Method and system for analyzing and early warning temperature change of bus of power distribution cabinet

Abstract

The invention discloses a power distribution cabinet bus temperature change analysis and early warning method and system, and relates to the technical field of power distribution cabinets, and the method comprises the steps: collecting the temperature data of a three-phase bus of a power distribution cabinet, dividing a temperature sequence into time-space units, and calculating the spatial characteristics and time characteristics of each time-space unit; performing frequency domain analysis on the temperature data of the space-time unit, decomposing the temperature data to obtain a low-frequency component, an intermediate-frequency component and a high-frequency component, extracting sub-components of different time scales, and calculating a time coupling feature, a space coupling feature and a physical coupling feature; fault-free data is collected to establish a normal mode library, historical fault data is collected to establish a fault mode library, the deviation degree of features and a baseline is calculated in real time, and marking and feedback updating are carried out on potential new faults; on the basis of distributed monitoring, temperature abnormal points are positioned, reconstruction errors of all scales are calculated, the abnormal origin scale is determined, and graded early warning is triggered, so that the problem that a traditional monitoring method is insufficient in fault recognition capability is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution cabinets, and in particular to a method and system for analyzing and warning temperature changes of a busbar of a power distribution cabinet. Background Art

[0002] As an important part of the power system, the safe and stable operation of the distribution network is of vital importance. The busbar in the distribution cabinet undertakes the core task of power transmission, and its temperature change directly reflects the operating status of the equipment. In the existing technology, the temperature monitoring of the distribution cabinet busbar is mainly achieved by installing a limited number of temperature sensors, such as thermocouples or infrared temperature measuring devices. These sensors are usually arranged at key positions on the busbar, and the early warning function is realized by periodically collecting temperature data combined with threshold judgment.

[0003] However, since only a small number of sensors are deployed at key locations on the busbar, it is impossible to capture the temperature gradient changes along the entire length of the busbar in real time, which may lead to missed detection of local abnormal points. Secondly, the early warning mechanism relies on static thresholds and cannot adapt to the impact of dynamic load changes and ambient temperature fluctuations. For example, under high load or high temperature environments, fixed thresholds are prone to false alarms or missed alarms, reducing the accuracy of early warnings. In addition, existing technologies mainly rely on time domain threshold judgments and lack frequency domain feature analysis of temperature changes. They cannot effectively distinguish normal fluctuations from potential faults such as poor contact and insulation aging. At the same time, the temperature data is not decomposed at multiple scales, making it difficult to identify abnormal conduction patterns at different time scales. Summary of the Invention

[0004] (1) Technical problems solved In response to the shortcomings of the existing technology, the present invention provides a method and system for analyzing and warning temperature changes of the busbar of a distribution cabinet. Real-time monitoring of the temperature gradient along the entire length of the busbar is achieved through distributed optical fiber sensors, and the dynamic noise threshold adaptive adjustment technology is combined to solve the false alarm problem of the static threshold under load and environmental changes; frequency domain analysis and multi-scale wavelet decomposition technology are used to accurately extract the frequency domain characteristics and time scale characteristics of temperature data, effectively distinguishing normal fluctuations from potential faults; and by establishing a dynamically updated normal mode library and fault mode library, combined with cluster analysis and a hierarchical warning mechanism, the accuracy and timeliness of anomaly detection are significantly improved, solving the problems of local missed detection, high false alarm rate and insufficient fault identification ability in traditional monitoring methods.

[0005] (2) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for analyzing and warning temperature changes of busbars in a power distribution cabinet, comprising: Collect temperature data on the three-phase busbars of the power distribution cabinet, preprocess the temperature data, output a smoothed temperature sequence, divide the temperature sequence into space-time units, and calculate the spatial and temporal characteristics of each space-time unit; Frequency domain analysis is performed on the temperature data of the time-space unit. The frequency band boundaries are determined through fast Fourier transform. Low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions are constructed. The temperature data is decomposed into low-frequency components, medium-frequency components, and high-frequency components. Multi-layer wavelet decomposition is performed on each component to extract sub-components at different time scales. The temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics are calculated. Collect fault-free data to establish a normal pattern library, calculate the initial baseline range of each feature, update the baseline range according to the load interval, collect historical fault data to establish a fault pattern library, extract the typical range of fault features through cluster analysis, calculate the deviation between the feature and the baseline in real time, trigger the fault pattern library search, mark potential new faults and provide feedback updates; Based on distributed monitoring, the temperature anomaly points are located, the abnormal area is expanded and the duration is recorded, the reconstruction error at each scale is calculated, the scale of the anomaly origin is determined, and graded warnings are triggered according to the abnormal conduction range, energy growth multiples and duration.

[0006] Furthermore, collecting the temperature data, current and cabinet door status data of the three-phase busbar of the power distribution cabinet includes the following steps: Distributed fiber optic sensing points are laid every 0.5 meters along the length of the A / B / C three-phase busbars of the distribution cabinet, and both ends of the fiber optic cables are connected to the DTS host; temperature and humidity sensors are installed on the top, middle and bottom of the distribution cabinet; current transformers are installed at the three-phase incoming line end to collect current signals; magnetic proximity switches are installed on the cabinet door hinges to output cabinet door status signals.

[0007] Furthermore, the preprocessing includes the following steps: Dynamically adjust the noise threshold based on load current and ambient temperature: When the load current does not exceed 80% of the rated current and the ambient temperature does not exceed 35°C, the noise threshold is 3°C; when the load current exceeds 80% of the rated current, the noise threshold increases by 0.2°C for every 10 amperes above the rated current; when the ambient temperature exceeds 35°C, the noise threshold increases by 0.1°C for every 1°C above the ambient temperature; For each temperature data point, the data within 10 seconds before and after it are taken to form an analysis window. The weight of each data in the window is calculated by the Gaussian kernel function, and the temperature change curve in the window is fitted using the weighted least squares method. If the deviation between any data point and the temperature change curve exceeds the noise threshold, it is determined to be a noise point and replaced by the value of the temperature change curve at the time of the data point. The processed data is secondary smoothed using the sliding average method, and the smoothed temperature series is output.

[0008] Furthermore, the temperature data of each time-space unit is subjected to fast Fourier transform to obtain the power spectrum, and the boundary between low frequency and medium frequency is preliminarily set. , and the boundary between mid-frequency and high-frequency , calculate the Pearson correlation coefficient between the power in the mid-frequency band and the load current; If the absolute value of the correlation coefficient is not less than 0.7, the current boundary is retained. and ; If the absolute value of the correlation coefficient is less than 0.7, the boundary is adjusted in steps of 0.0005 Hz until the absolute value of the correlation coefficient is no less than 0.7.

[0009] Further, based on and Construct empirical wavelet basis functions, including low-frequency basis functions, medium-frequency basis functions and high-frequency basis functions; The temperature data of phases A / B / C are decomposed using the empirical wavelet basis function to obtain the low-frequency component, intermediate-frequency component, and high-frequency component of each phase. The instantaneous phase difference of the high-frequency components of the three phases is calculated. If the phase difference between any two phases exceeds 30°, the center frequency of the empirical wavelet basis function is adjusted and re-decomposed until all phase differences do not exceed 30°.

[0010] Furthermore, the db4 wavelet basis is used to perform three-layer wavelet decomposition on the low-frequency, medium-frequency and high-frequency components respectively. The low-frequency component is decomposed into day scale, hour scale and minute scale, the medium-frequency component is decomposed into 30-minute scale, 10-minute scale and 1-minute scale, and the high-frequency component is decomposed into 30-second scale, 10-second scale and 1-second scale.

[0011] Furthermore, fault-free data for several months after the equipment is put into operation is collected, and the 95% confidence interval of each feature is calculated as the initial baseline range; for load-related features, the baseline is calculated separately according to the load interval; with a 7-day window, the baseline range is updated using the incremental mean method.

[0012] Furthermore, historical fault data are collected, features are extracted and standardized; the Euclidean distance, local density and relative distance between samples are calculated; cluster centers are selected by sorting the product of local density and relative distance, and the samples are divided into M categories; the typical range of features is extracted for each type of fault.

[0013] Furthermore, the physical coordinates of the temperature anomaly point are located based on distributed monitoring. If the spatial gradient of the anomaly point exceeds 5°C / meter, the anomaly area is expanded to the adjacent 0.5-meter range. The duration of the anomaly is recorded. The reconstruction error at each scale is calculated to determine the scale of the anomaly's origin. A graded warning is triggered based on the anomaly's conduction range, energy growth multiple, and duration. If the anomaly occurs in a single phase, is transmitted to the 1-minute scale, and the energy increases by 2 to 5 times within 15 minutes and lasts for more than 15 minutes, a level 1 warning is triggered; If the anomaly occurs in two phases and is transmitted to the 10-minute scale, and the energy increases by 5 to 10 times within 15 minutes and lasts for more than 15 minutes, a second-level warning is triggered; If anomalies occur in the three phases and are transmitted to the hourly scale, the energy will increase more than 10 times within 15 minutes, triggering a level 3 warning.

[0014] An analysis and early warning system for busbar temperature changes in a power distribution cabinet, comprising: The data acquisition module collects temperature data on the three-phase busbar of the power distribution cabinet, preprocesses the temperature data, outputs a smoothed temperature sequence, divides the temperature sequence into time and space units, and calculates the spatial and temporal characteristics of each time and space unit; The frequency domain analysis module performs frequency domain analysis on the temperature data of time and space units, determines the frequency band boundaries through fast Fourier transform, constructs low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions, decomposes the temperature data into low-frequency components, medium-frequency components, and high-frequency components, performs multi-layer wavelet decomposition on each component, extracts sub-components at different time scales, and calculates temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics; The pattern library management module collects fault-free data to establish a normal pattern library, calculates the initial baseline range of each feature, updates the baseline range according to the load interval, collects historical fault data to establish a fault pattern library, extracts the typical range of fault features through cluster analysis, calculates the deviation between features and baselines in real time, triggers fault pattern library retrieval, and marks and updates potential new faults with feedback. The abnormal warning module locates temperature anomalies based on distributed monitoring, expands the abnormal area and records the duration, calculates the reconstruction error at each scale, determines the scale of the anomaly origin, and triggers graded warnings based on the abnormal conduction range, energy growth multiple and duration.

[0015] (3) Beneficial effects The present invention provides a method and system for analyzing and warning temperature changes of busbars in a power distribution cabinet, which has the following beneficial effects: (1) By dynamically adjusting the noise threshold, false alarms and missed alarms caused by load fluctuations and ambient temperature changes are effectively reduced. The weighted least squares method is used to fit the temperature curve and combined with the sliding average method for secondary smoothing, which significantly improves the accuracy and stability of the temperature data. The time-space unit is divided and the cabinet door status correction coefficient is introduced, which enhances the comprehensiveness and reliability of the monitoring and provides a high-quality data basis for subsequent analysis.

[0016] (2) Through frequency domain analysis and adaptive empirical wavelet decomposition, the low-frequency trend, medium-frequency load fluctuation and high-frequency transient abnormal characteristics of temperature data are accurately extracted. Combined with multi-layer wavelet decomposition, sub-components of different time scales are refined to fully capture the abnormal conduction law, calculate the spatiotemporal coupling and physical coupling characteristics, and deeply integrate multi-dimensional information such as time correlation, spatial distribution and current change, significantly improving the accuracy of fault identification and early warning capabilities, and effectively distinguishing normal fluctuations from potential faults.

[0017] (3) By dynamically establishing and updating the normal mode library, the adaptability of the baseline to load and environmental changes is significantly improved, misjudgment is reduced, cluster analysis is used to accurately extract the typical feature range of historical faults, build a fault mode library, calculate the feature deviation in real time and trigger the fault library retrieval, quickly match known fault modes, mark potential new faults and feedback closed-loop updates, and continuously optimize the coverage and accuracy of the fault library, thereby greatly improving the accuracy of fault identification and the system's self-learning ability.

[0018] (4) By dynamically expanding the abnormal area to avoid local missed detection, recording the duration of the abnormality and drawing the characteristic change curve, a quantitative basis is provided for the fault evolution, the reconstruction error of each scale is calculated, the origin scale of the abnormality is accurately identified, the root cause of the fault is quickly located, and a graded warning is triggered based on the abnormal conduction range, energy growth multiple and duration, so as to achieve accurate assessment of the severity of the fault and targeted response, significantly improving the timeliness of the warning and the efficiency of operation and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the steps of the method for analyzing and warning the temperature change of the busbar of the power distribution cabinet of the present invention; Figure 2 This is a structural diagram of the analysis and early warning system for busbar temperature changes in a power distribution cabinet of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1 The present invention provides a method for analyzing and warning the temperature change of a distribution cabinet busbar, comprising the following steps: Step 1: Collect temperature data, current, and cabinet door status data on the three-phase busbar of the distribution cabinet, preprocess the temperature data, output a smoothed temperature sequence, divide the temperature sequence into time and space units, and calculate the spatial and temporal characteristics of each time and space unit; The step 1 includes the following contents: Step 101: On the three-phase busbars A / B / C of the power distribution cabinet, a distributed fiber optic sensing point is laid every 0.5 m along the length. Both ends of the fiber optic cable are connected to the DTS host. A temperature and humidity sensor is installed at the top, middle, and bottom of the power distribution cabinet. The top of the power distribution cabinet is 0.3 m from the top of the cabinet, the middle is the vertical center of the cabinet, and the bottom is 0.3 m from the bottom of the cabinet. A current transformer is installed at the three-phase incoming line end to collect current signals through the secondary side. A magnetic proximity switch is installed on the cabinet door hinge to output a "1" (off) or "0" (on) status signal. Step 102: Set the rated current as a reference (e.g., 630 amps), and the real-time load current and ambient temperature as variables. The dynamic noise threshold is calculated as follows: When the load current does not exceed 80% of the rated current and the ambient temperature does not exceed 35°C, the noise threshold is 3°C (the noise threshold is the noise threshold of the temperature data, in °C); When the load current exceeds 80% of the rated current, the noise threshold increases by 0.2°C for every 10 amperes above rated current; When the ambient temperature exceeds 35°C, the noise threshold increases by 0.1°C for every 1°C increase; Step 103: For each temperature data point, take all the data within 10 seconds before and after it (a total of 21 data points) to form an analysis window, and calculate the weight of each data in the window using the Gaussian kernel function (the closer to the current point, the higher the weight, specifically through Calculate, where w represents the weight, x The temperature curve within the window is fitted using the weighted least squares method. If the deviation between any data point and the temperature curve exceeds the dynamic noise threshold, it is considered a noise point and replaced with the value of the temperature curve at that data point. The processed data is then quadratically smoothed using the sliding average method to eliminate high-frequency jitter caused by inherent sensor noise, and the smoothed temperature series is output. Step 104: Divide the temperature series into spatiotemporal units according to the "10 minutes × 0.5 meters" standard. Each unit contains temperature data for 120 time points (collected once every 5 seconds within 10 minutes) × 1 spatial point (monitoring point every 0.5 meters). Spatial and temporal features are calculated for each unit. When calculating the spatial feature, the temperature gradient is obtained by dividing the temperature difference between two adjacent 0.5-meter data points by 0.5 meters. When calculating the temporal feature, the rate of change is obtained by dividing the difference between the maximum and minimum temperatures in the unit by 600 seconds. If the cabinet door status corresponding to the unit is "open" (the status signal is 0), the ambient temperature data of the unit is multiplied by a weight coefficient of 1.2 to compensate for the accelerated heat dissipation caused by the cabinet door opening. The unit label is marked as "cabinet door open".

[0022] When using, combine the contents of steps 101 to 104: By dynamically adjusting the noise threshold, false alarms and missed alarms caused by load fluctuations and ambient temperature changes are effectively reduced. The weighted least squares method is used to fit the temperature curve and combined with the sliding average method for secondary smoothing, which significantly improves the accuracy and stability of temperature data. The division of time and space units and the introduction of cabinet door status correction coefficients enhance the comprehensiveness and reliability of monitoring, providing a high-quality data foundation for subsequent analysis.

[0023] Step 2: Perform frequency domain analysis on the temperature data of the time-space unit. Determine the frequency band boundaries through fast Fourier transform, construct low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions, decompose the temperature data into low-frequency components, medium-frequency components, and high-frequency components. Perform multi-layer wavelet decomposition on each component, extract sub-components at different time scales, and calculate the temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics. The second step includes the following contents: Step 201: Perform a fast Fourier transform on the temperature data of each time-space unit to obtain a power spectrum in the frequency range of 0-0.1 Hz, and preliminarily set the boundary between low frequency and medium frequency from the power spectrum. , and the boundary between mid-frequency and high-frequency , such as 0.001 Hz and 0.01 Hz, calculate the mid-frequency band (boundary and Pearson correlation coefficient of power and load current: If the absolute value of the correlation coefficient is not less than 0.7, the current boundary is retained. and ; If the absolute value of the correlation coefficient is less than 0.7, the boundaries are adjusted in steps of 0.0005 Hz (the boundary between low frequency and medium frequency is limited to 0.0005~0.002 Hz, and the boundary between medium frequency and high frequency is limited to 0.008~0.02 Hz) until the absolute value of the correlation coefficient is not less than 0.7; Step 202: Based on and Three empirical wavelet basis functions are constructed, including low-frequency basis functions, medium-frequency basis functions, and high-frequency basis functions. The low-frequency basis function covers the frequency range from 0 to the low-frequency / medium-frequency boundary and is used to extract long-term temperature trends; the medium-frequency basis function covers the frequency range from the medium-frequency / low-frequency boundary to the medium-frequency / high-frequency boundary and is used to extract temperature changes related to load fluctuations; the high-frequency basis function covers the frequency range from the medium-frequency / high-frequency boundary to 0.1 Hz and is used to extract rapid fluctuations related to contact status, such as whether the contact is tight or whether there is local sparking; Step 203: Decompose the temperature data of the three phases A / B / C using the empirical wavelet basis function to obtain the low-frequency component, intermediate-frequency component, and high-frequency component of each phase. Calculate the instantaneous phase difference of the high-frequency components of the three phases. For example, the phase difference between phases A and B is the absolute value of their phase angles. If the phase difference between any two phases exceeds 30°, adjust the center frequency of the empirical wavelet basis function (by 0.0001 Hz each time) and re-decompose until all phase differences do not exceed 30°. Step 204: Using the db4 wavelet basis, perform three-layer wavelet decomposition on the low-frequency, medium-frequency, and high-frequency components, respectively. The low-frequency component is decomposed into day scale, hour scale, and minute scale, the medium-frequency component is decomposed into 30-minute scale, 10-minute scale, and 1-minute scale, and the high-frequency component is decomposed into 30-second scale, 10-second scale, and 1-second scale. Step 205: Within the space-time unit, record the 10-second peak occurrence time of the high-frequency component and the 1-minute peak occurrence time of the medium-frequency component, calculate the time difference (the absolute value of the difference between the two time points), and use it as the temporal coupling feature. For any monitoring point on phase A and the symmetrical point on phase B (at the same distance from the cabinet top), calculate the temperature difference between the low-frequency components. Then divide the temperature difference by 0.5 meters (the phase-to-phase distance is approximately 0.5 meters) to obtain the spatial gradient as the spatial coupling feature. Within the space-time unit, calculate the maximum fluctuation amplitude of the medium-frequency component and the current change in the corresponding time period. The ratio of the two is the coupling coefficient, which serves as the physical coupling feature.

[0024] When using, combine the contents of steps 201 to 205: Through frequency domain analysis and adaptive empirical wavelet decomposition, the low-frequency trends, medium-frequency load fluctuations and high-frequency transient anomaly characteristics of temperature data are accurately extracted. Combined with multi-layer wavelet decomposition, sub-components of different time scales are refined to comprehensively capture the abnormal conduction laws, calculate the spatiotemporal coupling and physical coupling characteristics, and deeply integrate multi-dimensional information such as time correlation, spatial distribution and current changes. This significantly improves the accuracy of fault identification and early warning capabilities, and effectively distinguishes normal fluctuations from potential faults.

[0025] Step 3: Collect fault-free data to establish a normal pattern library, calculate the initial baseline range of each feature, update the baseline range according to the load range, collect historical fault data to establish a fault pattern library, extract the typical range of fault features through cluster analysis, calculate the deviation between the feature and the baseline in real time, trigger the fault pattern library search, mark potential new faults, and provide feedback updates; The step three includes the following contents: Step 301: Collect fault-free data for several months after the equipment is put into operation. During this period, there are no warning records and the equipment is confirmed to be normal by operation and maintenance. Process the data according to the methods of steps 1 and 2 to obtain a three-dimensional "component-scale-feature" data set. For each feature, calculate its 95% confidence interval as the initial baseline range. For example, the normal range of the time coupling feature is 0 to 5 seconds. Step 302: For load-related features (such as time difference, spatial gradient, and coupling coefficient), baselines are calculated for each load range (0-30%, 30%-60%, and 60%-100% rated current). (For example, when the load is between 30% and 60%, the normal range of the coupling coefficient is 0.2-0.5°C / ampere). Using a 7-day window, 24 hours of normal data are added daily. The baseline range is updated using the incremental mean (with a decay coefficient of 0.9). For example, when the baseline is updated with the newly added 24 hours of normal data, the historical baseline and the new data mean are weighted using a decay coefficient of 0.9. The new baseline = historical baseline × 0.9 + new data mean × 0.1. If the mean of a feature increases from 2°C to 2.1°C, the upper limit of the 95% range is adjusted upwards. Step 303: Collect historical fault records, such as poor contact, insulation aging, and short circuits. Each fault must include complete data from "one hour before occurrence → occurrence → after processing". Extract features according to the methods of steps 1 and 2. Use cross-scale coupling features as input and normalize them to the interval [0, 1]. Calculate the Euclidean distance between samples. Preset a cutoff distance (such as 0.3) and calculate the local density and relative distance. The relative distance is the minimum distance to the sample with higher density. For each sample, calculate the Euclidean distance between the sample and all other samples. Divide each distance by the cutoff distance and take the square. Then take the negative of the square value as the exponent to calculate the natural exponential function value. Finally, add the natural exponential function values ​​corresponding to all samples. The sum obtained is the local density of the sample. Step 304: Sort by the product of local density and relative distance, take the top M peaks as cluster centers, and divide the samples into M categories. For each type of fault, extract the typical range of its characteristics. For example, poor contact: time difference exceeds 10 seconds, spatial gradient exceeds 5°C / meter, and the high-frequency component fluctuation frequency on the 10-second scale is 5-10 times / minute; insulation aging: daily offset of the low-frequency component exceeds 2°C (compared with the previous week), and the coupling coefficient increases by 0.1°C / ampere per month; short circuit precursor: 1-second high-frequency component jump exceeds 5°C, spatial gradient exceeds 10°C / meter, and duration exceeds 30 seconds. Step 305: The normal pattern library calculates the deviation between the current feature and the baseline every 10 minutes. The deviation is the Euclidean distance divided by the baseline standard deviation. For example, for the feature "time difference", assuming that its normal data fluctuates within the range of 0 to 5 seconds, the baseline standard deviation of the time difference is obtained by calculating the square root of the average of the squares of the deviations of each data point from the mean. When the deviation exceeds the deviation threshold, the fault pattern library search is triggered: The fault library uses the K-nearest neighbor algorithm, takes K nearest neighbor samples, calculates the similarity between the current feature and the samples in the library, and obtains the similarity by subtracting the ratio of the distance between samples to the maximum distance from 1. If the similarity is less than 50%, it is marked as a "potential new fault" and pushed to the operation and maintenance APP through the server, including the characteristic curve and positioning information. The operation and maintenance personnel must provide feedback of "confirmed fault" or "false alarm" within 48 hours. After confirmation, the new fault feature is automatically added to the fault library and the cluster center is updated. If it is a false alarm, the baseline range of the normal mode library will be expanded by 10%.

[0026] When using, combine the contents of steps 301 to 305: By dynamically establishing and updating the normal mode library, the baseline's adaptability to load and environmental changes is significantly improved, misjudgments are reduced, cluster analysis is used to accurately extract the typical feature range of historical faults, build a fault mode library, calculate feature deviations in real time and trigger fault library retrieval, quickly match known fault modes, mark potential new faults and provide feedback for closed-loop updates, and continuously optimize the coverage and accuracy of the fault library, thereby significantly improving the accuracy of fault identification and the system's self-learning capabilities.

[0027] Step 4: Locate temperature anomalies based on distributed monitoring, expand the anomaly area and record the duration, calculate the reconstruction error at each scale, determine the scale of the anomaly origin, and trigger graded warnings based on the anomaly conduction range, energy growth multiples, and duration.

[0028] The fourth step includes the following contents: Step 401: Based on distributed monitoring using optical fiber sensors, locate the physical coordinates of the temperature anomaly point, such as point A 2.3 meters from the cabinet top. If the spatial gradient of the anomaly point exceeds 5°C / meter, expand the anomaly region to an adjacent 0.5-meter range. Record the moment when the anomaly feature first exceeds the baseline, calculate its duration, and plot a curve of the feature value over time, such as a curve showing the coupling coefficient rising from 0.4°C / ampere to 0.8°C / ampere. Step 402: Calculate the reconstruction error (the difference between the original signal and the reconstructed signal) at each decomposition scale. The scale with the highest error percentage is the scale at which the anomaly originates. For example, if the error percentage at the 10-second scale is 65%, the anomaly originates from high-frequency rapid fluctuations. If the anomaly originates at the 1-second scale, check whether anomalies occur synchronously at the 10-second scale and the 1-minute scale, i.e., exceed the baseline, and the time difference between the occurrence of anomalies at each scale does not exceed 3 minutes. The original signal refers to the "smoothed temperature series" output after secondary smoothing processing, that is, the original monitoring data of the busbar temperature in the space-time unit.

[0029] The reconstructed signal refers to the temperature signal synthesized by reverse synthesis of the decomposition results of each scale after multi-scale decomposition: for the three-layer decomposition results of low-frequency, medium-frequency, and high-frequency components, according to the reconstruction rules of wavelet transform, the detail components of each layer are superimposed with the approximate components to synthesize the complete signals of the corresponding components respectively. Then, the reconstructed signals of low-frequency, medium-frequency, and high-frequency are added together to obtain a reconstructed signal with the same length as the original signal. Energy is defined as the sum of the amplitude of each feature multiplied by its duration. For example, if the high-frequency component has a fluctuation amplitude of 0.5°C on a 10-second scale and lasts for 2 minutes, the energy is 0.5°C × 120 seconds = 60°C·seconds. Step 403: If a single phase has an anomaly, and the anomaly is transmitted from the fine scale to the 1-minute scale, and the energy increases by 2 to 5 times within 15 minutes, and the duration exceeds 15 minutes, a level 1 warning is triggered, and the level 1 warning is notified to the operation and maintenance team; If two phases have an abnormality, such as phase A and phase B, and the abnormality is transmitted to the 10-minute scale, the energy increases by 5 to 10 times within 15 minutes, and the duration exceeds 15 minutes, a level 2 warning is triggered, and the level 2 warning notifies the operation and maintenance person in charge; If abnormalities occur in all three phases and are transmitted to the hourly scale, the energy will increase more than 10 times within 15 minutes, triggering a level 3 warning. The level 3 warning will notify the operation and maintenance person in charge and the emergency repair team.

[0030] When using, combine the contents of step 401 to step 403: By dynamically expanding the abnormal area to avoid local missed detection, recording the duration of the abnormality and drawing the characteristic change curve, a quantitative basis is provided for fault evolution, the reconstruction error at each scale is calculated, the origin scale of the abnormality is accurately identified, and the root cause of the fault is quickly located. Based on the abnormal conduction range, energy growth multiples and duration, graded warnings are triggered to achieve accurate assessment of the severity of the fault and targeted response, significantly improving the timeliness of warnings and operation and maintenance efficiency.

[0031] See also Figure 2 The present invention provides an analysis and early warning system for busbar temperature changes in a power distribution cabinet, comprising: a data acquisition module, a frequency domain analysis module, a pattern library management module, and an abnormality early warning module, wherein: The data acquisition module collects temperature data on the three-phase busbar of the power distribution cabinet, preprocesses the temperature data, outputs a smoothed temperature sequence, divides the temperature sequence into time and space units, and calculates the spatial and temporal characteristics of each time and space unit; The frequency domain analysis module performs frequency domain analysis on the temperature data of time and space units, determines the frequency band boundaries through fast Fourier transform, constructs low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions, decomposes the temperature data into low-frequency components, medium-frequency components, and high-frequency components, performs multi-layer wavelet decomposition on each component, extracts sub-components at different time scales, and calculates temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics; The pattern library management module collects fault-free data to establish a normal pattern library, calculates the initial baseline range of each feature, updates the baseline range according to the load interval, collects historical fault data to establish a fault pattern library, extracts the typical range of fault features through cluster analysis, calculates the deviation between features and baselines in real time, triggers fault pattern library retrieval, and marks and updates potential new faults with feedback. The abnormal warning module locates temperature anomalies based on distributed monitoring, expands the abnormal area and records the duration, calculates the reconstruction error at each scale, determines the scale of the anomaly origin, and triggers graded warnings based on the abnormal conduction range, energy growth multiple and duration.

[0032] In the application, the several formulas involved are all calculated by taking their numerical values ​​after removing the dimensions, and the formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the latest real situation. The coefficients in the formula are set by technical personnel in this field according to actual conditions.

[0033] The above embodiments can be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented using a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0034] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0035] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for analyzing and warning temperature changes of busbars in a power distribution cabinet, characterized by: include: Collect temperature data on the three-phase busbars of the power distribution cabinet, preprocess the temperature data, output a smoothed temperature sequence, divide the temperature sequence into space-time units, and calculate the spatial and temporal characteristics of each space-time unit; Frequency domain analysis is performed on the temperature data of the time-space unit. The frequency band boundaries are determined through fast Fourier transform. Low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions are constructed. The temperature data is decomposed into low-frequency components, medium-frequency components, and high-frequency components. Multi-layer wavelet decomposition is performed on each component to extract sub-components at different time scales. The temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics are calculated. Collect fault-free data to establish a normal pattern library, calculate the initial baseline range of each feature, update the baseline range according to the load interval, collect historical fault data to establish a fault pattern library, extract the typical range of fault features through cluster analysis, calculate the deviation between the feature and the baseline in real time, trigger the fault pattern library search, mark potential new faults and provide feedback updates; Based on distributed monitoring, the temperature anomaly points are located, the abnormal area is expanded and the duration is recorded, the reconstruction error at each scale is calculated, the scale of the anomaly origin is determined, and graded warnings are triggered according to the abnormal conduction range, energy growth multiples and duration.

2. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 1, characterized in that: Collecting the temperature data, current and cabinet door status data of the three-phase busbar of the power distribution cabinet includes the following steps: Distributed fiber optic sensing points are laid every 0.5 meters along the length of the A / B / C three-phase busbars of the distribution cabinet, and both ends of the fiber optic cables are connected to the DTS host; temperature and humidity sensors are installed on the top, middle and bottom of the distribution cabinet; current transformers are installed at the three-phase incoming line end to collect current signals; magnetic proximity switches are installed on the cabinet door hinges to output cabinet door status signals.

3. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 2, characterized in that: Preprocessing includes the following steps: Dynamically adjust the noise threshold based on load current and ambient temperature: When the load current does not exceed 80% of the rated current and the ambient temperature does not exceed 35°C, the noise threshold is 3°C; when the load current exceeds 80% of the rated current, the noise threshold increases by 0.2°C for every 10 amperes above the rated current; when the ambient temperature exceeds 35°C, the noise threshold increases by 0.1°C for every 1°C above the ambient temperature; For each temperature data point, the data within 10 seconds before and after it are taken to form an analysis window. The weight of each data in the window is calculated by the Gaussian kernel function, and the temperature change curve in the window is fitted using the weighted least squares method. If the deviation between any data point and the temperature change curve exceeds the noise threshold, it is determined to be a noise point and replaced by the value of the temperature change curve at the time of the data point. The processed data is secondary smoothed using the sliding average method, and the smoothed temperature series is output.

4. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 1, characterized in that: Perform fast Fourier transform on the temperature data of each time-space unit to obtain the power spectrum and preliminarily set the boundary between low frequency and medium frequency. , and the boundary between mid-frequency and high-frequency , calculate the Pearson correlation coefficient between the power in the mid-frequency band and the load current; If the absolute value of the correlation coefficient is not less than 0.7, the current boundary is retained. and ;If the absolute value of the correlation coefficient is less than 0.7, adjust the boundary in steps of 0.0005 Hz until the absolute value of the correlation coefficient is not less than 0.

7.

5. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 4, characterized in that: based on and Construct empirical wavelet basis functions, including low-frequency basis functions, medium-frequency basis functions and high-frequency basis functions; The temperature data of phases A / B / C are decomposed using the empirical wavelet basis function to obtain the low-frequency component, intermediate-frequency component, and high-frequency component of each phase. The instantaneous phase difference of the high-frequency components of the three phases is calculated. If the phase difference between any two phases exceeds 30°, the center frequency of the empirical wavelet basis function is adjusted and re-decomposed until all phase differences do not exceed 30°.

6. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 5, characterized in that: The db4 wavelet basis is used to perform three-layer wavelet decomposition on the low-frequency, medium-frequency and high-frequency components respectively. The low-frequency component is decomposed into day scale, hour scale and minute scale, the medium-frequency component is decomposed into 30-minute scale, 10-minute scale and 1-minute scale, and the high-frequency component is decomposed into 30-second scale, 10-second scale and 1-second scale.

7. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 1, characterized in that: Fault-free data from several months after the equipment is put into operation is collected, and the 95% confidence interval of each feature is calculated as the initial baseline range. For load-related features, the baseline is calculated separately according to the load interval. The baseline range is updated using the incremental mean method with a 7-day window.

8. The method for analyzing and warning temperature changes of a distribution cabinet busbar according to claim 7, characterized in that: Collect historical fault data, extract features and standardize them; calculate the Euclidean distance, local density and relative distance between samples; select cluster centers by sorting the product of local density and relative distance, and divide the samples into M categories; extract the typical range of features for each type of fault.

9. The method for analyzing and warning of busbar temperature changes in a power distribution cabinet according to claim 8, characterized in that: Based on distributed monitoring, the physical coordinates of the temperature anomaly point are located; if the spatial gradient of the anomaly point exceeds 5°C / meter, the anomaly area is expanded to the adjacent 0.5-meter range; and the duration of the anomaly is recorded; Calculate the reconstruction error at each scale and determine the scale of the anomaly origin; Trigger graded warnings based on abnormal conduction range, energy growth multiples, and duration: If the anomaly occurs in a single phase, is transmitted to the 1-minute scale, and the energy increases by 2 to 5 times within 15 minutes and lasts for more than 15 minutes, a level 1 warning is triggered; If the anomaly occurs in two phases and is transmitted to the 10-minute scale, and the energy increases by 5 to 10 times within 15 minutes and lasts for more than 15 minutes, a second-level warning is triggered; If anomalies occur in the three phases and are transmitted to the hourly scale, the energy will increase more than 10 times within 15 minutes, triggering a level 3 warning.

10. An analysis and early warning system for busbar temperature changes in a power distribution cabinet, used to implement the method according to any one of claims 1 to 9, characterized in that: include: The data acquisition module collects temperature data on the three-phase busbar of the power distribution cabinet, preprocesses the temperature data, outputs a smoothed temperature sequence, divides the temperature sequence into time and space units, and calculates the spatial and temporal characteristics of each time and space unit; The frequency domain analysis module performs frequency domain analysis on the temperature data of time and space units, determines the frequency band boundaries through fast Fourier transform, constructs low-frequency, medium-frequency, and high-frequency empirical wavelet basis functions, decomposes the temperature data into low-frequency components, medium-frequency components, and high-frequency components, performs multi-layer wavelet decomposition on each component, extracts sub-components at different time scales, and calculates temporal coupling characteristics, spatial coupling characteristics, and physical coupling characteristics; The pattern library management module collects fault-free data to establish a normal pattern library, calculates the initial baseline range of each feature, updates the baseline range according to the load interval, collects historical fault data to establish a fault pattern library, extracts the typical range of fault features through cluster analysis, calculates the deviation between features and baselines in real time, triggers fault pattern library retrieval, and marks and updates potential new faults with feedback. The abnormal warning module locates temperature anomalies based on distributed monitoring, expands the abnormal area and records the duration, calculates the reconstruction error at each scale, determines the scale of the anomaly origin, and triggers graded warnings based on the abnormal conduction range, energy growth multiple and duration.

Images