Bus duct anti-high temperature, anti-damp and anti-shake detection system

Abstract

The invention relates to a bus duct anti-high temperature, anti-damp and anti-shake detection system, and belongs to the technical field of monitoring and protection of power distribution equipment of a power system. The system comprises a multi-parameter collaborative early warning module, a dynamic threshold value self-adaption module, an abnormal linkage response and tracing module and a shake recognition and temperature and humidity coupling protection module. Obtaining a temperature parameter, a humidity parameter and a shake related parameter of the bus duct, and generating abnormal early warning information; adjusting a temperature safety threshold value, a humidity safety threshold value and a shaking safety judgment condition, and outputting an abnormal judgment conclusion; recording abnormal whole-process key information, and constructing an abnormal information database; extracting the shaking characteristics of the bus duct, constructing a shaking characteristic model, and outputting an abnormal shaking judgment result; integrating the abnormal shaking judgment result and the abnormal judgment conclusion, and outputting a comprehensive detection report; the system realizes multi-parameter cooperative monitoring, dynamic adaptation and protection, and improves the operation safety and operation and maintenance efficiency of the bus duct.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system distribution equipment monitoring and protection, and particularly relates to a bus duct high-temperature, moisture and swing prevention detection system. BACKGROUND

[0002] As the core distribution equipment of power system power transmission and distribution, the bus duct is widely used in large power consumption scenarios such as high-rise buildings, data centers and industrial plants due to its strong current-carrying capacity, flexible wiring and excellent safety performance, and is a key component to ensure the stable operation of the power system. With the continuous growth of modern power load and the complex application environment, the safety hazards faced by the bus duct in long-term operation are increasingly prominent. High temperature easily leads to insulation layer aging and metal component deformation; environmental moisture reduces insulation performance and causes the risk of electric leakage, which is more significant in underground machine rooms and coastal high-humidity areas; mechanical swing exacerbates joint loosening and fixed structure failure, further inducing temperature and humidity abnormalities and other chain risks. Therefore, real-time monitoring and accurate protection of the temperature, humidity and swing state of the bus duct are key requirements to ensure the continuous and reliable operation of the power system.

[0003] The existing bus duct safety monitoring scheme has many technical defects. Most of the schemes only monitor a single parameter independently and do not consider the coupling and correlation effects between temperature, humidity and swing, which easily leads to risk misjudgment or omission and is difficult to fully cover the safety hazards in complex operating scenarios. The traditional monitoring scheme uses fixed safety thresholds or judgment conditions, which cannot be dynamically adjusted according to the historical operation law of the bus duct and real-time environmental changes. When the application scenario or operating state changes, frequent false alarms or missed alarms are easily caused, reducing the practicability of the system. The existing swing-related monitoring mechanism is simple and only uses a single index for judgment, without considering the coupling effects of swing duration, fluctuation accumulation and temperature and humidity parameters, making it difficult to distinguish between normal slight vibration and harmful swing, and the judgment accuracy is insufficient. At the same time, the existing scheme can only realize simple abnormal warning and lacks a hierarchical response mechanism for abnormal risks, and a complete abnormal information tracing system is not established, which cannot record the abnormal evolution process, response action details and rectification effect, and is not conducive to subsequent fault troubleshooting and system optimization. SUMMARY

[0004] To solve the above problems in the prior art, the application provides a bus duct high-temperature, moisture and swing prevention detection system. The purpose of the application can be achieved by the following technical scheme: a bus duct high-temperature, moisture and swing prevention detection system, comprising: a multi-parameter cooperative warning module, a dynamic threshold self-adaptive module, an abnormal linkage response and tracing module, and a swing identification and temperature and humidity coupling protection module. The multi-parameter collaborative early warning module acquires the temperature parameters, humidity parameters, and sway-related parameters of the bus trunking, as well as the load current parameters and ambient temperature parameters of the bus trunking. These parameters are compared with preset temperature safety thresholds, humidity safety thresholds, and sway safety judgment conditions. Based on the load current parameters and ambient temperature parameters of the bus trunking, the comparison results are coupled and analyzed to generate abnormal early warning information. The dynamic threshold adaptive module acquires the historical operating parameters, real-time environmental correlation parameters, and real-time status parameters of the bus trunking, constructs a historical data model, quantifies the influence of load current and ambient temperature on the bus trunking temperature, analyzes the trend of historical operating parameters and the influence weight of real-time environmental correlation parameters, dynamically adjusts the temperature safety threshold, the humidity safety threshold, and the sway safety judgment condition, verifies them with the real-time status parameters, and outputs anomaly judgment conclusions. The abnormal linkage response and tracing module, based on the abnormal early warning information and the abnormal judgment conclusion, monitors the temperature change trend, humidity accumulation status and sway fluctuation status of the bus trunking in real time, executes the preset protection and collaborative response logic, and synchronously records key information of the entire abnormal process to build an abnormal information database. The sway recognition and temperature and humidity coupled protection module extracts busbar sway features based on the sway-related parameters, constructs a sway feature model, compares the real-time extracted sway features with a preset safety sway feature baseline for similarity, and outputs an abnormal sway judgment result; it integrates the abnormal sway judgment result with the abnormal judgment conclusion and outputs a comprehensive detection report.

[0005] Specifically, the sway safety determination conditions include: adopting a dual-dimensional collaborative determination rule of duration and fluctuation amplitude, taking the allowable range of sway fluctuation as the benchmark, and making a comprehensive determination by combining the duration of sway range and the cumulative effect of fluctuation amplitude.

[0006] Specifically, the process for generating the abnormal warning information is as follows: the temperature parameter, the humidity parameter, and the swaying-related parameter are respectively judged as abnormal, and then coupled calculations are performed in combination with preset parameter correlation weights. Finally, the abnormal parameter identifiers and coupling analysis conclusions are integrated to generate the abnormal warning information.

[0007] Specifically, the process of analyzing the trends of historical operating parameters and the influence weights of real-time environmental parameters is as follows: First, the historical operating parameters are decomposed to extract trend terms that reflect long-term change patterns. Evolution characteristics of historical operating parameters are generated through trend fitting modeling. The initial influence weights of the real-time environmental parameters are calculated using a correlation degree grading method. The initial influence weights are dynamically calibrated by combining the influence ratio of environmental parameters in historical anomaly cases. Finally, influence weights of real-time environmental parameters that are adapted to the current operating scenario are generated.

[0008] Specifically, the process of dynamically adjusting the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is as follows: based on the evolution characteristics of historical operating parameters, the initial adjustment direction of the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is calculated. Then, combined with the influence weight of the calibrated real-time environmental correlation parameters, a multi-objective optimization mechanism is used to allocate the adjustment range of the thresholds. The coupling constraint relationship between temperature, humidity, and sway parameters is analyzed simultaneously, and verified through a constraint feedback mechanism. Finally, the adjustment result is output.

[0009] Specifically, the process of outputting the anomaly determination conclusion is as follows: First, the real-time state parameters are compared with the adjusted temperature safety threshold and humidity safety threshold respectively. At the same time, the real-time swaying state is verified for compliance based on the adjusted swaying safety determination conditions to generate a preliminary single-parameter determination result. Then, the synergistic effect between the single-parameter determination results is analyzed. Subsequently, a statistical confidence verification mechanism is used to verify the parameter data corresponding to potential anomalies. Finally, the anomaly determination conclusion is generated by integrating the data.

[0010] Specifically, the protection and collaborative response logic includes: first, using risk transmission path analysis technology, identifying the core risk points of busbar operation based on the abnormal early warning information and the abnormal judgment conclusion; then, using a response priority ranking mechanism, prioritizing temperature and humidity control, load adjustment, and fault location notification response actions according to the importance of the risk points, the rate of parameter change, and the range of coupling influence; and simultaneously recording the action triggering conditions, execution sequence, and real-time effect feedback data.

[0011] Specifically, the construction process of the abnormal information database is as follows: a hierarchical clustering storage architecture is adopted, and the recorded data is classified hierarchically according to the abnormality level and occurrence scenario. At the same time, an index system based on the core features of the abnormality is established, and the index dimensions include abnormal parameter type, risk transmission path and response action type. Through an incremental learning mechanism, newly generated abnormal records and rectification verification results are included in the database in real time. Simultaneously, a cosine similarity algorithm is used to compare the features of new records with historical cases and associate the processing solutions and effect data of similar abnormalities.

[0012] Specifically, the process of extracting busbar sway features is as follows: First, a time-domain-frequency-domain joint extraction mechanism is used to extract basic features from the sway-related parameters; then, a feature importance evaluation algorithm is used to calculate the contribution of the basic features to the busbar safety status determination, a contribution threshold is set to filter out the core feature set, normalization processing is performed, and finally, a core feature system is generated.

[0013] Specifically, the process of constructing the sway feature model is as follows: Based on the sway features of the busbar trunking, the model is labeled and filtered according to normal operation, single abnormality, and temperature and humidity coupled abnormality scenarios to generate a feature sample set; using the entropy weight method and the analytic hierarchy process, the weight of the core features in the judgment of abnormal sway is quantified; a model framework is built based on the support vector machine; the sample set is divided into a training set and a validation set according to the proportion; the decision boundary is optimized through training iteration; and the recognition accuracy is tested using the validation set.

[0014] Specifically, the process of generating the abnormal sway judgment result is as follows: input the busbar sway characteristics into the sway characteristic model, use the cosine similarity algorithm to calculate the similarity between the real-time characteristics and the preset safety baseline, combine the dynamically adjusted threshold to screen potential abnormal samples, and weight and fuse the samples according to the core feature weights. Through a continuous duration verification mechanism, the potential abnormal state is monitored for multiple cycles. At the same time, the judgment criteria are adjusted according to the temperature and humidity status, and finally the abnormal sway judgment result is generated.

[0015] Specifically, the comprehensive test report generation process is as follows: First, the anomaly judgment conclusion and the abnormal shaking judgment result are integrated using a data fusion mechanism to extract key information and establish a multi-dimensional information correlation map; then, the map is analyzed through a preset coupling effect evaluation model to quantify the comprehensive risk index; next, a cross-scenario cross-validation mechanism is initiated to calibrate the judgment deviation using an error correction algorithm; finally, a structured report generation framework is used to generate the comprehensive test report.

[0016] This invention, achieved through multi-module collaborative design and innovative technology, breaks through the core deficiencies of existing busbar trunking monitoring technologies and has the following beneficial effects: Enhancing the comprehensiveness and accuracy of multi-dimensional risk monitoring: This invention integrates multiple state parameters such as temperature, humidity, and sway to construct a multi-parameter coupling analysis mechanism. It fully considers the correlation effects and chain risks between various parameters, avoids misjudgment or omission caused by monitoring a single parameter, and can cover the core safety hazards in busbar operation, thereby improving the accuracy and completeness of anomaly identification.

[0017] Enhance the system's dynamic adaptability to complex scenarios: By using a dynamic threshold adaptive module, combining the influence weights of historical operating patterns and real-time environmental parameters, the safety judgment criteria are dynamically adjusted, breaking away from the limitations of traditional fixed thresholds. This enables the monitoring system to flexibly adapt to different application scenarios, operating states, and environmental changes, effectively reducing false alarms and missed alarms under extreme operating conditions or environmental fluctuations, and improving the system's practicality and reliability in complex scenarios.

[0018] Achieving abnormal sway identification and coupled protection: Innovatively adopting sway feature modeling and multi-dimensional judgment mechanism, extracting core sway features and combining temperature and humidity coupling effect for comprehensive analysis, it can accurately distinguish between normal slight vibration and harmful sway, solving the problems of the rudimentary and inaccurate judgment of existing sway monitoring mechanisms, providing a reliable basis for the prevention and control of sway-related risks, and further ensuring the structural stability and operational safety of busbar trunking.

[0019] Enhance the timeliness and effectiveness of anomaly response: Through an anomaly linkage response mechanism, collaborative protection actions are executed based on risk classification and sorting to ensure that appropriate protection strategies can be quickly triggered for different levels and types of anomalies, reducing the risk of fault escalation; at the same time, an anomaly information database is built to fully record the anomaly evolution, response execution and rectification effect, providing data support for subsequent fault investigation and system optimization, and improving operation and maintenance efficiency and fault traceability.

[0020] Ensuring the continuity and reliability of busbar operation: This invention, through a complete process design of precise monitoring, dynamic adaptation, rapid response, and traceability closed loop, forms a complete safety assurance system from risk perception, early warning, judgment to protection and traceability. It effectively reduces the probability of equipment failure caused by high temperature, moisture, and vibration, extends the service life of busbars, provides solid support for the stable operation of the power system, and reduces power outage losses caused by equipment failure.

[0021] Optimize the scientific and efficient nature of operation and maintenance management: The comprehensive detection report and anomaly information database can provide operation and maintenance personnel with comprehensive and accurate equipment operation status data and historical anomaly traceability information, helping them to quickly locate the root cause of the fault, formulate targeted rectification plans, reduce operation and maintenance workload and costs, and improve the scientific and timely nature of operation and maintenance decisions. Attached Figure Description

[0022] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0023] Fig. 1 This is a schematic diagram of the process of a busbar trunking anti-high temperature, anti-moisture, and anti-sway detection system according to the present invention; Fig. 2 This is a structural block diagram of a busbar trunking anti-high temperature, anti-moisture, and anti-sway detection system according to the present invention. Detailed Implementation

[0024] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0025] Please see Figs. 1-2A bus trunking anti-high temperature, anti-moisture, and anti-sway detection system includes: a multi-parameter collaborative early warning module, a dynamic threshold adaptive module, an abnormal linkage response and tracing module, and a sway recognition and temperature and humidity coupling protection module. The multi-parameter collaborative early warning module acquires the temperature parameters, humidity parameters, and sway-related parameters of the bus trunking, as well as the load current parameters and ambient temperature parameters of the bus trunking. These parameters are compared with preset temperature safety thresholds, humidity safety thresholds, and sway safety judgment conditions. Based on the load current parameters and ambient temperature parameters of the bus trunking, the comparison results are coupled and analyzed to generate abnormal early warning information. The dynamic threshold adaptive module acquires the historical operating parameters, real-time environmental correlation parameters, and real-time status parameters of the bus trunking, constructs a historical data model, quantifies the influence of load current and ambient temperature on the bus trunking temperature, analyzes the trend of historical operating parameters and the influence weight of real-time environmental correlation parameters, dynamically adjusts the temperature safety threshold, the humidity safety threshold, and the sway safety judgment condition, verifies them with the real-time status parameters, and outputs anomaly judgment conclusions. The abnormal linkage response and tracing module, based on the abnormal early warning information and the abnormal judgment conclusion, monitors the temperature change trend, humidity accumulation status and sway fluctuation status of the bus trunking in real time, executes the preset protection and collaborative response logic, and synchronously records key information of the entire abnormal process to build an abnormal information database. The sway recognition and temperature and humidity coupled protection module extracts busbar sway features based on the sway-related parameters, constructs a sway feature model, compares the real-time extracted sway features with a preset safety sway feature baseline for similarity, and outputs an abnormal sway judgment result; it integrates the abnormal sway judgment result with the abnormal judgment conclusion and outputs a comprehensive detection report.

[0026] Specifically, the sway safety determination conditions include: adopting a dual-dimensional collaborative determination rule of duration and fluctuation amplitude, taking the allowable range of sway fluctuation as the benchmark, and making a comprehensive determination by combining the duration of sway range and the cumulative effect of fluctuation amplitude.

[0027] Specifically, the process for generating the abnormal warning information is as follows: the temperature parameter, the humidity parameter, and the swaying-related parameter are respectively judged as abnormal, and then coupled calculations are performed in combination with preset parameter correlation weights. Finally, the abnormal parameter identifiers and coupling analysis conclusions are integrated to generate the abnormal warning information.

[0028] In this embodiment, a busbar trunking system used in an industrial application is taken as the monitoring object, and the specific working process is as follows: Parameter Acquisition: The multi-parameter collaborative early warning module collects three core parameters in real time through sensors deployed at key joints, housings, and fixed supports of the busbar trunking: temperature parameter T (unit: °C), humidity parameter H (unit: %RH), and sway-related parameters S (including real-time sway fluctuation amplitude S1 and instantaneous sway peak value S2, unit: m / s). 2 ).

[0029] Preset judgment criteria: The preset temperature safety thresholds are T0 (upper limit threshold) and T1 (lower limit threshold), meaning the normal range of temperature parameters is [T1, T0]. The preset humidity safety thresholds are H0 (upper limit threshold) and H1 (lower limit threshold), meaning the normal range of humidity parameters is [H1, H0]. Preset sway safety judgment conditions: The allowable range of sway fluctuation is [S 01 ,S 02 (Normal fluctuation range), duration threshold is t0, and cumulative fluctuation effect threshold is S. 01 ×t0 (i.e., the cumulative upper limit of fluctuation amplitude within a unit of time); Simultaneously, set the parameter correlation weight: temperature parameter weight W. t Humidity parameter weight W h Shaking parameter weight W s And satisfy W t +W h +W s =1.

[0030] Single parameter anomaly detection: Temperature parameter determination: If T > T0 or T < T1, then mark the temperature parameter as abnormal as A. t =1; if T∈[T1,T0], then A t =0; Humidity parameter determination: If H > H0 or H < H1, then mark the humidity parameter as abnormal as A. h =1; if H∈[H1,H0], then A h =0; Shaking parameter determination: Calculate whether the real-time shaking fluctuation amplitude S1 exceeds [S 01 ,S 02 Simultaneously record the duration t1 of this fluctuation state; if S1∉[S 01 ,S 02 ], and t1>t0, or S1×t1>S 01 If ×t0 (cumulative effect exceeds the limit), then the abnormal sway parameter is marked as A. s =1; otherwise, A s =0.

[0031] Coupling Calculation: Based on preset parameter correlation weights, the abnormal coupling coefficient C is calculated using the formula: C = A t ×W t +A h ×W h +A s ×W s Simultaneously analyze the correlation effects between parameters, for example, when A t =1 and A s When A = 1, the risk of joint loosening caused by vibration may be exacerbated by abnormal temperature, so "temperature vibration coupling risk" needs to be marked in the coupling analysis; when A h =1 and A s When =1, mark "wet sway coupling risk".

[0032] Anomaly warning information generation: Integrating anomaly parameter identifiers (A t A h A s The specific values ​​of the coupling coefficient C and the coupling analysis conclusions are used to generate anomaly warning information. An example warning message is as follows: "Bus trunking operating status abnormal: Abnormal parameter identifier (T: A)" t =1, H:A h =0, S:A s =1), abnormal coupling coefficient C=W t +W s Coupling analysis conclusion: Temperature swaying coupling risk (temperature exceeds the safety threshold T0, swaying fluctuation amplitude S1 > S) 02 Furthermore, the duration t1 > t0, indicating that the cumulative effect exceeds the standard. Please focus on checking the heating at the joint and the stability of the fixing structure.

[0033] Specifically, the process of analyzing the trends of historical operating parameters and the influence weights of real-time environmental parameters is as follows: First, the historical operating parameters are decomposed to extract trend terms that reflect long-term change patterns. Evolution characteristics of historical operating parameters are generated through trend fitting modeling. The initial influence weights of the real-time environmental parameters are calculated using a correlation degree grading method. The initial influence weights are dynamically calibrated by combining the influence ratio of environmental parameters in historical anomaly cases. Finally, influence weights of real-time environmental parameters that are adapted to the current operating scenario are generated.

[0034] Specifically, the process of dynamically adjusting the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is as follows: based on the evolution characteristics of historical operating parameters, the initial adjustment direction of the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is calculated. Then, combined with the influence weight of the calibrated real-time environmental correlation parameters, a multi-objective optimization mechanism is used to allocate the adjustment range of the thresholds. The coupling constraint relationship between temperature, humidity, and sway parameters is analyzed simultaneously, and verified through a constraint feedback mechanism. Finally, the adjustment result is output.

[0035] Specifically, the process of outputting the anomaly determination conclusion is as follows: First, the real-time state parameters are compared with the adjusted temperature safety threshold and humidity safety threshold respectively. At the same time, the real-time swaying state is verified for compliance based on the adjusted swaying safety determination conditions to generate a preliminary single-parameter determination result. Then, the synergistic effect between the single-parameter determination results is analyzed. Subsequently, a statistical confidence verification mechanism is used to verify the parameter data corresponding to potential anomalies. Finally, the anomaly determination conclusion is generated by integrating the data.

[0036] In this embodiment, continuing the scenario of the previous embodiment, a busbar operating in an industrial setting is used as the monitoring object. The specific workflow is as follows: The module synchronously collects three types of core data, which are synchronized with the parameters of the multi-parameter collaborative early warning module to ensure data consistency. Historical operating parameters: Extract the historical temperature dataset T_hist, historical humidity dataset H_hist, and historical sway parameter set S_hist (including sway fluctuation amplitude and duration data in each cycle) for the bus trunking of this industrial scenario for nearly N operating cycles. Real-time environmental parameters: Collect four types of parameters in real-time within the industrial plant: ambient temperature E_T, ambient humidity E_H, vibration intensity of surrounding production equipment E_S, and production load fluctuation coefficient E_L (adapted to the characteristics of dense equipment and significant load fluctuation in industrial scenarios). Real-time status parameters: Obtain the current busbar's real-time temperature T, humidity H, and sway-related parameters S (fluctuation amplitude S1, duration t1), and keep them updated in sync with the real-time data collected by the sensors.

[0037] Based on the historical operating parameters I_hist (historical load current), E_T_hist (historical ambient temperature), and T_hist (historical busbar temperature), a historical data model of "load current - ambient temperature - busbar temperature" is constructed.

[0038] The influence is quantified using a multiple linear regression algorithm. The model expression is as follows: T_pred=k1×I+k2×E_T+k0 in: T_pred is the predicted temperature of the busbar trunking; I is the load current, and E_T is the ambient temperature; k1 is the influence coefficient of load current on temperature (quantifying the temperature fluctuation caused by current change; in industrial scenarios, k1 > 0, meaning that the temperature rises as the current increases). k2 is the influence coefficient of ambient temperature on temperature (quantifying the transmission weight of ambient temperature to busbar temperature, k2∈(0,1)). k0 is the baseline correction coefficient (a fixed error compensation that adapts to the busbar material and installation method in this industrial scenario). By fitting historical data using the least squares method, the specific values ​​of k1, k2, and k0 are obtained.

[0039] Analysis of the influence weights of historical operating parameter trends and environmental parameters: The historical temperature dataset T_hist was decomposed using a trend decomposition method to separate the trend term T_trend, the periodic fluctuation term T_cycle, and the random disturbance term T_rand, which reflect the long-term variation pattern. The core trend term T_trend was extracted and linear fitting model was performed to obtain the evolution feature model T_trend=a×t+b (t is the running time, and a and b are the fitting coefficients). The model shows that the temperature of the busbar in this industrial scenario slowly increases with the running time due to long-term high-load operation. Trend decomposition and fitting were performed simultaneously on the historical humidity dataset H_hist and the historical sloshing parameter set S_hist to obtain the humidity evolution feature H_trend=c×t+d and the sloshing evolution feature S_trend=e×t+f. The results show that the humidity has no obvious long-term fluctuations, and the sloshing parameter trend is stable with no significant drift.

[0040] The correlation degree classification method is adopted. Based on the degree of correlation between each environmental parameter and the bus trunking operation status in the industrial scenario, the initial influence weights are assigned as follows: ambient temperature weight W1, ambient humidity weight W2, surrounding equipment vibration intensity weight W3, and production load fluctuation coefficient weight W4, and W1+W2+W3+W4=1. Retrieve the historical anomaly case database of the bus trunking in this industrial scenario and count the percentage of influence of various environmental parameters in past anomaly events: the percentage of influence of ambient temperature K1, the percentage of influence of ambient humidity K2, the percentage of influence of vibration of surrounding equipment K3, and the percentage of influence of production load fluctuation K4 (in industrial scenarios, the percentage of influence of production load fluctuation and ambient temperature is usually higher). The initial weights are dynamically calibrated based on the historical influence ratio. The calibration formula is: W1'=W1×K1 / (W1×K1+W2×K2+W3×K3+W4×K4). Similarly, the final calibrated weights W1', W2', W3', and W4' are calculated to generate a weight system adapted to the current industrial scenario.

[0041] Dynamically adjust safety thresholds and shaking safety judgment conditions: Based on the evolution characteristics of historical operating parameters, the temperature shows a slow upward trend, and the initial adjustment direction is to raise the upper limit of the temperature threshold; the humidity does not show significant long-term fluctuations, and the initial adjustment direction is to make small bidirectional fine adjustments; the oscillation parameter shows a stable trend, and the initial adjustment direction is to keep the basic judgment range unchanged.

[0042] Based on the calibrated weighting system (assuming that W1' (ambient temperature) and W4' (production load fluctuation) have the highest weighting in the industrial scenario), a multi-objective optimization mechanism is used to allocate the adjustment range: The original temperature safety threshold [T1,T0] is adjusted to [T1,T0+ΔT] (ΔT is the adjustment amount calculated jointly based on W1' and W4'). The original humidity safety threshold [H1,H0] is adjusted to [H1-ΔH1,H0+ΔH2] (ΔH1 and ΔH2 are small fine-tuning amounts calculated based on W2'). The permissible fluctuation range in the original sway safety judgment conditions [S] 01 ,S 02 [S] 01 -ΔS,S 02 +ΔS] (ΔS is the adjustment amount calculated based on W3'), and the duration threshold t0 remains unchanged.

[0043] The coupling constraint relationship between temperature, humidity, and sway parameters is analyzed simultaneously. For example, it verifies whether "raising the upper limit of temperature will lead to an additional insulation risk with a high humidity environment" and whether "adjusting the sway amplitude will exacerbate the risk of joint loosening caused by abnormal temperature." Through a constraint feedback mechanism, it verifies that the adjusted threshold combination meets the safe operation requirements of industrial scenarios. Finally, the adjusted standards are output: temperature threshold [T1,T0'], humidity threshold [H1',H0'], and sway safety judgment condition [S]. 01 ',S 02 ']+t0.

[0044] Anomaly determination conclusion output: Compare the real-time temperature T with the adjusted temperature threshold [T1,T0']. If T∉[T1,T0'], mark A_t'=1 (abnormal); otherwise, A_t'=0. The real-time humidity H is compared with [H1',H0'], and A_h' is marked as 1 (abnormal) or 0; Verify S1 and t1 based on the adjusted sway safety judgment conditions. If S1∉[S 01 ',S 02 If t1 > t0, then mark A_s' = 1 (abnormal); otherwise, mark A_s' = 0.

[0045] If A_t'=1 and A_s'=1, the analysis concludes that "the synergy between abnormal temperature and abnormal vibration in industrial scenarios may exacerbate the risk of loosening of busbar joints (this risk is transmitted faster under high load operation)"; if A_h'=1 and A_t'=1, it is marked that "temperature and humidity coupling leads to a decrease in insulation performance, and the risk amplification effect is adapted to the high dust environment in industry".

[0046] A statistical confidence verification mechanism was adopted, and a confidence interval [α,β] was set to verify the real-time temperature data T and the sway data S1 corresponding to A_t'=1 and A_s'=1. The data were confirmed to fall within the confidence interval, thus eliminating misjudgments caused by random disturbances caused by equipment start-up and shutdown in industrial scenarios.

[0047] Integrating the single-parameter judgment results (A_t'=1, A_h'=0, A_s'=1), the synergistic impact analysis conclusions, and the confidence verification results, an anomaly judgment conclusion is generated: "The real-time temperature T of the busbar in the current industrial scenario is greater than the adjusted upper limit threshold T0' (A_t'=1), and the fluctuation amplitude S1 is greater than the adjusted upper limit S." 02 "And the duration t1 > t0 (A_s' = 1), the confidence level meets the requirements, there is a risk of temperature fluctuation anomaly under high industrial load conditions, it is judged as a level 2 anomaly, and it is recommended to start load adjustment and joint inspection response actions."

[0048] Specifically, the protection and collaborative response logic includes: first, using risk transmission path analysis technology, identifying the core risk points of busbar operation based on the abnormal early warning information and the abnormal judgment conclusion; then, using a response priority ranking mechanism, prioritizing temperature and humidity control, load adjustment, and fault location notification response actions according to the importance of the risk points, the rate of parameter change, and the range of coupling influence; and simultaneously recording the action triggering conditions, execution sequence, and real-time effect feedback data.

[0049] Specifically, the construction process of the abnormal information database is as follows: a hierarchical clustering storage architecture is adopted, and the recorded data is classified hierarchically according to the abnormality level and occurrence scenario. At the same time, an index system based on the core features of the abnormality is established, and the index dimensions include abnormal parameter type, risk transmission path and response action type. Through an incremental learning mechanism, newly generated abnormal records and rectification verification results are included in the database in real time. Simultaneously, a cosine similarity algorithm is used to compare the features of new records with historical cases and associate the processing solutions and effect data of similar abnormalities.

[0050] In this embodiment, consistent with the scenarios of the multi-parameter collaborative early warning module and the dynamic threshold adaptive module, the busbar operating in a certain industrial scenario is used as the monitoring object, focusing on the "temperature fluctuation collaborative level 2 anomaly" scenario. The specific workflow of the module is as follows: Input information reception: The module synchronously receives two types of core input data to ensure consistency with the data from the preceding module: The abnormal warning information output by the multi-parameter collaborative early warning module: "Abnormal parameter identifier (T:A)" t =1, H:A h =0, S:A s =1), abnormal coupling coefficient C=W t +W sCoupling analysis conclusion: "Temperature sway coupling risk under high-load industrial conditions"; The anomaly detection conclusion output by the dynamic threshold adaptive module is: "The real-time temperature T of the busbar in the current industrial scenario is greater than the adjusted upper limit threshold T0' (A_t'=1), and the fluctuation amplitude S1 is greater than the adjusted upper limit S." 02 "And the duration t1 > t0 (A_s' = 1), the confidence level meets the requirements, there is a risk of temperature sway synergy anomaly under high industrial load conditions, and it is judged as a level 2 anomaly."

[0051] Execution of protection and collaborative response logic: Using risk transmission path analysis technology, and considering the characteristics of high load and dense equipment in industrial scenarios, the risk transmission chain is broken down as follows: "abnormal swaying (S1 exceeds the standard) → loose busbar trunking joints → increased contact resistance → local temperature rise (T exceeds the standard) → accelerated aging of insulation layer under high load → short circuit fault". The core risk point is identified as "the risk of temperature swaying coupling caused by loose joints", and the secondary risk point is "the risk of power outage in the production line caused by the expansion of the fault under high load conditions".

[0052] A response priority ranking mechanism is adopted, taking into account the production continuity requirements of industrial scenarios, and prioritizing response actions according to the importance of risk points, the rate of parameter change, and the scope of coupled impact: Level 1 Priority (P1): Load adjustment action (directly suppresses the temperature rise trend, blocks the risk transmission, and ensures the power supply to the production base). Level 2 Priority (P2): Fault location notification action (rapidly linking with operations and maintenance to shorten fault handling time and reduce production impact); Level 3 Priority (P3): Temperature and humidity control actions (to help improve the operating environment, mitigate secondary risks, and adapt to the ventilation conditions of industrial plants).

[0053] Action triggering conditions: The triggering benchmark is "the anomaly judgment conclusion is level two anomaly" and "the temperature change rate V_T in the industrial scene is greater than V0 (preset high load condition exclusive rate threshold)". Execution sequence: Trigger P1 action: Send a load adjustment command to the industrial power distribution control system associated with the bus trunking, reduce the current production load ratio from the rated value to the preset safe ratio (adapting to the load adjustment characteristics of the industrial production line), and record the trigger time T1, adjustment range ΔL, power distribution system command number, and executing entity (power distribution control cabinet). After an interval of Δt1 (a short delay in industrial scenarios to avoid load fluctuations affecting other equipment), the P2 action is triggered: a fault location notification is pushed to the on-duty personnel through the industrial operation and maintenance management platform, specifying the abnormal busbar section number (e.g., busbar No. 3 in workshop A area), the core risk point (loose joint), real-time parameter data (specific values ​​of T and S1), and high load condition risk warnings, and recording the notification recipient, push time T2, and operation and maintenance feedback receipt status; Synchronous trigger of P3 action: Start local temperature and humidity control equipment in industrial plants (such as workshop-specific industrial air conditioners and dehumidifiers) to reduce the ambient temperature of the abnormal section by ΔE_T and maintain the humidity within the safe range of industrial scenarios [E_H1, E_H0]. Record the control equipment number, start time T3, and real-time control effect data (collect environmental parameters once every Δt2 to adapt to the temperature and humidity fluctuation characteristics of industrial environments). Effect feedback record: Continuously monitor parameter changes after response execution, record the time T4 for the temperature to drop from T to below T0', and the amplitude of the fluctuation S1. 01 ',S 02 The time T5, and the stable operating status data of the busbar after load adjustment (such as current and voltage fluctuation values).

[0054] Abnormal information log: Synchronously record key information throughout the entire process to form a complete record dataset adapted to industrial scenarios: Anomaly start time: T_start (the time when the temperature fluctuation parameter is first detected to exceed the standard and meet the confidence level requirement under high industrial load conditions, accurate to the second, to meet the accuracy requirements of industrial equipment fault tracing). State parameter evolution curve: With time as the horizontal axis, record the real-time changes of T (temperature), S1 (sway amplitude), H (humidity) and industrial production load value from T_start to abnormal termination, generate a continuous evolution curve, and mark the time nodes of key actions such as load adjustment and temperature and humidity control. Response action execution details: including the triggering conditions, execution sequence (T1-T3), instruction content, execution subject, adjustment parameters, industrial equipment number, etc. for each priority action; Abnormal termination condition: when T∈[T1,T0'], S1∈[S 01 ',S 02 If the duration reaches t_end (the preset stable duration for industrial scenarios, adapted to the continuous operation requirements of the production line), it is determined to be abnormally terminated, and the termination time T_end is recorded.

[0055] Construction of a structured anomaly information database: A hierarchical clustering storage architecture is adopted to classify the current anomaly record according to "anomaly level (level 2 anomaly) - occurrence scenario (industrial scenario - workshop A area - high load condition)" and store it in the corresponding level of storage unit. This forms a cluster group with historical anomaly records of the same level and industrial scenario, which facilitates subsequent traceability analysis according to production scenario.

[0056] A multi-dimensional index is built based on the core features of anomalies to adapt to the operation and maintenance needs of industrial scenarios: Abnormal parameter type index: Mark "temperature + sway" dual-parameter abnormality (high-frequency abnormality type in industrial scenarios); Risk transmission path index: Mark the dedicated path "shaking → loose joint → temperature rise (high load)"; Response action type index: Mark the combined action of "industrial load adjustment + fault location notification + industrial temperature and humidity control".

[0057] Incremental inclusion: Through the incremental learning mechanism, the complete record dataset of this anomaly and the subsequent rectification and verification results of the operation and maintenance personnel (such as joint fastening process parameters, rectification time, industrial scenario retest parameters, and production line recovery status) are written into the database in real time; Feature comparison and association: The cosine similarity algorithm is used to extract the core features of this anomaly (high-load industrial scenario, temperature and vibration dual-parameter anomaly, risk of loose joints, and load adjustment response), and perform feature similarity calculation with historical industrial scenario cases in the database. If there are historical cases with similarity higher than the threshold γ, the details of the industrial scenario handling solution of the case (such as the joint tightening standard adapted to high load) and long-term operation effect feedback data are automatically associated to provide a reference for the subsequent optimization of this anomaly and the handling of similar industrial scenario failures.

[0058] Specifically, the process of extracting busbar sway features is as follows: First, a time-domain-frequency-domain joint extraction mechanism is used to extract basic features from the sway-related parameters; then, a feature importance evaluation algorithm is used to calculate the contribution of the basic features to the busbar safety status determination, a contribution threshold is set to filter out the core feature set, normalization processing is performed, and finally, a core feature system is generated.

[0059] Specifically, the process of constructing the sway feature model is as follows: Based on the sway features of the busbar trunking, the model is labeled and filtered according to normal operation, single abnormality, and temperature and humidity coupled abnormality scenarios to generate a feature sample set; using the entropy weight method and the analytic hierarchy process, the weight of the core features in the judgment of abnormal sway is quantified; a model framework is built based on the support vector machine; the sample set is divided into a training set and a validation set according to the proportion; the decision boundary is optimized through training iteration; and the recognition accuracy is tested using the validation set.

[0060] Specifically, the process of generating the abnormal sway judgment result is as follows: input the busbar sway characteristics into the sway characteristic model, use the cosine similarity algorithm to calculate the similarity between the real-time characteristics and the preset safety baseline, combine the dynamically adjusted threshold to screen potential abnormal samples, and weight and fuse the samples according to the core feature weights. Through a continuous duration verification mechanism, the potential abnormal state is monitored for multiple cycles. At the same time, the judgment criteria are adjusted according to the temperature and humidity status, and finally the abnormal sway judgment result is generated.

[0061] Specifically, the comprehensive test report generation process is as follows: First, the anomaly judgment conclusion and the abnormal shaking judgment result are integrated using a data fusion mechanism to extract key information and establish a multi-dimensional information correlation map; then, the map is analyzed through a preset coupling effect evaluation model to quantify the comprehensive risk index; next, a cross-scenario cross-validation mechanism is initiated to calibrate the judgment deviation using an error correction algorithm; finally, a structured report generation framework is used to generate the comprehensive test report.

[0062] In this embodiment, consistent with the previous module scenario, the monitoring object is the No. 3 high-load busbar trunking in area A of an industrial workshop. The specific workflow of the module is as follows: Obtaining shaking-related parameters: The module collects sway-related parameters in real time, including sway acceleration, vibration frequency, displacement amplitude, and impact coefficient, through vibration sensors deployed on the busbar trunking fixed supports, key joints, and adjacent sections of industrial equipment (adapting to the characteristics of scenarios with dense industrial equipment and complex vibration sources), forming a raw data sequence to provide a basic data source for feature extraction.

[0063] Busbar sway feature extraction: Employing a joint time-domain and frequency-domain extraction mechanism: Temporal dimension: Extract basic features of sway parameters such as peak value, mean, variance, kurtosis, impulse factor, and waveform factor to meet the feature capture needs of impact vibration in industrial scenarios; Frequency domain dimension: The original data is converted to the frequency domain through fast Fourier transform, and basic features such as main frequency, spectral energy, harmonic component ratio, and frequency band power are extracted, with a focus on capturing specific frequency vibration signals generated by the linkage of industrial equipment; finally, an initial feature set containing M basic features is formed.

[0064] Core feature system generation: The contribution of each basic feature to the determination of the safety status of busbar trunking in industrial scenarios is calculated by the feature importance evaluation algorithm. A contribution threshold θ is set (to adapt to the determination accuracy requirements of high-load industrial conditions). P core features with a contribution greater than θ (such as main frequency ratio, acceleration peak, impact coefficient, and spectrum energy concentration coefficient) are selected. The core features are normalized (mapped to the [0,1] interval) and finally the core feature system F=[F1,F2,...,F_P) is generated.

[0065] Shaking feature model construction: Based on the historical sway characteristics data of the busbar trunking in this industrial scenario, the data was filtered according to three scenario categories: Normal operating scenario (industrial low load, equipment running smoothly); A single abnormal shaking scenario (excessive vibration caused by loose fixed structure, without abnormal temperature or humidity); Abnormal temperature and humidity coupling scenarios (high temperature-sway coupling under high load, high humidity environment-sway coupling); after removing noise samples generated by the start-up and shutdown of industrial equipment, a feature sample set containing Q valid samples is generated.

[0066] The objective weights of the core features are calculated using the entropy weight method (based on the distribution of sample data in industrial scenarios), and the subjective weights are determined by combining the analytic hierarchy process (referencing industrial operation and maintenance experience). The weighted fusion yields the comprehensive weights of each core feature, W=[W1,W2,...,W_P], among which the main frequency ratio (F1) and the peak acceleration (F2) have the highest weight ratios (adapting to the core causes of shaking faults in industrial scenarios), satisfying ΣW_i=1.

[0067] The model framework is built based on support vector machine (SVM), and the feature sample set is divided into training set and validation set in a 7:3 ratio. The model decision boundary is optimized iteratively through the training set, focusing on adapting to the complex feature distribution of "high load vibration superposition" in industrial scenarios. The model recognition accuracy is tested using the validation set until the accuracy is ≥96% (meeting the high precision requirements of industrial equipment safety monitoring), thus completing the construction of the sway feature model.

[0068] Abnormal shaking judgment result generation: The core feature system F generated in real time is input into the sway feature model in real time. The cosine similarity algorithm is used to calculate the similarity Sim between F and the preset industrial scenario safety sway feature baseline F0. Combined with the similarity threshold Sim0 adjusted by the dynamic threshold adaptive module (a dynamic standard adapted to high-load industrial conditions), if Sim < Sim0, it is marked as a potential abnormal sample.

[0069] The feature values ​​of potential abnormal samples are weighted and fused according to the comprehensive weight W of the core features to obtain the fused feature value F_merge=Σ(F_i×W_i); a continuous duration verification mechanism is started, taking into account the continuous needs of industrial production, K monitoring cycles are set (the duration of each cycle is adapted to the rhythm of production line operation), potential abnormal states are continuously monitored, and the F_merge value of each cycle and the corresponding real-time temperature, humidity and production load data are recorded.

[0070] Based on the criteria for adjusting temperature, humidity, and load conditions in industrial settings: If the real-time humidity H > H0' (adjusted industrial high humidity threshold), then tighten the judgment threshold (Sim0 is increased by ΔSim) to avoid the superposition effect of insulation risk and vibration in high humidity environment. If the real-time temperature T > T0' (adjusted industrial high-load temperature threshold), the monitoring period is extended to K+ΔK to adapt to the slower vibration decay under high load. When F_merge meets the anomaly judgment condition for ≥K1 periods within K periods, and Sim is continuously lower than the adjusted threshold, the final output of the abnormal sway judgment result is: "Abnormal sway state of bus trunking in industrial scenario, anomaly type is high temperature-sway coupling anomaly (adapted to high load conditions), core feature exceeding the standard: F1 (main frequency ratio), F2 (acceleration peak), similarity Sim=Sim_real<Sim0, associated production load: rated load ratio X%".

[0071] Comprehensive test report generation: By integrating the above abnormal sway judgment results with the "secondary abnormality of temperature and sway co-existence under industrial high load conditions" judgment conclusion output by the dynamic threshold adaptive module using the data fusion mechanism, key information such as abnormal parameter type, risk level, production load correlation status, and temperature and humidity coupling relationship is extracted to establish a multi-dimensional information correlation map (nodes include No. 3 busbar in workshop A area, high load conditions, temperature and sway dual parameter abnormalities, etc., and edges represent risk transmission relationships).

[0072] By analyzing the correlation graph through the pre-set industrial scenario coupling effect assessment model, the comprehensive risk index R=7.2 (∈[0,10]) was quantified and determined to be medium risk (adapted to the industrial scenario risk classification standard, 5≤R<8 is medium risk). The core risk was identified as "the joint loosening caused by temperature sway coupling under high load was aggravated".

[0073] A cross-scenario verification mechanism is initiated, which calls historical data of similar scenarios such as "abnormal temperature fluctuation under high load in industrial workshops" in the database for verification and uses an error correction algorithm to calibrate the judgment deviation. Finally, a structured report generation framework adapted for industrial operation and maintenance is adopted to integrate anomaly details, comprehensive risk index, industrial scenario coupling impact analysis, and rectification suggestions (such as "prioritize strengthening joints according to industrial fastening standards, simultaneously reduce the production load of this section to a safe proportion, and continuously monitor the temperature drop trend") to generate a comprehensive test report.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A busbar trunking anti-high temperature, anti-moisture, and anti-sway detection system, characterized in that, include: Multi-parameter collaborative early warning module, dynamic threshold adaptive module, abnormal linkage response and tracing module, and shaking recognition and temperature and humidity coupling protection module; The multi-parameter collaborative early warning module acquires the temperature parameters, humidity parameters, and sway-related parameters of the bus trunking, as well as the load current parameters and ambient temperature parameters of the bus trunking. These parameters are compared with preset temperature safety thresholds, humidity safety thresholds, and sway safety judgment conditions. Based on the load current parameters and ambient temperature parameters of the bus trunking, the comparison results are coupled and analyzed to generate abnormal early warning information. The dynamic threshold adaptive module acquires the historical operating parameters, real-time environmental correlation parameters, and real-time status parameters of the bus trunking, constructs a historical data model, quantifies the influence of load current and ambient temperature on the bus trunking temperature, analyzes the trend of historical operating parameters and the influence weight of real-time environmental correlation parameters, dynamically adjusts the temperature safety threshold, the humidity safety threshold, and the sway safety judgment condition, verifies them with the real-time status parameters, and outputs anomaly judgment conclusions. The abnormal linkage response and tracing module, based on the abnormal early warning information and the abnormal judgment conclusion, monitors the temperature change trend, humidity accumulation status and sway fluctuation status of the bus trunking in real time, executes the preset protection and collaborative response logic, and synchronously records key information of the entire abnormal process to build an abnormal information database. The sway recognition and temperature and humidity coupled protection module extracts busbar sway features based on the sway-related parameters, constructs a sway feature model, compares the real-time extracted sway features with a preset safety sway feature baseline for similarity, and outputs an abnormal sway judgment result; it integrates the abnormal sway judgment result with the abnormal judgment conclusion and outputs a comprehensive detection report.

2. The system according to claim 1, characterized in that, The specific conditions for determining the safety of swaying include: adopting a dual-dimensional collaborative determination rule of duration and amplitude of fluctuation, taking the allowable range of swaying fluctuation as the benchmark, and making a comprehensive determination by combining the duration of swaying range and the cumulative effect of amplitude of fluctuation.

3. The system according to claim 1, characterized in that, The specific process for generating the abnormal warning information is as follows: the temperature parameter, the humidity parameter, and the swaying-related parameter are respectively judged as abnormal, and then coupled calculation is performed in combination with the preset parameter correlation weight. Finally, the abnormal parameter identifier and the coupling analysis conclusion are integrated to generate the abnormal warning information.

4. The system according to claim 1, characterized in that, The specific process for analyzing the influence weights of historical operating parameters and real-time environmental parameters is as follows: First, the historical operating parameters are decomposed to extract trend terms that reflect long-term change patterns. Then, the evolution characteristics of the historical operating parameters are generated through trend fitting modeling. The initial influence weights of the real-time environmental correlation parameters are calculated using a correlation degree grading method. The initial influence weights are then dynamically calibrated by combining the influence ratio of environmental parameters in historical anomaly cases, and finally, real-time environmental correlation parameter influence weights adapted to the current operating scenario are generated.

5. The system according to claim 1, characterized in that, The specific process of dynamically adjusting the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is as follows: Based on the evolution characteristics of historical operating parameters, the initial adjustment direction of the temperature safety threshold, humidity safety threshold, and sway safety judgment condition is calculated. Then, combined with the influence weight of the calibrated real-time environmental correlation parameters, a multi-objective optimization mechanism is used to allocate the adjustment range of the thresholds. The coupling constraint relationship between temperature, humidity, and sway parameters is analyzed simultaneously, and the result is verified through a constraint feedback mechanism. Finally, the adjustment result is output.

6. The system according to claim 1, characterized in that, The specific process of the output anomaly judgment conclusion is as follows: First, the real-time status parameters are compared with the adjusted temperature safety threshold and humidity safety threshold respectively. At the same time, the real-time swaying status is verified for compliance based on the adjusted swaying safety judgment conditions to generate a preliminary judgment result for a single parameter. Then, the synergistic influence between the single parameter judgment results is analyzed. Subsequently, a statistical confidence verification mechanism was used to verify the parameter data corresponding to potential anomalies; finally, the anomaly determination conclusion was generated by integrating the data.

7. The system according to claim 1, characterized in that, The protection and collaborative response logic specifically includes: first, using risk transmission path analysis technology, identifying the core risk points of busbar operation based on the abnormal early warning information and the abnormal judgment conclusion; then, using a response priority ranking mechanism, prioritizing temperature and humidity control, load adjustment, and fault location notification response actions according to the importance of the risk points, the rate of parameter change, and the range of coupling influence; and simultaneously recording the action triggering conditions, execution sequence, and real-time effect feedback data.

8. The system according to claim 1, characterized in that, The construction process of the abnormal information database is as follows: a hierarchical clustering storage architecture is adopted, and the recorded data is classified hierarchically according to the abnormality level and occurrence scenario. At the same time, an index system based on the core features of the abnormality is established. The index dimensions include abnormal parameter type, risk transmission path and response action type. Through an incremental learning mechanism, newly generated abnormal records and rectification verification results are included in the database in real time. Simultaneously, the cosine similarity algorithm is used to compare the features of new records with historical cases and associate the handling schemes and effect data of similar abnormalities.

9. The system according to claim 1, characterized in that, The specific process for extracting busbar sway features is as follows: First, a time-domain-frequency-domain joint extraction mechanism is used to extract basic features from the sway-related parameters; then, a feature importance evaluation algorithm is used to calculate the contribution of the basic features to the busbar safety status determination, a contribution threshold is set to filter out the core feature set, normalization processing is performed, and finally, a core feature system is generated.

10. The system according to claim 1, characterized in that, The process of constructing the sway feature model is as follows: Based on the sway features of the busbar trunking, the model is labeled and filtered according to normal operation, single abnormality, and temperature and humidity coupled abnormality scenarios to generate a feature sample set; using the entropy weight method and the analytic hierarchy process, the weight of the core features in the judgment of abnormal sway is quantified; a model framework is built based on the support vector machine; the sample set is divided into a training set and a validation set according to the proportion; the decision boundary is optimized through training iteration; and the recognition accuracy is tested using the validation set.

11. The system according to claim 1, characterized in that, The process of generating the abnormal sway judgment result is as follows: inputting the busbar sway characteristics into the sway characteristic model, calculating the similarity between the real-time characteristics and the preset safety baseline using the cosine similarity algorithm, filtering potential abnormal samples in combination with dynamically adjusted thresholds, and weighting and fusing the samples according to the core feature weights. Through a continuous duration verification mechanism, the potential abnormal state is monitored for multiple cycles. At the same time, the judgment criteria are adjusted according to the temperature and humidity conditions, and finally the abnormal sway judgment result is generated.

12. The system according to claim 1, characterized in that, The specific process for generating the comprehensive inspection report is as follows: First, the anomaly judgment conclusion and the abnormal shaking judgment result are integrated using a data fusion mechanism to extract key information and establish a multi-dimensional information correlation map; then, the map is analyzed through a preset coupling effect evaluation model to quantify the comprehensive risk index; next, a cross-scenario cross-validation mechanism is initiated to calibrate the judgment deviation using an error correction algorithm; finally, a structured report generation framework is used to generate the comprehensive inspection report.

Images