Intelligent detection method and system for operation state of enclosed bus

Abstract

The invention relates to the field of enclosed bus detection, in particular to an intelligent detection method and system for the operation state of an enclosed bus. The method comprises the following steps: acquiring a multi-source real-time monitoring data set of an isolated-phase bus, and based on the multi-source real-time monitoring data set, performing time-frequency domain transformation processing on a three-axis vibration acceleration signal to separate and extract vibration characteristic frequency components dominated by electrodynamic pulsation, and generating a bus vibration characteristic spectrum set; on the basis, operation modal analysis and electrodynamic force frequency identification are carried out, the dominant inherent frequency and the electrodynamic force main excitation frequency of the bus structure are solved in real time, the approaching degree and the energy coupling degree between the dominant inherent frequency and the electrodynamic force main excitation frequency are calculated, and a mechanical resonance risk index set is generated; based on this, multi-dimensional evaluation of the resonance risk state is performed, and a closed bus resonance risk state report for guiding operation and maintenance intervention is generated accordingly. According to the invention, in the enclosed bus operation state detection process, the safety and reliability of power grid key equipment are significantly improved.

Description

Technical Field

[0001] This application relates to the field of enclosed busbar detection, and in particular to an intelligent detection method and system for the operating status of enclosed busbars. Background Technology

[0002] In the field of operation and maintenance of key equipment in large power systems, the phase-separated enclosed busbar, as the core hub for power transmission between power plants and substations, is directly related to the reliability of the main grid, the security of regional power supply, and even the safety of the generator units themselves in terms of its long-term safe and stable operation. It is a vital primary equipment for ensuring the safe and efficient operation of modern large power grids.

[0003] However, existing busbar condition monitoring methods mainly rely on periodic manual inspections or simple vibration amplitude threshold alarms. They lack a mechanism for real-time synchronous sensing and deep coupling analysis of multiple physical quantities. This not only makes it difficult to achieve accurate early warning in the early stages of mechanical resonance risk accumulation, but may also only provide passive alarms after structural damage occurs, missing the best intervention opportunity and thus posing a major safety hazard that could lead to electrical short circuits or forced shutdowns. Summary of the Invention

[0004] This application provides an intelligent detection method and system for the operating status of a closed busbar to solve the above-mentioned technical problems.

[0005] Firstly, this application provides an intelligent detection method for the operating status of a closed busbar, the method comprising:

[0006] A multi-source real-time monitoring dataset of the isolated closed busbar is acquired. Based on this dataset, the triaxial vibration acceleration signal is processed by time-frequency domain transformation to separate and extract the vibration characteristic frequency components dominated by electrodynamic pulsation, generating a busbar vibration characteristic spectrum set. Based on this spectrum set, operational modal analysis and electrodynamic frequency identification are performed to calculate the dominant natural frequency and main electrodynamic excitation frequency of the busbar structure in real time, and the convergence and energy coupling degree between them are calculated to generate a mechanical resonance risk index set. Based on this index set, a multi-dimensional assessment of the resonance risk status is conducted, and a closed busbar resonance risk status report is generated to guide operation and maintenance intervention.

[0007] The above technical solutions enable early and accurate warning and quantitative assessment of mechanical resonance risks in enclosed busbars, driving a shift in operation and maintenance models from reactive maintenance to predictive maintenance. This effectively avoids sudden faults caused by resonance, significantly improves the safety and reliability of critical power grid equipment, reduces unplanned downtime losses, and ensures continuous and stable power supply.

[0008] Optionally, generating the busbar vibration characteristic spectrum set includes: acquiring a multi-source real-time monitoring dataset arranged at specific structural locations on the isolated closed busbar shell, the multi-source real-time monitoring dataset including the synchronously acquired triaxial vibration acceleration signal, three-phase current waveform signal, and busbar shell temperature signal; calculating the fundamental frequency of the electrodynamic force and its harmonic components based on the three-phase current waveform signal as the theoretical excitation spectrum; performing time-frequency domain transformation on the triaxial vibration acceleration signal to obtain the measured vibration spectrum; and decoupling the measured vibration spectrum from the theoretical excitation spectrum based on the busbar shell temperature signal to filter out broadband noise caused by environmental background vibration and internal particle impact, and separating the narrowband resonance risk frequency component with structural mode modulation characteristics caused by the conductor electrodynamic force transmitted to the shell through the insulator support structure, thereby generating the busbar vibration characteristic spectrum set.

[0009] Optionally, acquiring the multi-source real-time monitoring dataset arranged at specific structural locations on the outer shell of the phase-separated enclosed busbar includes: based on the structural dynamic characteristics and vibration transmission path of the phase-separated enclosed busbar, simultaneously deploying sensor groups at the following key locations on the outer shell to collect signals: deploying support point monitoring positions at the mechanical anchoring points corresponding to the internal support insulators penetrating the outer shell; deploying mid-span monitoring positions at the mid-span location of the outer shell between two adjacent support insulator anchoring points; deploying expansion joint monitoring positions at the bellows or expansion joint connection sections on the outer shell used to compensate for thermal expansion and contraction; and simultaneously acquiring the multi-source real-time monitoring dataset at the support point monitoring positions, mid-span monitoring positions, and expansion joint monitoring positions.

[0010] Optionally, separating the narrowband resonance risk frequency components with structural mode modulation characteristics caused by the transmission of conductor electrodynamic force to the shell via the insulator support structure includes: constructing a multi-point vibration spectrum matrix based on vibration signals synchronously collected from the support point monitoring position, the mid-span monitoring position, and the expansion joint monitoring position; performing collaborative analysis on the multi-point vibration spectrum matrix to identify narrowband frequency components with significant energy at the support point monitoring position, exhibiting an amplitude amplification effect at the mid-span monitoring position, and showing a specific phase lag relationship at the expansion joint monitoring position, as candidate structural mode frequencies; matching the candidate structural mode frequencies with the theoretical excitation spectrum to screen out frequency components that simultaneously have an integer or fractional multiple relationship with the fundamental frequency or harmonic component of the electrodynamic force, and whose energy changes synchronously with the bus load current; performing dynamic stiffness compensation correction on the screened frequency components based on the bus shell temperature signal, and finally separating the narrowband resonance risk frequency components characterizing the mechanical resonance excited by the electrodynamic force through the insulator-shell path.

[0011] Optionally, generating the mechanical resonance risk index set includes: based on the narrowband resonance risk frequency components separated from the bus vibration characteristic spectrum set and their spatial distribution characteristics, using operational modal analysis technology to identify and track the dominant natural frequency drift trajectory of the phase-separated enclosed bus shell structure in real time; based on the theoretical excitation spectrum and the narrowband energy characteristics that change synchronously with the current, identifying the main conductive dynamic excitation frequency under the current operating condition in real time; analyzing the frequency convergence between the dominant natural frequency and the main conductive dynamic excitation frequency and the electro-mechanical energy coupling degree characterizing the energy transfer efficiency between the two, and generating the mechanical resonance risk index set containing the frequency convergence and the electro-mechanical energy coupling degree.

[0012] Optionally, the real-time identification and tracking process of the dominant natural frequency drift trajectory includes: using the synchronous vibration signals of the support point monitoring position, the mid-span monitoring position, and the expansion joint monitoring position as input, analyzing the multi-order working mode shapes and frequencies of the busbar shell under operating conditions based on random subspace identification or frequency domain decomposition algorithms; integrating the preset offline impact test results of the key bolt connection points of the busbar shell to establish a benchmark reference value for the structure's natural frequency, and selecting the dominant natural frequencies related to the overall bending or torsional main vibration modes accordingly; performing thermal stiffness correction on the dominant natural frequencies based on the busbar shell temperature signal, recording its dynamic changes in time sequence, and forming the dominant natural frequency drift trajectory.

[0013] Optionally, the analysis process of the frequency convergence and electromechanical energy coupling degree includes: comparing the absolute value of the difference between the main conductor dynamic excitation frequency and the dominant natural frequency with a safety margin threshold set according to the structural damping characteristics to obtain a normalized frequency convergence; extracting the vibration energy in the bus vibration characteristic spectrum set and the corresponding electrodynamic excitation energy in the theoretical excitation spectrum within the narrow-band neighborhood of the main conductor dynamic excitation frequency, and using the ratio of the two as the electromechanical energy coupling degree; and combining the normalized frequency convergence and the electromechanical energy coupling degree with a vibration energy margin coefficient reflecting the current overall vibration level, performing weighted fusion to generate a single comprehensive index for characterizing the instantaneous risk of mechanical resonance, thus constituting the mechanical resonance risk index set.

[0014] Optionally, the process of generating the closed bus resonance risk status report includes: assessing the immediate risk level of resonance based on the frequency convergence, the electromechanical energy coupling degree, and the vibration energy margin coefficient; assessing the risk development rate and accumulation trend based on the immediate risk level and the temporal change trend of the dominant natural frequency drift trajectory; and performing a multi-dimensional resonance risk status assessment based on the immediate risk level and the risk development rate and accumulation trend, and generating the closed bus resonance risk status report.

[0015] Optionally, the implementation of multi-dimensional resonance risk status assessment includes: if the immediate risk level indicates the presence of significant electrodynamic excitation frequency energy, and the risk accumulation trend is stable and the dominant natural frequency drift trajectory is stable, then it is determined to be an observation-level warning; if the immediate risk level continues to rise, and the risk accumulation trend shows that the dominant natural frequency drift trajectory continues to approach the main conductive electrodynamic excitation frequency, causing the frequency convergence of the two to be lower than a preset safety dynamic margin, then it is determined to be a warning level warning; if the immediate risk level exceeds the absolute safety threshold, then it is determined to be an action-level alarm.

[0016] Secondly, this application provides an intelligent detection system for the operating status of a closed busbar, the system comprising:

[0017] The vibration feature analysis module is used to acquire a multi-source real-time monitoring dataset of the phase-separated closed busbar. Based on the multi-source real-time monitoring dataset, the triaxial vibration acceleration signal is processed by time-frequency domain transformation to separate and extract the vibration feature frequency components dominated by electrodynamic pulsation, and generate a busbar vibration feature spectrum set.

[0018] The resonance risk analysis module is used to perform operational mode analysis and electrodynamic frequency identification based on the bus vibration characteristic spectrum set, calculate the dominant natural frequency and main electrodynamic excitation frequency of the bus structure in real time, calculate the convergence degree and energy coupling degree between the two, and generate a set of mechanical resonance risk indices.

[0019] The status report generation module is used to perform a multi-dimensional assessment of the resonance risk status based on the mechanical resonance risk index set, and generate a closed bus resonance risk status report to guide operation and maintenance intervention. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application;

[0022] Figure 2 A flowchart illustrating an intelligent detection method for the operating status of a closed busbar, as provided in an embodiment of this application;

[0023] Figure 3 This is a schematic diagram of the structure of an intelligent detection system for the operating status of a closed busbar, provided in one embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0025] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0026] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0027] Existing busbar condition monitoring methods mainly rely on periodic manual inspections or simple vibration amplitude threshold alarms. They lack a mechanism for real-time synchronous sensing and deep coupling analysis of multiple physical quantities. This not only makes it difficult to achieve accurate early warning in the early stages of mechanical resonance risk accumulation, but may also only provide passive alarms after structural damage has occurred, missing the best intervention opportunity and thus posing a major safety hazard that could lead to electrical short circuits or forced shutdowns.

[0028] Based on this, this application provides an intelligent detection method and system for the operating status of enclosed busbars. First, time-frequency analysis technology is used to accurately separate the characteristic frequency components dominated by electrodynamics from the vibration signal. Second, based on operational modal analysis, the drift trajectory of the structure's natural frequency is identified in real time, and the excitation frequency of the main conductive force is locked in conjunction with current data. Finally, the frequency convergence and energy coupling degree between the two are innovatively calculated, and a dynamic mechanical resonance risk index is generated. Based on this, a multi-dimensional assessment is performed, automatic graded early warning is issued, and an enclosed busbar resonance risk status report is output to the inspection personnel. This solution achieves early and accurate early warning and quantitative assessment of the mechanical resonance risk of enclosed busbars, promoting the transformation of operation and maintenance models from post-maintenance to predictive maintenance. It can effectively avoid sudden faults caused by resonance, significantly improve the safety and reliability of key power grid equipment, reduce unplanned downtime losses, and ensure the continuous and stable power supply.

[0029] Figure 1 This is a schematic diagram illustrating an application scenario provided by this application. In the process of detecting the operating status of a closed busbar, the method provided in this application significantly improves the safety and reliability of key power grid equipment, while reducing unplanned downtime losses and ensuring a continuous and stable power supply.

[0030] Specifically, the method of this application is applied to any server that communicates with a monitoring sensor array. The server acquires multi-source real-time monitoring datasets provided by the monitoring sensor array. First, time-frequency analysis technology is used to accurately separate the characteristic frequency components dominated by electrodynamics from the vibration signal. Second, based on operational modal analysis, the drift trajectory of the structure's natural frequency is identified in real time, and the excitation frequency of the main conductive force is locked in conjunction with current data. Finally, the frequency convergence and energy coupling degree between the two are innovatively calculated, and a dynamic mechanical resonance risk index is generated. Based on this, a multi-dimensional assessment is performed, automatic graded early warning is issued, and a closed bus resonance risk status report is output to the inspection personnel.

[0031] For specific implementation details, please refer to the following examples.

[0032] Figure 2 This is a flowchart illustrating an intelligent detection method for the operating status of a closed busbar, provided as an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0033] S201. Obtain the multi-source real-time monitoring dataset of the isolated phase closed bus. Based on the multi-source real-time monitoring dataset, perform time-frequency domain transformation processing on the triaxial vibration acceleration signal to separate and extract the vibration characteristic frequency components dominated by electrodynamic pulsation, and generate the bus vibration characteristic spectrum set.

[0034] A phase-separated enclosed busbar refers to a busbar system where each phase conductor is encased in an independent metal shell. It is primarily used in large power plants and substations to transmit high currents, with the metal shell serving to isolate phases, prevent electric shock, and reduce electromagnetic interference. A multi-source real-time monitoring dataset refers to a collection of time-series data synchronously acquired from multiple sensors deployed at different physical locations and monitoring dimensions along the busbar to comprehensively perceive its operating status. The data originates from an array of monitoring sensors deployed on the busbar shell. A triaxial vibration acceleration signal refers to the sensor output signal capable of simultaneously measuring the vibration acceleration of the busbar shell in three orthogonal directions (e.g., axial, radial, and tangential) in space, used to comprehensively capture the complex vibration patterns of the shell. Time-frequency domain transformation processing refers to converting vibration signals from simple time-domain waveforms (amplitude varying with time) to the time-frequency domain (e.g., through short-time Fourier transform, wavelet transform, etc.) to simultaneously observe the frequency components of the signal and their evolution over time. Electrodynamic pulsation refers to the repulsive or attractive forces generated when a large current flows through adjacent conductors, based on electromagnetic principles (e.g., Ampere's law). A bus vibration characteristic spectrum set can refer to one or more sets of spectrum (or time spectrum) data that are obtained after specific signal processing and can reflect the structural vibration characteristics caused by electrodynamic excitation and have potential resonance risks.

[0035] Specifically, the long-term safe and stable operation of isolated phase busbars, as key equipment for transmitting large currents in power plants and substations, is crucial. Busbars vibrate under strong electrodynamic forces, and continuous abnormal vibration is a major cause of structural fatigue, loose connections, and even insulation failures. Traditional detection methods mainly rely on manual periodic inspections or simple vibration amplitude exceeding-limit alarms. These methods have significant time lags and cannot distinguish whether the vibration is caused by harmful electrodynamic resonance or environmental interference (such as the operation of nearby equipment or wind load), often resulting in false alarms or missed alarms. To address this problem, a method is proposed that simultaneously collect multi-source data such as current, vibration, and temperature, and perform in-depth time-frequency domain processing on the vibration signals. The core objective is to use the current signal as a "cause" label to accurately separate and extract the unique vibration characteristic frequency component excited by the specific physical cause of electrodynamic pulsation from the complex mixed vibration "effect," thereby generating a pure and clearly directional busbar vibration characteristic spectrum set. This is equivalent to equipping the system with a keen eye, achieving a fundamental shift from monitoring all vibrations to focusing on high-risk vibrations, laying an irrefutable data foundation for subsequent accurate risk assessment. The aforementioned process of generating feature spectrum sets from multi-source data can be used as a signal preprocessing and feature extraction service within the entire intelligent detection system, providing high-quality, high signal-to-noise ratio input to upper-level analysis modules through standardized data interfaces.

[0036] S202. Based on the bus vibration characteristic spectrum set, perform operational modal analysis and electrodynamic frequency identification, calculate the dominant natural frequency and main electrodynamic excitation frequency of the bus structure in real time, and calculate the convergence degree and energy coupling degree between the two to generate a mechanical resonance risk index set.

[0037] Operational modal analysis is a technique that identifies the dynamic characteristics (such as natural frequencies, damping ratios, and mode shapes) of a structure based solely on its vibration response data under normal operating conditions (energized and loaded). Electrodynamic frequency identification accurately determines the specific frequency value of the dominant electrodynamic pulsation under the current operating condition by correlating electrical signals and vibration characteristics. The dominant natural frequency can be the first or several natural vibration frequencies that are most easily and likely to be excited in the bus structure under operating conditions and have the greatest impact on the overall structural safety; it usually corresponds to the low-order principal mode shape of overall bending or torsion. The main electrodynamic excitation frequency can be the frequency component with the strongest energy and the most significant excitation effect on the structure in the electrodynamic pulsation generated under the current operating current (which may include fundamental and harmonic frequencies). The degree of convergence can be a quantitative indicator of the closeness between the dominant natural frequency and the main electrodynamic excitation frequency, usually characterized by the frequency difference or normalized ratio between the two. Energy coupling can be a quantitative indicator of the efficiency of converting electrodynamic excitation energy into structural vibration energy at a specific frequency, reflecting the sensitivity of the structure to excitation at that frequency. The mechanical resonance risk index set can be a collection of one or more quantitative risk indicators such as convergence degree and energy coupling degree, used to comprehensively and quantitatively characterize the instantaneous risk level of mechanical resonance.

[0038] Specifically, the key to identifying the risk of mechanical resonance lies in capturing the dynamic relationship between the excitation frequency and the structure's natural frequency. Traditional methods rely on fixed natural frequency values ​​in design manuals or offline impact test results during shutdown maintenance. However, in actual operation, the natural frequency of the busbar will "drift" due to temperature changes, material stress relaxation, and changes in bolt preload. Static frequency data cannot reflect this dynamic process, making it impossible to warn of the risks that accumulate silently due to the slow convergence of frequencies. At the same time, judging only whether the frequencies are close is one-sided, because even if the frequencies coincide, if the structural damping is large enough or the energy transfer path is obstructed, it will not cause serious resonance. To address these core issues, modal analysis technology is used to "diagnose" the current, "live" dominant natural frequency of the structure in real time online, and correlation analysis is used to confirm the real-time main electrodynamic excitation frequency. More importantly, a dual quantitative index of convergence degree and energy coupling degree is innovatively introduced: the former scientifically assesses the possibility of frequency coincidence, and the latter quantitatively assesses the structure's energy amplification efficiency for the excitation. By calculating and fusing these two indicators to generate a set of mechanical resonance risk indices, a leap has been achieved in assessing resonance risk from "qualitative guessing" to "quantitative calculation," and from "single-condition judgment" to "multi-factor comprehensive evaluation," making the risk assessment results both physically rigorous and engineering practical. The calculation process of the aforementioned core risk indices can be encapsulated as a real-time risk calculation engine service for the entire system. It receives feature spectrum sets and outputs standardized risk indicators, serving as the brain and core judgment unit of the entire method.

[0039] S203. Based on the mechanical resonance risk index set, conduct a multi-dimensional assessment of the resonance risk status and generate a closed bus resonance risk status report to guide operation and maintenance intervention.

[0040] A closed bus resonance risk status report can be a comprehensive output document generated by structuring and visualizing the assessment results, including risk level, key parameters, changing trends, risk positioning, and specific action recommendations.

[0041] Specifically, the ultimate value of intelligent detection lies not in displaying complex curves and numbers, but in driving effective operation and maintenance actions. Traditional enclosed busbar monitoring systems often only provide raw data or simple exceedance alarms. Maintenance personnel must rely on their own experience to interpret the data, assess the severity, and formulate measures. This decision-making process is slow and dependent on individual skill, easily leading to delays when risks evolve rapidly. To address this last-mile obstacle "from data to decision," a multi-dimensional risk index assessment mechanism is designed. This mechanism not only focuses on the instantaneous magnitude of the risk but also comprehensively analyzes its changing trends (acceleration, stabilization, or mitigation), spatial distribution (which location is most severe), and dominant factors (whether frequency convergence or increased coupling), thus providing a panoramic and three-dimensional depiction of the risk status. Based on this, the system can automatically generate an enclosed busbar resonance risk status report that can directly guide on-site work. This report, in language understandable even to non-technical experts, clearly indicates the current risk level (e.g., observation, alert, action), key evidence, and suspected locations. This achieves a closed loop from "passive alarm" to "proactive decision support," directly transforming advanced algorithmic analysis capabilities into frontline maintenance productivity. The aforementioned assessment and report generation process can be structured as a systematic decision support and report generation service. It receives risk indices, invokes rules and knowledge bases, and ultimately outputs structured action guidelines, serving as a bridge connecting intelligent analysis and on-site practice.

[0042] The method provided in this embodiment firstly uses time-frequency analysis technology to accurately separate the characteristic frequency components dominated by electrodynamics from the vibration signal; secondly, based on operational modal analysis, the drift trajectory of the structure's natural frequency is identified in real time, and the excitation frequency of the main conductor is locked in combination with current data; finally, the frequency convergence and energy coupling degree between the two are innovatively calculated, and a dynamic mechanical resonance risk index is generated by fusing them. Based on this, a multi-dimensional assessment is performed, automatic graded early warning is issued, and a resonance risk status report of the enclosed busbar is output to the inspection personnel. This solution achieves early and accurate early warning and quantitative assessment of the mechanical resonance risk of enclosed busbars, promoting the transformation of operation and maintenance mode from post-maintenance to predictive maintenance. It can effectively avoid sudden failures caused by resonance, significantly improve the safety and reliability of key power grid equipment, reduce unplanned downtime losses, and ensure the continuous and stable power supply.

[0043] In some embodiments, a multi-source real-time monitoring dataset is acquired at specific structural locations of the isolated busbar enclosure. The multi-source real-time monitoring dataset includes synchronously acquired triaxial vibration acceleration signals, three-phase current waveform signals, and busbar enclosure temperature signals. Based on the three-phase current waveform signals, the fundamental frequency of the electrodynamic force and its harmonic components are calculated as the theoretical excitation spectrum. The triaxial vibration acceleration signals are transformed in the time and frequency domain to obtain the measured vibration spectrum. Based on the busbar enclosure temperature signals, the measured vibration spectrum and the theoretical excitation spectrum are decoupled and analyzed to filter out broadband noise caused by environmental background vibration and internal particle impacts, and to separate the narrowband resonance risk frequency components with structural mode modulation characteristics caused by the conductor electrodynamic force transmitted to the enclosure through the insulator support structure, thereby generating a busbar vibration characteristic spectrum set.

[0044] The three-phase current waveform signal can reflect the time-varying three-phase current inside the isolated busbar, and its waveform characteristics directly determine the magnitude and frequency of the electrodynamic force. The busbar shell temperature signal can be temperature data collected from the surface of the isolated busbar shell, reflecting the thermal state of the busbar during operation. Temperature changes affect the structural stiffness of the busbar, and thus its natural frequency. The fundamental frequency of the electrodynamic force can be the fundamental frequency of the electrodynamic force generated by the interaction of the three-phase currents inside the busbar, and is the core frequency component that excites the busbar structure. The theoretical excitation spectrum can be a frequency-energy distribution spectrum constructed with the fundamental frequency of the electrodynamic force and its harmonic components as the core, and is a theoretical reference benchmark characterizing the excitation characteristics of the electrodynamic force on the busbar structure. The measured vibration spectrum can be the actual vibration frequency-energy distribution spectrum obtained by performing time-frequency domain transformation on the triaxial vibration acceleration signal, which includes all frequency components of the busbar shell vibration (including effective signals and noise signals). Environmental background vibration can be the busbar shell vibration caused by external environmental factors such as the operation of equipment around the busbar and site vibration. Its frequency distribution is relatively wide and does not reflect the essential characteristics of the busbar structure and the electrodynamic force interaction. Internal particle impaction can be caused by a small amount of impurities (such as metal particles, dust, etc.) remaining inside the isolated busbar impacting the outer casing during busbar operation due to vibration or airflow. The signal manifests as irregular broadband noise. The insulator support structure can be a structural component connecting the busbar conductor and the outer casing, providing insulation and mechanical support. Narrowband resonance risk frequency components can be vibrational frequency components caused by the conductor's electrodynamic force transmitted to the outer casing via the insulator support structure, with frequencies concentrated within a specific range and related to the structural modes.

[0045] Specifically, traditional enclosed busbar vibration monitoring often only collects a single vibration signal, ignoring key influencing factors such as three-phase current and temperature. It cannot distinguish between effective vibrations caused by electrodynamics and environmental background vibrations (such as 8-25Hz vibrations generated by surrounding transformers) and internal particle impact noise. It often leads to misjudgment of characteristic frequencies. For example, a power station once misjudged broadband noise generated by airflow disturbance as a resonance signal because it did not combine the current signal to construct an excitation reference, resulting in ineffective operation and maintenance. Moreover, it lacks a temperature correction mechanism. When the temperature of the busbar shell rises from 25°C to 40°C, the natural frequency drift caused by the change in structural stiffness will be misjudged as a risk signal, which seriously affects the accuracy of monitoring. To address the above issues, this step first, based on the dynamic characteristics of the busbar structure, simultaneously deploys triaxial accelerometers, current sensors, and temperature sensors at the support point monitoring location (insulator anchorage point), mid-span monitoring location (between two insulators), and expansion joint monitoring location. This simultaneously collects triaxial vibration acceleration signals (e.g., instantaneous peak value of 3.2 m / s² on the Y-axis), three-phase current waveform signals (e.g., line current fundamental frequency of 50 Hz), and busbar shell temperature signals (e.g., real-time 38°C). Then, the fundamental frequency of the electrodynamic force (e.g., 100 Hz) is calculated based on the three-phase current waveform. The theoretical excitation spectrum is constructed by taking the second and third harmonic components (such as 200Hz and 300Hz). The vibration signal is then subjected to Fourier transform for time-frequency domain transformation to obtain the measured vibration spectrum containing broadband noise. The influence of structural stiffness is then corrected by combining the temperature signal. The measured spectrum is decoupled from the theoretical excitation spectrum by comparison, and environmental background vibration (such as broadband noise near 20Hz) and particle impact interference are filtered out. Finally, the narrowband resonance risk frequency component with structural mode modulation characteristics near 100Hz is separated to generate the bus vibration characteristic spectrum set.

[0046] The method provided in this embodiment clarifies the excitation frequency range through current signals, corrects frequency offsets with temperature signals, and accurately filters out irrelevant noise. This ensures that the extracted narrowband frequency components truly reflect the interaction between electrodynamics and structural modes, which is the core data foundation for subsequent resonance risk assessment and directly determines the reliability of the test results.

[0047] In some embodiments, based on the structural dynamics characteristics and vibration transmission path of the phase-separated enclosed busbar, sensor groups are simultaneously deployed at the following key locations on the outer shell to collect signals: support point monitoring positions are deployed at the mechanical anchoring points corresponding to the internal support insulators penetrating the outer shell; mid-span monitoring positions are deployed at the mid-span of the outer shell between two adjacent support insulator anchoring points; expansion joint monitoring positions are deployed at the bellows or expansion joint connection sections on the outer shell used to compensate for thermal expansion and contraction; and multi-source real-time monitoring datasets are simultaneously collected at the support point monitoring positions, mid-span monitoring positions, and expansion joint monitoring positions.

[0048] Structural dynamics characteristics can be inherent properties of the isolated enclosed busbar shell, including stiffness distribution, natural frequencies, and mode shapes, which determine the vibration response of the shell when subjected to electrodynamic excitation. The vibration transmission path can be the path by which vibrations induced by conductor electrodynamics are transmitted from the source (conductor) through the insulator support structure to the shell, and then propagate at different locations within the shell. Support point monitoring positions can be monitoring locations located at the mechanical anchor points where internal support insulators penetrate the shell; these are key nodes for electrodynamic vibration transmission, and the vibration signals are directly related to the connection state between the insulator and the shell. Mid-span monitoring positions can be monitoring locations located in the middle of the shell between two adjacent support insulator anchor points; this location is the amplitude amplification area of ​​the shell vibration and is most sensitive to resonance response. Expansion joint monitoring positions can be monitoring locations located at the bellows or expansion joint connection sections of the shell used to compensate for thermal expansion and contraction; the structural stiffness at this location is relatively weak, and the vibration phase characteristics are unique.

[0049] Specifically, traditional vibration monitoring of enclosed busbars often employs random sensor placement or only single-location sensor deployment, failing to consider the busbar's structural dynamics and vibration transmission path, resulting in unrepresentative data. For example, one substation only deployed sensors at support points; due to the high stiffness and suppressed vibration amplitude at these locations, it failed to capture the three-fold amplitude vibration caused by resonance at the mid-span, ultimately leading to long-term resonance and busbar shell cracking. Another substation overlooked expansion joint monitoring locations, failing to detect the structural loosening issues reflected by phase anomalies at these locations. To address these issues, this step first uses finite element simulation technology to analyze the structural dynamics (such as stiffness distribution) and vibration transmission path of the phase-separated enclosed busbar, clarifying the response patterns of vibration signals at different locations. Subsequently, sensor groups are deployed at three key locations: the support point monitoring location is set at the anchor point where the insulator penetrates the shell; the mid-span monitoring location is selected at 4 meters in the middle of the distance between two insulators (e.g., 8 meters); and the expansion joint monitoring location is fixed at both ends of the corrugated pipe, avoiding the core area of ​​the expansion joint. The sensor group includes a triaxial vibration acceleration sensor (measurement range ±50m / s², sampling frequency 2000Hz), a Rogowski coil current sensor, and a temperature sensor with an accuracy of ±0.1℃. All sensors are connected through a standardized data acquisition module and a unified timestamp synchronization mechanism is set up (synchronization accuracy ≤1ms). Finally, triaxial vibration acceleration signals (such as X-axis instantaneous value 2.8m / s²), three-phase current waveform signals, and busbar shell temperature signals (such as real-time 35℃) are synchronously acquired at three monitoring positions and integrated to form a multi-source real-time monitoring dataset.

[0050] In the method provided in this embodiment, the support point is the starting point of vibration transmission, and the signal reflects the transmission efficiency of the insulator; the mid-span is the vibration amplitude amplification zone, which sensitively captures resonance; the expansion joint has weak stiffness and unique phase characteristics. Synchronous point acquisition can completely cover the entire vibration transmission chain, making up for the one-sidedness of traditional single-point data, ensuring the spatial correlation and integrity of multi-source data, and providing a reliable foundation for subsequent multi-point spectrum matrix construction and resonance feature separation.

[0051] In some embodiments, a multi-point vibration spectrum matrix is ​​constructed based on vibration signals synchronously collected from support point monitoring positions, mid-span monitoring positions, and expansion joint monitoring positions. The multi-point vibration spectrum matrix is ​​then analyzed collaboratively to identify narrowband frequency components that exhibit significant energy at the support point monitoring positions, amplitude amplification at the mid-span monitoring positions, and specific phase lag at the expansion joint monitoring positions, serving as candidate structural modal frequencies. These candidate structural modal frequencies are matched with the theoretical excitation spectrum to screen out frequency components that simultaneously have an integer or fractional multiple relationship with the fundamental frequency or harmonic components of the electrodynamic force, and whose energy changes synchronously with the bus load current. Based on the bus shell temperature signal, the screened frequency components are dynamically stiffened and corrected, ultimately separating out the narrowband resonance risk frequency components characterizing mechanical resonance excited by the electrodynamic force through the insulator-shell path.

[0052] A multi-point vibration spectrum matrix can be a three-dimensional data matrix formed by integrating the vibration signals synchronously collected from the support point monitoring position, mid-span monitoring position, and expansion joint monitoring position, after time-frequency domain transformation, into the spectrum data of each monitoring point according to the measurement point dimension, frequency dimension, and energy dimension. This matrix is ​​used to visually present the distribution characteristics of vibration frequency components at different spatial locations. The amplitude amplification effect can be seen in the significant increase in vibration amplitude of the same frequency component at the mid-span monitoring position compared to the support point monitoring position. This is because the mid-span position is the amplitude-sensitive area of ​​shell vibration, where structural modal vibrations are easily amplified. A specific phase lag relationship can be seen in the fixed lag angle (e.g., 30°-60°) of the vibration phase of the same frequency component at the expansion joint monitoring position compared to the support point monitoring position. Candidate structural modal frequencies can be narrowband frequency components selected through co-analysis that exhibit significant energy (support point), amplitude amplification (mid-span), and phase lag (expansion joint) characteristics. An integer multiple relationship can be seen in the ratio of the candidate structural modal frequency to the fundamental frequency or harmonic component of the electrodynamics being an integer (e.g., 1, 2, 3, etc.). Fractional ratios can be the ratio of the candidate structural modal frequency to the fundamental frequency or harmonic component of the electrodynamic force (e.g., 1 / 2, 1 / 3, etc.). Dynamic stiffness compensation correction can be based on the correlation between the busbar shell temperature signal and structural stiffness, adjusting the selected frequency components numerically to offset the impact of temperature-induced changes in structural stiffness on the frequency. Narrowband resonance risk frequency components can be the vibration signal components within a specific frequency range that are ultimately separated, transmitted from the conductor's electrodynamic force through the insulator support structure to the shell, coupled with structural modal characteristics, have a fixed correlation with the electrodynamic excitation frequency, and have undergone temperature correction.

[0053] Specifically, traditional methods for extracting vibration characteristics of enclosed busbars often rely on single-point signal analysis, neglecting the spatial distribution of vibration, failing to establish a correlation with electrodynamic excitation, and lacking a temperature compensation mechanism, which easily leads to misjudgments. For example, one substation misjudged the 25Hz broadband noise generated by the operation of the ambient fan as a resonance risk frequency based solely on the mid-span monitoring signal, resulting in unnecessary shutdowns for maintenance. Another substation, lacking temperature compensation, misjudged the frequency drift caused by changes in structural stiffness when the busbar casing temperature was raised from 25°C to 45°C as an increased resonance risk. To address the above issues, this step first collects triaxial vibration acceleration signals (e.g., the instantaneous Y-axis vibration value of the support point is 3.5 m / s²) simultaneously from three monitoring locations: the support point, mid-span, and expansion joint. After Fourier transform to complete the time-frequency domain conversion, a multi-point vibration spectrum matrix is ​​constructed according to the dimensions of "3 measuring points - 0-1000 Hz frequency range - vibration energy value". The matrix is ​​then subjected to collaborative analysis to screen out narrowband frequency components (e.g., 90 Hz) that exhibit energy values ​​exceeding the average energy threshold by 3 times at the support point, amplitude values ​​at the mid-span that are more than twice that at the support point, and phase lags the expansion joint by 30°-60°. z and 180Hz were selected as candidate structural modal frequencies. The candidate frequencies were matched with the theoretical excitation spectrum (electrodynamic fundamental frequency 100Hz, second harmonic 200Hz) to screen out components with a 0.9-fold correlation and whose energy changes with the bus load current (energy increases by 40% when it increases from 800A to 1200A). Combined with the real-time temperature of the bus shell of 42℃, dynamic stiffness compensation correction was performed according to the rule that the structural stiffness decreases by 5% for every 10℃ increase in temperature (e.g., 90Hz is corrected to 92Hz). Finally, the narrowband resonance risk frequency components were separated.

[0054] By constructing a multi-point vibration spectrum matrix and using the spatial characteristics of vibration at different monitoring locations to screen candidate frequencies, and then matching the electrodynamic spectrum and performing temperature correction, the shortcomings of traditional technologies, such as one-sidedness, lack of correlation verification, and lack of temperature compensation, can be overcome. This ensures that the separated frequency components are structural modal vibrations caused by electrodynamic forces through the insulator-shell path, providing accurate core data for subsequent resonance risk assessment.

[0055] In some embodiments, based on the narrowband resonance risk frequency components separated from the bus vibration characteristic spectrum and their spatial distribution characteristics, the dominant natural frequency drift trajectory of the isolated closed bus shell structure is identified and tracked in real time using operational modal analysis technology; based on the theoretical excitation spectrum and the narrowband energy characteristics that change synchronously with the current, the main conductive dynamic excitation frequency under the current operating condition is identified in real time; the frequency convergence between the dominant natural frequency and the main conductive dynamic excitation frequency and the electromechanical energy coupling degree characterizing the energy transfer efficiency between the two are analyzed to generate a mechanical resonance risk index set that includes the frequency convergence and the electromechanical energy coupling degree.

[0056] Operational modal analysis (EMA) is a technique that identifies structural modal parameters (natural frequencies, mode shapes, and damping) by collecting vibration response signals during normal equipment operation. It requires no additional excitation and is suitable for online real-time monitoring scenarios. The isolated enclosed busbar shell structure serves as the external protection and support structure for the isolated enclosed busbar; its natural frequencies and stiffness characteristics directly influence the generation and development of resonance risk. The dominant natural frequency drift trajectory is the continuous trajectory of the core natural frequency of the isolated enclosed busbar shell structure (related to the overall bending and torsional main modes) dynamically changing over time under the influence of factors such as temperature changes and load fluctuations, reflecting the real-time state of structural stiffness. The main conductor dynamic excitation frequency is the frequency component with the strongest energy and most significant excitation effect on the busbar structure's vibration among the fundamental frequency and its harmonic components under the current operating conditions. Frequency convergence is the degree of proximity between the main conductor dynamic excitation frequency and the dominant natural frequency, a quantified index obtained after normalization, reflecting the probability of frequency overlap between the two. Electro-mechanical energy coupling degree can be a quantitative indicator that characterizes the efficiency of the transfer of electrodynamic excitation energy to the vibration energy of the bus structure, reflecting the intensity of energy interaction between electrodynamic and structural modes.

[0057] Specifically, traditional closed bus resonance risk assessment has three major drawbacks: static measurement of natural frequency ignores dynamic drift. For example, a power station set a threshold based on an initial natural frequency of 100Hz, failing to detect that the frequency drifted to 105Hz due to the temperature rising to 60℃ during operation, coinciding with the electrodynamic excitation frequency and causing resonance; blindly analyzing all electrodynamic frequencies without focusing on the dominant component increases the probability of misjudgment; and assessing only through the single dimension of frequency difference. For example, a substation was misjudged due to a frequency difference of 3Hz, when in fact the energy coupling degree was only 0.2 and there was no resonance risk. To address the above issues, this step first extracts the narrowband resonance risk frequency components (such as 95Hz and 190Hz) and their spatial distribution characteristics (mid-span amplitude is 2.5 times that of the support point, and the expansion joint phase lags by 40°) from the busbar vibration characteristic spectrum set. Using a random subspace identification algorithm with synchronous vibration signals from three monitoring positions as input, the dominant natural frequency (initially 100Hz) related to the overall bending mode shape is identified. Combined with the real-time busbar shell temperature of 55℃, the natural frequency is corrected to 103.5Hz according to the rule that the natural frequency shifts by 1Hz for every 10℃ increase in temperature. Recording at a frequency of once every 15 minutes forms a 24-hour drift trajectory of the dominant natural frequency from 99Hz to 104Hz. Then, the theoretical excitation spectrum (electrodynamic fundamental frequency 100Hz, harmonics 200Hz) is retrieved. Based on the characteristic that the energy of the 95Hz component increases by 55% synchronously with the current increasing from 900A to 1300A, 100Hz is determined as the conducting electrodynamic excitation frequency. Finally, the frequency convergence (|100-103.5| compared with the safety margin threshold of 5Hz, normalized to 0.7) and the electro-mechanical energy coupling degree (the ratio of vibration energy in the narrow-band neighborhood of 95-105Hz to the corresponding electrodynamic excitation energy is 0.65) are calculated and integrated to generate a set of mechanical resonance risk indices.

[0058] The method provided in this embodiment tracks the drift trajectory of the dominant natural frequency in real time based on operational modal analysis, accurately identifies the excitation frequency of the main conductive force that is strongly correlated with the current, and then calculates the frequency convergence and energy coupling degree. It quantifies the risk from three dimensions: "structural side", "excitation side" and "coupling side", perfectly making up for the static and one-sided defects of traditional technology, ensuring accurate risk assessment, and providing a reliable basis for subsequent early warning.

[0059] In some embodiments, synchronous vibration signals from support point monitoring positions, mid-span monitoring positions, and expansion joint monitoring positions are used as input. Based on random subspace identification or frequency domain decomposition algorithms, the multi-order working mode shapes and frequencies of the busbar shell under operating conditions are analyzed. The results of preset offline impact tests on key bolt connection points of the busbar shell are integrated to establish a reference value for the structure's natural frequencies, and the dominant natural frequencies related to the overall bending or torsional main vibration modes are selected accordingly. The dominant natural frequencies are thermally stiffened based on the busbar shell temperature signal, and their dynamic changes are recorded in time sequence to form the drift trajectory of the dominant natural frequencies.

[0060] Random subspace identification can be a modal analysis algorithm based on vibration response signals. By constructing a random state space model, it identifies the modal parameters (natural frequencies, mode shapes, damping) of the structure from measured signals with noise pollution, making it suitable for online real-time monitoring scenarios. Frequency domain decomposition algorithms can be another type of modal analysis algorithm. By performing singular value decomposition on the power spectral density matrix, it separates the frequency components corresponding to different modes, thereby extracting structural modal parameters. It is highly adaptable to stationary vibration signals. Multi-order operating mode shapes and frequencies can refer to multiple vibration modes (mode shapes) and corresponding natural frequencies exhibited by the enclosed busbar shell during operation. Different order modes reflect the vibration characteristics of different parts of the structure. Critical bolt connection points can be bolt connections on the enclosed busbar shell that play a crucial role in structural stiffness and vibration transmission; their connection state directly affects the structure's natural frequencies. Preset offline impact test results can be obtained by manually impacting key bolt connection points and other structural parts during the installation and commissioning phase of the enclosed busbar, collecting vibration response signals to obtain modal parameter data as an initial reference for natural frequencies. The benchmark reference value can be a standard value of the natural frequency determined based on the results of a preset offline impact test and combined with structural design parameters. This value is used to screen and verify the modal frequencies identified under operational conditions. The dominant natural frequency can be the core natural frequency related to the overall bending or torsional mode shape. It is a key structural parameter affecting the resonance risk of the enclosed busbar, and its changes directly affect the likelihood of resonance. Thermally induced stiffness correction can be performed by numerically adjusting the identified natural frequencies based on the correlation between the busbar shell temperature signal and structural stiffness to offset the influence of temperature changes on the natural frequencies.

[0061] Specifically, traditional measurements of the natural frequency of enclosed busbars often rely on static offline impact tests during the installation phase. These tests obtain fixed baseline values ​​and use them for extended periods, ignoring frequency drift caused by temperature and load fluctuations during operation. Furthermore, the lack of baseline screening makes it prone to misjudging local modes. For example, a power station measured a natural frequency of 100Hz offline. During operation, the busbar casing temperature rose to 65℃, causing a decrease in structural stiffness and a frequency drift to 106Hz. This frequency coincided with the electrodynamic excitation frequency, triggering resonance. Because the drift was not tracked in real time, no timely warning was issued. In another power station, an 80Hz vibration mode corresponding to a localized bulge in the casing was mistakenly identified as the dominant natural frequency, leading to a complete deviation in risk assessment. To address the above issues, this step uses triaxial vibration acceleration signals (e.g., instantaneous Y-axis value of 3.1 m / s² at the support point, peak Z-axis value of 3.8 m / s² at the mid-span) synchronously collected from the support point, mid-span, and expansion joint monitoring locations as input. If the signal exhibits non-stationary characteristics caused by load fluctuations, a random subspace identification algorithm is employed to extract multiple modes by constructing a Hankel matrix, performing singular value decomposition, and estimating a state-space model. If the signal is stationary, a frequency domain decomposition algorithm is used to obtain multiple frequencies such as 82 Hz, 101 Hz, and 153 Hz and their corresponding mode shapes through singular value decomposition of the power spectral density matrix. The results of a preset offline impact test (a reference value of 100 Hz ± 3 Hz obtained from impacting key bolt connection points) are then integrated to select 101 Hz, which is related to the overall bending mode shape, as the dominant natural frequency. Based on the real-time temperature of the busbar casing at 60℃, and following the rule that the structural stiffness decreases by 4% and the natural frequency shifts by 0.8Hz for every 10℃ increase in temperature, the frequency was corrected to 104.2Hz. The above process was repeated every 20 minutes, and the values ​​of the frequency gradually drifting from 98.5Hz to 105.3Hz over 24 hours were recorded to form the dominant natural frequency drift trajectory.

[0062] The method provided in this embodiment, through multi-point synchronous signal input, combined with the adaptation algorithm to identify multi-mode, integrates offline benchmarks to screen the core dominant frequency, and then uses temperature correction to track the drift trajectory, perfectly makes up for the shortcomings of traditional technology such as static, unscreened, and uncompensated, ensuring that the obtained dominant natural frequency matches the real-time operating state, and providing accurate structural parameter support for subsequent frequency convergence calculation and resonance risk assessment.

[0063] In some embodiments, the absolute value of the difference between the main conductor dynamic excitation frequency and the dominant natural frequency is compared with a safety margin threshold set according to the structural damping characteristics to obtain a normalized frequency convergence. The vibration energy in the bus vibration characteristic spectrum set and the corresponding electrodynamic excitation energy in the theoretical excitation spectrum are extracted within the narrow-band neighborhood of the main conductor dynamic excitation frequency, and the ratio of the two is used as the electro-mechanical energy coupling degree. The normalized frequency convergence and the electro-mechanical energy coupling degree are combined with the vibration energy margin coefficient reflecting the current overall vibration level and weighted to generate a single comprehensive index for characterizing the instantaneous risk of mechanical resonance, thus forming a mechanical resonance risk index set.

[0064] Structural damping characteristics, encompassing the energy dissipation capacity of the isolated busbar shell material and the structure itself, determine the structure's ability to resist resonance and serve as the core basis for setting safety margins. Safety margin thresholds are critical values ​​within a safe frequency range set based on structural damping characteristics, used to assess whether there is a risk associated with frequency approaches. Different busbars with different damping characteristics correspond to different thresholds. The narrowband neighborhood can be a range (typically ±5Hz) centered on the main conductor's excitation frequency, covering a certain frequency range nearby, used to accurately extract the vibration energy and excitation energy corresponding to that excitation frequency. Vibration energy can be the structural vibration energy value extracted from the busbar vibration characteristic spectrum set within the narrowband neighborhood, triggered by the main conductor's excitation frequency, reflecting the structure's response intensity to the excitation. Electrodynamic excitation energy can be the excitation energy value extracted from the theoretical excitation spectrum within the narrowband neighborhood, corresponding to the main conductor's excitation frequency, reflecting the energy intensity of the excitation source. The vibration energy margin coefficient reflects the ratio of the current overall busbar vibration level to the safe vibration threshold, characterizing the redundancy of the structure's current vibration state and serving as an important dimension for supplementing risk assessment.

[0065] Specifically, traditional methods for quantifying resonance risks in enclosed busbars have significant shortcomings: they only calculate the absolute frequency difference without normalizing it according to structural damping characteristics; using a uniform threshold to judge different damping buses lacks scientific rigor; for example, a power station set a 3Hz threshold for buses with damping ratios of 0.03 and 0.05, leading to missed risks for buses with low damping; they also ignore energy coupling, misjudging situations with similar frequencies but inefficient energy transfer as high-risk, such as a substation blindly shutting down due to a 2Hz frequency difference, when in fact the motor-machine energy coupling was only 0.1, with no resonance hazard; and they fail to consider the overall vibration level, easily overlooking potential risks. To address the above issues, this step first obtains the excitation frequency of the main conductor (e.g., 100Hz) and the corrected dominant natural frequency (e.g., 102Hz). Based on the damping characteristics of the busbar structure (e.g., damping ratio 0.04), a safety margin threshold of 5Hz is set, and the absolute value of the difference between the two is calculated to be 2Hz. Comparing this with the threshold yields a normalized frequency convergence of 0.4. A narrow band neighborhood of 95-105Hz is defined centered on the excitation frequency of the main conductor, and the vibration energy in this range (e.g., 102Hz) is extracted from the busbar vibration characteristic spectrum. Extracting the corresponding electrodynamic excitation energy from the theoretical excitation spectrum (e.g.) The ratio of the two is 0.6, which is taken as the electro-mechanical energy coupling degree. Then, the vibration energy margin coefficient (such as 0.8) reflecting the overall vibration level is obtained. The vibration energy margin coefficient is then weighted and fused according to the weights of frequency convergence degree 0.4, electro-mechanical energy coupling degree 0.4, and vibration energy margin coefficient 0.2, that is, 0.4×0.4+0.6×0.4+0.8×0.2=0.56. Finally, the frequency convergence degree, electro-mechanical energy coupling degree and single comprehensive index 0.56 are integrated to form the mechanical resonance risk index set.

[0066] The method provided in this embodiment achieves a unified evaluation standard based on the normalized frequency convergence, calculates the energy coupling degree to overcome the misconception that "close frequency equals risk", incorporates the vibration energy margin coefficient to improve the evaluation dimensions, and generates a weighted comprehensive index. This completely solves the problems of traditional technology being one-sided, lacking a unified standard, and having inefficient decision-making, ensuring that risk quantification is comprehensive and accurate, and providing a reliable basis for subsequent operation and maintenance intervention.

[0067] In some embodiments, the immediate risk level of resonance is assessed based on frequency convergence, electromechanical energy coupling degree, and vibration energy margin coefficient; based on the immediate risk level, combined with the temporal change trend of the dominant natural frequency drift trajectory, the risk development rate and accumulation trend are assessed; according to the immediate risk level and the risk development rate and accumulation trend, a multi-dimensional resonance risk status assessment is implemented, and a closed bus resonance risk status report is generated.

[0068] The immediate risk level can be determined based on frequency convergence, electromechanical energy coupling degree, and vibration energy margin coefficient, comprehensively assessing the resonance risk level of the enclosed busbar at the current moment, directly reflecting the instantaneous safety status. Multi-dimensional resonance risk assessment can be a method that comprehensively determines the resonance risk status by considering three core dimensions: "immediate risk intensity," "risk development speed," and "risk accumulation degree," combined with structural vibration characteristics and excitation features.

[0069] Specifically, traditional closed busbar risk assessment focuses only on the immediate state, ignoring risk trends and cumulative effects. Furthermore, the assessment dimensions are singular, the outputs are fragmented, and there is no operational guidance, easily leading to misjudgments or omissions. For example, a substation failed to intervene because its immediate risk level of 0.4 did not reach the emergency threshold, yet it failed to notice that the dominant natural frequency was approaching the excitation frequency daily at 0.5Hz, ultimately causing resonance and cracking of the casing welds. Another substation blindly shut down based solely on a frequency convergence of 0.7, when in fact the energy coupling degree was only 0.2, posing no substantial risk. To address these issues, this step first retrieves the frequency convergence (e.g., 0.5), motor-machine energy coupling degree (e.g., 0.6), and vibration energy margin coefficient (e.g., 0.7), and divides the range by weighted summation of these three factors (0-0.3 low risk, 0.3-0.7 medium risk, and above 0.7 high risk), assessing the immediate risk as medium risk. Then, the dominant inherent frequency drift trajectory (e.g., drifting from 100Hz to 102Hz over 24 hours, with the excitation frequency stabilizing at 103Hz) is retrieved. Combined with time-series data showing the immediate risk increasing from 0.4 to 0.5 within 8 hours, a gradual development rate of 0.0125 per hour is calculated, indicating an escalating risk accumulation trend. Referring to preset rules, because the frequency convergence is below the safety dynamic margin of 0.3, a warning level alert is issued. Finally, by integrating key parameters, trend analysis, and operational recommendations for load adjustments, a closed bus resonance risk status report is generated.

[0070] The method provided in this embodiment integrates real-time risks, development rates, and cumulative trends to achieve multi-dimensional assessments. This not only addresses the static and one-sided shortcomings of traditional technologies but also transforms scattered parameters into standardized reports containing risk levels, causes, and operational recommendations, making operational decisions more systematic. This is the key to achieving precise control of resonance risks.

[0071] In some embodiments, if the immediate risk level indicates the presence of significant electrodynamic excitation frequency energy, and the risk accumulation trend is stable with a stable dominant natural frequency drift trajectory, it is determined to be an observation-level warning; if the immediate risk level continues to rise, and the risk accumulation trend shows that the dominant natural frequency drift trajectory continues to approach the main conductive electrodynamic excitation frequency, causing the frequency convergence of the two to be lower than a preset safety dynamic margin, it is determined to be a warning level warning; if the immediate risk level exceeds the absolute safety threshold, it is determined to be an action-level alarm.

[0072] Observation-level warnings can be identified as low-risk states after multi-dimensional assessments. These are characterized by significant electrodynamic excitation energy, but with stable risk accumulation and a stable dominant natural frequency drift trajectory. Immediate intervention is not required, but continuous monitoring is necessary. Alert-level warnings can be identified as medium-risk states after multi-dimensional assessments. These are characterized by a continuously increasing immediate risk level, a dominant natural frequency drift trajectory continuously approaching the excitation frequency, and a frequency convergence rate lower than the preset safety dynamic margin. Timely targeted operational and maintenance interventions are required. Action-level alarms can be identified as high-risk states after multi-dimensional assessments. These are characterized by an immediate risk level exceeding the absolute safety threshold, with resonance already occurring or about to occur. Immediate mandatory measures such as emergency shutdowns and load adjustments are required.

[0073] Specifically, traditional closed busbar risk assessment simply categorizes risks as "safe" or "dangerous" without any grading standards, which can easily lead to inappropriate operation and maintenance decisions. For example, a power station was shut down directly because its electro-mechanical energy coupling degree reached 0.45 (significant excitation energy) without considering the risk trend. In reality, the dominant natural frequency only changed by 0.1Hz in 24 hours, indicating a stable risk that required no intervention. Another power station had an immediate risk level of 0.6 that did not exceed the absolute threshold of 0.8, but its continuously increasing trend and frequency approach characteristics were ignored, ultimately leading to resonance. To address the above issues, this step first retrieves relevant parameters for the immediate risk level. If the electro-mechanical energy coupling degree is 0.45 (indicating significant electrodynamic excitation frequency energy), and the dominant natural frequency drift trajectory changes from 100Hz to 100.1Hz (stable) over 24 hours, with a stable risk accumulation trend, it is judged as an observation-level warning, and it is recommended to update the monitoring data every 15 minutes. If the immediate risk level rises from 0.5 to 0.72 (continuously increasing), and the dominant natural frequency drifts from 100.1Hz to 102.3Hz over 72 hours (continuously approaching the excitation frequency of 102.5Hz), with a frequency convergence of 0.25 lower than the preset safety dynamic margin of 0.3, it is judged as a warning-level warning, and it is recommended to adjust the load to 80% and check the insulators. If the immediate risk level of 0.85 exceeds the absolute safety threshold of 0.8, it is directly judged as an action-level alarm, and an immediate shutdown and maintenance order is issued.

[0074] The method provided in this embodiment accurately covers "potential stability risks, continuously escalating risks, and critical failure risks" through three levels of early warning: observation level, alert level, and action level. This solves the shortcomings of traditional technologies that rely on a single judgment and cannot adapt to different operation and maintenance needs, allowing operation and maintenance personnel to take corresponding measures according to the risk level and achieve refined management.

[0075] Figure 3 This is a schematic diagram of the structure of an intelligent detection system for the operating status of a closed busbar, as provided in an embodiment of this application. Figure 3 As shown, the intelligent detection system 300 for the operating status of a closed busbar in this embodiment includes: a vibration characteristic analysis module 301, a resonance risk analysis module 302, and a status report generation module 303.

[0076] The vibration feature analysis module 301 is used to acquire a multi-source real-time monitoring dataset of the phase-separated closed busbar. Based on the multi-source real-time monitoring dataset, the triaxial vibration acceleration signal is processed by time-frequency domain transformation to separate and extract the vibration feature frequency components dominated by electrodynamic pulsation and generate a busbar vibration feature spectrum set.

[0077] The resonance risk analysis module 302 is used to perform operational mode analysis and electrodynamic frequency identification based on the bus vibration characteristic spectrum set, calculate the dominant natural frequency and main electrodynamic excitation frequency of the bus structure in real time, calculate the convergence degree and energy coupling degree between the two, and generate a mechanical resonance risk index set.

[0078] The status report generation module 303 is used to perform a multi-dimensional assessment of the resonance risk status based on the mechanical resonance risk index set, and generate a closed bus resonance risk status report to guide operation and maintenance intervention.

[0079] Optionally, when generating the bus vibration characteristic spectrum set, the vibration characteristic analysis module 301 is specifically used for:

[0080] A multi-source real-time monitoring dataset is acquired at specific structural locations on the outer casing of the phase-separated enclosed busbar. The multi-source real-time monitoring dataset includes the synchronously acquired triaxial vibration acceleration signal, three-phase current waveform signal, and busbar casing temperature signal.

[0081] Based on the three-phase current waveform signal, the fundamental frequency of the electrodynamic force and its harmonic components are calculated as the theoretical excitation spectrum;

[0082] The measured vibration spectrum is obtained by performing a time-frequency domain transformation on the triaxial vibration acceleration signal.

[0083] Based on the busbar shell temperature signal, the measured vibration spectrum and the theoretical excitation spectrum are decoupled and analyzed to filter out broadband noise caused by environmental background vibration and internal particle impact, and to separate the narrowband resonance risk frequency component with structural mode modulation characteristics caused by the conductor electrodynamic force transmitted to the shell through the insulator support structure, thereby generating the busbar vibration characteristic spectrum set.

[0084] Optionally, when the vibration feature analysis module 301 acquires the multi-source real-time monitoring dataset based on the specific structural locations of the isolated phase enclosed busbar casing, it is specifically used for:

[0085] Based on the structural dynamics characteristics and vibration transmission path of the aforementioned phase-separated closed busbar, sensor groups are synchronously deployed at the following key locations on the outer shell to collect signals:

[0086] At the mechanical anchor points corresponding to the internal support insulators penetrating the outer shell, support point monitoring positions are set up;

[0087] At the mid-span position of the outer shell between two adjacent anchor points of the supporting insulators, a mid-span monitoring point is set up;

[0088] Expansion joint monitoring points are installed on the outer casing of the bellows or expansion joint connection section used to compensate for thermal expansion and contraction.

[0089] The multi-source real-time monitoring dataset is simultaneously collected at the support point monitoring position, mid-span monitoring position, and expansion joint monitoring position.

[0090] Optionally, when the vibration characteristic analysis module 301 separates the narrowband resonance risk frequency components with structural mode modulation characteristics caused by the transmission of conductor electrodynamic force through the insulator support structure to the shell, it is specifically used for:

[0091] Based on the vibration signals synchronously collected from the support point monitoring position, the mid-span monitoring position and the expansion joint monitoring position, a multi-point vibration spectrum matrix is ​​constructed;

[0092] Collaborative analysis of the vibration spectrum matrix at multiple measurement points identifies narrowband frequency components that exhibit significant energy at the support point monitoring position, amplitude amplification at the mid-span monitoring position, and specific phase lag at the expansion joint monitoring position, which are then used as candidate structural modal frequencies.

[0093] The candidate structural modal frequencies are matched with the theoretical excitation spectrum to screen out frequency components that have an integer or fractional multiple relationship with the fundamental frequency or harmonic components of the electrodynamics, and whose energy changes synchronously with the bus load current.

[0094] Based on the busbar shell temperature signal, the selected frequency components are dynamically stiffened and corrected, and finally the narrowband resonance risk frequency components that characterize the mechanical resonance excited by the electrodynamic force through the insulator-shell path are separated.

[0095] Optionally, when generating the mechanical resonance risk index set, the resonance risk analysis module 302 is specifically used for:

[0096] Based on the narrowband resonance risk frequency components separated from the bus vibration characteristic spectrum and their spatial distribution characteristics, the dominant natural frequency drift trajectory of the phase-separated closed bus shell structure is identified and tracked in real time using operational modal analysis technology.

[0097] Based on the theoretical excitation spectrum and the narrowband energy characteristics that change synchronously with the current, the main conductor power excitation frequency under the current operating condition is identified in real time.

[0098] The frequency convergence between the dominant natural frequency and the main conductive dynamic excitation frequency, as well as the electromechanical energy coupling degree characterizing the energy transfer efficiency between the two, are analyzed to generate the mechanical resonance risk index set containing the frequency convergence and the electromechanical energy coupling degree.

[0099] Optionally, the resonance risk analysis module 302, during the real-time identification and tracking process based on the dominant natural frequency drift trajectory, is specifically used for:

[0100] Using the synchronous vibration signals of the support point monitoring position, the mid-span monitoring position and the expansion joint monitoring position as input, the multi-order working mode vibration mode and frequency of the busbar shell under the operating state are analyzed based on random subspace identification or frequency domain decomposition algorithm;

[0101] By integrating the results of offline impact tests on key bolt connection points of the busbar shell, a reference value for the structure's natural frequency is established, and the dominant natural frequencies related to the overall bending or torsional mode are selected accordingly.

[0102] The dominant natural frequency is thermally stiffened based on the busbar shell temperature signal, and its dynamic changes are recorded in time sequence to form the drift trajectory of the dominant natural frequency.

[0103] Optionally, the resonance risk analysis module 302, during the analysis process based on the frequency convergence and electro-mechanical energy coupling, is specifically used for:

[0104] The absolute value of the difference between the main conductive power excitation frequency and the dominant natural frequency is compared with the safety margin threshold set according to the structural damping characteristics to obtain the normalized frequency convergence.

[0105] The vibration energy in the bus vibration characteristic spectrum set and the corresponding electrodynamic excitation energy in the theoretical excitation spectrum are extracted within the narrow band neighborhood of the main conductor's excitation frequency, and the ratio of the two is used as the electro-mechanical energy coupling degree.

[0106] The normalized frequency convergence and the electromechanical energy coupling degree are combined with the vibration energy margin coefficient, which reflects the overall level of current vibration, and weighted and fused to generate a single comprehensive index for characterizing the instantaneous risk of mechanical resonance, thus forming the mechanical resonance risk index set.

[0107] Optionally, the status report generation module 303, during the generation process of the closed bus resonance risk status report, is specifically used for:

[0108] The immediate risk level of resonance is assessed based on the frequency convergence, the electromechanical energy coupling degree, and the vibration energy margin coefficient.

[0109] Based on the aforementioned immediate risk level, and combined with the temporal variation trend of the dominant inherent frequency drift trajectory, the risk development rate and accumulation trend are assessed.

[0110] Based on the real-time risk level and the risk development rate and accumulation trend, a multi-dimensional resonance risk status assessment is implemented, and a closed bus resonance risk status report is generated.

[0111] Optionally, the status report generation module 303, when conducting the multi-dimensional resonance risk status assessment, is specifically used for:

[0112] If the instantaneous risk level indicates the presence of significant electrodynamic excitation frequency energy, and the risk accumulation trend is stable and the dominant natural frequency drift trajectory is stable, then it is determined to be an observation-level warning.

[0113] If the immediate risk level continues to rise, and the risk accumulation trend shows that the dominant inherent frequency drift trajectory continues to approach the main conductive power excitation frequency, causing the frequency convergence of the two to be lower than the preset safety dynamic margin, then it is determined to be a warning level warning.

[0114] If the immediate risk level exceeds the absolute safety threshold, it is determined to be an action-level alert.

[0115] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. An intelligent detection method for the operating status of a closed busbar, characterized in that, include: A multi-source real-time monitoring dataset of the isolated phase closed busbar is obtained. Based on the multi-source real-time monitoring dataset, the triaxial vibration acceleration signal is processed by time-frequency domain transformation to separate and extract the vibration characteristic frequency components dominated by electrodynamic pulsation, and generate a busbar vibration characteristic spectrum set. Based on the aforementioned bus vibration characteristic spectrum set, operational modal analysis and electrodynamic frequency identification are performed. The dominant natural frequency and main electrodynamic excitation frequency of the bus structure are calculated in real time, and the convergence and energy coupling between the two are calculated to generate a set of mechanical resonance risk indices. Based on the aforementioned set of mechanical resonance risk indices, a multi-dimensional assessment of the resonance risk status is conducted, and a closed bus resonance risk status report is generated to guide operation and maintenance intervention.

2. The method according to claim 1, characterized in that, The generated bus vibration characteristic spectrum set includes: A multi-source real-time monitoring dataset is acquired at specific structural locations on the outer casing of the phase-separated enclosed busbar. The multi-source real-time monitoring dataset includes the synchronously acquired triaxial vibration acceleration signal, three-phase current waveform signal, and busbar casing temperature signal. Based on the three-phase current waveform signal, the fundamental frequency of the electrodynamic force and its harmonic components are calculated as the theoretical excitation spectrum; The measured vibration spectrum is obtained by performing a time-frequency domain transformation on the triaxial vibration acceleration signal. Based on the busbar shell temperature signal, the measured vibration spectrum and the theoretical excitation spectrum are decoupled and analyzed to filter out broadband noise caused by environmental background vibration and internal particle impact, and to separate the narrowband resonance risk frequency component with structural mode modulation characteristics caused by the conductor electrodynamic force transmitted to the shell through the insulator support structure, thereby generating the busbar vibration characteristic spectrum set.

3. The method according to claim 2, characterized in that, The acquisition of multi-source real-time monitoring datasets arranged at specific structural locations on the outer casing of the phase-separated enclosed busbar includes: Based on the structural dynamics characteristics and vibration transmission path of the aforementioned phase-separated closed busbar, sensor groups are synchronously deployed at the following key locations on the outer shell to collect signals: At the mechanical anchor points corresponding to the internal support insulators penetrating the outer shell, support point monitoring positions are set up; At the mid-span position of the outer shell between two adjacent anchor points of the supporting insulators, a mid-span monitoring point is set up; Expansion joint monitoring points are installed on the outer casing of the bellows or expansion joint connection section used to compensate for thermal expansion and contraction. The multi-source real-time monitoring dataset is simultaneously collected at the support point monitoring position, mid-span monitoring position, and expansion joint monitoring position.

4. The method according to claim 3, characterized in that, The narrowband resonant risk frequency components with structural mode modulation characteristics, which are separated from the conductor's electrodynamic force transmitted to the outer shell through the insulator support structure, include: Based on the vibration signals synchronously collected from the support point monitoring position, the mid-span monitoring position and the expansion joint monitoring position, a multi-point vibration spectrum matrix is ​​constructed; Collaborative analysis of the vibration spectrum matrix at multiple measurement points identifies narrowband frequency components that exhibit significant energy at the support point monitoring position, amplitude amplification at the mid-span monitoring position, and specific phase lag at the expansion joint monitoring position, which are then used as candidate structural modal frequencies. The candidate structural modal frequencies are matched with the theoretical excitation spectrum to screen out frequency components that have an integer or fractional multiple relationship with the fundamental frequency or harmonic components of the electrodynamics, and whose energy changes synchronously with the bus load current. Based on the busbar shell temperature signal, the selected frequency components are dynamically stiffened and corrected, and finally the narrowband resonance risk frequency components that characterize the mechanical resonance excited by the electrodynamic force through the insulator-shell path are separated.

5. The method according to claim 4, characterized in that, The generated set of mechanical resonance risk indices includes: Based on the narrowband resonance risk frequency components separated from the bus vibration characteristic spectrum and their spatial distribution characteristics, the dominant natural frequency drift trajectory of the phase-separated closed bus shell structure is identified and tracked in real time using operational modal analysis technology. Based on the theoretical excitation spectrum and the narrowband energy characteristics that change synchronously with the current, the main conductor power excitation frequency under the current operating condition is identified in real time. The frequency convergence between the dominant natural frequency and the main conductive dynamic excitation frequency, as well as the electromechanical energy coupling degree characterizing the energy transfer efficiency between the two, are analyzed to generate the mechanical resonance risk index set containing the frequency convergence and the electromechanical energy coupling degree.

6. The method according to claim 5, characterized in that, The real-time identification and tracking process of the dominant natural frequency drift trajectory includes: Using the synchronous vibration signals of the support point monitoring position, the mid-span monitoring position and the expansion joint monitoring position as input, the multi-order working mode vibration mode and frequency of the busbar shell under the operating state are analyzed based on random subspace identification or frequency domain decomposition algorithm; By integrating the results of offline impact tests on key bolt connection points of the busbar shell, a reference value for the structure's natural frequency is established, and the dominant natural frequencies related to the overall bending or torsional mode are selected accordingly. The dominant natural frequency is thermally stiffened based on the busbar shell temperature signal, and its dynamic changes are recorded in time sequence to form the drift trajectory of the dominant natural frequency.

7. The method according to claim 6, characterized in that, The analysis process of frequency convergence and electro-mechanical energy coupling includes: The absolute value of the difference between the main conductive power excitation frequency and the dominant natural frequency is compared with the safety margin threshold set according to the structural damping characteristics to obtain the normalized frequency convergence. The vibration energy in the bus vibration characteristic spectrum set and the corresponding electrodynamic excitation energy in the theoretical excitation spectrum are extracted within the narrow band neighborhood of the main conductor's excitation frequency, and the ratio of the two is used as the electro-mechanical energy coupling degree. The normalized frequency convergence and the electromechanical energy coupling degree are combined with the vibration energy margin coefficient, which reflects the overall level of current vibration, and weighted and fused to generate a single comprehensive index for characterizing the instantaneous risk of mechanical resonance, thus forming the mechanical resonance risk index set.

8. The method according to claim 7, characterized in that, The process of generating the closed busbar resonance risk status report includes: The immediate risk level of resonance is assessed based on the frequency convergence, the electromechanical energy coupling degree, and the vibration energy margin coefficient. Based on the aforementioned immediate risk level, and combined with the temporal variation trend of the dominant inherent frequency drift trajectory, the risk development rate and accumulation trend are assessed. Based on the real-time risk level and the risk development rate and accumulation trend, a multi-dimensional resonance risk status assessment is implemented, and a closed bus resonance risk status report is generated.

9. The method according to claim 8, characterized in that, The implementation of the multi-dimensional resonance risk status assessment includes: If the instantaneous risk level indicates the presence of significant electrodynamic excitation frequency energy, and the risk accumulation trend is stable and the dominant natural frequency drift trajectory is stable, then it is determined to be an observation-level warning. If the immediate risk level continues to rise, and the risk accumulation trend shows that the dominant inherent frequency drift trajectory continues to approach the main conductive power excitation frequency, causing the frequency convergence of the two to be lower than the preset safety dynamic margin, then it is determined to be a warning level warning. If the immediate risk level exceeds the absolute safety threshold, it is determined to be an action-level alert.

10. An intelligent detection system for the operating status of a closed busbar, characterized in that, The method applied to any one of claims 1-9 includes: The vibration feature analysis module is used to acquire a multi-source real-time monitoring dataset of the phase-separated closed busbar. Based on the multi-source real-time monitoring dataset, the triaxial vibration acceleration signal is processed by time-frequency domain transformation to separate and extract the vibration feature frequency components dominated by electrodynamic pulsation, and generate a busbar vibration feature spectrum set. The resonance risk analysis module is used to perform operational mode analysis and electrodynamic frequency identification based on the bus vibration characteristic spectrum set, calculate the dominant natural frequency and main electrodynamic excitation frequency of the bus structure in real time, calculate the convergence degree and energy coupling degree between the two, and generate a set of mechanical resonance risk indices. The status report generation module is used to perform a multi-dimensional assessment of the resonance risk status based on the mechanical resonance risk index set, and generate a closed bus resonance risk status report to guide operation and maintenance intervention.

Images