Device and method for monitoring running state of electrical cabinet

Abstract

The invention discloses a device and a method for monitoring the running state of an electrical cabinet, relates to the technical field of electrical data processing, and can solve the technical problems of fault early warning lag and insufficient early abnormality recognition capability in the monitoring process of the running state of the electrical cabinet. Comprising the following steps: acquiring internal temperature data, external temperature data and electrical parameter data of an electrical cabinet; determining a first judgment feature according to the internal temperature data and the external temperature data of the electrical cabinet; wherein the first judgment feature is used for representing the similarity between the spatial distribution of the temperature and Gaussian distribution; determining a second judgment feature according to the internal temperature data and the external temperature data of the electrical cabinet; wherein the second judgment feature is used for representing the stability of the temperature in the spatial dimension and the time dimension; determining a temperature state index according to the first judgment feature and the second judgment feature; and judging whether the operation state of the electrical cabinet is abnormal according to the temperature state index and the electrical parameter data.

Description

Technical Field

[0001] This invention relates to the field of electrical data processing technology, and specifically to a device and method for monitoring the operating status of an electrical cabinet. Background Technology

[0002] As a core component of power distribution and control systems, the operational stability of electrical cabinets directly affects the continuity and safety of the entire industrial production process. To ensure reliable operation, real-time monitoring of key parameters such as internal temperature, current, and voltage is typically required to assess equipment status and provide early warnings of faults. Current technologies for electrical cabinet status monitoring primarily rely on the collection and analysis of single or multiple temperature data points and electrical parameters, using fixed thresholds or historical statistical methods for anomaly detection. However, these methods often suffer from response lag, making it difficult to promptly detect early fault signs and resulting in unsatisfactory early warning effects, thus limiting the overall performance and reliability of status monitoring to some extent. Summary of the Invention

[0003] To address the technical problems of delayed fault warnings and insufficient early anomaly identification capabilities in electrical cabinet operation status monitoring caused by relying on a single temperature parameter and fixed threshold for fault diagnosis, the present invention aims to provide an electrical cabinet operation status monitoring device and method, the specific technical solution of which is as follows: In a first aspect, the present invention provides a method for monitoring the operating status of an electrical cabinet, comprising: acquiring internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet; determining a first judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein the first judgment feature is used to characterize the similarity between the spatial distribution of temperature and a Gaussian distribution; determining a second judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein the second judgment feature is used to characterize the stability of temperature in both spatial and temporal dimensions; determining a temperature state index based on the first judgment feature and the second judgment feature; and determining whether the operating status of the electrical cabinet is abnormal based on the temperature state index and the electrical parameter data.

[0004] In one possible implementation, acquiring the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet specifically includes: synchronously acquiring the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet at a preset sampling period; wherein, the internal temperature data is acquired through a temperature monitoring array arranged inside the electrical cabinet, the external temperature data is acquired through an external temperature sensor arranged outside the electrical cabinet, and the electrical parameter data includes voltage data and current data, with each sampling period corresponding to a sampling moment.

[0005] In one possible implementation, after acquiring the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet, the method further includes: determining a true temperature signal based on the internal temperature data and external temperature data of the electrical cabinet; wherein the true temperature signal is used to characterize the temperature of the heating element inside the electrical cabinet after eliminating the influence of ambient temperature.

[0006] In one possible implementation, the first judgment feature is determined based on the internal and external temperature data of the electrical cabinet. Specifically, this includes: constructing a Gaussian kernel centered on each real temperature signal value based on the real temperature signal; determining the correlation between each Gaussian kernel and the corresponding local temperature signal matrix; wherein the local temperature signal matrix is ​​a set of temperature signals within a local area centered on the current real temperature signal value, and the correlation is used to characterize the degree of agreement between the local temperature distribution and the Gaussian distribution; and determining the first judgment feature based on the correlation and the number of invalid signals in the local temperature signal matrix.

[0007] In one possible implementation, a second judgment feature is determined based on the internal and external temperature data of the electrical cabinet. Specifically, this includes: determining a spatial variation feature based on the actual temperature signal; wherein the spatial variation feature is used to characterize the smoothness of the overall temperature distribution change between adjacent sampling times; determining a temporal variation feature based on the actual temperature signal; wherein the temporal variation feature is used to characterize the stability of the temperature change of a single temperature sensor over time; and determining the second judgment feature based on the spatial variation feature and the temporal variation feature.

[0008] In one possible implementation, spatial variation characteristics are determined based on the real temperature signal. Specifically, this includes: arranging the real temperature signal at each sampling time into a spatial sequence according to the sensor location, and calculating the overall rate of change between the spatial sequences at adjacent sampling times; performing trend enhancement processing on the spatial sequence at the current sampling time based on the overall rate of change, and calculating the similarity between adjacent spatial sequences after trend enhancement, using the similarity as a spatial variation characteristic.

[0009] In one possible implementation, determining the time variation characteristics based on the real temperature signal specifically includes: dividing the real temperature signal from each internal temperature sensor in the temperature monitoring array at different sampling times into multiple time series of equal length; calculating the first similarity between adjacent time series as a first time feature; predicting the current time series based on historical time series, and calculating the second similarity between the predicted series and the current time series as a second time feature; and determining the time variation characteristics based on the first and second time features.

[0010] In one possible implementation, the temperature state index is determined based on the first judgment feature and the second judgment feature, specifically including: when the cumulative duration corresponding to the current sampling time is less than the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the spatial change feature; when the cumulative duration corresponding to the current sampling time is greater than or equal to the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the second judgment feature.

[0011] In one possible implementation, the determination of whether the operating status of the electrical cabinet is abnormal is based on the temperature status index and electrical parameter data. Specifically, this includes: acquiring historical data from the same period in the past; the historical data includes temperature status index and electrical parameter data from the same historical period; determining the normal fluctuation range of the temperature status index and electrical parameter data based on the historical data; comparing the temperature status index and electrical parameter data at the current sampling time with the corresponding normal fluctuation range; if the current temperature status index or any electrical parameter data exceeds the corresponding normal fluctuation range, the operating status of the electrical cabinet is determined to be abnormal.

[0012] Secondly, the present invention provides an operating status monitoring device for an electrical cabinet, comprising: an acquisition unit and a processing unit; the acquisition unit is used to acquire internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet; the processing unit is used to determine a first judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein the first judgment feature is used to characterize the similarity between the spatial distribution of temperature and a Gaussian distribution; the processing unit is further used to determine a second judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein the second judgment feature is used to characterize the stability of temperature in both spatial and temporal dimensions; the processing unit is further used to determine a temperature state index based on the first judgment feature and the second judgment feature; the processing unit is further used to determine whether the operating status of the electrical cabinet is abnormal based on the temperature state index and the electrical parameter data.

[0013] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer-executable instructions, and when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the electrical cabinet operation status monitoring method as described in the first aspect and any possible implementation thereof.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by an electronic device of the present invention, cause the electronic device to perform the method for monitoring the operating status of an electrical cabinet as described in the first aspect and any possible implementation thereof.

[0015] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the electronic device of the present invention to perform the method for monitoring the operating status of an electrical cabinet as described in the first aspect and any possible implementation thereof.

[0016] In a sixth aspect, the present invention provides a chip system applied to an electrical cabinet operation status monitoring device; the chip system includes one or more interface circuits and one or more processors. The interface circuits and the processors are interconnected via lines; the interface circuits are used to receive signals from the memory of the electrical cabinet operation status monitoring device and send the signals to the processors, the signals including computer instructions stored in the memory. When the processor executes the computer instructions, the electrical cabinet operation status monitoring device performs the electrical cabinet operation status monitoring method as described in the first aspect and any possible design of the present invention.

[0017] The present invention has the following beneficial effects: by constructing a temperature monitoring array and combining it with environmental temperature compensation, a multi-dimensional temperature anomaly identification mechanism based on spatial distribution characteristics and temporal evolution law is established. The first judgment feature is used to accurately capture Gaussian distribution distortion caused by local overheating. The second judgment feature is used to synchronously perceive the abrupt changes in the spatial sequence and the abnormal fluctuations in the temporal sequence of the temperature field. A phased weighted fusion strategy is adopted to generate a dynamically evolving temperature state index. Finally, combined with the multi-parameter collaborative criteria of electrical parameters, sensitive detection and reliable early warning of early faults in electrical cabinets are realized, which significantly improves the timeliness and accuracy of status monitoring and effectively overcomes the technical defects of traditional single threshold criteria, such as high lag and high false alarm rate. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for monitoring the operating status of an electrical cabinet according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating another method for monitoring the operating status of an electrical cabinet, provided in one embodiment of the present invention. Figure 3 This is a flowchart illustrating another method for monitoring the operating status of an electrical cabinet, provided in one embodiment of the present invention. Figure 4 This is a schematic diagram of the structure of an electrical cabinet operation status monitoring device provided in one embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of another electrical cabinet operation status monitoring device provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] The specific solution of the electrical cabinet operation status monitoring device and method provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0023] For example, such as Figure 1 The diagram shown is a flowchart illustrating a method for monitoring the operational status of an electrical cabinet according to an embodiment of the present invention, comprising the following steps: S101. Obtain the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet.

[0024] Optionally, the electrical parameter data includes voltage data and current data.

[0025] In one possible implementation, the electrical cabinet's operational status monitoring device (hereinafter referred to as the monitoring device) synchronously acquires internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet at a preset sampling period. The internal temperature data is acquired through a temperature monitoring array located inside the electrical cabinet, while the external temperature data is acquired through an external temperature sensor located outside the electrical cabinet. It should be noted that each sampling period corresponds to a sampling moment.

[0026] For example, nine temperature sensors are arranged in a 3×3 grid pattern along the horizontal direction inside the electrical cabinet to form an internal temperature monitoring array to collect internal temperature data of the electrical cabinet; one external temperature sensor is arranged at the center of the outer side of the top of the electrical cabinet to collect external temperature data; voltage sensors and current sensors are arranged at the inlet and outlet ends of the electrical cabinet to collect voltage and current data, respectively.

[0027] It is understandable that the monitoring device sets a preset sampling period when collecting the above data in order to ensure that all data at the same timestamp are collected at the same time, thus eliminating the impact of time deviation on subsequent analysis.

[0028] Optionally, the monitoring device can also generate a synchronous trigger signal through a global sampling time module (with an accuracy of ±1ms). When the trigger signal arrives, all the above sensors simultaneously start digital-to-analog conversion and send the corresponding data to the monitoring device.

[0029] Furthermore, after collecting the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet, the monitoring device preprocesses these raw data to eliminate outliers. For example, filtering can be performed on the aforementioned three categories of data, specifically using a third-order moving average filtering algorithm.

[0030] Therefore, the monitoring device can achieve high-precision, synchronized acquisition and preprocessing of internal and external temperatures and key electrical parameters of the electrical cabinet, providing a reliable data foundation for subsequent operational status judgment.

[0031] Optionally, before executing S102 and S103, the monitoring device first determines the true temperature signal based on the internal and external temperature data of the electrical cabinet. The true temperature signal characterizes the temperature of the heating elements inside the electrical cabinet after eliminating the influence of ambient temperature.

[0032] It should be noted that the core objective of determining the true temperature signal is to eliminate the interference of external ambient temperature on the internal temperature of the electrical cabinet, retaining only the temperature characteristics generated by the internal heating elements during operation. For example, the monitoring device obtains the true temperature signal by calculating the difference between preprocessed internal and external temperature data. For instance, at the same sampling time, the nine temperature signals collected by the temperature monitoring array are subtracted from the ambient temperature signals collected by external temperature sensors to obtain nine true temperature signals for the internal components of the electrical cabinet.

[0033] It is understandable that, since the internal components of the electrical cabinet continuously generate heat during operation, their temperature is usually higher than the external ambient temperature. Therefore, the actual temperature signal is the result of subtracting the external temperature data from the internal temperature data, and this result is always non-negative. If a negative value appears, it indicates that there is an anomaly in the data (such as sensor failure or transient interference). In this case, it is corrected to 0 to ensure that subsequent calculations are not affected by invalid values.

[0034] S102. Based on the internal and external temperature data of the electrical cabinet, determine the first judgment feature. The first judgment feature is used to characterize the degree of similarity between the spatial temperature distribution and the Gaussian distribution.

[0035] Optionally, the monitoring device determines the first judgment feature through the following steps: First, the monitoring device constructs a Gaussian kernel centered on each real temperature signal value based on the real temperature signal; then, the monitoring device determines the correlation between each Gaussian kernel and the corresponding local temperature signal matrix; finally, the monitoring device determines the first judgment feature based on the correlation and the number of invalid signals in the local temperature signal matrix. Here, the local temperature signal matrix is ​​a set of temperature signals within a local region centered on the current real temperature signal value, and the correlation is used to characterize the degree of agreement between the local temperature distribution and the Gaussian distribution. It should be noted that the specific process of the monitoring device determining the first judgment feature according to the aforementioned sub-steps can be found in S201-S203 below, and will not be repeated here.

[0036] In another possible implementation, when determining the first judgment feature, the monitoring device can also use the local temperature distribution of historical normal periods as a dynamic reference. That is, the average Gaussian distribution feature of the period is obtained by statistically analyzing historical data, which replaces the fixed Gaussian kernel; the cosine similarity between the current local temperature signal matrix and the dynamic reference distribution is calculated as the degree of agreement; and the invalid signal is repaired by interpolation and the similarity is calculated again to obtain the first judgment feature.

[0037] Therefore, the monitoring device, based on real temperature signals, transforms the Gaussian similarity of the temperature spatial distribution into quantifiable feature values. This process reflects both a reference to normal temperature distribution patterns and takes into account interference from anomalous signals in the actual data, ultimately achieving a preliminary judgment on the rationality of the temperature spatial distribution.

[0038] S103. Based on the internal and external temperature data of the electrical cabinet, determine the second judgment feature. The second judgment feature characterizes the stability of temperature in both spatial and temporal dimensions; a larger value indicates greater temperature stability in both dimensions, meaning a lower degree of anomaly.

[0039] For example, the monitoring device determines the second judgment feature based on the internal temperature data and external temperature data of the electrical cabinet, specifically including the following steps: (1) Determine the spatial variation characteristics based on the actual temperature signal. The spatial variation characteristics are used to characterize the smoothness of the overall temperature distribution change between adjacent sampling times. The larger the value, the smoother the temperature distribution change, that is, the lower the degree of change.

[0040] Optionally, the monitoring device determines spatial variation characteristics based on the actual temperature signal, specifically including: first, arranging the actual temperature signal at each sampling time into a spatial sequence according to the sensor location, and calculating the overall rate of change between spatial sequences at adjacent sampling times; then, performing trend enhancement processing on the spatial sequence at the current sampling time based on the overall rate of change, and calculating the similarity between adjacent spatial sequences after trend enhancement, using the similarity as the spatial variation characteristic. It should be noted that the specific process of the monitoring device determining spatial variation characteristics based on the aforementioned two sub-steps can be found in S301-S302 below, and will not be repeated here.

[0041] In another possible implementation, when determining spatial change characteristics, the monitoring device may not arrange the data according to physical location, but instead construct a spatial sequence based on the gradient direction of the real temperature signal, making the sequence more consistent with the physical laws of heat diffusion; then, the local rate of change of the sequence at adjacent time points is calculated, that is, the amplitude of temperature change at each location; finally, after smoothing the noise through Gaussian filtering, the dynamic time warping distance of the sequence is calculated, and the distance value is used to characterize the degree of change.

[0042] Alternatively, the monitoring device can divide the interior of the electrical cabinet into several functional areas, such as the incoming line area, the outgoing line area, and the circuit breaker area, using the actual temperature signals of each area to form a sub-space sequence. Then, the rate of change of each sub-sequence at adjacent time points is calculated and assigned regional weights, and the overall rate of change is obtained by weighting. Finally, after thresholding to retain significant changes, the average cosine similarity of the sub-sequences is calculated to obtain the spatial change characteristics.

[0043] Therefore, by analyzing the spatial distribution of real temperature signals, the monitoring device transforms the drastic spatial changes in temperature distribution at adjacent times into quantifiable feature values. This process preserves the spatial correlation of temperature distribution while amplifying key change information, ultimately achieving a precise characterization of dynamic spatial temperature changes.

[0044] (2) Determine the time variation characteristics based on the actual temperature signal. The time variation characteristics are used to characterize the stability of temperature changes of a single temperature sensor over time. The larger the value, the more stable the temperature is over time, that is, the lower the degree of anomaly.

[0045] Optionally, the monitoring device determines the time variation characteristics based on the real temperature signals, specifically including: first, dividing the real temperature signals from each internal temperature sensor in the temperature monitoring array at different sampling times into multiple time series of equal length; second, calculating the first similarity between adjacent time series as the first time feature; then, predicting the current time series based on historical time series, and calculating the second similarity between the predicted series and the current time series as the second time feature; finally, determining the time variation characteristics based on the first and second time features. It should be noted that the specific process of the monitoring device determining the time variation characteristics based on the aforementioned four sub-steps can be found in S303-S306 below, and will not be repeated here.

[0046] In another possible implementation, when determining the time-varying characteristics, the monitoring device can also use wavelet decomposition to decompose the original time series data into subsequences of different time scales, corresponding to short-term fluctuations and long-term trends, respectively. Then, the subsequences of each scale are divided into fixed lengths, and the correlation coefficient between adjacent subsequences is calculated as the first time feature of that scale. Furthermore, a correlation model is constructed based on historical subsequences of the same scale to predict the current subsequence, and the prediction deviation is calculated as the second time feature of that scale. Finally, the time-varying characteristics are obtained by weighting the features of each scale.

[0047] Therefore, by establishing a time series prediction model based on historical data and combining it with a feature contribution analysis mechanism, the monitoring device has achieved in-depth mining and quantitative evaluation of temperature time series change patterns. It can not only effectively capture abnormal fluctuation trends of temperature data in the time dimension, but also accurately locate the contribution of temperature data at each sampling time to the overall judgment result. This significantly improves the time series perception capability of fault early warning and the transparency of judgment basis, providing more reliable time dimension feature support for early fault diagnosis.

[0048] (3) Determine the second judgment feature based on the spatial and temporal change characteristics.

[0049] For example, taking the t-th sampling time as an example, the monitoring device equally weights the spatial change features and temporal change features at the t-th sampling time, and uses this weighted average as the second judgment feature at each sampling time. The specific process can be found in S307 below, and will not be repeated here.

[0050] Based on the above steps (1) to (3), the monitoring device can construct a comprehensive judgment basis that can simultaneously capture the spatial anomaly pattern and temporal evolution law of the temperature field of the electrical cabinet by collaboratively analyzing the distribution change characteristics of temperature data in the spatial dimension and the sequence fluctuation characteristics in the temporal dimension. This enables multi-dimensional and forward-looking identification of early faults, effectively overcomes the limitations of single-dimensional monitoring, and significantly improves the sensitivity of the condition monitoring system to potential faults and the reliability of early warning.

[0051] S104. Determine the temperature state index based on the first judgment feature and the second judgment feature.

[0052] It should be noted that when constructing the second judgment feature, since the generation of the first time feature requires at least 20 seconds of data accumulation and the generation of the second time feature requires at least 60 seconds of data accumulation, the "time change feature" part of the second judgment feature adopts a phased construction strategy: when the cumulative duration is less than 60 seconds, the time change feature consists only of the first time feature; when the cumulative duration reaches 60 seconds, the time change feature is composed of both the first and second time features. However, when calculating the temperature state index, the system uses 20 seconds as the judgment threshold, and decides whether to use the spatial change feature or the complete second judgment feature for fusion calculation based on whether the cumulative duration reaches 20 seconds. For example, when the cumulative duration is 70 seconds, the system can generate the complete second judgment feature (where the time change feature is composed of both the first and second time features) and directly use this feature to calculate the temperature state index.

[0053] For example, when the monitoring device determines the temperature state index based on the first judgment feature and the second judgment feature, it specifically follows the following rules: (1) If the cumulative duration corresponding to the current sampling time is less than the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the spatial change feature. (2) If the cumulative duration corresponding to the current sampling time is greater than or equal to the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the second judgment feature.

[0054] Specifically, the monitoring device determines the temperature state index according to the following formula: In the above formula, This represents the temperature state index at the t-th sampling time. This represents the first judgment feature at the t-th sampling time. This represents the spatial characteristics at the t-th sampling time. This represents the second judgment feature at the t-th sampling time. and These are the weighting coefficients. . The larger the value, the higher the degree of agreement between the spatial distribution of the temperature inside the electrical cabinet and the Gaussian distribution at the t-th sampling time, the smoother the change in the space-time dimension, and the greater the probability of normal operation. The smaller the value, the greater the probability of abnormal temperature distribution, drastic changes, and abnormal operating conditions. Based on this step, the monitoring device calculates the temperature status index in real time for each sampling moment.

[0055] It should be noted that the time parameters of 20 seconds, 60 seconds, etc., given in this embodiment are illustrative examples and can be adjusted according to monitoring needs and system performance in actual applications. The specific values ​​of the weighting coefficients α and β can be determined through experimental data optimization. Different weighting ratios can be used for different application scenarios. For example, empirical values ​​of α = 0.4 and β = 0.6 are given here. In addition, the calculation order and combination method of features can also be adaptively adjusted according to the characteristics of the actual data.

[0056] S105. Based on the temperature status index and electrical parameter data, determine whether the operating status of the electrical cabinet is abnormal.

[0057] For example, the monitoring device determines whether the operating status of the electrical cabinet is abnormal based on the temperature status index and electrical parameter data, specifically including the following steps: (1) Obtain historical data for the same period. The historical data for the same period includes temperature status index and electrical parameter data for the same period in history.

[0058] Specifically, the monitoring device reads historical data with the same operating conditions as the current monitoring period from the storage unit of the electrical cabinet, including historical temperature status index sequences and historical electrical parameter data sequences, where the electrical parameter data includes voltage data and current data.

[0059] (2) Determine the normal fluctuation range of temperature status index and electrical parameter data based on historical data of the same period.

[0060] In this step, the monitoring device establishes the normal fluctuation range of each parameter based on historical data from the same period and employs statistical process control methods. Specifically, the mean and standard deviation are calculated for the historical temperature state index sequence and the historical electrical parameter data sequence, and the mean plus or minus three times the standard deviation is taken as the normal fluctuation range of each parameter.

[0061] (3) Compare the temperature status index and electrical parameter data at the current sampling time with the corresponding normal fluctuation range. If the current temperature status index or any electrical parameter data exceeds the corresponding normal fluctuation range, the electrical cabinet is judged to be in an abnormal operating state.

[0062] For example, the monitoring device compares the temperature status index and electrical parameter data at the current sampling time with the corresponding normal fluctuation range determined in the aforementioned steps. If the current temperature status index or any electrical parameter data exceeds its corresponding normal fluctuation range, the device determines that the electrical cabinet is in an abnormal operating state and generates a fault warning signal; otherwise, the device determines that the electrical cabinet is in a normal operating state.

[0063] It should be noted that the statistical process control method used in this embodiment is not limited to the three-standard-deviation criterion. In actual implementation, other statistical judgment methods with the same function can also be used, such as the interquartile range method based on box plots, the control chart method based on moving range, or nonparametric statistical methods based on percentiles. The selection criteria for historical data from the same period can also be adjusted according to the specific application scenario, including but not limited to the same seasonal period, the same workday type, or the same load condition. These alternative solutions can all achieve effective judgment of the operating status.

[0064] Based on the above technical solution, this invention establishes a multi-dimensional temperature anomaly identification mechanism based on spatial distribution characteristics and temporal evolution by constructing a temperature monitoring array and combining it with environmental temperature compensation. It accurately captures Gaussian distribution distortion caused by local overheating using a first judgment feature, and synchronously senses abrupt changes in the spatial sequence and abnormal fluctuations in the temporal sequence of the temperature field through a second judgment feature. It generates a dynamically evolving temperature state index using a phased weighted fusion strategy, and finally combines it with multi-parameter collaborative criteria of electrical parameters to achieve sensitive detection and reliable early warning of early faults in electrical cabinets. This significantly improves the timeliness and accuracy of status monitoring and effectively overcomes the technical defects of traditional single threshold criteria, such as high lag and high false alarm rate.

[0065] For example, in combination Figure 1 ,like Figure 2 The diagram shown is a flowchart illustrating another method for monitoring the operating status of an electrical cabinet according to an embodiment of the present invention. In this method, a first judgment feature is determined based on the internal and external temperature data of the electrical cabinet, specifically including the following steps: S201. Based on the real temperature signals, construct a Gaussian kernel centered on each real temperature signal value.

[0066] For example, referring to the example of the real temperature signal in S101 above, let's take the real temperature signal at the t-th sampling time (a 3×3 matrix, denoted as...) as an example. For example: Set the Gaussian kernel size to 3×3, and... Centered on each real temperature signal, a Gaussian kernel for each real temperature signal is calculated using a two-dimensional Gaussian distribution. Specifically, taking... The internal coordinates are Taking the real temperature signal as an example, the Gaussian kernel centered on this real temperature signal is denoted as... .

[0067] S202. Determine the correlation between each Gaussian kernel and its corresponding local temperature signal matrix. The local temperature signal matrix is ​​a set of temperature signals within a local region centered on the current true temperature signal value. The correlation is used to characterize the degree of agreement between the local temperature distribution and the Gaussian distribution.

[0068] In this step, in conjunction with S201 The internal coordinates are Taking the real temperature signal as an example, the Gaussian kernel centered on the real temperature signal is denoted as... The temperature signal matrix centered on this real temperature signal is denoted as... .

[0069] Optionally, before determining the correlation between each Gaussian kernel and the corresponding local temperature signal matrix, the monitoring device uses a min-max algorithm or a Z-score algorithm to... and Normalize the values ​​within. The value inside is related to All Gaussian kernels within the calculation are normalized. The value inside is only Normalization calculations are performed within the window.

[0070] For example, the monitoring device uses the correlation coefficient method to calculate and The correlation between them. The correlation coefficient method can be Pearson correlation coefficient, cross-correlation coefficient, cosine similarity, etc. The specific methods for determining the correlation based on these algorithms are existing technologies and will not be elaborated here.

[0071] S203. Determine the first judgment feature based on the correlation and the number of invalid signals in the local temperature signal matrix.

[0072] For example, the monitoring device will The number of invalid signals is used as a weight to calculate The similarity between the actual temperature signal distribution within the sample and the Gaussian distribution, used as the first judgment feature at the t-th sampling time, can be calculated using the following formula: In the above formula, This represents the first judgment feature at the t-th sampling time. express The internal coordinates are The actual temperature signal, This indicates that the calculation is performed using the correlation coefficient method. This indicates taking the maximum value. and They represent The internal coordinates are A Gaussian kernel centered on the actual temperature signal and a temperature signal matrix. express The internal coordinates are The weights of the actual temperature signal, express The number of actual temperature signals within. express The number of invalid signals within. It should be noted that if... no The center coordinates of , then There may be invalid signals, i.e., 0; The more invalid signals there are, the more likely it is to cause The lower the probability that the temperature signal within the signal follows a Gaussian distribution, the fewer invalid signals there are. Get the number of zero values.

[0073] It should be noted that the above formula uses a reciprocal weighting mechanism to handle the influence of invalid signals: First, the correlation between the Gaussian kernel and the local temperature signal matrix at each coordinate point is calculated to characterize the degree of fit between the local temperature distribution and the ideal Gaussian distribution; then, the correlation value is penalized according to the proportion of invalid signals in the local window, with a larger penalty weight for more invalid signals (correlation value divided by...). This process suppresses the contribution of data missing or sensor faulty regions. Finally, the maximum correlation value after penalty is selected from all coordinate points as the first judgment feature to ensure that the feature can accurately and robustly characterize the similarity between the temperature distribution and the Gaussian distribution, thus avoiding misjudgment due to sensor faults.

[0074] Based on the above technical solution, this embodiment of the invention performs local correlation analysis between the real temperature signal at each sampling moment and a Gaussian kernel constructed with the sample as the center, and introduces the number of invalid signals as a weight adjustment factor, thereby realizing a quantitative evaluation of the fit between the spatial distribution pattern of the temperature field and the typical fault thermal diffusion model. This approach enables the system to capture spatial distribution anomalies that occur before the absolute temperature exceeds the limit, thus advancing the fault identification time from the traditional temperature exceeding stage to the early stage of thermal diffusion pattern distortion, significantly improving the timeliness of early warning and providing key spatial dimension feature basis for subsequent comprehensive judgment.

[0075] For example, in combination Figure 1 ,like Figure 3The diagram shown is a flowchart illustrating another method for monitoring the operating status of an electrical cabinet according to an embodiment of the present invention. In this method, a second judgment feature is determined based on the internal and external temperature data of the electrical cabinet, specifically including the following steps: S301. Arrange the real temperature signals at each sampling time into a spatial sequence according to the sensor location, and calculate the overall rate of change between the spatial sequences at adjacent sampling times.

[0076] For example, the monitoring device is based on the real temperature signal matrix at the t-th sampling time. The matrix is ​​arranged according to the rule of sensor serial numbers from largest to smallest. Unfolding into a one-dimensional real temperature space sequence .

[0077] Furthermore, the true temperature spatial sequence at adjacent times of sampling time t is denoted as... ,calculate and The mean of the absolute values ​​of the differences between the corresponding actual temperatures is used as and The rate of change between them is taken as the overall rate of change between the spatial sequences at adjacent sampling times.

[0078] S302. Perform trend enhancement processing on the spatial sequence at the current sampling time according to the overall rate of change, and calculate the similarity between adjacent spatial sequences after trend enhancement, and use the similarity as a spatial change feature.

[0079] Specifically, the monitoring device records and The sign of each corresponding true temperature difference between two sequences is represented by a sequence T consisting of "1, 0, -1". If the true temperature difference between the two sequences is negative, -1 is recorded at the corresponding position; if it is positive, 1 is recorded at the corresponding position; otherwise, 0 is recorded. Furthermore, based on sequence T... Add or subtract the corresponding temperature data and rate of change: If If the value of a temperature data point at the corresponding position in sequence T is -1, then... The internal temperature data is replaced with the temperature data + (-1 × overall rate of change), thus obtaining the spatial sequence at sampling time t after trend enhancement. Similarly, the spatial sequence at sampling time t-1 can be obtained. .

[0080] Furthermore, the monitoring device utilizes a similarity algorithm suitable for equal-length non-temporal sequences to calculate... and The similarity between the samples is used as the spatial feature at the t-th sampling time. For example, similarity algorithms can employ Euclidean distance, Manhattan distance, Spearman's rank correlation coefficient, etc., all of which are well-known techniques, and their specific calculation methods will not be elaborated further. It should be noted that the true temperature spatial sequence at the 1st sampling time cannot be used for trend enhancement; therefore, spatial similarity calculation is not performed on the true temperature spatial sequence at the 1st sampling time.

[0081] S303. Divide the real temperature signals of each internal temperature sensor in the temperature monitoring array at different sampling times into multiple time series of equal length.

[0082] For example, dividing multiple time series of equal length can be done according to the following process: taking the internal temperature sensor with serial number a in the temperature monitoring array as an example, the real temperature signal collected by temperature sensor a within 100s is divided into 10 real temperature signal sequences with an interval of 10s. These 10 real temperature sequences are multiple time series of equal length.

[0083] S304. Calculate the first similarity between adjacent time series as the first time feature.

[0084] For example, referring to the example in S303, the monitoring device uses a similarity algorithm suitable for equal-length time series to calculate the similarity between the 9th and 10th real temperature signal sequences, as the first time feature of the 10th real temperature signal sequence. The similarity algorithm can be Dynamic Time Warping (DTW) distance, cosine similarity, etc., all of which are well-known technologies, and their specific calculation methods will not be elaborated further. For each real temperature signal sequence, its corresponding first time feature is calculated.

[0085] S305. Predict the current time series based on the historical time series, and calculate the second similarity between the predicted sequence and the current time series as the second time feature. Here, the current time series is one of multiple time series of equal length, and the historical time series are sequences preceding the current time series.

[0086] For example, referring to the examples in S303-S304, in the process of constructing the predicted sequence of real temperature signals, the monitoring device first concatenates the real temperature signal sequences (i.e., historical time series) of the first nine consecutive time periods into an extended sequence of length 90 as input data. The period parameter is set to 10, which corresponds to the number of sampling points contained in each independent time period. A time-series decomposition algorithm is used to decompose the extended sequence, obtaining three component sequences: a trend component sequence, a periodic component sequence, and a residual component sequence, each with a length of 90.

[0087] Furthermore, the monitoring device uses the trend component sequence as input and employs a linear fitting algorithm to predict the trend, resulting in a trend prediction sequence of length 10, which reflects the long-term trend characteristics of temperature changes. The last complete cycle (i.e., the last 10 data points) is extracted from the periodic component sequence as a periodic prediction sequence, which retains the periodic characteristics of temperature changes. The values ​​at corresponding positions in the trend prediction sequence and the periodic prediction sequence are then added point-by-point to generate a true temperature signal prediction sequence of length 10.

[0088] Finally, the second temporal feature is obtained by calculating the similarity between the predicted sequence of the real temperature signal and the currently acquired 10th segment of the real temperature signal sequence (i.e., the current time series). Similarity calculation can employ various similarity measurement methods suitable for equal-length sequences, including but not limited to dynamic time warping algorithms or cosine similarity algorithms.

[0089] It should be noted that the sequence length of 90 and the period parameter of 10 used in this embodiment are only illustrative examples. In actual implementation, the corresponding parameter settings can be adjusted according to the specific application scenario. At the same time, the time series decomposition algorithm can also use other decomposition algorithms with equivalent functions besides the STL algorithm, such as the X11 decomposition algorithm or the SEATs decomposition algorithm; the linear fitting algorithm can also use regression algorithms with trend prediction functions such as the least squares method and the Huber regression algorithm.

[0090] S306. Determine the time change characteristics based on the first time characteristics and the second time characteristics.

[0091] For example, referring to the aforementioned steps, in determining the time characteristics, the monitoring device first performs a weighted calculation of the first and second time characteristics of the 10th real temperature signal sequence according to a preset weight ratio to obtain the initial time characteristics of the sequence. This weighted calculation process and its dependent feature extraction process (i.e., the calculation flow of the first and second time characteristics described in S304 and S305) are encapsulated together into a complete feature calculation model. This model takes the current time sequence (i.e., the 10th real temperature signal sequence) as input and internally accesses its dependent historical sequences (the 9th real temperature signal sequence and earlier historical sequence data). Internally, the model first calculates the first and second time characteristics based on the input sequence and the historical sequences, then sums the two features according to preset weights, and finally outputs the corresponding initial time characteristics (one-dimensional scalar).

[0092] Furthermore, the monitoring device takes the arithmetic mean P of all temperature signal values ​​in the 10th segment of the real temperature signal sequence and constructs a reference sequence of the same length as this segment, where each element has the value P. Using the 10th segment of the real temperature signal sequence, the aforementioned feature calculation model, and the reference sequence as input, and setting the sampling count to 100, the SHAP interpreter calculates the contribution sequence of the same length as the 10th segment of the real temperature signal sequence. This contribution sequence accurately records the contribution of each sensor in the temperature monitoring array to the final time feature value at each sampling moment of the 10th segment of the sequence. Calculating the contribution using the SHAP interpreter is a well-known technique and will not be elaborated upon here.

[0093] Finally, based on the contribution sequence, the monitoring device calculates the arithmetic mean of the contributions of all temperature sensors at the same sampling time, and uses this average value as the final time variation characteristic at that sampling time.

[0094] S307. Determine the second judgment feature based on the spatial and temporal change characteristics.

[0095] For example, taking the t-th sampling time as an example, the spatial features and temporal features of that time are weighted and fused according to equal weights, and the result is the second judgment feature of that sampling time.

[0096] It should be noted that the sampling count of 100 used in this embodiment is an empirical value and can be adjusted according to the required calculation accuracy in actual implementation. The weight ratio in the weighted calculation can also be configured according to the actual application scenario. In addition, contribution calculation can also use other feature importance evaluation methods besides the SHAP interpreter, such as the LIME interpreter or feature attribution methods based on integral gradients, etc. These alternatives can all achieve the same technical effect.

[0097] Based on the above technical solution, this embodiment of the invention constructs a comprehensive judgment basis that can simultaneously capture the spatial anomaly pattern and temporal evolution law of the temperature field of electrical cabinets by collaboratively analyzing the distribution and change characteristics of temperature data in the spatial dimension and the sequence fluctuation characteristics in the temporal dimension. This enables multi-dimensional and forward-looking identification of early faults, effectively overcomes the limitations of single-dimensional monitoring, and significantly improves the sensitivity of the condition monitoring system to potential faults and the reliability of early warning.

[0098] In this embodiment of the invention, the operating status monitoring device of the electrical cabinet can be divided into functional modules or functional units according to the above method example. For example, each function can be divided into its own functional modules or functional units, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or in software functional modules or functional units. The module or unit division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used.

[0099] For example, such as Figure 4 The diagram shown is a possible structural schematic of an electrical cabinet operation status monitoring device according to an embodiment of the present invention. The electrical cabinet operation status monitoring device 400 includes: an acquisition unit 401 and a processing unit 402; The acquisition unit 401 is used to acquire the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet.

[0100] The processing unit 402 is used to determine a first judgment feature based on the internal temperature data and external temperature data of the electrical cabinet. The first judgment feature is used to characterize the degree of similarity between the spatial distribution of temperature and the Gaussian distribution.

[0101] The processing unit 402 is further configured to determine a second judgment feature based on the internal temperature data and external temperature data of the electrical cabinet. The second judgment feature characterizes the stability of temperature in both spatial and temporal dimensions.

[0102] The processing unit 402 is also used to determine the temperature state index based on the first judgment feature and the second judgment feature.

[0103] The processing unit 402 is also used to determine whether the operating status of the electrical cabinet is abnormal based on the temperature status index and electrical parameter data.

[0104] Optionally, the acquisition unit 401 is also used to synchronously acquire internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet at a preset sampling period. The internal temperature data is acquired through a temperature monitoring array arranged inside the electrical cabinet, the external temperature data is acquired through an external temperature sensor arranged outside the electrical cabinet, and the electrical parameter data includes voltage and current data. Each sampling period corresponds to one sampling moment.

[0105] Optionally, the processing unit 402 is further configured to determine the true temperature signal based on the internal temperature data and external temperature data of the electrical cabinet. The true temperature signal characterizes the temperature of the heating elements inside the electrical cabinet after eliminating the influence of ambient temperature.

[0106] Optionally, the processing unit 402 is also configured to construct a Gaussian kernel centered on each real temperature signal value based on the real temperature signal.

[0107] Optionally, the processing unit 402 is further configured to determine the correlation between each Gaussian kernel and the corresponding local temperature signal matrix. Here, the local temperature signal matrix is ​​a set of temperature signals within a local region centered on the current true temperature signal value, and the correlation is used to characterize the degree of agreement between the local temperature distribution and the Gaussian distribution.

[0108] Optionally, the processing unit 402 is further configured to determine a first judgment feature based on the correlation and the number of invalid signals in the local temperature signal matrix.

[0109] Optionally, the processing unit 402 is further configured to determine spatial variation characteristics based on the actual temperature signal. These spatial variation characteristics characterize the smoothness of the overall temperature distribution change between adjacent sampling times.

[0110] Optionally, the processing unit 402 is further configured to determine time-varying characteristics based on the actual temperature signal. These time-varying characteristics characterize the stability of temperature changes over time for a single temperature sensor.

[0111] Optionally, the processing unit 402 is further configured to determine a second judgment feature based on spatial change features and temporal change features.

[0112] Optionally, the processing unit 402 is also used to arrange the real temperature signal at each sampling time into a spatial sequence according to the sensor position, and calculate the overall rate of change between the spatial sequences at adjacent sampling times.

[0113] Optionally, the processing unit 402 is further configured to perform trend enhancement processing on the spatial sequence at the current sampling time according to the overall rate of change, and calculate the similarity between adjacent spatial sequences after trend enhancement, and use the similarity as a spatial change feature.

[0114] Optionally, the processing unit 402 is also used to divide the real temperature signals of each internal temperature sensor in the temperature monitoring array at different sampling times into multiple time series of equal length.

[0115] Optionally, the processing unit 402 is also used to calculate a first similarity between adjacent time series as a first time feature.

[0116] Optionally, the processing unit 402 is further configured to predict the current time series based on the historical time series, and calculate a second similarity between the predicted sequence and the current time series as a second time feature.

[0117] Optionally, the processing unit 402 is further configured to determine time change characteristics based on the first time characteristics and the second time characteristics.

[0118] Optionally, the processing unit 402 is further configured to determine the temperature state index based on the weighted sum of the first judgment feature and the spatial change feature when the cumulative duration corresponding to the current sampling time is less than the first time threshold.

[0119] Optionally, the processing unit 402 is further configured to determine the temperature state index based on the weighted sum of the first judgment feature and the second judgment feature when the cumulative duration corresponding to the current sampling time is greater than or equal to the first time threshold.

[0120] Optionally, the acquisition unit 401 is also used to acquire historical data from the same period. This historical data includes temperature status indices and electrical parameter data for the same historical time period.

[0121] Optionally, the processing unit 402 is also used to determine the normal fluctuation range of the temperature status index and electrical parameter data based on historical data from the same period.

[0122] Optionally, the processing unit 402 is also used to compare the temperature status index and electrical parameter data at the current sampling time with the corresponding normal fluctuation range. If the current temperature status index or any electrical parameter data exceeds the corresponding normal fluctuation range, the operating status of the electrical cabinet is determined to be abnormal.

[0123] Optionally, the electrical cabinet's operating status monitoring device 400 may also include a storage unit ( Figure 4 (shown in dashed box) The storage unit stores a program or instruction. When the acquisition unit 401 and the processing unit 402 execute the program or instruction, the electrical cabinet operation status monitoring device can execute the electrical cabinet operation status monitoring method described in the above method embodiment.

[0124] also, Figure 4 The technical effects of the electrical cabinet operation status monitoring device can be referred to the technical effects of the electrical cabinet operation status monitoring method described in the above embodiments, and will not be repeated here.

[0125] For example, Figure 5 This is another possible structural schematic diagram of the electrical cabinet operation status monitoring device involved in the above embodiments. For example... Figure 5 As shown, the electrical cabinet operation status monitoring device 500 includes: processor 502.

[0126] The processor 502 is used to control and manage the operation of the electrical cabinet operation status monitoring device 400, for example, to execute the steps performed by the acquisition unit 401 and the processing unit 402 in the electrical cabinet operation status monitoring device 400, and / or to execute other processes of the technical solution described herein.

[0127] The processor 502 described above can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the present invention. The processor can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logic blocks, modules, and circuits described in connection with the present invention. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0128] Optionally, the electrical cabinet operation status monitoring device 500 may further include a communication interface 503, a memory 501, and a bus 504. The communication interface 503 supports communication between the electrical cabinet operation status monitoring device 500 and other network entities. The memory 501 stores the program code and data of the electrical cabinet operation status monitoring device.

[0129] The memory 501 may be a memory in the operating status monitoring device of the electrical cabinet. The memory may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as read-only memory, flash memory, hard disk or solid-state drive; the memory may also include a combination of the above types of memory.

[0130] Bus 504 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 504 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0131] Through the above description of the embodiments, those skilled in the art will clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and module described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0132] This invention provides a computer program product containing instructions. When the computer program product is run on the electronic device of this invention, it causes the computer to execute the electrical cabinet operation status monitoring method described in the above method embodiment.

[0133] This invention also provides a computer-readable storage medium storing instructions. When a computer executes these instructions, the electronic device of this invention performs each step of the operation status monitoring device of the electrical cabinet in the method flow shown in the above method embodiments.

[0134] The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), registers, hard disks, optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing, or any other form of computer-readable storage medium in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may reside in an application-specific integrated circuit (ASIC). In embodiments of the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0135] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0136] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for monitoring the operating status of an electrical cabinet, characterized in that, The method includes: Acquire internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet; Based on the internal and external temperature data of the electrical cabinet, a first judgment feature is determined; wherein, the first judgment feature is used to characterize the degree of similarity between the spatial distribution of temperature and the Gaussian distribution; Based on the internal and external temperature data of the electrical cabinet, a second judgment feature is determined; wherein, the second judgment feature is used to characterize the stability of temperature in the spatial and temporal dimensions. The temperature state index is determined based on the first judgment feature and the second judgment feature; Based on the temperature status index and the electrical parameter data, determine whether the operating status of the electrical cabinet is abnormal.

2. The method for monitoring the operating status of an electrical cabinet according to claim 1, characterized in that, Acquire internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet, specifically including: The internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet are acquired synchronously at a preset sampling period. The internal temperature data is acquired through a temperature monitoring array arranged inside the electrical cabinet, the external temperature data is acquired through an external temperature sensor arranged outside the electrical cabinet, and the electrical parameter data includes voltage data and current data. Each sampling period corresponds to a sampling time.

3. The method for monitoring the operating status of an electrical cabinet according to claim 2, characterized in that, After acquiring the internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet, the method further includes: Based on the internal and external temperature data of the electrical cabinet, a true temperature signal is determined; wherein, the true temperature signal is used to characterize the temperature of the heating element inside the electrical cabinet after eliminating the influence of ambient temperature.

4. The method for monitoring the operating status of an electrical cabinet according to claim 3, characterized in that, Based on the internal and external temperature data of the electrical cabinet, a first judgment feature is determined, specifically including: Based on the actual temperature signals, construct a Gaussian kernel centered at each actual temperature signal value; Determine the correlation between each Gaussian kernel and its corresponding local temperature signal matrix; wherein, the local temperature signal matrix is ​​a set of temperature signals within a local region centered on the current true temperature signal value, and the correlation is used to characterize the degree of agreement between the local temperature distribution and the Gaussian distribution; The first judgment feature is determined based on the correlation and the number of invalid signals in the local temperature signal matrix.

5. The method for monitoring the operating status of an electrical cabinet according to claim 3, characterized in that, Based on the internal and external temperature data of the electrical cabinet, the second judgment feature is determined, specifically including: Based on the actual temperature signal, spatial variation characteristics are determined; wherein, the spatial variation characteristics are used to characterize the smoothness of the overall temperature distribution change between adjacent sampling times; Based on the actual temperature signal, the time variation characteristics are determined; wherein, the time variation characteristics are used to characterize the stability of temperature variation of a single temperature sensor over time. The second judgment feature is determined based on the spatial change feature and the temporal change feature.

6. The method for monitoring the operating status of an electrical cabinet according to claim 3, characterized in that, Based on the actual temperature signal, the spatial variation characteristics are determined, specifically including: The actual temperature signals at each sampling time are arranged into a spatial sequence according to the sensor location, and the overall rate of change between the spatial sequences at adjacent sampling times is calculated. The spatial sequence at the current sampling time is subjected to trend enhancement processing based on the overall rate of change, and the similarity between adjacent spatial sequences after trend enhancement is calculated. The similarity is used as the spatial change feature.

7. The method for monitoring the operating status of an electrical cabinet according to claim 3, characterized in that, Based on the actual temperature signal, the time-varying characteristics are determined, specifically including: The real temperature signals of each internal temperature sensor in the temperature monitoring array at different sampling times are divided into multiple time series of equal length. Calculate the first similarity between adjacent time series as a first temporal feature; The current time series is predicted based on the historical time series, and a second similarity between the predicted sequence and the current time series is calculated as a second time feature; wherein the current time series is one of the multiple time series of equal length, and the historical time series are the sequences preceding the current time series; The time change characteristics are determined based on the first time characteristic and the second time characteristic.

8. The method for monitoring the operating status of an electrical cabinet according to claim 5, characterized in that, Based on the first judgment feature and the second judgment feature, the temperature state index is determined, specifically including: If the cumulative duration corresponding to the current sampling time is less than the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the spatial change feature. If the cumulative duration corresponding to the current sampling time is greater than or equal to the first time threshold, the temperature state index is determined based on the weighted sum of the first judgment feature and the second judgment feature.

9. The method for monitoring the operating status of an electrical cabinet according to any one of claims 1-8, characterized in that, Based on the temperature status index and the electrical parameter data, determine whether the operating status of the electrical cabinet is abnormal, specifically including: Acquire historical data for the same period; wherein, the historical data for the same period includes temperature status index and electrical parameter data for the same historical period; Based on the historical data from the same period, the normal fluctuation range of the temperature state index and the electrical parameter data is determined; The temperature status index and electrical parameter data at the current sampling time are compared with the corresponding normal fluctuation range. If the current temperature status index or any of the electrical parameter data exceeds the corresponding normal fluctuation range, the operating status of the electrical cabinet is determined to be abnormal.

10. An electrical cabinet operation status monitoring device, characterized in that, The electrical cabinet operation status monitoring device includes: an acquisition unit and a processing unit; The acquisition unit is used to acquire internal temperature data, external temperature data, and electrical parameter data of the electrical cabinet. The processing unit is used to determine a first judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein, the first judgment feature is used to characterize the degree of similarity between the spatial distribution of temperature and the Gaussian distribution; The processing unit is further configured to determine a second judgment feature based on the internal temperature data and external temperature data of the electrical cabinet; wherein the second judgment feature is used to characterize the stability of temperature in the spatial and temporal dimensions; The processing unit is further configured to determine a temperature state index based on the first judgment feature and the second judgment feature; The processing unit is also used to determine whether the operating status of the electrical cabinet is abnormal based on the temperature status index and the electrical parameter data.

Images