Electrical power distribution cabinet and control method thereof

Abstract

The invention relates to the technical field of power distribution cabinet control, in particular to an electrical power distribution cabinet and a control method thereof. The method comprises the following steps: acquiring multi-dimensional data of the power distribution cabinet; performing data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a power distribution cabinet preprocessing data set; performing contribution degree sorting on the preprocessed data set of the power distribution cabinet to generate contribution degree sorting data of the power distribution cabinet; constructing an electric cabinet model according to the contribution degree sorting data of the power distribution cabinet, and generating an electric cabinet power consumption prediction model; acquiring real-time power distribution cabinet data; performing power distribution cabinet power utilization prediction on the real-time power distribution cabinet data based on the power utilization cabinet power utilization prediction model to generate a power distribution cabinet prediction result; therefore, by comprehensively integrating the multi-dimensional data of the power distribution cabinet, intelligent prediction and optimal scheduling are performed, the problems of low prediction precision, non-uniform resource distribution, response lag and the like in traditional power distribution cabinet management are solved, and the operation efficiency and the resource utilization rate of the electrical power distribution cabinet are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution cabinet control, and in particular to an electrical power distribution cabinet and a control method thereof. Background Art

[0002] In traditional distribution cabinet control methods, data processing is often limited to the monitoring of basic parameters such as current and voltage, and fails to make full use of environmental parameters (such as temperature, humidity, vibration, etc.) and historical operating data for comprehensive analysis. This makes it often impossible to fully optimize the scheduling decisions of the distribution cabinet when facing a complex operating environment. The lack of comprehensive understanding and integration of multi-dimensional data leads to insufficient flexibility and accuracy in strategy execution, especially when abnormal conditions such as high load and high temperature occur, there is a lack of effective dynamic adjustment mechanism. In the existing technology, the performance evaluation of the distribution cabinet relies on static indicators or simple rules for judgment, lacks a dynamic evaluation mechanism, and cannot timely reflect the actual performance of the distribution cabinet under different operating conditions. Although some existing technologies identify faults by monitoring current and voltage fluctuations, these methods often have weak recognition capabilities for complex fault modes, long-term accumulated small changes, and other relatively hidden faults in actual applications. Most of these technologies focus on early warning based on a single indicator, lack multi-dimensional data cross-analysis and linkage prediction, and it is difficult to accurately predict potential faults. Summary of the invention

[0003] Based on this, it is necessary to provide an electrical distribution cabinet and a control method thereof to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for controlling an electrical distribution cabinet is provided, the method comprising the following steps:

[0005] Step S1: Acquire multi-dimensional data of the power distribution cabinet; perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set of the power distribution cabinet;

[0006] Step S2: sorting the contribution of the distribution cabinet preprocessing data set to generate distribution cabinet contribution ranking data; constructing a power cabinet model based on the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model;

[0007] Step S3: Acquire real-time distribution cabinet data; perform distribution cabinet power consumption prediction on the real-time distribution cabinet data based on the power cabinet power consumption prediction model, and generate a distribution cabinet prediction result; perform judgment and analysis on the distribution cabinet prediction result based on a preset result judgment threshold, and generate a distribution cabinet power distribution strategy to generate a distribution cabinet power distribution strategy;

[0008] Step S4: execute the power distribution strategy of the distribution cabinet to generate a multi-dimensional scheduling result of the distribution cabinet; generate a report based on the multi-dimensional scheduling result of the distribution cabinet to generate a distribution cabinet control report, thereby completing the control operation of the electrical distribution cabinet.

[0009] The beneficial effect of the present invention is that by acquiring and preprocessing the multi-dimensional data of the distribution cabinet (including current, voltage, temperature and humidity, vibration and other data), noise and outliers can be eliminated to ensure the quality of the input data, thereby providing a reliable basis for subsequent feature analysis and model training. By sorting the data by contribution, the key factors affecting the power consumption of the distribution cabinet can be identified, providing a more accurate basis for the construction of the model. Based on these contribution-ranked data, the constructed power consumption prediction model can not only reflect the power consumption of the distribution cabinet under different environmental conditions, but also predict the load change trend and potential failure risks, and further optimize the distribution management strategy. Real-time data prediction and threshold-based strategy generation mechanism can dynamically adjust the load distribution and electrical environment control of the distribution cabinet, thereby realizing accurate scheduling and efficient management of the distribution cabinet. Finally, through strategy execution and scheduling result analysis, the generated control report provides data support and decision-making basis for the operation of the electrical distribution cabinet, which helps to improve the overall operation efficiency and reliability of the distribution cabinet. Scheduling optimization based on multidimensional data not only enhances the stability of the power distribution system, but also improves the energy utilization efficiency in the power distribution process, reduces the risk of failures, and ensures the rational allocation and efficient use of power resources. Therefore, the present invention solves the problems of low prediction accuracy, uneven resource allocation, and delayed response in traditional power distribution cabinet management by comprehensively integrating the multidimensional data of the power distribution cabinet for intelligent prediction and optimized scheduling, thereby improving the operating efficiency and resource utilization of the electrical power distribution cabinet.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: acquiring multidimensional data of the power distribution cabinet; performing data cleaning processing on the multidimensional data of the power distribution cabinet to generate multidimensional cleaning data of the power distribution cabinet, wherein the multidimensional cleaning data of the power distribution cabinet includes current cleaning data, voltage cleaning data, temperature cleaning data, humidity cleaning data and vibration cleaning data;

[0012] Step S12: performing feature processing on the multi-dimensional cleaning data of the power distribution cabinet to generate multi-dimensional feature data of the power distribution cabinet;

[0013] Step S13: Standardize the data set according to the multi-dimensional feature data of the power distribution cabinet to generate a pre-processed data set of the power distribution cabinet.

[0014] The present invention can effectively remove abnormal values, noise data and incomplete data through the cleaning process of the multidimensional data of the distribution cabinet, thereby reducing the interference of invalid information on the analysis results. The implementation of this link ensures the accuracy of multidimensional data such as current, voltage, temperature, humidity and vibration, and provides more reliable input data for subsequent analysis. In particular, the cleaning of current and voltage data can eliminate abnormal fluctuations caused by equipment failure or sensor error, thereby ensuring the stability and reliability of the power consumption model. In addition, the cleaning process of temperature, humidity and vibration data can eliminate errors caused by environmental factors or external interference, ensuring the true reflection of data in a complex environment. Secondly, the feature processing link can capture the key relationship between the operating state of the distribution cabinet and the external environment, such as the response of current changes to temperature or humidity, by extracting representative features from the cleaning data. These processed feature data not only reflect the operating law of the distribution cabinet, but also help reduce redundant information and improve the validity of the data. Finally, by standardizing the multidimensional feature data, the data of different dimensions and scales are unified within a standard range, which helps to eliminate the scale differences between the data and avoids the uneven impact of certain features on the model due to large or small values. This standardization step makes the data adapt to the needs of different analysis algorithms and improves the convergence speed and stability in the subsequent modeling process. Through these steps, the quality of the entire data set has been significantly improved, which can provide more accurate, stable and valuable information for the subsequent prediction model construction.

[0015] Preferably, step S12 comprises the following steps:

[0016] Step S121: Perform Fourier transform spectrum analysis on the current cleaning data and the voltage cleaning data to generate current-voltage frequency component data; perform mean value calculation on the current-voltage frequency component data to generate current-voltage characteristic data of the distribution cabinet;

[0017] Step S122: Perform time series analysis on the temperature cleaning data and the humidity cleaning data to generate temperature-humidity seasonal fluctuation data; perform temperature and humidity coordinated processing on the temperature-humidity seasonal fluctuation data to generate temperature and humidity characteristic data of the power distribution cabinet;

[0018] Step S123: performing acceleration analysis on the vibration cleaning data to generate acceleration analysis data; performing amplitude fluctuation range analysis on the acceleration analysis data to generate acceleration-amplitude characteristic data of the distribution cabinet;

[0019] Step S124: performing time stamp calibration on the current-voltage characteristic data of the distribution cabinet, the temperature and humidity characteristic data of the distribution cabinet, and the acceleration-amplitude characteristic data of the distribution cabinet to generate multi-dimensional characteristic data of the distribution cabinet.

[0020] The present invention can effectively identify the frequency components in the current and voltage signals by performing Fourier transform spectrum analysis on the current cleaning data and the voltage cleaning data, and reveal the changing rules of the electrical characteristics at different frequencies. This process can capture the fluctuation patterns of current and voltage, especially the high-frequency oscillations during load fluctuations or equipment failures, and then generate representative current-voltage frequency component data, which provides key features for subsequent power consumption prediction and fault diagnosis. Secondly, the current-voltage frequency component data is averaged, the frequency domain information is simplified from the data level, the core features of current and voltage are extracted, and the current-voltage characteristic data of the distribution cabinet is further generated, which provides a stable and representative input for subsequent analysis. At the same time, the time series analysis of the temperature cleaning data and the humidity cleaning data can reveal the regular fluctuations of temperature and humidity in seasonal changes, and generate temperature-humidity seasonal fluctuation data, which provides a deep analysis in the time dimension for evaluating the impact of environmental factors on the distribution cabinet. Through further temperature and humidity collaborative processing, the relationship between temperature and humidity can be integrated to generate temperature and humidity characteristic data of the distribution cabinet, which reflects the synergistic effect between environmental factors and electrical characteristics, and helps to improve the accurate prediction of the operating status of the distribution cabinet. Furthermore, by performing acceleration analysis on vibration cleaning data, the intensity and frequency changes of vibration signals can be identified, providing an effective monitoring basis for early warning of equipment failures. The amplitude fluctuation range analysis further reveals the abnormal vibration patterns that occur during the operation of the equipment. The generated acceleration-amplitude characteristic data of the distribution cabinet helps to determine the health status of the distribution cabinet and predict potential failure risks. Finally, these feature data from different dimensions are timestamped to ensure that the current, voltage, temperature, humidity and vibration data are aligned on the same time axis, eliminating the time deviation between different sensors or measurement points, and ensuring the consistency and comparability of the data. The multi-dimensional feature data of the distribution cabinet after timestamp calibration provides accurate time synchronization information for further analysis and prediction, which helps to improve the prediction accuracy of the model and the accuracy of decision-making.

[0021] Preferably, step S2 comprises the following steps:

[0022] Step S21: performing feature cross analysis on the distribution cabinet preprocessing data set to generate a distribution cabinet cross data set;

[0023] Step S22: Sorting the contribution of the distribution cabinets according to the cross-feature data set to generate distribution cabinet contribution ranking data;

[0024] Step S23: construct a power cabinet model according to the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model.

[0025] The present invention combines and interactively analyzes different features in the distribution cabinet preprocessing data set through feature cross analysis, thereby generating a cross data set containing multi-dimensional information. Through this cross analysis, the nonlinear relationship between each feature and the synergistic effect of multiple influencing factors can be captured. For example, the interaction between current and temperature, humidity, the coupling effect between voltage and vibration, etc., these cross feature data help to reveal the complex physical phenomena and their potential laws in the operation of the distribution cabinet, thereby providing a rich feature basis for further analysis and modeling. Secondly, based on the cross feature data set of the distribution cabinet, the contribution degree is sorted, and the key features that have the greatest impact on the power consumption behavior and system performance of the distribution cabinet can be identified. This sorting step screens out the features that have a greater impact on the power consumption model by evaluating the predictive ability of each feature, and gives priority to them for model construction. Contribution sorting not only improves the efficiency of the model and avoids the introduction of redundant data, but also enables the final power consumption prediction model to better reflect the actual operation of the distribution cabinet, thereby improving the accuracy and credibility of the prediction. Finally, the power consumption prediction model constructed based on the distribution cabinet contribution ranking data can not only effectively process multi-dimensional and multi-level data, but also accurately predict future power consumption trends, especially in a dynamically changing environment, and can adjust the power distribution strategy and resource allocation in a timely manner. Through these refined steps, the entire data processing process greatly improves the scientificity and practicality of the distribution cabinet power consumption prediction, and helps to realize the intelligent management and refined scheduling of the distribution cabinet system.

[0026] Preferably, step S21 includes the following steps:

[0027] Step S211: performing current-temperature product calculation on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the current-temperature product ratio influence data of the distribution cabinet; performing voltage-humidity interaction item fluctuation analysis on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the voltage-humidity fluctuation interaction item data of the distribution cabinet;

[0028] Step S212: Perform current-vibration frequency load correlation analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate current-vibration frequency related data; perform Pearson correlation coefficient calculation on the current-vibration frequency related data to generate current-vibration frequency characteristic data of the distribution cabinet; perform volt-voltage-acceleration linkage analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate voltage-acceleration linkage data of the distribution cabinet;

[0029] Step S213: Perform temperature-vibration amplitude thermal analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate temperature-vibration amplitude thermal analysis data; perform vibration friction analysis on the temperature-vibration amplitude thermal analysis data to generate temperature-vibration amplitude data of the distribution cabinet; perform humidity-acceleration physical property analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate humidity-acceleration physical property change data; perform dynamic analysis on the mechanical energy ratio of the humidity-acceleration physical property change data to generate humidity-acceleration change characteristic data of the distribution cabinet;

[0030] Step S214: Generate a multidimensional feature data matrix using the distribution cabinet current-temperature product ratio influence data, distribution cabinet voltage-humidity fluctuation interaction item data, distribution cabinet current-vibration frequency characteristic data, distribution cabinet voltage-acceleration linkage data, distribution cabinet temperature-vibration amplitude data and distribution cabinet humidity-acceleration change characteristic data to generate a distribution cabinet multidimensional data matrix; perform principal component analysis on the distribution cabinet multidimensional data matrix to reduce the dimensionality, and construct a data set to generate a distribution cabinet cross data set.

[0031] The present invention can reveal the correlation between current and temperature and the interaction between voltage and humidity through current-temperature product calculation and voltage-humidity interaction fluctuation analysis, thereby providing quantitative data for the dynamic relationship between electrical characteristics and environmental factors. This interactive analysis not only helps to capture the impact of environmental changes on the electrical performance of the distribution cabinet, but also enhances the explanatory power of the model by constructing interactive data, thereby improving the prediction accuracy of potential faults. Current-vibration frequency load correlation analysis and volt-voltage-acceleration linkage analysis further enhance the joint sensitivity of the model to electrical and mechanical systems. By analyzing the interaction between current, vibration frequency and acceleration, the relationship between electrical load and mechanical vibration can be revealed, providing a key basis for equipment fault prediction and health monitoring. The calculation of the Pearson correlation coefficient helps to quantify the degree of correlation between these variables, thereby providing accurate data support for the next step of model construction. Temperature-vibration amplitude thermal analysis and humidity-acceleration physical property analysis can further capture the dynamic characteristics of the distribution cabinet under different environmental conditions through in-depth analysis of the thermodynamic relationship between temperature, humidity and vibration amplitude and the change of physical properties. These features not only provide a multi-dimensional perspective on equipment status and environmental impact, but also optimize the understanding of physical phenomena through dynamic analysis of mechanical energy ratios, and enhance the time series characteristics and trend prediction capabilities of the data. All the above cross-feature data are matrixed and principal component analysis (PCA) is performed to reduce the dimension, so that redundant features are effectively reduced, core features are highlighted, and the data set is more concise and has a high information density, which can improve the efficiency and accuracy of subsequent modeling.

[0032] Preferably, step S23 includes the following steps:

[0033] Step S231: Sorting the distribution cabinet contribution ranking data to generate distribution cabinet contribution ranking data, wherein the distribution cabinet contribution ranking data includes distribution cabinet high contribution data and distribution cabinet low contribution data;

[0034] Step S232: eliminating redundancy of low contribution by using a preset contribution threshold to generate low contribution elimination data;

[0035] Step S233: Perform linear regression prediction on the high contribution data and the low contribution eliminated data to construct a power cabinet model, and generate a power cabinet power consumption prediction model.

[0036] The present invention sorts the contribution ranking data of the distribution cabinet, and classifies the data according to the degree of influence on the power consumption forecast, thereby forming high contribution data and low contribution data. This sorting process can help identify the key factors that have the greatest influence on the power consumption forecast of the distribution cabinet, and distinguish them from the relatively redundant low contribution data, so as to provide more accurate feature input for subsequent modeling. By eliminating the redundancy of the low contribution data, the interference of unimportant features on the model is further reduced. Redundancy elimination not only improves the calculation efficiency of the model, but also avoids the occurrence of data overfitting problems, thereby improving the generalization ability of the model, so that the model can maintain high accuracy and stability when facing applications in different distribution cabinet environments. In addition, the elimination of low contribution data also helps avoid the information noise problem caused by data redundancy, and further optimizes the quality of the data. The high contribution data and the data after eliminating the low contribution are merged, and the power cabinet model is constructed using a linear regression method. This method uses the significant features of high contribution data for predictive modeling, and at the same time ensures the accuracy and effectiveness of the model in a complex environment through optimized input data. Through linear regression calculations, the model can more accurately capture the linear relationship between each feature and electricity consumption, thereby providing more reliable prediction results. Overall, this process, from data screening, redundancy elimination to model construction, reflects a virtuous cycle of improving data quality and optimizing model accuracy, and effectively solves the problems of data redundancy, feature noise and low computational efficiency in traditional models.

[0037] Preferably, step S3 comprises the following steps:

[0038] Step S31: Acquire real-time distribution cabinet data;

[0039] Step S32: Based on the power consumption prediction model of the power cabinet, the real-time distribution cabinet data is used to predict the power consumption of the distribution cabinet, and a distribution cabinet prediction result is generated; the distribution cabinet prediction result is marked with a position, and a distribution cabinet position-prediction result is generated;

[0040] Step S33: Based on a preset result judgment threshold, the prediction result of the distribution cabinet is judged and analyzed to generate a prediction judgment result of the distribution cabinet; based on the prediction judgment result of the distribution cabinet, a power distribution strategy of the distribution cabinet is generated to generate a power distribution strategy of the distribution cabinet.

[0041] The present invention provides input data with strong timeliness and high correlation for subsequent analysis and prediction by acquiring real-time distribution cabinet data. These data include multi-dimensional information such as current, voltage, temperature and humidity of the distribution cabinet, which can truly reflect the current working state and operating environment of the equipment, and provide a basis for accurately predicting power demand. Through the power consumption prediction model of the power cabinet, the real-time data is processed and predicted to generate the power consumption prediction result of the distribution cabinet. The key to this step is to use the established prediction model to perform intelligent analysis on the real-time data, and to mine the laws and trends in the data through mathematical models to predict the power consumption of the distribution cabinet. The accuracy of the prediction result directly affects the rationality of the subsequent power distribution strategy and the efficient use of power resources. The prediction result is associated with the specific geographical location through location marking to generate the distribution cabinet location-prediction result data. Through location marking, the distribution cabinets at different locations can be managed differently, and corresponding power dispatching strategies can be formulated according to the different requirements of their locations. According to the preset threshold, the prediction result of the distribution cabinet is judged and analyzed to generate the prediction judgment result of the distribution cabinet. This analysis phase not only helps identify whether there are abnormal or high-load conditions, but also provides a basis for determining whether the distribution cabinet needs to be adjusted. Based on the judgment results, a distribution strategy for the distribution cabinet is generated to ensure that the operating status of the distribution cabinet matches the power demand and achieve optimal power distribution and scheduling. Through the reasonable operation of this series of steps, the system can flexibly generate a distribution strategy that best meets actual needs based on real-time monitoring and prediction, thereby improving the operating efficiency of the power system, reducing the risk of failures, and improving the service life and safety of the distribution cabinet.

[0042] Preferably, step S33 includes the following steps:

[0043] Step S331: judging and analyzing the prediction results of the distribution cabinet based on the preset result judgment threshold, and generating the prediction judgment results of the distribution cabinet; generating the distribution cabinet distribution strategy based on the prediction judgment results of the distribution cabinet, and generating the distribution cabinet distribution strategy, wherein the distribution cabinet distribution strategy includes the distribution cabinet load and environment distribution strategy and the fault prediction and load adjustment distribution strategy;

[0044] Step S332: When the prediction judgment result of the distribution cabinet is greater than or equal to the preset result judgment threshold, the real-time load analysis of the distribution cabinet is performed on the current-voltage characteristic data of the distribution cabinet to generate the real-time load capacity data of the distribution cabinet; the high-load area is marked on the real-time load capacity data of the distribution cabinet to generate the high-load area marking data of the distribution cabinet; the temperature and humidity analysis is performed according to the temperature and humidity characteristic data of the distribution cabinet, and the high-temperature humidity marking is performed to generate the high-temperature humidity marking data of the distribution cabinet; the distribution load is optimized based on the high-load area marking data of the distribution cabinet and the high-temperature humidity marking data of the distribution cabinet, so as to generate the load and environmental distribution strategy of the distribution cabinet;

[0045] Step S333: When the prediction result of the distribution cabinet is less than the preset result judgment threshold, the current fluctuation-amplitude excess analysis is performed on the current-voltage characteristic data of the distribution cabinet to generate abnormal fault warning data; based on the prediction result of the distribution cabinet, the abnormal fault warning data is monitored and analyzed, and a warning message is generated and uploaded to the cloud platform, thereby completing the fault prediction and load adjustment distribution strategy.

[0046] The present invention analyzes the prediction results of the distribution cabinet through a preset judgment threshold, and then generates the prediction judgment results of the distribution cabinet. This judgment basis helps to accurately identify the potential risks and abnormal changes in the operation of the system, and ensures that the power distribution of the distribution cabinet can adapt to the real-time demand changes more flexibly and intelligently. On this basis, the generated distribution cabinet distribution strategy includes two parts: load and environment distribution strategy, and fault prediction and load adjustment distribution strategy. The load and environment distribution strategy mainly focuses on adjusting the load and scheduling mode of the distribution cabinet to cope with the impact of environmental changes (such as temperature and humidity changes) on the load, while the fault prediction and load adjustment distribution strategy focuses on the early identification of fault risks and the corresponding adjustment of the load, so as to avoid faults caused by overload operation. When the prediction judgment result of the distribution cabinet is greater than or equal to the preset threshold, real-time load analysis is performed through the current-voltage characteristic data to generate real-time load force data, and the high-load area is marked to generate high-load area marking data. At the same time, the temperature and humidity characteristic data are analyzed to mark the high temperature and humidity area, and then the high temperature and humidity marking data is generated. Combining these two tagged data, we can optimize the power distribution of load and environment, generate the power distribution strategy of load and environment, effectively avoid the influence of environmental factors on load distribution, and improve the accuracy of power distribution and system stability. When the prediction result of the distribution cabinet is less than the preset threshold, we can generate abnormal fault warning data through the analysis of current fluctuation and amplitude exceeding the standard, and further monitor and analyze it, generate warning messages and upload them to the cloud platform to ensure that the power system can respond in real time and take preventive measures when abnormal load or fault occurs. In this way, the system can detect and respond to potential faults in a timely manner, greatly reduce the probability of faults, and realize intelligent adjustment of loads.

[0047] Preferably, step S4 comprises the following steps:

[0048] Step S41: executing the power distribution strategy of the power distribution cabinet to generate a multi-dimensional scheduling result of the power distribution cabinet;

[0049] Step S42: performing a distribution cabinet performance evaluation on the multi-dimensional scheduling result of the distribution cabinet to generate distribution cabinet scheduling evaluation data;

[0050] Step S43: Generate a report on the distribution cabinet scheduling evaluation data and generate a distribution cabinet control report, thereby completing the electrical distribution cabinet control operation.

[0051] The present invention uses the distribution strategy to execute the load and environment distribution strategy formulated in the early stage, and makes the power distribution more real-time and flexible through reasonable multi-dimensional scheduling. The multi-dimensional scheduling results of the distribution cabinet are evaluated for performance, and the distribution cabinet scheduling evaluation data is generated. This evaluation process can not only comprehensively analyze the performance of the distribution cabinet after executing the distribution strategy, but also identify the potential scheduling bottlenecks in the system, ensuring that the power system can maintain efficient operation when the load peak or emergency occurs. Through the feedback of the scheduling evaluation data, the operation effect of the distribution cabinet can be accurately evaluated, and whether the distribution strategy achieves the expected goal can be verified, thereby providing data support and decision-making basis for the optimization of subsequent strategies. By processing the scheduling evaluation data, a distribution cabinet control report is generated, thereby providing detailed power control operation status for the operation and maintenance team of the distribution cabinet. This report not only summarizes the performance of the distribution cabinet in operation, but also makes improvement suggestions based on the evaluation results to ensure that the scheduling operation of the distribution cabinet is effectively executed, and the distribution strategy is adjusted in time to cope with load fluctuations or equipment failures.

[0052] The beneficial effect of the present invention is that by acquiring and preprocessing the multi-dimensional data of the distribution cabinet (including current, voltage, temperature and humidity, vibration and other data), noise and outliers can be eliminated to ensure the quality of the input data, thereby providing a reliable basis for subsequent feature analysis and model training. By sorting the data by contribution, the key factors affecting the power consumption of the distribution cabinet can be identified, providing a more accurate basis for the construction of the model. Based on these contribution-ranked data, the constructed power consumption prediction model can not only reflect the power consumption of the distribution cabinet under different environmental conditions, but also predict the load change trend and potential failure risks, and further optimize the distribution management strategy. Real-time data prediction and threshold-based strategy generation mechanism can dynamically adjust the load distribution and electrical environment control of the distribution cabinet, thereby realizing accurate scheduling and efficient management of the distribution cabinet. Finally, through strategy execution and scheduling result analysis, the generated control report provides data support and decision-making basis for the operation of the electrical distribution cabinet, which helps to improve the overall operation efficiency and reliability of the distribution cabinet. Scheduling optimization based on multidimensional data not only enhances the stability of the power distribution system, but also improves the energy utilization efficiency in the power distribution process, reduces the risk of failures, and ensures the rational allocation and efficient use of power resources. Therefore, the present invention solves the problems of low prediction accuracy, uneven resource allocation, and delayed response in traditional power distribution cabinet management by comprehensively integrating the multidimensional data of the power distribution cabinet for intelligent prediction and optimized scheduling, thereby improving the operating efficiency and resource utilization of the electrical power distribution cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of a process flow of an electrical distribution cabinet control method;

[0054] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0055] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0056] Figure 4 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0057] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0059] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0060] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0061] To achieve this, please refer to Figures 1 to 4 , a method for controlling an electrical distribution cabinet, the method comprising the following steps:

[0062] Step S1: Acquire multi-dimensional data of the power distribution cabinet; perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set of the power distribution cabinet;

[0063] Step S2: sorting the contribution of the distribution cabinet preprocessing data set to generate distribution cabinet contribution ranking data; constructing a power cabinet model based on the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model;

[0064] Step S3: Acquire real-time distribution cabinet data; perform distribution cabinet power consumption prediction on the real-time distribution cabinet data based on the power cabinet power consumption prediction model, and generate a distribution cabinet prediction result; perform judgment and analysis on the distribution cabinet prediction result based on a preset result judgment threshold, and generate a distribution cabinet power distribution strategy to generate a distribution cabinet power distribution strategy;

[0065] Step S4: execute the power distribution strategy of the distribution cabinet to generate a multi-dimensional scheduling result of the distribution cabinet; generate a report based on the multi-dimensional scheduling result of the distribution cabinet to generate a distribution cabinet control report, thereby completing the control operation of the electrical distribution cabinet.

[0066] The beneficial effect of the present invention is that by acquiring and preprocessing the multi-dimensional data of the distribution cabinet (including current, voltage, temperature and humidity, vibration and other data), noise and outliers can be eliminated to ensure the quality of the input data, thereby providing a reliable basis for subsequent feature analysis and model training. By sorting the data by contribution, the key factors affecting the power consumption of the distribution cabinet can be identified, providing a more accurate basis for the construction of the model. Based on these contribution-ranked data, the constructed power consumption prediction model can not only reflect the power consumption of the distribution cabinet under different environmental conditions, but also predict the load change trend and potential failure risks, and further optimize the distribution management strategy. Real-time data prediction and threshold-based strategy generation mechanism can dynamically adjust the load distribution and electrical environment control of the distribution cabinet, thereby realizing accurate scheduling and efficient management of the distribution cabinet. Finally, through strategy execution and scheduling result analysis, the generated control report provides data support and decision-making basis for the operation of the electrical distribution cabinet, which helps to improve the overall operation efficiency and reliability of the distribution cabinet. Scheduling optimization based on multidimensional data not only enhances the stability of the power distribution system, but also improves the energy utilization efficiency in the power distribution process, reduces the risk of failures, and ensures the rational allocation and efficient use of power resources. Therefore, the present invention solves the problems of low prediction accuracy, uneven resource allocation, and delayed response in traditional power distribution cabinet management by comprehensively integrating the multidimensional data of the power distribution cabinet for intelligent prediction and optimized scheduling, thereby improving the operating efficiency and resource utilization of the electrical power distribution cabinet.

[0067] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of a step flow chart of a method for controlling an electrical distribution cabinet according to the present invention. In this example, the method for controlling an electrical distribution cabinet includes the following steps:

[0068] Step S1: Acquire multi-dimensional data of the power distribution cabinet; perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set of the power distribution cabinet;

[0069] In an embodiment of the present invention, multidimensional data of a power distribution cabinet is obtained, and these data include but are not limited to various physical quantities such as current, voltage, temperature, humidity and vibration. These data come from real-time monitoring of different sensors in the power distribution cabinet, and are a comprehensive reflection of the operating status of the power equipment. Next, data preprocessing is performed on these multidimensional data to ensure data quality and provide a reliable basis for subsequent analysis and modeling. The specific technical means of data preprocessing include data cleaning, missing value filling, outlier detection and elimination, noise removal and other methods. First, the multidimensional data of the power distribution cabinet is filled with missing values, and the missing data is filled with mean filling, interpolation or machine learning-based prediction methods to ensure the integrity of the data set. Secondly, an outlier detection algorithm, such as Z-score, box plot method, etc., is used to identify and eliminate data that does not conform to the actual operating rules, so as to avoid these abnormal data affecting the accuracy of the model. For noise data, high-frequency noise signals are removed by smoothing or filtering techniques, such as low-pass filters, Gaussian filters, etc., to ensure the stability and accuracy of the data. Finally, the original data is standardized or normalized, and data of different dimensions are converted into a unified standard so that they have the same influence in the subsequent analysis and modeling process. Commonly used methods in this process include Z-score standardization and minimum-maximum normalization.

[0070] Step S2: sorting the contribution of the distribution cabinet preprocessing data set to generate distribution cabinet contribution ranking data; constructing a power cabinet model based on the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model;

[0071] In an embodiment of the present invention, the contribution ranking of the distribution cabinet preprocessing data set is achieved by evaluating the influence or importance of each feature on the target variable. Specifically, the contribution ranking uses feature selection and evaluation techniques, such as information gain, variance analysis, Pearson correlation coefficient, mutual information and other methods. These methods help determine which features have a higher contribution to the power forecast of the distribution cabinet by quantifying the relationship between each feature and the target variable. For the contribution evaluation of continuous variables, the correlation coefficient can be used to measure the linear relationship between the feature and the target variable, while for categorical variables, information gain or chi-square test is often used to evaluate the impact of the feature on the target classification. According to the result of contribution ranking, the features in the data set are sorted according to their contribution to the power consumption prediction model to obtain the distribution cabinet contribution ranking data. Next, based on the contribution ranking data, a power cabinet model is constructed to predict power consumption. The technical means for constructing the power cabinet model adopts regression analysis or machine learning algorithm. In regression analysis, commonly used methods include linear regression, ridge regression, Lasso regression, etc., which are suitable for situations where there is a linear relationship between features and target variables. For complex nonlinear relationships, machine learning models such as decision trees, random forests, support vector machines, and neural networks can be selected. These models learn the relationship between the features in the preprocessed data set of the distribution cabinet and the power consumption, and generate a power cabinet power consumption prediction model that can predict the power consumption of the distribution cabinet. The model training process includes data segmentation (such as the division of training sets, validation sets, and test sets), hyperparameter optimization, cross-validation, and other steps to improve the generalization ability and prediction accuracy of the model. In the end, the constructed power cabinet power consumption prediction model can not only accurately predict future power consumption, but also provide support for subsequent distribution cabinet optimization scheduling, fault warning and other decisions.

[0072] Step S3: Acquire real-time distribution cabinet data; perform distribution cabinet power consumption prediction on the real-time distribution cabinet data based on the power cabinet power consumption prediction model, and generate a distribution cabinet prediction result; perform judgment and analysis on the distribution cabinet prediction result based on a preset result judgment threshold, and generate a distribution cabinet power distribution strategy to generate a distribution cabinet power distribution strategy;

[0073] In an embodiment of the present invention, multi-dimensional data such as current, voltage, temperature, humidity, vibration, etc. are collected from the distribution cabinet by real-time monitoring equipment or sensors. The acquisition of real-time data depends on an efficient data acquisition system and sensor network, and these data are transmitted to the central data processing platform through wireless communication technologies (such as LoRa, ZigBee, Wi-Fi, 4G / 5G, etc.). Next, the real-time distribution cabinet data is predicted based on the power consumption prediction model of the power cabinet to generate a distribution cabinet prediction result. The prediction model constructed in is a regression analysis model or a machine learning model. By inputting the real-time distribution cabinet data into the power consumption prediction model, the model generates a predicted value for the power consumption of the distribution cabinet in the future period of time based on the relationship between the features learned during the previous training and the target variable. Specifically, in actual operation, methods such as linear regression, random forest regression, support vector regression, neural network, etc. are used to generate prediction results based on the characteristic values ​​of the input data. The prediction result contains information such as the power demand and load changes of the distribution cabinet in a certain period of time in the future, which is the basis for subsequent decision-making. Further, the prediction result of the distribution cabinet is judged and analyzed based on the preset result judgment threshold to generate a distribution cabinet prediction judgment result. The core of this step is to compare the prediction results of the distribution cabinet with the preset threshold to determine whether the current prediction results meet the safe load range or energy efficiency standards of the system. If the prediction results exceed the preset load threshold, it indicates that the system has a potential overload risk or other abnormal conditions, and further optimization and adjustment or early warning signals are required. Finally, the distribution cabinet distribution strategy is generated based on the prediction and judgment results of the distribution cabinet. This process involves formulating appropriate power dispatching strategies and load management plans. This step can determine the appropriate distribution strategy through rule engines and optimization algorithms (such as linear programming, genetic algorithms, particle swarm optimization, etc.), including load distribution, fault isolation, load balancing and other operations.

[0074] Step S4: execute the power distribution strategy of the distribution cabinet to generate a multi-dimensional scheduling result of the distribution cabinet; generate a report based on the multi-dimensional scheduling result of the distribution cabinet to generate a distribution cabinet control report, thereby completing the control operation of the electrical distribution cabinet.

[0075] In an embodiment of the present invention, by real-time monitoring of various operating parameters of the distribution cabinet (such as current, voltage, load, temperature, etc.), the distribution cabinet control system can adjust the power flow according to the set scheduling strategy to ensure the stable distribution of electricity. For example, if the load of the distribution cabinet exceeds the predetermined threshold, the system will automatically adjust the load distribution and perform operations such as load transfer or load current limiting to prevent overload or insufficient power supply. In addition, the operating state of the distribution cabinet can be dynamically optimized through intelligent algorithms (such as deep learning, reinforcement learning, etc.), and the scheduling strategy can be automatically executed to generate multi-dimensional scheduling results of the distribution cabinet. These scheduling results are obtained after integrating multi-dimensional data (such as current, voltage, temperature, humidity, etc.), and include scheduling suggestions for parameters such as load, power, temperature and humidity, ensuring that the system can maintain an efficient and stable working state under different operating environments. Report generation is performed based on the multi-dimensional scheduling results of the distribution cabinet to generate a distribution cabinet control report. This process uses data visualization and report generation technology to display and visualize the scheduling results in detail through data analysis tools (such as Pandas, Matplotlib, Seaborn and other libraries in Python, or BI tools such as Tableau, Power BI, etc.). The control report not only contains the real-time operation status of the distribution cabinet, but also provides the execution status of the scheduling strategy, the abnormal conditions that occur and their treatment methods, and the performance evaluation data of the system after the strategy is executed. These reports can be used as a decision-making basis for operation and maintenance personnel, helping them to understand the working status of the distribution cabinet in a timely manner and take corresponding maintenance or optimization measures. This process ensures the automatic control and intelligent scheduling of the distribution cabinet, improves the stability and efficiency of the system operation, and provides a basis for subsequent decision-making, maintenance and improvement by generating detailed control reports.

[0076] Preferably, step S1 comprises the following steps:

[0077] Step S11: acquiring multidimensional data of the power distribution cabinet; performing data cleaning processing on the multidimensional data of the power distribution cabinet to generate multidimensional cleaning data of the power distribution cabinet, wherein the multidimensional cleaning data of the power distribution cabinet includes current cleaning data, voltage cleaning data, temperature cleaning data, humidity cleaning data and vibration cleaning data;

[0078] Step S12: performing feature processing on the multi-dimensional cleaning data of the power distribution cabinet to generate multi-dimensional feature data of the power distribution cabinet;

[0079] Step S13: Standardize the data set according to the multi-dimensional feature data of the power distribution cabinet to generate a pre-processed data set of the power distribution cabinet.

[0080] In an embodiment of the present invention, multidimensional data of a power distribution cabinet is obtained and data cleaning is performed on it. The technical means at this stage mainly rely on data cleaning algorithms to ensure the quality and reliability of the data. The core goal of data cleaning is to remove unnecessary noise, repair missing values, and eliminate outliers in the data. Current, voltage, temperature, humidity, and vibration data are key parameters of the power distribution cabinet and need to undergo precise cleaning operations. In this process, commonly used cleaning methods include using interpolation methods to fill missing values ​​(such as linear interpolation, spline interpolation, etc.), using statistical methods to identify and eliminate outliers (such as the 3σ principle or IQR method), and data normalization processing (such as normalization or standardization) to eliminate the influence of different dimensions and units on data analysis. These cleaned data form a multidimensional cleaned data set containing current cleaning data, voltage cleaning data, temperature cleaning data, humidity cleaning data, and vibration cleaning data, ensuring the accuracy and consistency of the data used for subsequent analysis. The multidimensional cleaned data of the power distribution cabinet is feature processed to extract key features and enhance the expressiveness of the data set. The core of the feature processing technical means is to extract features that have a greater impact on the operating status of the power distribution cabinet by analyzing the relationship between each variable. Common feature processing methods include time series feature extraction, frequency domain analysis, principal component analysis (PCA), cluster analysis, etc. For example, current and voltage data can be converted into frequency domain features through Fourier transform, temperature and humidity data can extract seasonal fluctuation patterns, and vibration data can be analyzed by amplitude or acceleration spectrum. Through these technical means, the multidimensional feature data of the distribution cabinet is refined into a set of more concise and information-rich feature sets, which play a key role in subsequent data modeling and prediction tasks. The multidimensional feature data of the distribution cabinet is standardized to generate a preprocessed data set of the distribution cabinet. The technical means of data standardization is to adjust all features in the data set to a unified standard scale through mathematical transformation. Common standardization methods include Z-Score standardization and Min-Max normalization. The standardized data makes each feature have the same dimension in numerical value, avoids the influence of different dimension data on model training and optimization, and ensures the effective input of data in the machine learning model. In addition, standardization can also improve the convergence speed of the algorithm, reduce computational complexity, and improve the stability and accuracy of the prediction model.

[0081] Preferably, step S12 comprises the following steps:

[0082] Step S121: Perform Fourier transform spectrum analysis on the current cleaning data and the voltage cleaning data to generate current-voltage frequency component data; perform mean value calculation on the current-voltage frequency component data to generate current-voltage characteristic data of the distribution cabinet;

[0083] Step S122: Perform time series analysis on the temperature cleaning data and the humidity cleaning data to generate temperature-humidity seasonal fluctuation data; perform temperature and humidity coordinated processing on the temperature-humidity seasonal fluctuation data to generate temperature and humidity characteristic data of the power distribution cabinet;

[0084] Step S123: performing acceleration analysis on the vibration cleaning data to generate acceleration analysis data; performing amplitude fluctuation range analysis on the acceleration analysis data to generate acceleration-amplitude characteristic data of the distribution cabinet;

[0085] Step S124: performing time stamp calibration on the current-voltage characteristic data of the distribution cabinet, the temperature and humidity characteristic data of the distribution cabinet, and the acceleration-amplitude characteristic data of the distribution cabinet to generate multi-dimensional characteristic data of the distribution cabinet.

[0086] In an embodiment of the present invention, Fourier transform spectrum analysis is performed on the current cleaning data and the voltage cleaning data to extract the frequency component information of the data. As a classic signal processing technology, Fourier transform can reveal the periodic components and frequency distribution in the signal by converting the time domain signal into the frequency domain signal. Through this technology, the current and voltage signals can be converted from the time domain to the frequency domain to obtain the current-voltage frequency component data, and then the frequency characteristics of the signal can be analyzed. Subsequently, for the obtained current-voltage frequency component data, the main frequency characteristics are extracted by calculating the mean value to generate the current-voltage characteristic data of the distribution cabinet. These characteristic data are of great significance in the operation status monitoring of power equipment, and can help analyze the stability of the power system and predict faults. Time series analysis is performed using temperature cleaning data and humidity cleaning data to generate temperature-humidity seasonal volatility data. Time series analysis is a method for studying the trend, periodicity and volatility of time series data. Commonly used analysis methods include autoregressive model (AR), moving average model (MA) and seasonal difference. Through time series analysis, the fluctuation law of temperature and humidity in different seasons or time periods can be effectively extracted to help identify the environmental impact caused by seasonal changes. Subsequently, the temperature and humidity collaborative processing method is adopted to combine the temperature and humidity data, consider the correlation and interaction between them, and generate the temperature and humidity characteristic data of the distribution cabinet. This collaborative processing technology not only improves the comprehensive expressiveness of the temperature and humidity data, but also provides more accurate environmental parameters for the subsequent equipment operation status analysis. Acceleration analysis is performed on the vibration cleaning data to extract the vibration characteristics generated during the operation of the equipment. Acceleration analysis quantifies the amplitude, frequency and change law of vibration by analyzing the vibration signal in the time domain or frequency domain. Acceleration analysis can reveal the vibration changes caused by various mechanical or electrical faults in the distribution cabinet and discover potential equipment problems in advance. Then, the amplitude fluctuation range of the acceleration analysis data is analyzed to evaluate the fluctuation amplitude of the vibration and its change trend, and generate the acceleration-amplitude characteristic data of the distribution cabinet. This process can help accurately identify the status and failure risk of the equipment. The characteristic data of the above three dimensions—current-voltage characteristic data, temperature and humidity characteristic data, and acceleration-amplitude characteristic data—are timestamped. This calibration technology synchronizes data from different sources by unifying the time base, so that each data dimension can correspond to the same time point. The timestamp calibration technology ensures the time consistency of each data source, so that they can be compared and analyzed in the same time frame. Finally, the data after timestamp calibration is integrated into a multidimensional feature data set to generate multidimensional feature data of the distribution cabinet. These multidimensional feature data contain multiple key parameters of the distribution cabinet operation status, providing comprehensive basic data for subsequent fault diagnosis, performance evaluation and predictive analysis.

[0087] As an example of the present invention, refer to Figure 2As shown, in this example, step S2 includes:

[0088] Step S21: performing feature cross analysis on the distribution cabinet preprocessing data set to generate a distribution cabinet cross data set;

[0089] Step S22: Sorting the contribution of the distribution cabinets according to the cross-feature data set to generate distribution cabinet contribution ranking data;

[0090] Step S23: constructing a power cabinet model according to the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model;

[0091] In an embodiment of the present invention, feature cross analysis is performed on the preprocessed data set of the power distribution cabinet. Feature cross analysis is an important means commonly used in data science to mine the relationship between features. By combining different features together, new combined features can be generated, thereby better capturing the potential interaction effects in the data. In the power distribution cabinet data, there are complex nonlinear relationships between different sensor data such as current, voltage, temperature, and humidity. These complex relationships can be revealed through feature crossover. For example, the crossover features of current and temperature better reflect the operating status of power equipment under different environments. The generated power distribution cabinet crossover data set not only provides a single feature perspective, but also increases the interactive information, which can improve the accuracy of subsequent analysis and modeling. Contribution ranking is performed based on the crossover feature data set of the power distribution cabinet. Contribution ranking is an important feature selection method used to evaluate the degree of influence of each feature on the target variable. By calculating the correlation or contribution between each feature and the power consumption prediction target, the most influential feature for predicting power consumption behavior can be identified. Common contribution ranking techniques include methods based on information gain, chi-square test, correlation coefficient analysis, and model-based feature importance evaluation. For example, using a decision tree model or a random forest algorithm, the importance of each feature in the decision-making process can be calculated to sort them, thereby identifying which features are most critical to the effect of the power consumption prediction model. Through contribution sorting, the generated distribution cabinet contribution ranking data provides a clear feature selection basis for subsequent modeling. According to the distribution cabinet contribution ranking data, a power cabinet model is constructed to generate a power cabinet power consumption prediction model. In this step, a regression model or a classification model (such as linear regression, support vector machine, random forest, etc.) is constructed by selecting feature data with high contribution to predict the power consumption of the distribution cabinet. Using the feature set sorted by contribution can improve the prediction accuracy of the model and reduce the interference of redundant features, thereby making the model more streamlined and efficient. The power consumption prediction model can predict the power consumption and load changes of the distribution cabinet based on the input real-time data, helping operation and maintenance personnel to perform load adjustment, fault warning and other tasks.

[0092] Preferably, step S21 includes the following steps:

[0093] Step S211: performing current-temperature product calculation on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the current-temperature product ratio influence data of the distribution cabinet; performing voltage-humidity interaction item fluctuation analysis on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the voltage-humidity fluctuation interaction item data of the distribution cabinet;

[0094] Step S212: Perform current-vibration frequency load correlation analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate current-vibration frequency related data; perform Pearson correlation coefficient calculation on the current-vibration frequency related data to generate current-vibration frequency characteristic data of the distribution cabinet; perform volt-voltage-acceleration linkage analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate voltage-acceleration linkage data of the distribution cabinet;

[0095] Step S213: Perform temperature-vibration amplitude thermal analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate temperature-vibration amplitude thermal analysis data; perform vibration friction analysis on the temperature-vibration amplitude thermal analysis data to generate temperature-vibration amplitude data of the distribution cabinet; perform humidity-acceleration physical property analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate humidity-acceleration physical property change data; perform dynamic analysis on the mechanical energy ratio of the humidity-acceleration physical property change data to generate humidity-acceleration change characteristic data of the distribution cabinet;

[0096] Step S214: Generate a multidimensional feature data matrix using the distribution cabinet current-temperature product ratio influence data, distribution cabinet voltage-humidity fluctuation interaction item data, distribution cabinet current-vibration frequency characteristic data, distribution cabinet voltage-acceleration linkage data, distribution cabinet temperature-vibration amplitude data and distribution cabinet humidity-acceleration change characteristic data to generate a distribution cabinet multidimensional data matrix; perform principal component analysis on the distribution cabinet multidimensional data matrix to reduce the dimensionality, and construct a data set to generate a distribution cabinet cross data set.

[0097] In an embodiment of the present invention, the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet are subjected to current-temperature product calculation to generate the current-temperature product ratio influence data of the distribution cabinet. This step extracts the influence of current change on temperature change by performing mathematical operations on the relationship between current and temperature, and helps to evaluate the influence of current load change on the thermal effect of the equipment. Next, the voltage and humidity characteristic data are subjected to interactive term fluctuation analysis to generate voltage-humidity fluctuation interactive term data. This step uses the correlation between voltage and humidity to analyze the influence of voltage fluctuation on humidity under different environmental conditions, and reveals the potential influence of temperature and humidity changes on the electrical performance of the distribution cabinet. The combination of these two analyses helps to understand the interaction between temperature and humidity and current and voltage, thereby providing more accurate equipment operation monitoring data. First, the current-vibration frequency load correlation analysis is performed on the current-voltage characteristic data and the acceleration-amplitude characteristic data to generate current-vibration frequency correlation data. This process reveals the correlation between current change and vibration frequency, which is used to study the influence of current fluctuation on the vibration characteristics of the equipment, thereby optimizing the vibration monitoring of electrical equipment. On this basis, the Pearson correlation coefficient of the current-vibration frequency related data is further calculated to generate the current-vibration frequency characteristic data of the distribution cabinet. The Pearson correlation coefficient can quantify the linear relationship between current and vibration frequency, providing a more scientific numerical basis. Finally, the current-voltage characteristic data and the acceleration-amplitude characteristic data are combined to perform a volt-to-acceleration linkage analysis to generate voltage-acceleration linkage data. This step reveals the mutual influence between voltage and acceleration, which helps to dynamically predict the operating status of the equipment when the load is large. The temperature-vibration amplitude thermal analysis is performed through the temperature and humidity characteristic data and the acceleration-amplitude characteristic data to generate the temperature-vibration amplitude thermal analysis data. This analysis provides theoretical support for understanding the vibration response of the distribution cabinet under different temperature environments. The temperature-vibration amplitude thermal analysis data is further processed by vibration friction analysis to generate the temperature-vibration amplitude data of the distribution cabinet, which provides a key basis for the prediction of the thermal load of the equipment. In addition, the analysis of the physical properties of humidity and acceleration reveals the impact of humidity changes on vibration characteristics, thereby generating humidity-acceleration physical property change data, and through dynamic analysis of mechanical energy ratio, generating distribution cabinet humidity-acceleration change characteristic data, optimizing the load change characteristics of the distribution cabinet under different environments. All generated characteristic data, such as current-temperature product ratio influence data, voltage-humidity fluctuation interaction data, current-vibration frequency characteristic data, voltage-acceleration linkage data, temperature-vibration amplitude data, and humidity-acceleration change characteristic data, are used to generate a multi-dimensional characteristic data matrix. This operation combines different features to form a highly correlated and rich feature space, thereby providing multi-dimensional information input for subsequent data analysis and modeling.Finally, principal component analysis (PCA) is used to reduce the dimensionality of the multidimensional data matrix to reduce data redundancy and retain the main information of the data, thereby constructing a distribution cabinet cross-dataset.

[0098] Preferably, step S23 includes the following steps:

[0099] Step S231: Sorting the distribution cabinet contribution ranking data to generate distribution cabinet contribution ranking data, wherein the distribution cabinet contribution ranking data includes distribution cabinet high contribution data and distribution cabinet low contribution data;

[0100] Step S232: eliminating redundancy of low contribution by using a preset contribution threshold to generate low contribution elimination data;

[0101] Step S233: Perform linear regression prediction on the high contribution data and the low contribution eliminated data to construct a power cabinet model, and generate a power cabinet power consumption prediction model.

[0102] In an embodiment of the present invention, the contribution ranking data of the distribution cabinets are sorted, and the contribution data of the distribution cabinets are arranged according to the contribution level through a certain sorting algorithm, such as quick sort or merge sort, so as to generate the contribution ranking data of the distribution cabinets. This operation can clarify which distribution cabinets contribute more to the overall system and which distribution cabinets contribute less, thereby providing a clear data basis for subsequent analysis. On this basis, the contribution ranking data of the distribution cabinets is further subdivided into high contribution data of the distribution cabinets and low contribution data of the distribution cabinets. High contribution data represents equipment that is critical to the operation of the system, and low contribution data is equipment data that has little impact on the overall performance. This process realizes the priority distinction of features at the data level and provides valuable information for subsequent data processing. Redundancy elimination is performed for low contribution data, and the distribution cabinet data below the threshold is screened and eliminated through a preset contribution threshold. This redundant elimination process is based on the set contribution threshold. Data below the threshold is considered to have little impact on the overall system prediction, so there is no need to continue processing. Through this step, the dimension of the data set is reduced, noise data and redundant information are reduced, and the subsequent modeling process is ensured to be more efficient and accurate. The setting of the threshold depends on the understanding of the system operation characteristics and is determined through the analysis of historical data or experimental optimization. The high contribution data and the low contribution data are combined, and the linear regression method is used for modeling to generate a power consumption prediction model for the power distribution cabinet. Linear regression is a classic regression analysis method. By analyzing the relationship between the high contribution data and the low contribution data after elimination, a power distribution cabinet power consumption prediction model is established. Specifically, the linear regression method fits the linear relationship between the power consumption of the distribution cabinet and its related characteristic data through the least squares method, thereby constructing a mathematical model that can predict future power consumption. This model can not only reflect the power consumption trend of the distribution cabinet, but also provide accurate prediction data for power dispatching and load management.

[0103] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0104] Step S31: Acquire real-time distribution cabinet data;

[0105] Step S32: Based on the power consumption prediction model of the power cabinet, the real-time distribution cabinet data is used to predict the power consumption of the distribution cabinet, and a distribution cabinet prediction result is generated; the distribution cabinet prediction result is marked with a position, and a distribution cabinet position-prediction result is generated;

[0106] Step S33: Based on a preset result judgment threshold, the prediction result of the distribution cabinet is judged and analyzed to generate a prediction judgment result of the distribution cabinet; based on the prediction judgment result of the distribution cabinet, a power distribution strategy of the distribution cabinet is generated to generate a power distribution strategy of the distribution cabinet.

[0107] In the embodiment of the present invention, obtaining real-time distribution cabinet data is the starting step of the whole process, and the data source is the operation data of the distribution cabinet collected by sensors or monitoring equipment. These data include multi-dimensional sensor data such as current, voltage, temperature, humidity, etc., which can reflect the operation status of the distribution cabinet in real time. The technical means of this step involve sensor data acquisition and real-time data stream processing, in which the data acquisition system needs to ensure the high frequency and accuracy of the data, and can effectively transmit the original data to the processing platform through the data transmission system for subsequent processing. The key technology at this stage is the integrated application of data acquisition and data transmission technology to ensure that the collected data can be sent to the subsequent analysis and processing links in a timely and accurate manner. Based on the power consumption prediction model of the power cabinet, the real-time distribution cabinet data is predicted to generate the distribution cabinet prediction result. This process relies on a trained power cabinet power consumption prediction model, which is a machine learning model obtained by training historical distribution cabinet data, and adopts machine learning methods such as regression analysis, decision tree, and support vector machine. The power cabinet power consumption prediction model predicts the power consumption in the future period of time by inputting various sensor data of the real-time distribution cabinet. The generated distribution cabinet prediction results include information such as the future power consumption and load changes of the distribution cabinet, and provide a prediction of the operating status of the distribution cabinet. The technical means involved in this step include feature engineering (such as data cleaning, data standardization, etc.) and the prediction and reasoning process of the machine learning model. Through location annotation, the prediction results are combined with specific geographical location, distribution cabinet number and other information to form the distribution cabinet location-prediction results. This step can be achieved through data association technology, in which the location identifier (such as ID or coordinates) of each distribution cabinet is matched with the corresponding prediction results to generate a prediction data set with location information. Location annotation not only facilitates the analysis and tracking of results, but also helps to accurately schedule according to the location of different distribution cabinets during actual operation and maintenance. The distribution cabinet prediction results are judged and analyzed based on the preset result judgment threshold. The technical means of this step involve threshold judgment and data classification analysis. According to business needs, certain power consumption prediction thresholds are set (such as too high or too low load). When the predicted power consumption of the distribution cabinet exceeds or falls below these thresholds, the system will classify according to different judgment conditions and generate distribution cabinet prediction judgment results. Through this analysis, the system can identify which distribution cabinets have abnormal power usage, thus providing a basis for subsequent power distribution strategies. Finally, the power distribution strategy for the distribution cabinet is generated based on the prediction and judgment results of the distribution cabinet.

[0108] Preferably, step S33 includes the following steps:

[0109] Step S331: judging and analyzing the prediction results of the distribution cabinet based on the preset result judgment threshold, and generating the prediction judgment results of the distribution cabinet; generating the distribution cabinet distribution strategy based on the prediction judgment results of the distribution cabinet, and generating the distribution cabinet distribution strategy, wherein the distribution cabinet distribution strategy includes the distribution cabinet load and environment distribution strategy and the fault prediction and load adjustment distribution strategy;

[0110] Step S332: When the prediction judgment result of the distribution cabinet is greater than or equal to the preset result judgment threshold, the real-time load analysis of the distribution cabinet is performed on the current-voltage characteristic data of the distribution cabinet to generate the real-time load capacity data of the distribution cabinet; the high-load area is marked on the real-time load capacity data of the distribution cabinet to generate the high-load area marking data of the distribution cabinet; the temperature and humidity analysis is performed according to the temperature and humidity characteristic data of the distribution cabinet, and the high-temperature humidity marking is performed to generate the high-temperature humidity marking data of the distribution cabinet; the distribution load is optimized based on the high-load area marking data of the distribution cabinet and the high-temperature humidity marking data of the distribution cabinet, so as to generate the load and environmental distribution strategy of the distribution cabinet;

[0111] Step S333: When the prediction result of the distribution cabinet is less than the preset result judgment threshold, the current fluctuation-amplitude excess analysis is performed on the current-voltage characteristic data of the distribution cabinet to generate abnormal fault warning data; based on the prediction result of the distribution cabinet, the abnormal fault warning data is monitored and analyzed, and a warning message is generated and uploaded to the cloud platform, thereby completing the fault prediction and load adjustment distribution strategy.

[0112] In an embodiment of the present invention, the prediction result of the distribution cabinet is judged and analyzed based on a preset result judgment threshold to generate a distribution cabinet prediction judgment result. The core technical means of this process involve threshold judgment and data classification methods. First, the prediction result of the distribution cabinet is judged by a preset threshold to determine whether the load condition of the distribution cabinet meets the predetermined safety standard. When the prediction result exceeds a preset threshold, the system will mark it as a high load or abnormal state; when the prediction result is lower than a threshold, it will be identified as insufficient load or low equipment operation efficiency. The generated prediction judgment result will serve as the basis for subsequent power distribution strategy adjustment to ensure the optimization and safe operation of the power system. Based on these judgment results, the system will further generate corresponding power distribution strategies, including load and environment distribution strategies and fault prediction and load adjustment distribution strategies. The technical means of this step mainly involve data classification, conditional judgment and the application of rule engines, which are used to formulate power distribution strategies that adapt to different loads and environmental conditions. When the prediction judgment result of the distribution cabinet is greater than or equal to the preset result judgment threshold, the system will perform real-time load analysis on the current-voltage characteristic data of the distribution cabinet. This process uses a load analysis algorithm to evaluate the current load of the distribution cabinet through real-time calculation and analysis based on the characteristic data of current and voltage, and generates real-time load capacity data of the distribution cabinet. Next, the system marks the high-load area according to the generated real-time load capacity data, thereby generating high-load area marking data. This analysis method can identify areas with high load pressure by comparing the historical load data and real-time load data of the distribution cabinet, and then formulate targeted optimization strategies. At the same time, the system will combine the temperature and humidity characteristic data of the distribution cabinet to perform temperature and humidity analysis, detect high temperature and humidity areas, mark them and generate high temperature and humidity marking data. The technical means of this process include sensor data acquisition, real-time data analysis, feature annotation and data fusion to ensure that areas with dual pressure of load and environmental conditions can be identified. Based on the high-load area marking data and high temperature and humidity marking data, the system optimizes the distribution load, thereby generating a distribution cabinet load and environmental distribution strategy, which helps to reduce the burden of high-load areas and improve the operating efficiency and stability of the distribution cabinet. When the prediction judgment result of the distribution cabinet is less than the preset result judgment threshold, the system will perform current fluctuation-amplitude excess analysis on the current-voltage characteristic data of the distribution cabinet. This analysis method involves time series data analysis and fluctuation detection, aiming to identify situations where current fluctuations are too large and to determine whether there are abnormal behaviors that exceed the normal amplitude range. The generated abnormal fault warning data can capture potential fault hazards in a timely manner by comparing and analyzing the real-time monitoring data. Subsequently, the system further monitors and analyzes the abnormal fault warning data based on the prediction results of the distribution cabinet, generates warning messages, and uploads the warning data to the cloud platform. This process involves data transmission, data storage and processing on the cloud computing platform, and can provide fault warning information to relevant operation and maintenance personnel in real time.

[0113] As an example of the present invention, refer to Figure 4 As shown, in this example, step S4 includes:

[0114] Step S41: executing the power distribution strategy of the power distribution cabinet to generate a multi-dimensional scheduling result of the power distribution cabinet;

[0115] Step S42: performing a distribution cabinet performance evaluation on the multi-dimensional scheduling result of the distribution cabinet to generate distribution cabinet scheduling evaluation data;

[0116] Step S43: Generate a report on the distribution cabinet scheduling evaluation data and generate a distribution cabinet control report, thereby completing the electrical distribution cabinet control operation.

[0117] In the embodiment of the present invention, the key technical means for executing the power distribution strategy of the distribution cabinet is to convert the predetermined power distribution strategy into actual operation instructions through the system control algorithm, and adjust the strategy according to the real-time power distribution data. First of all, as a high-level decision rule set, the power distribution strategy of the distribution cabinet needs to be dynamically executed with the support of real-time data, which includes the instruction scheduling of the switch and load distribution of the distribution cabinet based on information such as load prediction, environmental analysis and fault warning. At the technical level, this process involves optimization algorithms and scheduling algorithms, such as scheduling methods based on heuristic search, constraint satisfaction problems (CSP) or linear programming (LP), etc., through which the load distribution of the distribution cabinet can be ensured to be reasonable and meet the safety and stability requirements of the operation of the distribution cabinet. The process of executing the strategy requires real-time acquisition of the operating status data of the equipment (such as current, voltage, temperature, etc.), and dynamic adjustment according to the power distribution strategy to achieve the multi-dimensional scheduling results of the distribution cabinet. The performance evaluation of the multi-dimensional scheduling results of the distribution cabinet involves the analysis and comprehensive evaluation of multi-dimensional data. First of all, it is necessary to quantify the performance indicators according to the operating data of the distribution cabinet, such as current load balancing, energy efficiency, temperature control, etc. The technical means adopts data evaluation model to generate performance evaluation data of distribution cabinet through multi-dimensional data fusion, feature extraction and weight distribution. Common technical means include decision tree-based evaluation method, fuzzy logic evaluation method or multi-attribute decision analysis (MADA). These methods compare the operation data of distribution cabinet under different strategies, evaluate its various performance indicators, such as load balancing, response speed, fault tolerance, etc., and finally generate distribution cabinet scheduling evaluation data to help optimize the distribution strategy. The process of generating distribution cabinet control report is to organize and summarize the above-mentioned scheduling evaluation data, and generate specific control reports through data visualization and report generation technology. This process includes data formatting, screening, integration and chart display, and finally generates distribution cabinet control report. Technically, data report generation tools, such as BI (business intelligence) tools, are used to dynamically integrate data and output reports in different formats according to user needs. The report will include various operating indicators, strategy execution, performance evaluation results and other contents of the distribution cabinet, so as to provide decision-making basis for subsequent electrical distribution cabinet control operations.

[0118] In this specification, an electrical distribution cabinet is provided for executing the above-mentioned electrical distribution cabinet control method, wherein the electrical distribution cabinet:

[0119] The data acquisition and preprocessing module is used to obtain multi-dimensional data of the power distribution cabinet; perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set of the power distribution cabinet;

[0120] The feature contribution analysis and model building module is used to rank the contribution of the distribution cabinet preprocessing data set and generate the distribution cabinet contribution ranking data; build the power cabinet model according to the distribution cabinet contribution ranking data and generate the power cabinet power consumption prediction model;

[0121] The real-time data prediction and power distribution strategy generation module is used to obtain real-time power distribution cabinet data; based on the power consumption prediction model of the power consumption cabinet, the power distribution cabinet power consumption is predicted for the real-time power distribution cabinet data, and the power distribution cabinet prediction result is generated; based on the preset result judgment threshold, the power distribution cabinet prediction result is judged and analyzed, and the power distribution cabinet power distribution strategy is generated;

[0122] The strategy execution and report generation module is used to execute the power distribution strategy of the distribution cabinet and generate the multi-dimensional scheduling results of the distribution cabinet; generate reports based on the multi-dimensional scheduling results of the distribution cabinet and generate distribution cabinet control reports, thereby completing the control operation of the electrical distribution cabinet.

[0123] The beneficial effect of the present invention is that by acquiring and preprocessing the multi-dimensional data of the distribution cabinet (including current, voltage, temperature and humidity, vibration and other data), noise and outliers can be eliminated to ensure the quality of the input data, thereby providing a reliable basis for subsequent feature analysis and model training. By sorting the data by contribution, the key factors affecting the power consumption of the distribution cabinet can be identified, providing a more accurate basis for the construction of the model. Based on these contribution-ranked data, the constructed power consumption prediction model can not only reflect the power consumption of the distribution cabinet under different environmental conditions, but also predict the load change trend and potential failure risks, and further optimize the distribution management strategy. Real-time data prediction and threshold-based strategy generation mechanism can dynamically adjust the load distribution and electrical environment control of the distribution cabinet, thereby realizing accurate scheduling and efficient management of the distribution cabinet. Finally, through strategy execution and scheduling result analysis, the generated control report provides data support and decision-making basis for the operation of the electrical distribution cabinet, which helps to improve the overall operation efficiency and reliability of the distribution cabinet. Scheduling optimization based on multidimensional data not only enhances the stability of the power distribution system, but also improves the energy utilization efficiency in the power distribution process, reduces the risk of failures, and ensures the rational allocation and efficient use of power resources. Therefore, the present invention solves the problems of low prediction accuracy, uneven resource allocation, and delayed response in traditional power distribution cabinet management by comprehensively integrating the multidimensional data of the power distribution cabinet for intelligent prediction and optimized scheduling, thereby improving the operating efficiency and resource utilization of the electrical power distribution cabinet.

[0124] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0125] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for controlling an electrical distribution cabinet, characterized in that: The following steps are involved: Step S1: Acquire multi-dimensional data of the power distribution cabinet; Perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set for the power distribution cabinet; Step S2: sorting the distribution cabinet preprocessing data set by contribution to generate distribution cabinet contribution sorting data; The power cabinet model is constructed based on the contribution ranking data of the distribution cabinet, and the power consumption prediction model of the power cabinet is generated; Step S3: Acquire real-time distribution cabinet data; perform distribution cabinet power consumption prediction on the real-time distribution cabinet data based on the power cabinet power consumption prediction model, and generate a distribution cabinet prediction result; perform judgment and analysis on the distribution cabinet prediction result based on a preset result judgment threshold, and generate a distribution cabinet power distribution strategy to generate a distribution cabinet power distribution strategy; Step S4: executing the power distribution strategy of the power distribution cabinet to generate a multi-dimensional scheduling result of the power distribution cabinet; Generate reports based on the multi-dimensional dispatch results of the distribution cabinet and generate distribution cabinet control reports to complete the electrical distribution cabinet control operation.

2. The electrical distribution cabinet control method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: acquiring multidimensional data of the power distribution cabinet; performing data cleaning processing on the multidimensional data of the power distribution cabinet to generate multidimensional cleaning data of the power distribution cabinet, wherein the multidimensional cleaning data of the power distribution cabinet includes current cleaning data, voltage cleaning data, temperature cleaning data, humidity cleaning data and vibration cleaning data; Step S12: performing feature processing on the multi-dimensional cleaning data of the power distribution cabinet to generate multi-dimensional feature data of the power distribution cabinet; Step S13: Standardize the data set according to the multi-dimensional feature data of the power distribution cabinet to generate a pre-processed data set of the power distribution cabinet.

3. The electrical distribution cabinet control method according to claim 2, characterized in that: Step S12 includes the following steps: Step S121: Perform Fourier transform spectrum analysis on the current cleaning data and the voltage cleaning data to generate current-voltage frequency component data; perform mean value calculation on the current-voltage frequency component data to generate current-voltage characteristic data of the distribution cabinet; Step S122: Perform time series analysis on the temperature cleaning data and the humidity cleaning data to generate temperature-humidity seasonal volatility data; perform temperature and humidity collaborative processing on the temperature-humidity seasonal volatility data to generate temperature and humidity characteristic data of the distribution cabinet; Step S123: performing acceleration analysis on the vibration cleaning data to generate acceleration analysis data; performing amplitude fluctuation range analysis on the acceleration analysis data to generate acceleration-amplitude characteristic data of the distribution cabinet; Step S124: performing time stamp calibration on the current-voltage characteristic data of the distribution cabinet, the temperature and humidity characteristic data of the distribution cabinet, and the acceleration-amplitude characteristic data of the distribution cabinet to generate multi-dimensional characteristic data of the distribution cabinet.

4. The electrical distribution cabinet control method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing feature cross analysis on the distribution cabinet preprocessing data set to generate a distribution cabinet cross data set; Step S22: Sorting the contribution of the distribution cabinets according to the cross-feature data set to generate distribution cabinet contribution ranking data; Step S23: construct a power cabinet model according to the distribution cabinet contribution ranking data to generate a power cabinet power consumption prediction model.

5. The electrical distribution cabinet control method according to claim 4, characterized in that: Step S21 includes the following steps: Step S211: performing current-temperature product calculation on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the current-temperature product ratio influence data of the distribution cabinet; performing voltage-humidity interaction item fluctuation analysis on the current-voltage characteristic data of the distribution cabinet and the temperature and humidity characteristic data of the distribution cabinet to generate the voltage-humidity fluctuation interaction item data of the distribution cabinet; Step S212: Perform current-vibration frequency load correlation analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate current-vibration frequency related data; perform Pearson correlation coefficient calculation on the current-vibration frequency related data to generate current-vibration frequency characteristic data of the distribution cabinet; perform volt-voltage-acceleration linkage analysis on the current-voltage characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate voltage-acceleration linkage data of the distribution cabinet; Step S213: Perform temperature-vibration amplitude thermal analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate temperature-vibration amplitude thermal analysis data; perform vibration friction analysis on the temperature-vibration amplitude thermal analysis data to generate temperature-vibration amplitude data of the distribution cabinet; perform humidity-acceleration physical property analysis on the temperature and humidity characteristic data of the distribution cabinet and the acceleration-amplitude characteristic data of the distribution cabinet to generate humidity-acceleration physical property change data; perform dynamic analysis on the mechanical energy ratio of the humidity-acceleration physical property change data to generate humidity-acceleration change characteristic data of the distribution cabinet; Step S214: Generate a multidimensional feature data matrix using the distribution cabinet current-temperature product ratio influence data, distribution cabinet voltage-humidity fluctuation interaction item data, distribution cabinet current-vibration frequency characteristic data, distribution cabinet voltage-acceleration linkage data, distribution cabinet temperature-vibration amplitude data and distribution cabinet humidity-acceleration change characteristic data to generate a distribution cabinet multidimensional data matrix; perform principal component analysis on the distribution cabinet multidimensional data matrix to reduce the dimensionality, and construct a data set to generate a distribution cabinet cross data set.

6. The electrical distribution cabinet control method according to claim 4, characterized in that: Step S23 includes the following steps: Step S231: Sorting the distribution cabinet contribution ranking data to generate distribution cabinet contribution ranking data, wherein the distribution cabinet contribution ranking data includes distribution cabinet high contribution data and distribution cabinet low contribution data; Step S232: eliminating redundancy of low contribution by using a preset contribution threshold to generate low contribution elimination data; Step S233: Perform linear regression prediction on the high contribution data and the low contribution eliminated data to construct a power cabinet model, and generate a power cabinet power consumption prediction model.

7. The electrical distribution cabinet control method according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Acquire real-time distribution cabinet data; Step S32: Based on the power consumption prediction model of the power cabinet, the real-time distribution cabinet data is used to predict the power consumption of the distribution cabinet, and a distribution cabinet prediction result is generated; the distribution cabinet prediction result is marked with a position, and a distribution cabinet position-prediction result is generated; Step S33: Based on a preset result judgment threshold, the prediction result of the distribution cabinet is judged and analyzed to generate a prediction judgment result of the distribution cabinet; based on the prediction judgment result of the distribution cabinet, a power distribution strategy of the distribution cabinet is generated to generate a power distribution strategy of the distribution cabinet.

8. The electrical distribution cabinet control method according to claim 7, characterized in that: Step S33 includes the following steps: Step S331: judging and analyzing the prediction results of the distribution cabinet based on the preset result judgment threshold, and generating the prediction judgment results of the distribution cabinet; generating the distribution cabinet distribution strategy based on the prediction judgment results of the distribution cabinet, and generating the distribution cabinet distribution strategy, wherein the distribution cabinet distribution strategy includes the distribution cabinet load and environment distribution strategy and the fault prediction and load adjustment distribution strategy; Step S332: When the prediction judgment result of the distribution cabinet is greater than or equal to the preset result judgment threshold, the real-time load analysis of the distribution cabinet is performed on the current-voltage characteristic data of the distribution cabinet to generate the real-time load capacity data of the distribution cabinet; the high-load area is marked on the real-time load capacity data of the distribution cabinet to generate the high-load area marking data of the distribution cabinet; the temperature and humidity analysis is performed according to the temperature and humidity characteristic data of the distribution cabinet, and the high-temperature humidity marking is performed to generate the high-temperature humidity marking data of the distribution cabinet; the distribution load is optimized based on the high-load area marking data of the distribution cabinet and the high-temperature humidity marking data of the distribution cabinet, so as to generate the load and environmental distribution strategy of the distribution cabinet; Step S333: When the prediction result of the distribution cabinet is less than the preset result judgment threshold, the current fluctuation-amplitude excess analysis is performed on the current-voltage characteristic data of the distribution cabinet to generate abnormal fault warning data; based on the prediction result of the distribution cabinet, the abnormal fault warning data is monitored and analyzed, and a warning message is generated and uploaded to the cloud platform, thereby completing the fault prediction and load adjustment distribution strategy.

9. The electrical distribution cabinet control method according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: executing the power distribution strategy of the power distribution cabinet to generate a multi-dimensional scheduling result of the power distribution cabinet; Step S42: performing a distribution cabinet performance evaluation on the multi-dimensional scheduling result of the distribution cabinet to generate distribution cabinet scheduling evaluation data; Step S43: Generate a report on the distribution cabinet scheduling evaluation data and generate a distribution cabinet control report, thereby completing the electrical distribution cabinet control operation.

10. An electrical distribution cabinet, characterized in that: An identification controller is included for executing the electrical distribution cabinet control method according to claim 1, wherein the identification controller of the electrical distribution cabinet includes: Data acquisition and preprocessing module, used to obtain multi-dimensional data of the power distribution cabinet; Perform data preprocessing on the multi-dimensional data of the power distribution cabinet to generate a preprocessing data set for the power distribution cabinet; The feature contribution analysis and model building module is used to rank the contribution of the distribution cabinet preprocessing data set and generate the distribution cabinet contribution ranking data; build the power cabinet model according to the distribution cabinet contribution ranking data and generate the power cabinet power consumption prediction model; The real-time data prediction and power distribution strategy generation module is used to obtain real-time power distribution cabinet data; based on the power consumption prediction model of the power consumption cabinet, the power distribution cabinet power consumption is predicted for the real-time power distribution cabinet data, and the power distribution cabinet prediction result is generated; based on the preset result judgment threshold, the power distribution cabinet prediction result is judged and analyzed, and the power distribution cabinet power distribution strategy is generated; The strategy execution and report generation module is used to execute the power distribution strategy of the distribution cabinet and generate multi-dimensional scheduling results of the distribution cabinet; Generate reports based on the multi-dimensional dispatch results of the distribution cabinet and generate distribution cabinet control reports to complete the electrical distribution cabinet control operation.

Images