Multi-parameter online monitoring and fault early warning system based on intelligent busbar

Abstract

This invention discloses a multi-parameter online monitoring and fault early warning system based on intelligent busbars. The system includes a data acquisition module, a feature extraction module, a data analysis module, an early warning triggering module, and a cross-node data aggregation module. The data acquisition module collects temperature, current, vibration, acoustic, and environmental data in real time. The feature extraction module calculates the anomaly and deviation indices of various data types. The data analysis module uses a weighted fusion algorithm to perform spatiotemporal alignment and fusion of multi-source features to generate a local fault confidence score. The early warning triggering module outputs multi-level alarm signals based on the score and preset thresholds. The cross-node data aggregation module constructs a system-level spatiotemporal dataset to achieve global status monitoring. Through multi-parameter fusion analysis and intelligent early warning, this system improves the operational reliability and safety of busbars and is suitable for status monitoring and fault early warning of key equipment in power systems.

Description

Technical Field

[0001] This invention belongs to the field of power equipment condition monitoring and fault diagnosis technology, and specifically discloses a multi-parameter online monitoring and fault early warning system based on intelligent busbars. Background Technology

[0002] Busbars are key conductive components in power systems, and their operating status directly affects power supply reliability and system security; however, traditional monitoring methods have the following shortcomings:

[0003] 1. Single monitoring dimension: Existing systems usually only monitor one physical quantity, such as temperature or current, and lack the ability to simultaneously collect and comprehensively analyze multiple parameters, such as vibration, acoustic and environmental parameters. This makes it difficult to fully reflect the operating status of the equipment and is prone to false alarms or missed alarms.

[0004] 2. Lack of integrated analysis in fault diagnosis: It fails to achieve joint analysis and anomaly assessment of multi-dimensional characteristics such as temperature gradient, current harmonics, vibration spectrum and acoustic emission signals, and cannot accurately identify early faults.

[0005] Therefore, it is essential to invent a multi-parameter online monitoring and fault early warning system based on intelligent busbars to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies in the prior art, this invention provides a multi-parameter online monitoring and fault early warning system based on intelligent busbars. This system includes a data acquisition module, a feature extraction module, a data analysis module, an early warning triggering module, and a cross-node data aggregation module. The data acquisition module collects temperature, current, vibration, acoustic, and environmental data in real time. The feature extraction module calculates the anomaly and deviation indices of various data types. The data analysis module uses a weighted fusion algorithm to perform spatiotemporal alignment and fusion of multi-source features, generating a local fault confidence score. The early warning triggering module outputs multi-level alarm signals based on the score and preset thresholds. The cross-node data aggregation module constructs a system-level spatiotemporal dataset to achieve global status monitoring, effectively solving the problems mentioned in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a multi-parameter online monitoring and fault early warning system based on intelligent busbars, comprising: a data acquisition module, a feature extraction module, a data analysis module, an early warning triggering module, and a cross-node data aggregation module;

[0008] Data acquisition module: Real-time acquisition of temperature data, current data, vibration data, acoustic data, and environmental data;

[0009] Feature extraction module: Extracts data anomaly / deviation degree regarding fault confidence score based on temperature data, current data, vibration data and acoustic data collected by the data acquisition module.

[0010] Data analysis module: Based on the spatiotemporal alignment strategy, a weighted fusion algorithm is used to calculate the confidence score of local faults;

[0011] Early warning triggering module: Outputs an alarm signal based on preset threshold rules and confidence level;

[0012] Cross-node data aggregation module: Constructs a system-level spatiotemporal dataset through a time synchronization protocol, mapping discrete node data to the busbar physical topology to form a global monitoring graph;

[0013] The specific analysis method of the feature extraction module is as follows: based on the temperature data, current data, vibration data, and acoustic data collected by the data acquisition module, the following parameters are calculated: connection point surface temperature anomaly, axial temperature gradient anomaly, vibration acceleration RMS value anomaly, vibration acceleration peak value anomaly, spectral characteristic anomaly, sound pressure level anomaly, acoustic emission signal energy anomaly, current true RMS value deviation, harmonic content anomaly, and three-phase imbalance.

[0014] The surface temperature anomaly T at the connection point c The calculation formula is: In the formula, T cur T represents the surface temperature of the connection point. b The historical average temperature under normal operating conditions is given by k, which is the tolerance coefficient; the axial temperature gradient anomaly degree is T. g The calculation formula is: , ∇T in the formula imax The maximum measured value of the temperature gradient in each axial segment is ∇T. ref For the standard gradient allowed by the device design, ∇T ran The gradient fluctuation safety threshold; the anomaly degree V of the effective value of vibration acceleration. r The calculation formula is: In the formula, V mea V is the measured effective value of vibration acceleration. thr The peak anomaly of vibration acceleration V is the threshold value according to ISO 10816 standard. p The calculation formula is: In the formula, P mea P is the measured peak vibration value. thr The vibration peak threshold and the spectral anomaly degree V s The calculation formula is: In the formula, A f For the FFT energy distribution in the 0.1Hz–5kHz frequency band, A ref The reference spectrum; sound pressure level anomaly S s The calculation formula is: In the formula, SPL meaSPL is a wideband sound pressure level of 20Hz–40kHz. thr The acoustic emission signal energy anomaly S is the baseline value for silent operation of the equipment. a The calculation formula is: , E in the formula mea The wavelet packet energy at the characteristic frequencies of 8kHz / 16kHz, E bas Energy baseline under fault-free conditions; current true RMS deviation I r The calculation formula is: In the formula, I mea I is the measured true RMS value of the current. rat The rated current value, harmonic content anomaly degree I h The calculation formula is: In the formula, H n The amplitudes of the 2nd to 13th harmonic components, H thr The harmonic limit is the national standard value, where N is the harmonic order; the three-phase unbalance is I. u The calculation formula is: In the formula, I max I is the maximum value of the three-phase current. min I is the minimum value of the three-phase current. avg This represents the average value of the three-phase current.

[0015] Preferably, the temperature data includes the surface temperature of the busbar connection point and the axial temperature gradient of the busbar body; the current data includes the true RMS value of the current in each branch, the content of the 2nd to 13th harmonics, and the maximum, minimum, and average values ​​of the three-phase current; the vibration data includes the RMS value and peak value of the vibration acceleration, and the FFT energy distribution in the 0.1Hz to 5kHz frequency band; the acoustic data includes the sound pressure level in the 20Hz to 40kHz frequency band and the wavelet packet energy of the characteristic frequencies at 8kHz and 16kHz; and the environmental data includes the ambient temperature and relative humidity values.

[0016] Preferably, the specific analysis method of the data analysis module is as follows: a weighted average fusion algorithm is used to fuse temperature, current, vibration and acoustic characteristics at the same time stamp and physical location to calculate a comprehensive local fault confidence score, which is used to quantify the health status of the monitoring point.

[0017] Preferably, the formula for calculating the local fault confidence score is: In the formula, S is the local fault confidence score, η is the environmental correction factor, and ω T ω V ω A and ω I These are the weighting coefficients for temperature, vibration, acoustic, and current characteristics, respectively. The weighting values ​​are adjusted according to the actual application scenario. F TFor temperature feature scoring, F V F is the score for vibration characteristics. A For acoustic feature scoring, F I Score the current characteristics;

[0018] The formula for calculating the temperature feature score is as follows: The formula for calculating the vibration characteristic score is as follows: The formula for calculating the acoustic feature score is as follows: The formula for calculating the current characteristic score is as follows: The formula for calculating the environmental correction factor η is as follows: In the formula, RH is the relative humidity value and T is the ambient temperature.

[0019] Preferably, the specific analysis method of the early warning triggering module is as follows: receiving the local fault confidence score from the data analysis module and comparing it with the preset multi-level alarm thresholds; if the score exceeds the first level threshold, an early warning signal is triggered; if it exceeds the second level threshold, an alarm signal is triggered; if it exceeds the third level threshold, an emergency shutdown signal is triggered, and the signal is pushed to the visualization module and related control terminals.

[0020] The technical effects and advantages of this invention are as follows:

[0021] 1. The system integrates multiple sensors such as temperature, current, vibration, acoustics, and environment to achieve comprehensive monitoring of the busbar's operating status; it calculates the anomaly and deviation indices of various data through a feature extraction module, covering multiple fault dimensions such as electrical, mechanical, and acoustic aspects; it overcomes the limitations of traditional single-parameter monitoring and reduces the rates of missed and false alarms.

[0022] 2. The early warning triggering module adopts a multi-level threshold rule, which dynamically triggers different levels of response such as early warning, alarm, and emergency shutdown based on the confidence score, to ensure that the system can issue an effective warning before or in the early stage of a fault, so as to avoid equipment damage or the escalation of the accident.

[0023] 3. The data analysis module adopts a weighted fusion algorithm to fuse multi-source features from the same time and location to generate a local fault confidence score; it introduces an environmental correction factor to consider the impact of temperature and humidity on equipment status and improve the environmental adaptability of the score; and it uses a spatiotemporal alignment strategy to ensure the consistency of multi-node data in time and space, providing a data foundation for accurate diagnosis. Attached Figure Description

[0024] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the module connection of the present invention.

[0026] Figure 2 This is the overall flowchart of the present invention. Detailed Implementation

[0027] 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.

[0028] like Figure 1 As shown, the multi-parameter online monitoring and fault early warning system based on intelligent bus provided by the present invention includes a data acquisition module, a feature extraction module, a data analysis module, an early warning triggering module, and a cross-node data aggregation module.

[0029] The following will combine Figure 2 The present invention will be explained in detail as follows:

[0030] Furthermore, in the above technical solution, the data acquisition module is used to collect temperature data, current data, vibration data, acoustic data, and environmental data in real time;

[0031] Furthermore, in the above technical solution, the temperature data includes the surface temperature of the busbar connection point and the axial temperature gradient of the busbar body; the current data includes the true RMS value of the current in each branch, the content of the 2nd to 13th harmonics, and the maximum, minimum, and average values ​​of the three-phase current; the vibration data includes the RMS value and peak value of the vibration acceleration, and the FFT energy distribution in the 0.1Hz to 5kHz frequency band; the acoustic data includes the sound pressure level in the 20Hz to 40kHz frequency band and the wavelet packet energy of the characteristic frequencies at 8kHz and 16kHz; and the environmental data includes the ambient temperature and relative humidity values.

[0032] In a preferred embodiment of the present invention, the methods for collecting various types of data are as follows:

[0033] Temperature data: The surface temperature of the busbar connection point is obtained directly by infrared temperature sensor; the axial temperature gradient parameters of the busbar body are obtained by axially arranged multi-point temperature sensors, and the temperature parameters are processed by differential calculation method to obtain axial temperature gradient data.

[0034] Current data: The current waveform parameters of each branch are obtained by direct detection through current transformers, and the true RMS value algorithm is used to process the current waveform parameters to obtain the true RMS value data of the current; the 2nd to 13th harmonic content data is obtained by processing the current waveform parameters through Fourier transform; the three-phase current imbalance data is obtained by calculating the three-phase current sampling values.

[0035] Vibration data: Vibration acceleration waveform parameters are obtained directly by accelerometers. The effective value algorithm is used to process the vibration acceleration parameters to obtain the effective value of vibration. The peak value is obtained by processing the vibration acceleration parameters using the peak detection algorithm. The 0.1Hz–5kHz frequency band spectral characteristic data is obtained by processing the vibration acceleration parameters using fast Fourier transform.

[0036] Acoustic data: Wideband sound pressure signal parameters are obtained directly by sound pressure sensor, and sound pressure level data of 20Hz–40kHz are obtained by processing the sound pressure signal parameters using sound pressure level calculation method; acoustic emission signal energy data of characteristic frequencies of 8kHz and 16kHz are obtained by processing the sound pressure signal parameters by wavelet packet transform.

[0037] Environmental data: Ambient temperature and relative humidity parameters are obtained directly through temperature and humidity sensors.

[0038] The feature extraction module is used to extract the data anomaly / deviation degree of the fault confidence score based on the temperature data, current data, vibration data and acoustic data collected by the data acquisition module.

[0039] Furthermore, in the above technical solution, the specific analysis method of the feature extraction module is as follows: based on the temperature data, current data, vibration data and acoustic data collected by the data acquisition module, calculate the surface temperature anomaly of the connection point, the axial temperature gradient anomaly, the effective value anomaly of vibration acceleration, the peak value anomaly of vibration acceleration, the spectral characteristic anomaly, the sound pressure level anomaly, the acoustic emission signal energy anomaly, the true effective value deviation of current, the harmonic content anomaly and the three-phase imbalance.

[0040] Furthermore, in the above technical solution, the surface temperature anomaly T at the connection point... c The calculation formula is: In the formula, T cur T represents the surface temperature of the connection point. b The historical average temperature under normal operating conditions is given by k, which is the tolerance coefficient; the axial temperature gradient anomaly degree is T. g The calculation formula is: , ∇T in the formula imax The maximum measured value of the temperature gradient in each axial segment is ∇T. ref For the standard gradient allowed by the device design, ∇T ran The gradient fluctuation safety threshold; the anomaly degree V of the effective value of vibration acceleration.r The calculation formula is: In the formula, V mea V is the measured effective value of vibration acceleration. thr The peak anomaly of vibration acceleration V is the threshold value according to ISO 10816 standard. p The calculation formula is: In the formula, P mea P is the measured peak vibration value. thr The vibration peak threshold and the spectral anomaly degree V s The calculation formula is: In the formula, A f For the FFT energy distribution in the 0.1Hz–5kHz frequency band, A ref The reference spectrum; sound pressure level anomaly S s The calculation formula is: In the formula, SPL mea SPL is a wideband sound pressure level of 20Hz–40kHz. thr The acoustic emission signal energy anomaly S is the baseline value for silent operation of the equipment. a The calculation formula is: , E in the formula mea The wavelet packet energy at the characteristic frequencies of 8kHz / 16kHz, E bas Energy baseline under fault-free conditions; current true RMS deviation I r The calculation formula is: In the formula, I mea I is the measured true RMS value of the current. rat The rated current value, harmonic content anomaly degree I h The calculation formula is: In the formula, H n The amplitudes of the 2nd to 13th harmonic components, H thr The harmonic limit is the national standard value, where N is the harmonic order; the three-phase unbalance is I. u The calculation formula is: In the formula, I max I is the maximum value of the three-phase current. min I is the minimum value of the three-phase current. avg This represents the average value of the three-phase current.

[0041] It should be further noted that the tolerance coefficient is typically set between 0.2 and 0.3, and the gradient fluctuation safety threshold is an empirical value, usually ∇T. ref 10–20%, P thr This is an empirical value, usually V. thr 1.5–2 times that.

[0042] The data analysis module is used to calculate the confidence score of local faults using a weighted fusion algorithm based on a spatiotemporal alignment strategy.

[0043] Furthermore, in the above technical solution, the specific analysis method of the data analysis module is as follows: a weighted average fusion algorithm is used to fuse temperature, current, vibration, and acoustic characteristics at the same time stamp and physical location to calculate a comprehensive local fault confidence score, which is used to quantify the health status of the monitoring point.

[0044] Furthermore, in the above technical solution, the formula for calculating the local fault confidence score is as follows: In the formula, S is the local fault confidence score, η is the environmental correction factor, and ω T ω V ω A and ω I These are the weighting coefficients for temperature, vibration, acoustic, and current characteristics, respectively. The weighting values ​​are adjusted according to the actual application scenario. F T For temperature feature scoring, F V F is the score for vibration characteristics. A For acoustic feature scoring, F I The current characteristic score is given.

[0045] It should be further explained that the weighting coefficients of the temperature, vibration, acoustic, and current features are dynamically adjusted based on the sensitivity and importance of each parameter to the fault in actual applications, and are determined through training with historical fault data. In a preferred embodiment, the system uses historical fault data for machine learning training to determine the optimal weights. The specific process is as follows: collect a large amount of multi-parameter data under normal operation and known fault conditions; have domain experts label the health status of each data sample; use the random forest algorithm for training, analyze the contribution of each feature to the classification results, and normalize it as the initial value of the weighting coefficients.

[0046] Furthermore, the environmental correction factor η is used to compensate for the influence of temperature and humidity on the measurement of certain physical quantities, especially acoustic signals and insulation performance. High temperature and high humidity environments may lead to increased attenuation of sound wave propagation or condensation on the equipment surface, affecting the accuracy of measurement. This formula is calibrated through laboratory environmental simulation tests. Under the reference environment of 25°C and 50%RH, the environmental correction factor is 1. When the environment deviates from the reference, the fusion score is linearly corrected according to the degree of temperature and humidity deviation. This coefficient has been experimentally verified to effectively improve the monitoring consistency under different environments.

[0047] Furthermore, in the above technical solution, the formula for calculating the temperature feature score is: The formula for calculating the vibration characteristic score is as follows: The formula for calculating the acoustic feature score is as follows: The formula for calculating the current characteristic score is as follows: The formula for calculating the environmental correction factor η is as follows: In the formula, RH is the relative humidity value and T is the ambient temperature.

[0048] The early warning triggering module is used to output an alarm signal based on preset threshold rules and confidence level.

[0049] Furthermore, in the above technical solution, the specific analysis method of the early warning triggering module is as follows: receiving the local fault confidence score from the data analysis module and comparing it with the preset multi-level alarm thresholds; if the score exceeds the first level threshold, an early warning signal is triggered; if it exceeds the second level threshold, an alarm signal is triggered; if it exceeds the third level threshold, an emergency shutdown signal is triggered, and the signal is pushed to the visualization module and related control terminals.

[0050] In a preferred embodiment of the present invention, regression analysis of historical fault data reveals that the typical value of the first-level threshold is 0.7, the typical value of the second-level threshold is 1.5, and the typical value of the third-level threshold is 3.0.

[0051] The cross-node data aggregation module is used to build a system-level spatiotemporal dataset through a time synchronization protocol, mapping discrete node data to the busbar physical topology to form a global monitoring graph.

[0052] Furthermore, the specific process of constructing the spatiotemporal dataset by the cross-node data aggregation module is as follows:

[0053] The cross-node data aggregation module performs spatiotemporal alignment on temperature data, current data, vibration data, acoustic data, and alarm signals from different nodes within the same time window, and organizes the processed data into a spatiotemporal dataset. The spatiotemporal dataset includes: a time dimension, a spatial dimension, and a parameter dimension. The time dimension is the timestamp of the dataset collection, the spatial dimension is the location of data collection, and the parameter dimension is the collected multidimensional parameter data and alarm signals.

[0054] A graph structure model of the busbar system is constructed, with each monitoring point as a node and electrical or mechanical connection relationships as edges.

[0055] The spatiotemporal datasets of each node are associated and merged according to the topology to form a global data view.

[0056] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-parameter online monitoring and fault warning system based on intelligent bus, characterized in that, Comprise: a data acquisition module, a feature extraction module, a data analysis module, an early warning triggering module and a cross-node data aggregation module; the data acquisition module: real-time acquisition of temperature data, current data, vibration data, acoustic data and environmental data; the feature extraction module: based on the temperature data, current data, vibration data and acoustic data collected by the data acquisition module, the data anomaly degree / deviation degree of the fault confidence score is extracted; the data analysis module: based on the spatio-temporal alignment strategy, the local fault confidence score is calculated using a weighted fusion algorithm; the early warning triggering module: by pre-setting threshold rules, combined with confidence, output alarm signal; the cross-node data aggregation module: through the time synchronization protocol, the system-level spatio-temporal data set is constructed, the discrete node data is mapped to the busbar physical topology structure, and the global monitoring view is formed; The specific analysis method of the feature extraction module is: based on the temperature data, current data, vibration data and acoustic data collected by the data acquisition module, the surface temperature anomaly degree of the connection point, the axial temperature gradient anomaly degree, the vibration acceleration effective value anomaly degree, the vibration acceleration peak value anomaly degree, the frequency spectrum feature anomaly degree, the sound pressure level anomaly degree, the acoustic emission signal energy anomaly degree, the current true effective value deviation degree, the harmonic content anomaly degree and the three-phase imbalance degree are calculated; The connection point surface temperature abnormality degree T c The calculation formula is: , in the formula, T cur is the connection point surface temperature, T b is the historical average temperature of the normal working condition, and k is the tolerance coefficient; the calculation formula of the axial temperature gradient abnormality degree T g is: , in the formula, ∇T imax is the maximum measured value of the temperature gradient of each section in the axial direction, ∇T ref is the standard gradient allowed by the design of the equipment, and ∇T ran is the gradient fluctuation safety threshold; the calculation formula of the vibration acceleration effective value abnormality degree V r is: , in the formula, V mea is the measured value of the vibration acceleration effective value, V thr is the ISO 10816 standard threshold value, the calculation formula of the vibration acceleration peak value abnormality degree V p is: , in the formula, P mea is the measured value of the vibration peak value, P thr is the vibration peak value threshold value, the calculation formula of the spectrum feature abnormality degree V s is: , in the formula, A f is the FFT energy distribution in the frequency band of 0.1 Hz-5 kHz, and A ref is the reference spectrum; the calculation formula of the sound pressure level abnormality degree S s is: , in the formula, SPL mea is the wide-frequency sound pressure level in the frequency band of 20 Hz-40 kHz, and SPL thr is the reference value of the quiet operation of the equipment, the calculation formula of the acoustic emission signal energy abnormality degree S a is: , in the formula, E mea is the wavelet packet energy of the characteristic frequency of 8 kHz / 16 kHz, and E bas is the energy baseline of the fault-free state; the calculation formula of the current true effective value deviation degree I r is: , in the formula, I mea is the measured value of the current true effective value, I rat is the rated current value, the calculation formula of the harmonic content abnormality degree I h is: , in the formula, H n is the amplitude of the harmonic component of 2-13 times, H thr is the national standard harmonic limit value, and N is the harmonic number; the calculation formula of the three-phase unbalance degree I u is: , in the formula, I max I is the maximum value of the three-phase current min I is the minimum value of the three-phase current avg I is the average value of the three-phase current.

2. The smart bus based multi-parameter online monitoring and fault warning system as claimed in claim 1, wherein: The temperature data includes the surface temperature of the busbar connection point and the axial temperature gradient of the busbar body; The current data includes the true effective value of each branch current, the 2-13 harmonic content, the maximum, minimum and average values of three-phase current; The vibration data includes the effective value and peak value of vibration acceleration, and the 0.1Hz to 5kHz frequency band FFT energy distribution; The acoustic data includes the sound pressure level in the 20Hz to 40kHz frequency band, and the wavelet packet energy of 8kHz and 16kHz characteristic frequency; The environmental data includes the environmental temperature and relative humidity value.

3. The smart bus based multi-parameter online monitoring and fault warning system as claimed in claim 1, wherein: The specific analysis method of the data analysis module is: using a weighted average fusion algorithm, the temperature, current, vibration and acoustic feature quantities at the same time stamp and the same physical location are fused to calculate a comprehensive local fault confidence score.

4. The smart bus based multi-parameter online monitoring and fault warning system as claimed in claim 3, wherein: The calculation formula of the local fault confidence score is: , wherein S is the local fault confidence score, η is an environmental correction factor, ω T , ω V , ω A , and ω I are weight coefficients of temperature, vibration, acoustic, and current characteristics respectively, the weight values are adjusted according to actual application scenarios, F T is a temperature characteristic score, F V is a vibration characteristic score, F A is an acoustic characteristic score, and F I is a current characteristic score. The calculation formula of the temperature feature score is: The calculation formula of the vibration feature score is: The calculation formula of the acoustic feature score is: The calculation formula of the current feature score is: The calculation formula of the environmental correction factor η is: In the formula, RH is a relative humidity value, and T is an environmental temperature.

5. The smart bus based multi-parameter online monitoring and fault warning system as claimed in claim 1, wherein: The specific analysis method of the early warning triggering module is: receiving the local fault confidence score from the data analysis module, comparing it with the preset multi-level alarm threshold; if the score exceeds the first level threshold, the early warning signal is triggered; If it exceeds the second level threshold, the alarm signal is triggered; if it exceeds the third level threshold, the emergency shutdown signal is triggered, and the signal is pushed to the visualization module and related control terminal.

Images