Power plant electrical equipment abnormal state intelligent detection and early warning method

Abstract

The invention relates to the technical field of electrical equipment detection, in particular to a power plant electrical equipment abnormal state intelligent detection and early warning method, which comprises the following steps of synchronously acquiring electromagnetic radiation pulse signals, vibration acceleration waveforms and surface temperature gradient signals of an electrical equipment body and a connecting component thereof; constructing a time aligned multi-modal sensing data stream; time delay locking characteristics and cross-modal coupling strength indexes among the signals are extracted; and inputting the time delay locking feature and the cross-modal coupling strength index into a composite decision model, and triggering positioning early warning through a double-threshold cross validation mechanism. According to the method, a full-process closed-loop judgment system covering instantaneous interference suppression, trend anomaly tracking and redundancy verification is formed, and the early warning intelligence level and response adaptability of electrical equipment in a complex scene are improved.

Description

Technical Field

[0001] This invention relates to the field of electrical equipment testing technology, and in particular to a method for intelligent detection and early warning of abnormal conditions of electrical equipment in power plants. Background Technology

[0002] As the core component of power transmission and control systems, the long-term operating status of power plant electrical equipment directly affects the safety, stability, and continuity of power supply of the power generation system. With the increasing size and complexity of equipment structures, as well as the increasing uncertainty of load fluctuations and operating environments, traditional condition detection methods that rely on monitoring a single physical quantity (such as temperature, vibration, or current) are no longer able to accurately identify early complex faults. In particular, abnormal conduction phenomena caused by local insulation degradation, structural loosening, or electromagnetic interference often have complex characteristics such as cross-modal, multi-path, and gradual evolution, leading to problems such as high false alarm rate, response delay, and insufficient positioning accuracy.

[0003] In existing technologies, single-channel monitoring methods based on statistical rules or fixed thresholds are often used, which lack in-depth modeling of the coupling relationship between multiple physical quantities and make it difficult to identify the causal and synchronous change characteristics between multimodal signals. At the same time, there is a lack of physical constraint modeling of the device's own three-dimensional topology and signal propagation path, which makes it difficult to accurately locate structure-related anomalies. Summary of the Invention

[0004] This invention provides an intelligent detection and early warning method for abnormal states of electrical equipment in power plants. It integrates multimodal signal acquisition, structure-sensitive conduction analysis, intelligent coupling criterion modeling, and dynamic multi-level early warning strategy to improve the ability to identify early and weak anomalies, the accuracy of response to coupling conduction mechanisms, and the adaptability and reliability of the overall early warning system.

[0005] A method for intelligent detection and early warning of abnormal conditions in power plant electrical equipment includes the following steps: S1: Synchronously acquire electromagnetic radiation pulse signals, vibration acceleration waveforms, and surface temperature gradient signals of the electrical equipment body and its connecting components to construct a time-aligned multimodal sensing data stream; S2: Perform abnormal conduction phase synchronization analysis on the multimodal sensing data stream, and extract the time delay lock-in characteristics and cross-modal coupling strength index between each signal; S3: Input the aforementioned time delay locking feature and cross-modal coupling strength index into the composite decision model, and trigger a location warning through a dual-threshold cross-validation mechanism, wherein: A primary warning is activated when the time delay locking feature exceeds the inherent structural constraint threshold of the device. A final warning is triggered when the cross-modal coupling strength index synchronously exceeds the modal association threshold.

[0006] Optionally, S1 specifically includes: S11, deploy sensor arrays at key transmission nodes of the device body and its connecting components, with each node synchronously integrated: The electromagnetic radiation probe uses a wideband log-periodic antenna, and its operating frequency band is: ; The triaxial vibration sensor has a built-in temperature compensation module, and its measurement range is: ; The angle between the field of view of the infrared thermometer and the normal to its mounting surface satisfies: ; in, It is the antenna response frequency. The acceleration measurement value, It is the angle between the central axis of the infrared thermometer's field of view and the normal to the mounting surface of the vibration sensor.

[0007] S12, by sending synchronization pulse signals to all sensors through a GPS-disciplined clock source, a unified time base coordinate system is established, and the system time synchronization error satisfies: ,in, It is the maximum time synchronization error between the sensors.

[0008] S13, establish a sensor spatial location mapping table based on the device's three-dimensional structural topology map, and perform interpolation processing on the heterogeneous frequency sampling signals to generate a multimodal sensing data stream with the following indicators: Time resolution: ; Spatial positioning accuracy: ; in, The minimum time interval between adjacent sampling points in the data stream. This represents the maximum error range for sensor position reconstruction in three-dimensional space.

[0009] In power plant electrical equipment, critical transmission nodes refer to structural locations where electrical signals, mechanical vibrations, or heat transfer easily accumulate or are conducted. Anomalies at these nodes most readily reflect equipment failure or deterioration, including: High-voltage switch contact points: strong electromagnetic radiation, changes in contact resistance can easily cause local overheating or discharge.

[0010] The connection point between the transformer winding lead and the bushing: high current density, concentrated thermal and electrical stress. Cable connectors and terminal block interfaces: Vibration coupling is significant, and poor contact or breakdown is likely to occur here; The bearing connecting the circuit breaker operating mechanism and the moving contact is the core path for the transmission of mechanical impact and vibration.

[0011] Busbar suspension insulator support points: prone to structural stress concentration and temperature anomalies.

[0012] The method for constructing the sensor spatial location mapping table is as follows: 1. Equipment 3D structural modeling: Use 3D modeling tools or CAD drawings to reconstruct the 3D topology of electrical equipment, and establish the position, connection relationship and coordinate system of components.

[0013] 2. Sensor Deployment Point Calibration: Convert the coordinates of each sensor's installation node (such as connection bolt points, insulating sleeve bases, contact interfaces, etc.) into local coordinates in the 3D model. Record its component, normal direction, and installation type.

[0014] 3: Establish spatial mapping representation to generate sensor mapping table ,in, Number the sensor. For spatial coordinates, For sensor types (electromagnetic, vibration, temperature), This is the normal direction of the sensor mounting surface, used for attitude correction and thermal axis compensation.

[0015] Optionally, S2 specifically includes: S21: Perform bandpass filtering and baseline calibration on the multimodal sensing data stream; S22: The time delay lock value of the electromagnetic-vibration signal pair is calculated using the cross-correlation function of a sliding window. The window length is a multiple of the device's fundamental frequency period and is dynamically adjusted. The phase difference of the maximum cross-correlation peak within the sliding window is extracted as the time delay lock feature. S23: Construct the cross-modal covariance matrix, and calculate the cross-modal coupling strength index based on the entropy growth rate of the vibration signal Hilbert envelope and the temperature gradient signal. .

[0016] Optionally, S21 includes: retaining a preset frequency band for electromagnetic radiation pulse signals, filtering out environmental noise from vibration signals, and eliminating solar radiation interference from temperature gradient signals.

[0017] Optionally, the sliding window length in S22 is defined as: ,in, The length of the sliding window. The base frequency for equipment operation is used, and the window length is dynamically adjusted according to the operating conditions. The cross-correlation function is expressed as: ,in, The cross-correlation function representing the relationship between electromagnetic and vibration signals. This represents the amplitude of the electromagnetic signal at time t. This represents the amplitude of the vibration signal at time t. Represents the time lag variable. Indicates the vibration signal in time amplitude, Indicates the start time of the current sliding window; The phase difference of the maximum cross-correlation peak within the sliding window is expressed as: , This represents the time delay lockout value at the maximum value of the electromagnetic-vibration cross-correlation.

[0018] Optionally, in step S23, the normalized covariance between the vibration signal energy and the temperature gradient signal change rate is calculated to evaluate the synchronous coupling strength between the two within the time window; at the same time, the variance of the electromagnetic signal is introduced as a weighting factor, and the abnormal perception capability in the presence of discharge pulse electromagnetic disturbance is enhanced by the logarithmic function, thereby constructing a coupling strength index that reflects the linkage effect of multiple physical fields and assisting in the identification of the composite conduction path of mechanical, thermal, and electrical anomalies.

[0019] Optionally, S3 specifically includes: S31, calculate the theoretical time delay based on the structural spacing of the nodes in the three-dimensional topology of the equipment and the sound velocity of the material, and set the structural constraint time delay threshold range. ; S32, Based on the historical health data statistical distribution of the cross-modal coupling strength index, set the modal association threshold. ; S33, when real-time delay locking feature Exceeding the structural constraint delay threshold range At that time, activate the primary warning and execute: Increase the sampling rate of the abnormal path sensor to 1MHz; Activate the monitoring redundancy check of the 3-hop neighboring nodes in its topology.

[0020] S34, if, within multiple consecutive sampling periods after the activation of the primary warning, the cross-modal coupling strength index simultaneously satisfies: the cross-modal coupling strength index exceeds... Furthermore, if the exponential growth rate of the cross-modal coupling strength exceeds the growth threshold, a final warning will be triggered.

[0021] Optionally, S3 further includes a dynamic adjustment mechanism, specifically including: If the cumulative number of primary warnings triggered by a certain fixed transmission path exceeds the set value within any consecutive operating hour, the calculation of the structural constraint delay threshold will be adjusted to improve sensitivity.

[0022] The beneficial effects of this invention are: This invention deploys a multimodal sensor array on the electrical equipment body and key connecting components to collect electromagnetic radiation, mechanical vibration and temperature gradient signals. Combined with the three-dimensional topology of the equipment, it achieves accurate modeling and response location of abnormal transmission paths inside complex equipment, improving the spatial structure sensitivity of signal coupling strength calculation and the reliability of fault location.

[0023] This invention employs a coupling strength index model based on Hilbert envelope, entropy growth rate, and electromagnetic fluctuation weights to effectively capture the nonlinear linkage effects between physical quantities such as vibration, heat, and electricity. It combines a real-time sliding cross-correlation algorithm to extract time delay locking features and utilizes a dual threshold mechanism of structural constraints and growth trends for joint judgment. This enables the differentiation between random noise and continuous abnormal evolution trends in actual operation, improving the accuracy of anomaly identification and anti-interference capability under complex working conditions.

[0024] This invention proposes a dual-threshold early warning mechanism based on structural constraint statistical intervals and trend growth rates. This mechanism not only enables graded responses to abnormal states but also introduces a dynamic sensitivity adjustment strategy. It can automatically adjust the sensitivity range of the judgment criteria according to the frequency of early warnings, forming a closed-loop judgment system covering instantaneous interference suppression, trend anomaly tracking, and redundancy verification. This improves the intelligence level and response adaptability of electrical equipment in complex scenarios. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the time delay locking feature extraction and cross-modal coupling strength index calculation in an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0028] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0029] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0030] like Figures 1-2 As shown, the intelligent detection and early warning method for abnormal conditions of electrical equipment in power plants includes the following steps: S1: Synchronously acquire electromagnetic radiation pulse signals, vibration acceleration waveforms, and surface temperature gradient signals of the electrical equipment body and its connecting components to construct a time-aligned multimodal sensing data stream; S2: Perform abnormal propagation phase synchronization analysis on multimodal sensing data streams to extract the time delay lock-in characteristics and cross-modal coupling strength index between each signal; S3: Input the time-delay locking feature and the cross-modal coupling strength index into the composite decision model, and trigger the location warning through a dual-threshold cross-validation mechanism, wherein: A primary warning is activated when the time delay locking feature exceeds the inherent structural constraint threshold of the device. A final warning is triggered when the cross-modal coupling strength index synchronously exceeds the modal association threshold.

[0031] A primary warning is equivalent to "discovering abnormal clues," while a final warning is "confirming the fault mechanism." To avoid erroneous protection activation, a primary warning provides a buffer period for manual review, while a final warning triggers maintenance.

[0032] S1 specifically includes: S11, deploy sensor arrays at key transmission nodes of the device body and its connecting components, with each node synchronously integrated: The electromagnetic radiation probe uses a wideband log-periodic antenna, and its operating frequency band is: ; A triaxial vibration sensor with a built-in temperature compensation module; measurement range: ; The angle between the field of view of the infrared thermometer and the normal to its mounting surface satisfies: ; in, It is the antenna response frequency. The acceleration measurement value, It is the angle between the central axis of the infrared thermometer's field of view and the normal to the mounting surface of the vibration sensor.

[0033] S12, by sending synchronization pulse signals to all sensors through a GPS-disciplined clock source, a unified time base coordinate system is established, and the system time synchronization error satisfies: ,in, It is the maximum time synchronization error between the sensors.

[0034] S13, establish a sensor spatial location mapping table based on the device's three-dimensional structural topology map, and perform interpolation processing on the heterogeneous frequency sampling signals to generate a multimodal sensing data stream with the following indicators: Time resolution: ; Spatial positioning accuracy: ; in, The minimum time interval between adjacent sampling points in the data stream. This represents the maximum error range for sensor position reconstruction in three-dimensional space.

[0035] In power plant electrical equipment, critical transmission nodes refer to structural locations where electrical signals, mechanical vibrations, or heat transfer easily accumulate or are conducted. Anomalies at these nodes most readily reflect equipment failure or deterioration, including: High-voltage switch contact points: strong electromagnetic radiation, changes in contact resistance can easily cause local overheating or discharge.

[0036] The connection point between the transformer winding lead and the bushing: high current density, concentrated thermal and electrical stress. Cable connectors and terminal block interfaces: Vibration coupling is significant, and poor contact or breakdown is likely to occur here; The bearing connecting the circuit breaker operating mechanism and the moving contact is the core path for the transmission of mechanical impact and vibration.

[0037] Busbar suspension insulator support points: prone to structural stress concentration and temperature anomalies.

[0038] The method for constructing the sensor spatial location mapping table is as follows: 1. Equipment 3D structural modeling: Use 3D modeling tools or CAD drawings to reconstruct the 3D topology of electrical equipment, and establish the position, connection relationship and coordinate system of components.

[0039] 2. Sensor Deployment Point Calibration: Convert the coordinates of each sensor's installation node (such as connection bolt points, insulating sleeve bases, contact interfaces, etc.) into local coordinates in the 3D model. Record its component, normal direction, and installation type.

[0040] 3: Establish spatial mapping representation to generate sensor mapping table ,in, Number the sensor. For spatial coordinates, For sensor types (electromagnetic, vibration, temperature), This is the normal direction of the sensor mounting surface, used for attitude correction and thermal axis compensation.

[0041] S2 specifically includes: S21, Preprocessing and Filtering Calibration: Perform bandpass filtering and baseline calibration operations on the multimodal sensor data stream, as follows: Preset frequency bands are reserved for electromagnetic radiation pulse signals: , The frequency of the electromagnetic radiation signal; Filtering out low-frequency environmental noise from vibration signals: , Vibration signal frequency; To remove solar radiation interference from the temperature gradient signal and perform short-period thermal drift compensation, the following steps can be taken: a. Obtain the solar incidence angle: based on the normal vector of the surface where the sensor is located. With the current solar azimuth vector Calculate the angle of incidence: ; b. Estimate the radiated interference heat flux density (based on a simplified reflection model): ,in, For surface reflectivity, Solar radiation intensity; c. Calculate the corresponding temperature rise disturbance and remove this component (per unit time) from the temperature gradient signal: Subtracting this term from the original gradient completes the thermal drift compensation. For the specific heat capacity of the material, For density, To detect the thickness of the detection layer, this compensation is suitable for short-period high-frequency disturbances and can effectively avoid interference from false temperature rise signals caused by solar incidence.

[0042] S22, Sliding window cross-correlation time delay extraction: The time delay locking characteristics of the electromagnetic-vibration signal pair are calculated using the sliding window cross-correlation algorithm. Sliding window length Defined as: ,in, The base frequency for equipment operation is used, and the window length is dynamically adjusted according to the operating conditions. Calculate electromagnetic signals With vibration signal cross-correlation function : ,in, This represents the amplitude of the electromagnetic radiation signal at time t. This represents the amplitude of the vibration signal at time t. Represents the time lag variable. Indicates the vibration signal in time amplitude, Indicates the start time of the current sliding window; Extract the phase difference at the location of the maximum correlation peak as the time delay locking feature: , This represents the time delay lockout value at the maximum value of the electromagnetic-vibration cross-correlation. In the initial warning trigger state, the window sliding step size is shortened to 50% of the original size to improve the resolution of abnormal response.

[0043] S23, Calculation of Cross-Modal Coupling Strength Index: Construct the cross-modal covariance matrix, integrate the Hilbert envelope energy of the vibration signal with the rate of change and entropy increase characteristics of the temperature gradient signal, and define the coupling strength index. for: ; in, Indicates the first The vibration envelope energy of each sampling point (obtained by Hilbert transform), The rate of change of the temperature gradient, This represents the statistical variance of the electromagnetic signal within the corresponding time window. This represents the number of sampling points in the current sliding window.

[0044] The above definition of the coupling strength index is based on the principle of normalized covariance and measures vibration energy. With the rate of change of temperature gradient The linear coupling strength reflects the mechanical-thermal anomaly conduction relationship; the last logarithmic term By introducing the fluctuation weight of electromagnetic signals, the impact of discharge pulses on coupling strength is amplified. This allows for a comprehensive assessment of resonance anomalies between multiple modes, effectively supporting intelligent early warning mechanisms.

[0045] S3 specifically includes: S31, based on the structural spacing of the equipment's three-dimensional topological structure nodes and the material sound velocity, calculate the theoretical time delay (structural spacing ÷ material sound velocity), and set the structural constraint time delay threshold. : ,in This represents the theoretical average latency under healthy conditions. The standard deviation of historical operating conditions, i.e., the structural constraint delay threshold. The range is: .

[0046] S32, Based on the historical health data statistical distribution of the cross-modal coupling strength index, set the modal association threshold. : ,in, The upper quartile of health data, It is the interquartile range.

[0047] S33, when real-time delay locking feature Exceeding the structural constraint threshold range At that time, the primary warning will be activated and the following actions will be initiated: Increase the sampling rate of the abnormal path sensor to 1MHz; Activate the monitoring redundancy check of the 3-hop neighboring nodes in its topology.

[0048] S34, if the cross-modal coupling strength index is within 3 consecutive sampling periods after the primary warning is activated. Simultaneously satisfy and Then the final warning will be triggered and executed: The fault location is mapped based on the joint distribution of time delay and coupling strength. Generate an early warning report containing a topology diagram of abnormal propagation paths.

[0049] S35, the dynamic adjustment mechanism includes: if a fixed transmission path triggers primary warnings ≥ 5 times within any consecutive operating hour, the standard deviation of historical operating conditions in the structural constraint delay is compressed, and the time is automatically adjusted. Narrowed by 20% to improve sensitivity.

[0050] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0051] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent detection and early warning of abnormal conditions of electrical equipment in power plants, characterized in that, Includes the following steps: S1: Synchronously acquire electromagnetic radiation pulse signals, vibration acceleration waveforms, and surface temperature gradient signals of the electrical equipment body and its connecting components to construct a time-aligned multimodal sensing data stream; S2: Perform abnormal conduction phase synchronization analysis on the multimodal sensing data stream, and extract the time delay lock-in characteristics and cross-modal coupling strength index between each signal; S3: Input the aforementioned time delay locking feature and cross-modal coupling strength index into the composite decision model, and trigger a location warning through a dual-threshold cross-validation mechanism, wherein: A primary warning is activated when the time delay locking feature exceeds the inherent structural constraint threshold of the device. A final warning is triggered when the cross-modal coupling strength index synchronously exceeds the modal association threshold.

2. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 1, characterized in that, S1 includes deploying a sensor array at key transmission nodes of the equipment body and connecting components. The sensor array includes an electromagnetic radiation probe, a vibration sensor, and an infrared thermometer. A GPS-disciplined clock source sends synchronization pulse signals to all sensors to establish a unified spatiotemporal coordinate system.

3. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 2, characterized in that, The electromagnetic radiation probe uses a wideband log-periodic antenna; the vibration sensor has a built-in temperature compensation unit; the field of view of the infrared thermometer forms a predetermined angle with the normal direction of the mounting surface of the vibration sensor.

4. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 3, characterized in that, The S1 further includes establishing a sensor spatial location mapping table based on the device's three-dimensional structural topology map, performing adaptive interpolation processing on the different frequency sampling signals, and generating a multimodal sensing data stream.

5. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 1, characterized in that, S2 specifically includes: S21: Perform bandpass filtering and baseline calibration on the multimodal sensing data stream; S22: The time delay lock value of the electromagnetic-vibration signal pair is calculated using the cross-correlation function of a sliding window. The window length is a multiple of the device's fundamental frequency period and is dynamically adjusted. The phase difference of the maximum cross-correlation peak within the sliding window is extracted as the time delay lock feature. S23: Construct the cross-modal covariance matrix, and calculate the cross-modal coupling strength index based on the entropy growth rate of the vibration signal Hilbert envelope and the temperature gradient signal. .

6. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 5, characterized in that, S21 includes: retaining a preset frequency band for electromagnetic radiation pulse signals, filtering out environmental noise from vibration signals, and eliminating solar radiation interference from temperature gradient signals.

7. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 5, characterized in that, The length of the sliding window in S22 is defined as follows: ,in, The length of the sliding window. The base frequency for equipment operation is used, and the window length is dynamically adjusted according to the operating conditions. The cross-correlation function is expressed as: ,in, The cross-correlation function representing the relationship between electromagnetic and vibration signals. This represents the amplitude of the electromagnetic signal at time t. This represents the amplitude of the vibration signal at time t. Represents the time lag variable. Indicates the vibration signal in time amplitude, Indicates the start time of the current sliding window; The phase difference of the maximum cross-correlation peak within the sliding window is expressed as: , This represents the time delay lockout value at the maximum value of the electromagnetic-vibration cross-correlation.

8. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 5, characterized in that, In step S23, the normalized covariance between the vibration signal energy and the temperature gradient signal change rate is calculated to evaluate the synchronous coupling strength between the two within the time window. At the same time, the variance of the electromagnetic signal is introduced as a weighting factor, and the abnormal perception capability in the presence of discharge pulse electromagnetic disturbance is enhanced by the logarithmic function. A coupling strength index reflecting the linkage effect of multiple physical fields is constructed to help identify the composite conduction path of mechanical, thermal and electrical anomalies.

9. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 1, characterized in that, S3 specifically includes: S31, calculate the theoretical time delay based on the structural spacing of the nodes in the three-dimensional topology of the equipment and the sound velocity of the material, and set the structural constraint time delay threshold range. ; S32, Based on the historical health data statistical distribution of the cross-modal coupling strength index, set the modal association threshold. ; S33, when real-time delay locking feature Exceeding the structural constraint delay threshold range At that time, activate the primary warning and execute: Increase the sampling rate of the abnormal path sensor to 1MHz; Activate the monitoring redundancy check of the 3-hop neighboring nodes in its topology; S34, if, within multiple consecutive sampling periods after the activation of the primary warning, the cross-modal coupling strength index simultaneously satisfies: the cross-modal coupling strength index exceeds... Furthermore, if the exponential growth rate of the cross-modal coupling strength exceeds the growth threshold, a final warning will be triggered.

10. The intelligent detection and early warning method for abnormal states of power plant electrical equipment according to claim 9, characterized in that, The S3 also includes a dynamic adjustment mechanism, specifically including: If the cumulative number of primary warnings triggered by a certain fixed transmission path exceeds the set value within any consecutive operating hour, the calculation of the structural constraint delay threshold will be adjusted to improve sensitivity.

Images