unlike traditional coarse judgment methods based on absolute temperature mutations, it can detect trend deviations at an early stage, preventing minor faults from evolving into system interference.

[0054] In this implementation scheme, this step constructs a hierarchical identification and response mechanism for trend anomalies by comparing the temperature trend comparison value with the set trend deviation threshold in real time, realizing dynamic determination of the consistency of the changing trends of the measuring point and its neighboring points. The influence of the measuring point in the control loop is automatically adjusted according to the degree of deviation, employing isolation control, neighboring point compensation, and trend monitoring strategies to ensure that abnormal measuring points do not affect the overall regulation stability. Compared with traditional judgment methods based on static temperature values, this method places greater emphasis on trend synchronization and local spatial matching relationships, enabling effective intervention in the early stages of instability and improving the system's ability to identify dynamic faults and its control robustness.

[0055] Specifically, the steps for implementing control participation and regulation by integrating the results of thermal deviation anomaly judgment and temperature trend comparison are as follows: Obtain the comprehensive value of temperature deviation and the temperature trend comparison value to construct the input basis reflecting the integrity of the thermal state of the measuring point; continuously calculate the comprehensive value of temperature difference, the temperature trend comparison value, and the control weight adjustment value of each measuring point during operation, and record their dynamic changes in chronological order to obtain a real-time scoring sequence reflecting the evolution of the measuring point's state, ensuring that the control response has temporal continuity and state tracking capability; obtain the control adjustment threshold by fitting historical operating data of normal measuring points in the real-time scoring sequence, enabling the threshold to have operating condition adaptability and dynamic self-adaptability; use the sum of the comprehensive value of temperature deviation and the temperature trend comparison value minus the control adjustment threshold as the numerator, the control smoothing coefficient as the denominator, divide the two and use the result as the input value of the exponential function to construct a nonlinear response curve of the control signal change; calculate the exponential value with the natural constant as the base to obtain the output of the exponential function, then add one to the output result as the denominator, and use one as the numerator to obtain the fractional result, finally obtaining the control weight adjustment value. The value is automatically obtained by fitting the correspondence between the control weight adjustment value and the rate of change of the measuring point state, with a recommended range of 0.1 to 0.5. This value serves as the dynamic weight basis for the measuring point's participation in the control calculation, reflecting its reliability and adjustment importance in the current state. By fusing multi-source anomaly features and constructing a nonlinear mapping function, this method achieves fine adjustment of the measuring point's control influence, effectively enhancing the system's resilience and feedback control capability to abnormal states.

[0056] The specific calculation method for the control weight adjustment value is as follows:

[0057]

[0058] In the formula, This indicates the control weight adjustment value. This represents the overall value of the temperature difference. This indicates a comparison value of temperature trends. Indicates the control and adjustment threshold. This represents the control smoothing coefficient.

[0059] In this implementation plan, this step integrates the comprehensive value of temperature deviation with the comparison value of temperature trend, and combines historical scoring data to construct a control weight adjustment value, thereby achieving dynamic quantitative adjustment of the participation degree of the measuring points in the control system. Compared with the traditional fixed weight and anomaly-based rejection approach, this method introduces an exponential function mapping mechanism to flexibly control the participation weight of the measuring points according to the degree of anomaly. This ensures both the system's responsiveness to anomalies and the ability to buffer boundary state points, significantly improving the stability, robustness, and adjustment accuracy of the thermal control system.

[0060] Specifically, the steps for adjusting the participation weight and response strategy of measurement points in the control loop are as follows: Real-time comparison of the control weight adjustment value and the weight evaluation interval. The weight evaluation interval includes a high confidence interval, a relatively high confidence interval, a low confidence interval, and an extremely low confidence interval, used to finely classify the control participation level of the measurement point based on its status. When the control weight adjustment value is greater than or equal to 0.9, the current measurement point is determined to be stable and reliable, and is classified as a high confidence category. This measurement point can fully participate in the control logic and alarm judgment. The controller directly uses its original adjustment value without additional processing, maintaining strong control over the main control target. When the control weight adjustment value is greater than or equal to 0.6 and less than 0.9, it is determined to be a relatively high confidence point. Its participation right is retained, but the adjustment result is appropriately scaled to weaken its dominance. At the same time, a trend tracking mechanism is activated to closely observe its subsequent changes. The alarm system appropriately raises the trigger threshold to reduce the probability of false alarms, ensuring a balance between response sensitivity and system stability. When the control weight adjustment value is greater than or equal to 0.3 and less than 0.6, it is judged as low confidence. The controller significantly reduces the control weight of this measurement point and instead activates the neighbor point interpolation substitution mechanism, using surrounding normal measurement points to provide auxiliary reference. At the same time, this measurement point enters the system observation state, and the alarm response strategy requires joint confirmation with other measurement points to improve fault tolerance. When the control weight adjustment value is less than 0.3, the state of this measurement point is considered extremely unreliable. It is completely removed from all control and alarm logic, and a substitution strategy is used to shield its output interference. It is marked with a red warning mark on the interface. Through this control participation adjustment strategy based on dynamic mapping of scoring values, a continuous hierarchical response mechanism for measurement point confidence is realized, improving control stability and fault tolerance, which is significantly better than the traditional one-size-fits-all rejection logic.

[0061] In this implementation scheme, this step establishes a hierarchical adjustment mechanism for the participation of measuring points in the control loop by comparing the control weight adjustment value with the preset confidence interval in real time, realizing dynamic adjustment of control weights and differentiated management of response strategies. Based on the confidence level of the measuring point status, they are divided into four categories: high confidence, relatively high confidence, low confidence, and extremely low confidence, corresponding to different control strategies ranging from full participation to complete exclusion. Furthermore, auxiliary mechanisms such as trend tracking, neighbor substitution, and joint alarms are introduced. This method breaks the rigid logic of the traditional "abnormality equals exclusion," improves the flexibility and intelligence of the adjustment strategy, and effectively enhances the adaptability and control stability of the thermal control system to measuring point anomalies under complex operating conditions.

[0062] Specifically, the process of organizing and recording the diagnostic results and processing status of measurement points, supporting operators in status confirmation and manual operations, and generating complete log information and structured reports involves the following steps: Organizing the diagnostic tags, scores, and processing status of each measurement point to generate structured diagnostic result data, ensuring that the diagnostic information is standardized, clear, and identifiable by the system; recording the diagnostic results and processing status of measurement points at different times, constructing a status trend sequence to achieve time-based tracking and horizontal comparative analysis of the evolution of measurement point operating status, assisting in judging long-term stability and abnormal fluctuation patterns; providing a manual operation interface to support operators in confirming the status of measurement points, modifying diagnostic tags, and adding remarks, establishing a flexible collaborative channel between system identification and manual intervention; automatically recording the diagnostic process, alarm trigger responses, and manual operation behaviors, generating complete and traceable log records for subsequent fault reconstruction and event analysis; and integrating and outputting the measurement point operating status, diagnostic results, and maintenance suggestions to generate a structured document with a unified format and detailed content for daily archiving management and support for operation and maintenance strategy optimization, ensuring that the system diagnostic information has completeness, operability, and decision-making value.

[0063] In this implementation plan, this step constructs a traceable, interactive, and decision-making data archiving closed loop by systematically organizing and recording the diagnostic results, processing status, and manual operations at monitoring points. It not only generates structured diagnostic results and status trend sequences but also retains log information for all diagnostic responses and manual interventions, and supports subsequent operation and maintenance management and analysis decisions through standardized report output. Compared to traditional result output methods, this mechanism improves information transparency and system interpretability, providing operators with comprehensive and clear decision-making basis and data assurance for fault review and optimization.

[0064] like Figure 3The figure shows a multi-index fusion comparison chart provided in this application embodiment. The chart uses bars to display the specific numerical differences of measuring points 1 to 5 under three indicators: comprehensive temperature difference value, temperature trend comparison value, and control adjustment value. Among them, the comprehensive temperature difference value of measuring point 2 reaches approximately 2.8, the highest among all measuring points, indicating that its current temperature state deviates significantly from the steam temperature, load temperature, and historical fluctuation levels; its corresponding temperature trend comparison value is 1.0, and its control adjustment value is 0.85, both at relatively high levels, indicating that the system has significantly intervened in its adjustment. In contrast, the comprehensive temperature difference value of measuring point 4 is approximately 0.9, the trend comparison value is approximately 0.6, and the control adjustment value is only 0.3, the lowest overall, reflecting its stable operating state and the absence of strong adjustment. This chart intuitively reveals the comprehensive performance of different measuring points in terms of thermal anomaly degree, trend consistency, and adjustment response intensity, providing a supporting basis for fault identification and control strategy optimization.

[0065] like Figure 4 The figure shows the dynamic change curve of the control weight adjustment value provided in the embodiment of this application. The figure shows the trend of the adjustment value of measuring point 1 and measuring point 2 over 10 consecutive time periods. The adjustment value of measuring point 1 rises from 0.20 to a maximum of 0.85 and then falls back slightly, with large overall fluctuations, indicating that the system responded quickly to its abnormal state; the adjustment value of measuring point 2 rises slowly from 0.30 to 0.71, with stable changes, indicating that its operating state is relatively stable and the adjustment response is relatively mild. This figure reflects the real-time adjustment process of the control weight of different measuring points by the system under the dynamic scoring mechanism.

[0066] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0067] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A self-diagnostic system for thermal control instruments based on multi-source data fusion, including: The module comprises a data collection and organization module, a feature construction and extraction module, a state intelligent assessment module, a diagnostic linkage control module, and a result display and archiving module, characterized by: The data collection and processing module is used to collect thermal control status data in real time and preprocess the thermal control status data. The feature construction and extraction module is used to determine thermal deviation anomalies in the preprocessed thermal control status data and extract key feature variables for status deviation identification. The specific steps for extracting key feature variables for state deviation identification are as follows: Real-time comparison of the comprehensive value of temperature difference with the deviation assessment threshold, wherein the deviation assessment threshold includes a primary deviation threshold and a secondary deviation threshold; When the comprehensive temperature difference is greater than or equal to the first-level deviation threshold, the measuring point is removed from the regulation loop and alarm logic, the adjacent point substitution sampling mechanism is activated and interpolation estimation compensation is performed, a maintenance prompt is triggered and a fault report is generated and written to the abnormal log, and the measuring point is marked as failed in the status panel. When the comprehensive temperature difference is greater than the secondary deviation threshold but less than the primary deviation threshold, the enhanced sampling mechanism is activated to shorten the sampling cycle and improve the data resolution. Temperature sequence smoothing is enabled to reduce occasional jump interference. The control weight of the measuring point is reduced to 50% to still participate in but not dominate the control. An early warning sign is set in the system and a trend tracking buffer period is entered. When the comprehensive temperature difference is less than or equal to the secondary deviation threshold, the default sampling period is maintained, no encrypted sampling or additional processing is required, the measurement point data participates in the full control and alarm logic, and the measurement point status is marked as normal and reliable. The intelligent status assessment module is used to compare the temperature trend of the thermal control status data after the thermal deviation anomaly is determined, and to analyze the matching degree between the change trend of the measuring point and the adjacent points. The specific steps for comparing the temperature trend of the thermal control status data after determining the thermal deviation anomaly are as follows: Obtain the current temperature at the measuring point and the number of adjacent points; The current temperature change is calculated by comparing the current temperature value with the previous time point. The temperature change at neighboring points is calculated by comparing the current temperature value with the previous value. The absolute value of the difference between the current temperature change and the temperature change at neighboring points is calculated as the numerator. The larger of the absolute values ​​of the current temperature change and the temperature change at neighboring points is calculated, and a minimum value correction term is added as the denominator. The numerator of each group is divided by the denominator to obtain the normalized difference. The above calculations are performed on all neighboring point temperature changes, and the sum is divided by the number of neighboring points to obtain the final temperature trend comparison value. The diagnostic linkage control module is used to integrate the results of thermal deviation anomaly judgment and temperature trend comparison to execute control participation and adjustment, and adjust the participation weight and response strategy of the measuring point in the control loop. The specific steps for implementing control and adjustment based on the results of the fusion thermal anomaly determination and temperature trend comparison are as follows: Obtain the comprehensive value of temperature deviation and the comparison value of temperature trend; A real-time scoring sequence is obtained by continuously calculating the comprehensive temperature difference, temperature trend comparison, and control weight adjustment value of each measuring point during operation, and recording their dynamic changes in chronological order. A control adjustment threshold is obtained by fitting historical operating data of normal measuring points in the real-time scoring sequence. The sum of the comprehensive temperature deviation value and the temperature trend comparison value, minus the control adjustment threshold, is used as the numerator. The control smoothing coefficient is used as the denominator, and the result of dividing the two is used as the input value of the exponential function. The exponential value is calculated with the natural constant as the base, yielding the output of the exponential function. One is added to the output of the exponential function, and one is used as the numerator to obtain a fractional result, ultimately yielding the control weight adjustment value. The results display and archiving module is used to organize and record the diagnostic results and processing status of measurement points, support operators in confirming the status and performing manual operations, and generate complete log information and structured reports.

2. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for real-time acquisition of thermal control status data are as follows: Thermal control status data includes: sliding window, current measuring point temperature, main steam temperature, load fitting temperature, and number of adjacent points; A sliding window is obtained by setting a local time series interval with the current time as the endpoint, a fixed time length back, and the corresponding number of data points; the current measuring point temperature is obtained by collecting data from thermocouple temperature sensors; the main steam temperature is obtained by collecting data from the main steam pipeline thermal control measuring point; the load fitting temperature is obtained by fitting a historical function between the boiler load and the current measuring point temperature; and the number of neighboring points is obtained by configuring the measuring point layout structure.

3. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for preprocessing the thermal control status data are as follows: The preprocessing steps include: interpolating data of different frequencies according to a set time step to correct for the main steam temperature and the current measuring point temperature, as well as variables with response lag; performing sliding interpolation on short-term missing data and retaining missing markers for long-term missing data; marking out-of-limit values ​​and abrupt change points as abnormal, not directly removing them, and providing them for subsequent module judgment; standardizing all continuous variables to unify their numerical scale and compressing their numerical range through normalization.

4. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for determining thermal deviation anomalies in the preprocessed thermal control status data are as follows: Obtain the sliding window, current measuring point temperature, main steam temperature, and load fitting temperature; The temperature moving average is obtained by calculating the temperature sequence of the current measuring point within a sliding window. The temperature fluctuation intensity is calculated by setting the standard deviation function within the sliding window; the main steam temperature is subtracted from the current measuring point temperature, the result is divided by the sum of the main steam temperature and the minimum correction term, and the quotient is squared to obtain the relative deviation term of the main steam temperature; the load-fitted temperature is subtracted from the current measuring point temperature, the result is divided by the sum of the current measuring point temperature and the minimum correction term, and the quotient is squared to obtain the relative deviation term of the load-fitted temperature; the temperature fluctuation intensity is divided by the sum of the temperature sliding mean and the minimum correction term to obtain the relative deviation term of the temperature fluctuation; finally, the above three relative deviation terms are added together to obtain the comprehensive temperature difference value.

5. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for analyzing the trend of changes in measurement points and the degree of matching with neighboring points are as follows: Real-time comparison of temperature trend comparison value with trend deviation threshold, wherein the trend deviation threshold includes a first-level deviation threshold and a second-level deviation threshold; When the temperature trend comparison value is greater than or equal to the first-level deviation threshold: the measuring point is automatically isolated and does not participate in the real-time control calculation. The spatial interpolation logic is activated to dynamically compensate this point with the trend of neighboring points. The trend abnormality signal is sent to the alarm system. It is recommended to manually check the wiring and sensors. When the temperature trend comparison value is greater than the secondary deviation threshold and less than the primary deviation threshold: the neighboring point trend monitoring program is activated, three or more adjacent measuring points are dynamically compared, the control weight is automatically reduced to 50%, the secondary adjustment role is retained, a trend fluctuation suspicious label is added to the system panel for the on-duty personnel to follow up, and the measuring point data is retained for subsequent analysis, but its participation scope is limited. When the temperature trend comparison value is less than or equal to the secondary deviation threshold: maintain the status mark of the current measuring point and the neighboring points being consistent with the trend, maintain the default trust level and control weight of the current measuring point, and do not perform additional comparison and intervention operations.

6. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for adjusting the participation weight and response strategy of the measuring points in the control loop are as follows: The control weight adjustment value is compared with the weight evaluation interval in real time. The weight evaluation interval includes a high confidence interval, a relatively high confidence interval, a low confidence interval, and an extremely low confidence interval. When the control weight adjustment value is greater than or equal to 0.9, the current measurement point is determined to be highly reliable, and the measurement point participates in all control logic and alarm judgment. The controller uses its original adjustment value without the need for enhanced sampling and replacement processing. When the control weight adjustment value is greater than or equal to 0.6 and less than 0.9, the current measurement point is judged to be highly reliable. The control system retains its participation right but appropriately scales the adjustment result, activates the trend tracking mechanism to closely observe the changes in the measurement point, and slightly increases the trigger threshold of the alarm system. When the control weight adjustment value is greater than or equal to 0.3 and less than 0.6, the current measurement point is judged as low confidence. The controller significantly reduces the control weight of the measurement point and starts the neighbor point interpolation replacement mechanism to use the surrounding normal measurement points to provide auxiliary reference. At the same time, the measurement point enters the system observation state, and the alarm response strategy requires joint confirmation with other measurement points. When the control weight adjustment value is less than 0.3, the current measurement point is judged as having extremely low confidence, and its participation in all control and alarm logic is eliminated. The control output is completely blocked using an alternative strategy, and the point is set to a red warning state in the interface.

7. The self-diagnostic system for thermal control instruments based on multi-source data fusion according to claim 1, characterized in that: The specific steps for organizing and recording the diagnostic results and processing status of the measurement points, supporting operators in confirming the status and performing manual operations, and generating complete log information and structured reports are as follows: Organize the diagnostic labels, scores, and processing status of each monitoring point to generate structured diagnostic result data; record the diagnostic results and processing status of monitoring points at different times to construct a status trend sequence for operational traceability and comparative analysis; provide a manual operation interface to support operators in confirming the status of monitoring points, modifying diagnostic labels, and adding remarks; automatically record the diagnostic process, alarm trigger responses, and manual operation behaviors to generate traceable log records; integrate and output the operational status of monitoring points, diagnostic results, and maintenance suggestions to generate structured documents for archiving management and operation and maintenance decision-making.

Images