Intelligent Electrical Equipment Early Warning and Monitoring System and Method Based on Multi-Sensor Interaction

Abstract

The present invention discloses an early warning and monitoring system and method for intelligent electrical equipment based on multi-sensor interaction, which relates to the technical field of early warning management of electrical equipment. The early warning and monitoring system for intelligent electrical equipment based on multi-sensor interaction includes: a data acquisition compliance judgment module, a data fusion effect judgment module, and a monitoring optimization compliance judgment module. The present invention evaluates the acquisition quality of the collected transformer aging monitoring data to judge whether to perform data fusion, then evaluates the effect of transformer early warning data fusion and judges whether to perform transformer early warning monitoring optimization, and finally obtains the transformer early warning optimization index after performing transformer early warning monitoring optimization to judge whether to perform dynamic priority scheduling of transformer early warning monitoring resources, realizing the improvement of the real-time performance of intelligent transformer aging performance early warning and solving the problem of untimely early warning and monitoring of the aging performance of intelligent transformers based on multi-sensor interaction in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of early warning management of electrical equipment, and particularly to an intelligent electrical equipment early warning monitoring system and method based on multi-sensor interaction. Background Art

[0002] The global manufacturing industry is undergoing a transformation and upgrading towards digitalization, networking, and intelligence. Intelligent manufacturing emphasizes the interconnection of devices, data-driven decision-making, and autonomous decision-making, and the intelligent early warning monitoring system based on multi-sensor interaction is one of the key technologies to achieve intelligent manufacturing. In the field of intelligent electrical equipment, with the development of technology, especially driven by technologies such as the Internet of Things, artificial intelligence, big data analysis, and machine learning, the management and maintenance of electrical equipment are undergoing profound changes. The traditional equipment management method mainly relies on manual inspection and regular maintenance, which has problems such as untimely response, unforeseeable faults, and high maintenance costs. To address these challenges, an intelligent electrical equipment early warning monitoring system based on multi-sensor interaction has emerged, becoming an important part of modern industrial production and intelligent manufacturing, and also an inevitable choice in dealing with equipment failures and improving production efficiency.

[0003] The existing intelligent electrical equipment early warning monitoring system based on multi-sensor interaction collects the status information of equipment in real time through various sensors, including physical quantities such as temperature, vibration, and current, and fuses multi-source data from different sensors to improve the accuracy and reliability of fault diagnosis. Machine learning algorithms (such as decision trees, support vector machines, neural networks, etc.) are used to train models through historical data to identify whether the equipment is in a fault mode, early warning of potential faults and visual display. By combining data analysis, sensor technology, and intelligent decision-making, the precision and timeliness of equipment management are improved, the impact of equipment failures on production is reduced, and enterprises are promoted to develop in a more efficient and intelligent direction.

[0004] For example, the building mechanical and electrical equipment management system based on Internet of Things technology disclosed in the patent application with the publication number of CN118735106A includes: an equipment monitoring module for dynamically monitoring the location, working status, and operating parameters of building mechanical and electrical equipment to determine the real-time data of building mechanical and electrical equipment based on Internet of Things technology; a data processing module for retrieving, sorting, and fusing the real-time data of building mechanical and electrical equipment based on Internet of Things technology to determine the characterization data of building mechanical and electrical equipment based on Internet of Things technology; an analysis and evaluation module for analyzing and evaluating the characterization data of building mechanical and electrical equipment based on the standard data of building mechanical and electrical equipment to determine the analysis and evaluation results of building mechanical and electrical equipment based on Internet of Things technology; and an early warning and control module for formulating an early warning and control plan for building mechanical and electrical equipment and performing early warning and control on building mechanical and electrical equipment based on the early warning and control plan for building mechanical and electrical equipment.

[0005] For example, a method and system for remote maintenance of electromechanical equipment for industrial cloud services announced in the invention patent announcement with the announcement number of CN115640860B includes: collecting basic data during the operation of electromechanical equipment, obtaining the operation data of electromechanical equipment through automatic identification and analysis of the basic data, obtaining the fault information of electromechanical equipment through the operation data of electromechanical equipment, constructing a matrix with the operation position data of electromechanical equipment, and obtaining the optimal maintenance distance through the matrix, obtaining the fault information of electromechanical equipment by calculating the weight analysis ratio of the temperature value affecting the current value, and combining with a vision sensor to judge the fault information of electromechanical equipment to obtain a more accurate analysis of the fault information of electromechanical equipment.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, it is found that the above technology has at least the following technical problems:

[0007] Intelligent electrical equipment (such as intelligent transformers) is usually in a relatively complex and changeable environment (such as power substations, remote areas far from urban areas, etc.). During the data collection process by sensors, data loss or data inconsistency may occur, resulting in the inability to obtain comprehensive monitoring information in a timely manner. At the same time, there is usually a strong electromagnetic field around intelligent transformers, which may interfere with certain types of sensors (such as current and voltage sensors), affecting the accuracy of data collection.

[0008] It should also be considered that multi-sensor data usually needs to be integrated through fusion algorithms. These algorithms usually have a large amount of calculations. Especially when there are many types of sensors and complex data dimensions, the data fusion process may cause delays, resulting in the problem of untimely early warning and monitoring of the aging performance of intelligent transformers based on multi-sensor interaction. Summary of the Invention

[0009] By providing an early warning and monitoring system and method for intelligent electrical equipment based on multi-sensor interaction in the embodiments of the present application, the problem of untimely early warning and monitoring of the aging performance of intelligent transformers based on multi-sensor interaction in the prior art is solved, and the real-time performance of the early warning of the aging performance of intelligent transformers is improved.

[0010] The embodiment of this application provides an early warning and monitoring system for intelligent electrical equipment based on multi-sensor interaction, including: a data acquisition compliance judgment module, a data fusion effect judgment module, and a monitoring optimization compliance judgment module; the data acquisition compliance judgment module is used to evaluate the acquisition quality of the transformer aging monitoring data collected to obtain a transformer acquisition index, and judge whether to perform data fusion based on the transformer acquisition index, and the transformer acquisition index is used to quantitatively evaluate the compliance degree of the sensor in collecting the transformer aging monitoring data quality; the data fusion effect judgment module is used to evaluate the effect of the transformer early warning data fusion to obtain a transformer fusion index after performing data fusion, and judge whether to perform transformer early warning monitoring optimization based on the transformer fusion index, and the transformer fusion index is used to comprehensively quantify the fusion effect of the transformer aging monitoring data; the monitoring optimization compliance judgment module is used to obtain the transformer early warning optimization index after performing the transformer early warning monitoring optimization, and judge whether to perform the dynamic priority scheduling of the transformer early warning monitoring resources based on the transformer early warning optimization index, and the transformer early warning optimization index is used to comprehensively quantify the compliance degree of the transformer early warning monitoring optimization.

[0011] Further, the specific steps of evaluating the acquisition quality of the collected transformer aging monitoring data to obtain a transformer acquisition index are as follows: statistically analyze the transformer aging monitoring data collected by the monitoring sensor within a preset time interval to obtain the corresponding average transformer monitoring data; obtain the transformer acquisition evaluation data within the preset time interval, and the transformer acquisition evaluation data includes the total amount of transformer acquisition monitoring data, the amount of lost transformer acquisition monitoring data, and the transformer acquisition redundancy correlation coefficient; obtain the transformer acquisition index according to the average transformer monitoring data, the transformer acquisition evaluation data, and the transformer acquisition related reference data obtained from the preset database; the monitoring sensor includes a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, and a gas sensor; the average transformer monitoring data includes the average transformer monitoring temperature data, the average transformer monitoring current data, the average transformer monitoring voltage data, the average transformer vibration amplitude data, and the average transformer methane concentration data; the transformer acquisition related reference data includes the transformer monitoring reference value, the transformer monitoring reference deviation, the reference acquisition redundancy correlation coefficient, and the transformer monitoring acquisition correction factor.

[0012] Further, the specific process of judging whether to perform data fusion based on the transformer acquisition index is as follows: compare the transformer acquisition index with the preset warning acquisition threshold range obtained from the preset database; if the transformer acquisition index is within the preset warning acquisition threshold range, perform data fusion, and evaluate the fusion effect of the transformer early warning monitoring data to obtain a transformer fusion index; if the transformer acquisition index exceeds the preset warning acquisition threshold range, do not perform data fusion, and at the same time remind the preset personnel to re-collect the transformer aging monitoring data.

[0013] Further, the specific steps for evaluating the fusion effect of the transformer early warning monitoring data to obtain the transformer fusion index are as follows: Obtain the transformer fusion evaluation data within a preset time interval after performing data fusion; the transformer fusion evaluation data includes the transformer acquisition index, the early warning monitoring fusion data volume, the total acquisition early warning monitoring data volume, the early warning monitoring fusion time, the predicted performance aging value, and the early warning monitoring fusion signal-to-noise ratio; Combine the transformer fusion evaluation data with the transformer fusion reference data obtained from the preset database to obtain the transformer fusion index; the transformer fusion reference data includes the reference optimal transformer acquisition index, the reference maximum fusion time, the reference predicted performance aging value, and the reference monitoring signal-to-noise ratio.

[0014] Further, the method for obtaining the transformer fusion index is as follows:

[0015] ;

[0016] In the formula, represents the transformer fusion index, represents the numbering of the data fusion times within the preset time interval, , represents the total number of data fusion times within the preset time interval, represents the transformer acquisition index of the r-th data fusion of the transformer aging monitoring data, represents the reference optimal transformer acquisition index, represents the early warning monitoring fusion data volume of the r-th data fusion of the transformer aging monitoring data, represents the total acquisition early warning monitoring data volume within the preset time interval, represents the reference maximum fusion time, represents the early warning monitoring fusion time of the r-th data fusion of the transformer aging monitoring data, represents the predicted performance aging value of the r-th data fusion of the transformer aging monitoring data, represents the reference predicted performance aging value, represents the early warning monitoring fusion signal-to-noise ratio of the preset time interval, represents the reference monitoring signal-to-noise ratio.

[0017] Further, the specific process of judging whether to perform transformer early warning monitoring optimization based on the transformer fusion index is as follows: If the transformer fusion index is within the preset early warning fusion threshold range, the transformer early warning monitoring optimization is not performed. At the same time, the transformer aging monitoring data is optimized and then uploaded to the cloud for storage; the optimization process includes abnormal detection of transformer aging monitoring data, feature extraction of transformer aging monitoring data, and compression processing of transformer aging monitoring data; If the transformer fusion index exceeds the preset early warning fusion threshold range, the transformer early warning monitoring optimization is performed, and the transformer early warning optimization index after performing the transformer early warning monitoring optimization is obtained.

[0018] Further, the specific process of the transformer early warning monitoring optimization is as follows: A1. Optimize the monitoring data collection of the transformer. When the transformer fusion index after the monitoring data collection optimization is within the preset early warning fusion threshold range, stop the transformer early warning monitoring optimization; otherwise, execute A2; A2. Optimize the data fusion of the transformer. Judge whether the transformer fusion index after the data fusion optimization is within the preset early warning fusion threshold range. If so, stop the transformer early warning monitoring optimization; otherwise, execute A3; A3. Optimize the early warning monitoring architecture of the transformer, and at the same time obtain the transformer early warning optimization index after performing the transformer early warning monitoring optimization.

[0019] Further, the specific steps for obtaining the transformer early warning optimization index after performing the transformer early warning monitoring optimization are as follows: Obtain the transformer index to be optimized and the transformer optimization evaluation data after performing the transformer early warning monitoring optimization; the transformer index to be optimized includes the acquisition index of the transformer to be optimized and the fusion index of the transformer to be optimized; the transformer optimization evaluation data includes the optimized sampling frequency, the total amount of early warning monitoring data collected after optimization, the amount of lost data in the early warning monitoring data collected after optimization, and the aging value of the average prediction performance of the early warning monitoring fusion after optimization; The transformer early warning optimization index is obtained according to the transformer index to be optimized, the transformer optimization evaluation data, and the transformer optimization reference data obtained from the preset database; the transformer optimization reference data includes the reference sampling frequency, the reference optimal transformer early warning fusion index, and the reference aging value of the early warning monitoring prediction performance.

[0020] Further, the specific process of judging whether to perform dynamic priority scheduling of transformer early warning monitoring resources based on the transformer early warning optimization index is as follows: If the transformer early warning optimization index is within the preset early warning optimization interval, do not perform dynamic priority scheduling of transformer early warning monitoring resources, and continuously monitor whether the transformer fusion index is within the preset early warning fusion threshold range; If the transformer early warning optimization index exceeds the preset early warning optimization interval, perform dynamic priority scheduling of transformer early warning monitoring resources; the dynamic priority scheduling of transformer early warning monitoring resources includes algorithm priority scheduling and monitoring resource scheduling.

[0021] The embodiments of the present application provide an early warning and monitoring method for intelligent electrical equipment based on multi-sensor interaction, including the following steps: S1, evaluate the acquisition quality of the collected transformer aging monitoring data to obtain a transformer acquisition index, and determine whether to perform data fusion based on the transformer acquisition index. The transformer acquisition index is used to quantitatively evaluate the compliance degree of the sensor in collecting the transformer aging monitoring data; S2, after performing data fusion, evaluate the effect of the transformer early warning data fusion to obtain a transformer fusion index, and determine whether to perform transformer early warning monitoring optimization based on the transformer fusion index. The transformer fusion index is used to comprehensively quantify the fusion effect of the transformer aging monitoring data; S3, obtain the transformer early warning optimization index after performing the transformer early warning monitoring optimization, and determine whether to perform dynamic priority scheduling of the transformer early warning monitoring resources based on the transformer early warning optimization index. The transformer early warning optimization index is used to comprehensively quantify the compliance degree of the transformer early warning monitoring optimization.

[0022] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0023] 1. By evaluating the acquisition quality of the collected transformer aging monitoring data to determine whether to perform data fusion, then evaluating the effect of the transformer early warning data fusion and determining whether to perform transformer early warning monitoring optimization, and finally obtaining the transformer early warning optimization index after performing the transformer early warning monitoring optimization to determine whether to perform dynamic priority scheduling of the transformer early warning monitoring resources, the evaluation of the transformer early warning data fusion and monitoring optimization is realized, and then the real-time performance of the intelligent transformer aging performance early warning is improved, effectively solving the problem of untimely early warning and monitoring of the aging performance of intelligent transformers based on multi-sensor interaction in the prior art.

[0024] 2. By obtaining the transformer fusion evaluation data within a preset time interval after performing data fusion, and combining the transformer fusion evaluation data with the transformer fusion reference data obtained from the preset database to obtain the transformer fusion index, the numerical evaluation of the transformer aging monitoring data fusion effect is realized, and then the more accurate evaluation of the transformer aging monitoring data fusion effect is realized.

[0025] 3. By obtaining the transformer to-be-optimized index and the transformer optimization evaluation data after performing the transformer early warning monitoring optimization, and obtaining the transformer early warning optimization index according to the transformer to-be-optimized index, the transformer optimization evaluation data and the transformer optimization reference data obtained from the preset database, the numerical evaluation of the compliance degree of the transformer early warning monitoring optimization is realized, and then the more precise evaluation of the compliance degree of the transformer early warning monitoring optimization is realized. Description of the Drawings

[0026] Figure 1Schematic diagram of the structure of the intelligent electrical equipment early warning monitoring system based on multi-sensor interaction provided by the embodiments of the present application;

[0027] Figure 2 Schematic diagram of the change of the transformer fusion index with the transformer acquisition index provided by the embodiments of the present application;

[0028] Figure 3 Schematic diagram of the change of the transformer fusion index with the early warning monitoring fusion signal-to-noise ratio provided by the embodiments of the present application;

[0029] Figure 4 Flowchart of the intelligent electrical equipment early warning monitoring method based on multi-sensor interaction provided by the embodiments of the present application. Specific implementation manners

[0030] In the embodiments of the present application, by providing an intelligent electrical equipment early warning monitoring system and method based on multi-sensor interaction, the problem that the aging performance of an intelligent transformer cannot be timely warned and monitored based on multi-sensor interaction in the prior art is solved. The acquisition quality of the collected transformer aging monitoring data is evaluated to obtain a transformer acquisition index, and it is judged whether to perform data fusion based on the transformer acquisition index. Then, after performing data fusion, the effect of the transformer early warning data fusion is evaluated to obtain a transformer fusion index and it is judged whether to perform transformer early warning monitoring optimization. Finally, the transformer early warning optimization index after performing transformer early warning monitoring optimization is obtained, and it is judged whether to perform dynamic priority scheduling of transformer early warning monitoring resources based on the transformer early warning optimization index, realizing the improvement of the real-time performance of the aging performance warning of the intelligent transformer.

[0031] The technical solutions in the embodiments of the present application are to solve the problem that the aging performance of an intelligent transformer cannot be timely warned and monitored based on multi-sensor interaction. The general idea is as follows:

[0032] By evaluating the acquisition quality of the collected transformer aging monitoring data to judge whether to perform data fusion, then evaluating the effect of the transformer early warning data fusion and judging whether to perform transformer early warning monitoring optimization, and finally judging whether to perform dynamic priority scheduling of transformer early warning monitoring resources, the effect of improving the real-time performance of the aging performance warning of the intelligent transformer is achieved.

[0033] To better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0034] As Figure 1As shown in the figure, it is a schematic structural diagram of an intelligent electrical equipment warning and monitoring system based on multi-sensor interaction provided by an embodiment of the present application. The intelligent electrical equipment warning and monitoring system based on multi-sensor interaction provided by an embodiment of the present application includes: a data acquisition compliance judgment module, a data fusion effect judgment module, and a monitoring optimization compliance judgment module; the data acquisition compliance judgment module is used to evaluate the acquisition quality of the transformer aging monitoring data collected to obtain a transformer acquisition index, and judge whether to perform data fusion based on the transformer acquisition index. The transformer acquisition index is used to quantitatively evaluate the compliance degree of the sensor in collecting the transformer aging monitoring data quality; the data fusion effect judgment module is used to evaluate the effect of the transformer warning data fusion to obtain a transformer fusion index after performing data fusion, and judge whether to perform transformer warning monitoring optimization based on the transformer fusion index. The transformer fusion index is used to comprehensively quantify the fusion effect of the transformer aging monitoring data; the monitoring optimization compliance judgment module is used to obtain a transformer warning optimization index after performing transformer warning monitoring optimization, and judge whether to perform transformer warning monitoring resource dynamic priority scheduling based on the transformer warning optimization index. The transformer warning optimization index is used to comprehensively quantify the compliance degree of the transformer warning monitoring optimization.

[0035] Among them, the transformer aging monitoring data includes transformer monitoring temperature data, transformer monitoring current data, transformer monitoring voltage data, transformer vibration amplitude data, and transformer methane concentration data.

[0036] Specifically, the transformer monitoring temperature data is obtained through temperature sensors deployed at key parts of the transformer (such as the windings and oil tanks of the transformer), the transformer monitoring current data and the transformer monitoring voltage data are obtained through current sensors and voltage sensors deployed at the output end of the transformer, the transformer vibration amplitude data is obtained through vibration sensors deployed at the center point of the transformer base, and the transformer methane concentration data is obtained through gas sensors (methane gas sensors) deployed at the ventilation openings of the transformer oil tank.

[0037] By using sensors to obtain transformer aging monitoring data, the evaluation of the acquisition quality of the collected transformer aging monitoring data is realized, and further, the improvement of the data reliability in the process of intelligent transformer aging performance warning and monitoring through multi-sensor interaction is realized. By improving the data reliability in the process of intelligent transformer aging performance warning and monitoring, the real-time performance of intelligent transformer aging performance warning is improved.

[0038] Further, the specific steps for evaluating the acquisition quality of the collected transformer aging monitoring data to obtain the transformer acquisition index are as follows: statistically analyze the transformer aging monitoring data collected by the monitoring sensors within a preset time interval to obtain the corresponding average transformer monitoring data; obtain the transformer acquisition evaluation data within the preset time interval, where the transformer acquisition evaluation data includes the total amount of transformer acquisition monitoring data, the amount of lost data in transformer acquisition monitoring, and the transformer acquisition redundancy correlation coefficient; obtain the transformer acquisition index based on the average transformer monitoring data, the transformer acquisition evaluation data, and the transformer acquisition-related reference data obtained from the preset database.

[0039] Among them, the monitoring sensors include current sensors, voltage sensors, temperature sensors, vibration sensors, and gas sensors; the average transformer monitoring data includes the average transformer monitoring temperature data, the average transformer monitoring current data, the average transformer monitoring voltage data, the average transformer vibration amplitude data, and the average transformer methane concentration data; the transformer acquisition-related reference data includes the transformer monitoring reference value, the transformer monitoring reference deviation, the reference acquisition redundancy correlation coefficient, and the transformer monitoring acquisition correction factor.

[0040] The method for obtaining the transformer acquisition index is as follows:

[0041] ;

[0042] In the formula, represents the transformer acquisition index, represents the type number of the monitoring sensor, , represents the total number of types of monitoring sensors, represents the average transformer monitoring data corresponding to the nth type of monitoring sensor within the preset time interval, represents the transformer monitoring reference value corresponding to the nth type of monitoring sensor, represents the transformer monitoring reference deviation corresponding to the nth type of monitoring sensor, represents the total amount of transformer acquisition monitoring data within the preset time interval, represents the amount of lost data in transformer acquisition monitoring within the preset time interval, represents the transformer acquisition redundancy correlation coefficient between the nth type of monitoring sensor and the jth type of monitoring sensor within the preset time interval, , represents the reference acquisition redundancy correlation coefficient, represents the transformer monitoring acquisition correction factor.

[0043] In this embodiment, the total number of types of monitoring sensors is 5 (i.e., = 5), among which, They respectively correspond to a current sensor, a voltage sensor, a temperature sensor, a vibration sensor, and a gas sensor. The average transformer monitoring data is the average value of the transformer aging monitoring data at preset moments within a preset time interval. The corresponding average transformer monitoring data is obtained by statistically analyzing the average transformer monitoring data corresponding to the preset time interval using mathematical functions in LabVIEW (such as the Mean function).

[0044] The total amount of transformer acquisition monitoring data is obtained through data acquisition software (such as LabVIEW, MATLAB, PI System, etc.). The amount of lost data in transformer acquisition monitoring is obtained through a data loss detection algorithm (such as using a time series analysis algorithm to detect data loss in programming environments such as MATLAB and Python). Specifically, in MATLAB, data loss can be detected using the isnan function. The redundancy correlation coefficient of transformer acquisition is obtained through a correlation coefficient calculation tool (such as the corr function in MATLAB). Specifically, the redundancy correlation coefficient of transformer acquisition is calculated by constructing a matrix. Each column represents a time series, and the corr function returns a correlation coefficient matrix, where each element represents the redundancy correlation coefficient of transformer acquisition between the corresponding two time series.

[0045] The aforementioned database is a database established before the design of the intelligent electrical equipment warning and monitoring system based on multi-sensor interaction provided in the embodiments of the present application for storing various types of set data. The database includes, but is not limited to, transformer monitoring temperature data, transformer monitoring current data, transformer monitoring voltage data, transformer vibration amplitude data, and transformer methane concentration data, etc. The various values therein are directly set by technicians. For example, the result of summing and averaging the transformer aging monitoring data corresponding to the collected historical monitoring sensors represents the corresponding transformer monitoring reference value. The result of summing and averaging the difference between the transformer aging monitoring data corresponding to the collected historical monitoring sensors and the corresponding transformer monitoring reference value represents the transformer monitoring reference deviation. The result of summing and averaging the collected historical redundancy correlation coefficients of transformer acquisition represents the reference redundancy correlation coefficient of acquisition.

[0046] The transformer monitoring acquisition correction factor is obtained from a preset database and is used to reflect the influence degree of factors affecting the quality of transformer aging monitoring data acquisition (such as the intensity of electromagnetic interference) on the acquisition indicators of the transformer. The real-time factors affecting the quality of transformer aging monitoring data acquisition (such as the intensity of electromagnetic interference) are input into a preset mapping set in the database to obtain the corresponding correction factor.

[0047] The transformer acquisition index is used to quantitatively evaluate the compliance degree of the sensor in acquiring the aging monitoring data of the transformer. Among them, the average transformer monitoring data, the amount of lost data in transformer acquisition monitoring, and the correlation coefficient of transformer acquisition redundancy all affect the transformer acquisition index. Specifically, as the deviation degree between the average transformer monitoring data and the corresponding transformer monitoring reference value decreases, the transformer acquisition index increases; as the amount of lost data in transformer acquisition monitoring decreases and the correlation coefficient of transformer acquisition redundancy increases, the transformer acquisition index increases accordingly.

[0048] In addition, the transformer acquisition index includes parameters in multiple aspects, and there are connections between the parameters and they do not exist independently. For example, as the amount of lost data in transformer acquisition monitoring increases, that is, due to data loss, the deviation degree between the average transformer monitoring data and the corresponding transformer monitoring reference value increases, and then the transformer acquisition index decreases accordingly; in addition, as the correlation coefficient of transformer acquisition redundancy decreases, it indicates that the data difference between sensors is large, which may lead to the distortion of the average transformer monitoring data, that is, the deviation degree between the average transformer monitoring data and the corresponding transformer monitoring reference value increases, and the transformer acquisition index also decreases; therefore, considering the relevance and mutual influence between various factors, the transformer acquisition index is obtained through comprehensive analysis, realizing the numerical evaluation of the compliance degree of the sensor in acquiring the aging monitoring data of the transformer, and judging the compliance degree of the sensor in acquiring the aging monitoring data of the transformer through numerical evaluation, thereby improving the accuracy of the evaluation of the compliance degree of the sensor in acquiring the aging monitoring data of the transformer.

[0049] Furthermore, the specific process of judging whether to perform data fusion based on the transformer acquisition index is as follows: compare the transformer acquisition index with the preset warning acquisition threshold range obtained from the preset database; if the transformer acquisition index is within the preset warning acquisition threshold range, perform data fusion and evaluate the fusion effect of the transformer warning monitoring data to obtain the transformer fusion index; if the transformer acquisition index exceeds the preset warning acquisition threshold range, do not perform data fusion, and at the same time remind the preset personnel to re-acquire the aging monitoring data of the transformer.

[0050] In this embodiment, specifically, the preset warning acquisition threshold range is set by professionals according to the standards in the field. For example, the preset warning acquisition threshold range is set to be from 0.857 to 0.9. Among them, data fusion is achieved through data fusion methods such as weighted average method, Kalman filter or Bayesian fusion method. Specifically, for example, the Bayesian inference data fusion method based on probability theory fuses data from different sources according to the calculation of the posterior probability distribution of sensor data. Among them, the prior distribution and likelihood function in the Bayesian inference data fusion method calculate the posterior distribution by combining the data of all sensors, and the optimal estimate after fusion is obtained according to the maximum posterior probability or the calculation of the expected value of the posterior distribution. By combining the preset warning acquisition threshold range for judgment and performing data fusion, the accuracy of judging the compliance degree of the quality of the transformer aging monitoring data collected by the sensor is improved.

[0051] Further, the specific steps for evaluating the fusion effect of the transformer warning monitoring data to obtain the transformer fusion index are as follows: Obtain the transformer fusion evaluation data within the preset time interval after performing data fusion; The transformer fusion evaluation data includes the transformer acquisition index, the warning monitoring fusion data volume, the total acquisition warning monitoring data volume, the warning monitoring fusion time, the predicted performance aging value, and the warning monitoring fusion signal-to-noise ratio; Combine the transformer fusion evaluation data with the transformer fusion reference data obtained from the preset database to obtain the transformer fusion index; The transformer fusion reference data includes the reference optimal transformer acquisition index, the reference maximum fusion time, the reference predicted performance aging value, and the reference monitoring signal-to-noise ratio.

[0052] The method for obtaining the transformer fusion index is as follows:

[0053] ;

[0054] In the formula, represents the transformer fusion index, represents the numbering of the data fusion times within the preset time interval, , represents the total number of data fusion times within the preset time interval, represents the transformer acquisition index of the r-th data fusion of the transformer aging monitoring data, represents the reference optimal transformer acquisition index, represents the warning monitoring fusion data volume of the r-th data fusion of the transformer aging monitoring data, represents the total acquisition warning monitoring data volume within the preset time interval, represents the reference maximum fusion time, represents the warning monitoring fusion time of the r-th data fusion of the transformer aging monitoring data, represents the predicted performance aging value of the r-th data fusion of the transformer aging monitoring data, represents the reference predicted performance aging value, represents the warning monitoring fusion signal-to-noise ratio for a preset time interval, represents the reference monitoring signal-to-noise ratio.

[0055] In this embodiment, the warning monitoring fusion data volume and the total warning monitoring data volume are obtained through a data acquisition system (such as a SCADA, Supervisory Control and Data Acquisition system). Specifically, by determining the data acquisition frequency (for example, once per second or once per minute), the data volume collected by all sensors within a specified time is obtained. The warning monitoring fusion time is obtained through a big data processing framework (such as Apache Kafka, Spark, etc.). The aging state of the transformer (such as the performance degradation value) is predicted through a machine learning algorithm (such as regression analysis, neural network, etc.) to obtain the predicted performance aging value. Specifically, the aging state of the transformer is predicted by training a model using historical data. The warning monitoring fusion signal-to-noise ratio is obtained through a signal analysis tool (such as MATLAB, LabVIEW, or NumPy in the Python library). The warning monitoring fusion signal-to-noise ratio is the signal-to-noise ratio of the transformer aging monitoring data after data fusion.

[0056] The reference optimal transformer acquisition index is generally set as the maximum value of the preset warning acquisition threshold range. The reference maximum fusion time is represented by the maximum value of the collected historical warning monitoring fusion time. The reference predicted performance aging value is represented by the result of summing and averaging the collected historical predicted performance aging values. The reference monitoring signal-to-noise ratio is represented by the result of summing and averaging the collected historical warning monitoring fusion signal-to-noise ratios.

[0057] Where it is set that: the total number of data fusions within the preset time interval is 1, the reference optimal transformer acquisition index is 0.9, the warning monitoring fusion data volume is 400 bytes, the total warning monitoring data volume is 500 bytes, the reference maximum fusion time is 300 seconds, the warning monitoring fusion time is 150 seconds, the predicted performance aging value is 0.6, the reference predicted performance aging value is 0.5, and the reference monitoring signal-to-noise ratio is 45; as Figure 2 shown, it is a schematic diagram of the change of the transformer fusion index with the transformer acquisition index provided by the embodiment of the present application. Specifically, the warning monitoring fusion signal-to-noise ratio is set to 40. As the deviation degree between the transformer acquisition index and the reference optimal transformer acquisition index decreases, the transformer fusion index increases; as Figure 3 shown, it is a schematic diagram of the change of the transformer fusion index with the warning monitoring fusion signal-to-noise ratio provided by the embodiment of the present application. Specifically, the transformer acquisition index is set to 0.8. As the warning monitoring fusion signal-to-noise ratio increases, the transformer fusion index also increases.

[0058] The transformer fusion index is used to comprehensively quantify the fusion effect of transformer aging monitoring data. Specifically, the transformer fusion index includes parameters in multiple aspects. Among them, the transformer acquisition index, the amount of early warning monitoring fusion data, the early warning monitoring fusion time, the aging value of prediction performance, and the signal-to-noise ratio of early warning monitoring fusion all affect the transformer fusion index. Specifically, as the transformer acquisition index, the amount of early warning monitoring fusion data, and the signal-to-noise ratio of early warning monitoring fusion increase, the transformer fusion index increases accordingly; as the early warning monitoring fusion time decreases, the transformer fusion index increases; as the deviation degree between the aging value of prediction performance and the reference aging value of prediction performance decreases, the transformer fusion index also increases accordingly.

[0059] It should be added that there are connections among the various parameters in the transformer fusion index and they do not exist independently. For example, as the transformer acquisition index increases, the amount of early warning monitoring fusion data increases accordingly. And the increase in the amount of early warning monitoring fusion data can help improve the signal-to-noise ratio because more data can average out the noise and enhance the signal, that is, the signal-to-noise ratio of early warning monitoring fusion increases accordingly, and then the transformer fusion index also increases accordingly. In addition, the transformer acquisition index also affects the aging value of prediction performance. Specifically, as the transformer acquisition index increases, the deviation degree between the aging value of prediction performance and the reference aging value of prediction performance decreases, and then the transformer fusion index increases accordingly. At the same time, as the amount of early warning monitoring fusion data increases, the signal-to-noise ratio of early warning monitoring fusion increases accordingly, and then the transformer fusion index also increases accordingly. Therefore, through quantification, the relevance and mutual influence among various factors are considered, and the transformer fusion index is obtained through comprehensive analysis, realizing the numerical evaluation of the fusion effect of transformer aging monitoring data. By judging the fusion effect of transformer aging monitoring data through numerical evaluation, a more accurate evaluation of the fusion effect of transformer aging monitoring data is realized.

[0060] Furthermore, the specific process of judging whether to execute the optimization of transformer early warning monitoring based on the transformer fusion index is as follows: Compare the transformer fusion index with the preset early warning fusion threshold range obtained from the preset database; if the transformer fusion index is within the preset early warning fusion threshold range, do not execute the optimization of transformer early warning monitoring, and at the same time, optimize the transformer aging monitoring data and upload it to the cloud storage after optimization; the optimization process includes abnormal detection of transformer aging monitoring data, feature extraction of transformer aging monitoring data, and compression processing of transformer aging monitoring data; if the transformer fusion index exceeds the preset early warning fusion threshold range, execute the optimization of transformer early warning monitoring and obtain the transformer early warning optimization index after executing the optimization of transformer early warning monitoring.

[0061] In this embodiment, the preset warning fusion threshold range is set by professionals according to the standards in the field. For example, the acquisition index range of the transformer is set to 0.8 to 0.9, the warning monitoring fusion data volume range is 80% to 100% of the total warning monitoring data volume, the total warning monitoring data volume range is 400 bytes to 900 bytes, the warning monitoring fusion time range is 100 seconds to 300 seconds, the predicted performance aging value range is 0 to 1, and the warning monitoring fusion signal-to-noise ratio range is 40 to 60, so as to determine the preset warning fusion threshold range.

[0062] Among them, the abnormal detection of the transformer aging monitoring data is realized through an abnormal detection algorithm (such as abnormal detection based on the mean and standard deviation). Among them, the values exceeding the mean plus or minus three times the standard deviation are abnormal values; the feature extraction of the transformer aging monitoring data is realized by using numpy in the Python library to calculate the time-domain features and the scipy or pywt library to calculate the wavelet transform. Specifically, the kurtosis and skewness of the transformer aging monitoring data are obtained by using numpy in the Python library to realize the feature extraction of the transformer aging monitoring data; the compression processing of the transformer aging monitoring data is realized through data coding compression (such as Huffman coding). Specifically, the generated Huffman coding is used to encode the original data, and each symbol is replaced with the corresponding binary coding to form the compressed data; by combining the preset warning fusion threshold range for judgment and optimizing the transformer warning monitoring, a more accurate judgment of the transformer aging monitoring data fusion effect is realized.

[0063] It should be added that the specific process of optimizing the transformer warning monitoring is as follows: A1, optimize the monitoring data acquisition of the transformer. When the fusion index of the transformer after optimizing the monitoring data acquisition is within the preset warning fusion threshold range, stop optimizing the transformer warning monitoring; otherwise, execute A2; A2, optimize the data fusion of the transformer. Judge whether the fusion index of the transformer after executing the data fusion optimization is within the preset warning fusion threshold range. If so, stop optimizing the transformer warning monitoring; otherwise, execute A3; A3, optimize the warning monitoring architecture of the transformer, and at the same time obtain the transformer warning optimization index after executing the transformer warning monitoring optimization.

[0064] In this embodiment, the optimization of monitoring data acquisition is achieved through an adaptive sampling algorithm (such as an event-driven sampling strategy). Specifically, for example, static thresholds are set (such as the temperature exceeding 80 degrees Celsius and the current exceeding 300 A), and the sampling condition is triggered when the static threshold is exceeded; the optimization of data fusion is achieved through a time series analysis method (such as ARIMA, Autoregressive Integrated Moving Average). Specifically, for example, the ARIMA model is implemented through the statsmodels library in Python, where the time series data is made stationary through differencing operations. Usually, the ADF (Augmented Dickey-Fuller) test is used to check whether the data is stationary. If the data is non-stationary, differencing is performed until the data becomes stationary; the optimization of the early warning monitoring architecture is achieved through the cloud platform and the Internet of Things. Specifically, for example, a stream processing framework (such as Apache Kafka, Apache Flink) is used for real-time processing of data streams. Among them, Flink provides exactly-once and at-least-once semantics to ensure that no data is lost during the real-time data processing; through the optimization of transformer early warning monitoring, the reliability of early warning monitoring in the case of abnormal data fusion effect of transformer aging monitoring is improved.

[0065] Furthermore, the specific steps to obtain the transformer early warning optimization index after the execution of transformer early warning monitoring optimization are as follows: Obtain the transformer index to be optimized and the transformer optimization evaluation data after the execution of transformer early warning monitoring optimization; The transformer index to be optimized includes the acquisition index of the transformer to be optimized and the fusion index of the transformer to be optimized; The transformer optimization evaluation data includes the optimized sampling frequency, the total amount of data for optimized acquisition early warning monitoring, the amount of lost data for optimized acquisition early warning monitoring, and the aging value of the average prediction performance of optimized early warning monitoring fusion; The transformer early warning optimization index is obtained based on the transformer index to be optimized, the transformer optimization evaluation data, and the transformer optimization reference data obtained from the preset database; The transformer optimization reference data includes the reference sampling frequency, the reference optimal transformer early warning fusion index, and the aging value of the reference early warning monitoring prediction performance.

[0066] The method to obtain the transformer early warning optimization index is as follows:

[0067] ;

[0068] In the formula, represents the transformer early warning optimization index, represents the acquisition index of the transformer to be optimized, represents the optimized sampling frequency, represents the reference sampling frequency, represents the total amount of data for optimized acquisition early warning monitoring, Indicates the optimized acquisition warning monitoring lost data volume, Indicates the transformer fusion index to be optimized, Indicates the reference optimal transformer warning fusion index, Indicates the optimized warning monitoring fusion average prediction performance aging value, Indicates the reference warning monitoring prediction performance aging value.

[0069] In this embodiment, the transformer acquisition index to be optimized refers to the transformer acquisition index after performing the transformer warning monitoring optimization, and the transformer fusion index to be optimized refers to the transformer fusion index after performing the transformer warning monitoring optimization.

[0070] The optimized sampling frequency is obtained through a wavelet transform tool (such as the cwt (continuous wavelet transform) function in MATLAB). Specifically, the maximum frequency component in the signal is obtained through the time-frequency diagram of the wavelet transform. The total optimized acquisition warning monitoring data volume refers to the total transformer acquisition monitoring data volume after performing the transformer warning monitoring optimization, the optimized acquisition warning monitoring lost data volume refers to the transformer acquisition monitoring lost data volume after performing the transformer warning monitoring optimization, and the optimized warning monitoring fusion average prediction performance aging value refers to the prediction performance aging value after performing the transformer warning monitoring optimization.

[0071] The result of summing and averaging the collected historical optimized sampling frequencies represents the reference sampling frequency, and the result of summing and averaging the transformer warning fusion indexes within the collected historical preset warning fusion thresholds represents the reference optimal transformer warning fusion index.

[0072] Specific assumptions: The transformer acquisition index to be optimized is 0.9, the reference sampling frequency is 50 Hz, the total optimized acquisition warning monitoring data volume is 300 bytes, the reference optimal transformer warning fusion index is 0.9, and the reference warning monitoring prediction performance aging value is 0.98; The transformer warning optimization index can be calculated and obtained through the above method, and the change statistical table of the transformer warning optimization index is shown in Table 1:

[0073] Table 1 Change statistical table of transformer warning optimization index

[0074]

[0075] As can be seen from the first group of data and the second group of data in Table 1, as the optimized sampling frequency increases, the amount of lost data in the optimized acquisition early warning monitoring decreases accordingly, and then the transformer early warning optimization index increases accordingly. In addition, as can be seen from the second group of data, the third group of data and the fourth group of data, as the fusion index of the transformer to be optimized increases, the deviation degree between the average predicted performance aging value of the optimized early warning monitoring fusion and the predicted performance aging value of the reference early warning monitoring decreases, and then the transformer early warning optimization index also increases accordingly.

[0076] The transformer early warning optimization index is used to comprehensively quantify the compliance degree of the transformer early warning monitoring optimization. Among them, the transformer early warning optimization index includes multiple aspects of parameters, and there are connections between the parameters and they do not exist independently. For example, the optimized sampling frequency directly determines the fineness of data acquisition. The increase in the optimized sampling frequency helps to capture more detailed information and reduce the loss of key signals, thereby reducing the amount of lost data in the optimized acquisition early warning monitoring, and then the transformer early warning optimization index increases accordingly; in addition, as the optimized sampling frequency increases and the amount of lost data in the optimized acquisition early warning monitoring decreases, the fusion index of the transformer to be optimized increases accordingly, and then the transformer early warning optimization index also increases accordingly. At the same time, as the fusion index of the transformer to be optimized increases, the deviation degree between the average predicted performance aging value of the optimized early warning monitoring fusion and the predicted performance aging value of the reference early warning monitoring decreases; therefore, through quantification, the relevance and mutual influence between various factors are considered, and the transformer early warning optimization index is obtained through comprehensive analysis, realizing the numerical evaluation of the compliance degree of the transformer early warning monitoring optimization. Through the numerical evaluation, the compliance degree of the transformer early warning monitoring optimization is judged, and then the early warning monitoring of the aging performance of the intelligent transformer is better carried out.

[0077] Furthermore, the specific process of judging whether to execute the dynamic priority scheduling of the transformer early warning monitoring resources based on the transformer early warning optimization index is as follows: compare the transformer early warning optimization index with the preset early warning optimization threshold range obtained from the preset database; if the transformer early warning optimization index is within the preset early warning optimization interval, do not execute the dynamic priority scheduling of the transformer early warning monitoring resources, and at the same time continuously monitor whether the transformer fusion index is within the preset early warning fusion threshold range; if the transformer early warning optimization index exceeds the preset early warning optimization interval, execute the dynamic priority scheduling of the transformer early warning monitoring resources; the dynamic priority scheduling of the transformer early warning monitoring resources includes algorithm priority scheduling and monitoring resource scheduling.

[0078] In this embodiment, specifically, the preset warning optimization threshold range is set by professionals according to the standards in the field. For example, the range of the acquisition indicators of the transformer to be optimized is set to 0.8 to 0.9, the range of the optimized sampling frequency is set to 30 Hz to 100 Hz, the range of the total amount of data for optimized acquisition warning monitoring is set to 100 bytes to 900 bytes, the range of the lost data amount for optimized acquisition warning monitoring is set to 20 bytes to 50 bytes, the range of the fusion indicators of the transformer to be optimized is set to 0.7 to 1, and the range of the aging value of the optimized average prediction performance of warning monitoring fusion is set to 0.3 to 1, so as to determine the preset warning optimization threshold range.

[0079] Among them, the algorithm priority scheduling is implemented through a multi-level priority algorithm (such as shortest job first, shortest remaining time first, etc.). Specifically, for example, when a task is scheduled to execute, the task with the shortest execution time is selected; the monitoring resource scheduling is implemented through a multi-task scheduling optimization algorithm (such as scheduling based on a priority queue). Specifically, for example, the monitoring tasks are put into the queue according to the priority, and the system allocates resources according to the priority order of the tasks. Specifically, the priority queue scheduling can optimize the resource allocation by processing high-priority tasks first (such as when the temperature monitoring sensor shows that the oil temperature exceeds the safety threshold, that is, 65 °C to 80 °C), and then processing low-priority tasks (such as archiving and storing the historical operation data); by combining the preset warning optimization threshold range for judgment and performing dynamic priority scheduling of the transformer warning monitoring resources, a more accurate judgment of the compliance degree of the transformer warning monitoring optimization is achieved.

[0080] Such as Figure 4 As shown in the figure, it is a flowchart of the intelligent electrical equipment warning monitoring method based on multi-sensor interaction provided by the embodiment of the present application. The method includes the following steps: S1, evaluate the acquisition quality of the acquired transformer aging monitoring data to obtain the transformer acquisition indicators, and judge whether to perform data fusion based on the transformer acquisition indicators. The transformer acquisition indicators are used to quantitatively evaluate the compliance degree of the quality of the sensor-acquired transformer aging monitoring data; S2, after performing data fusion, evaluate the effect of the transformer warning data fusion to obtain the transformer fusion indicators, and judge whether to perform transformer warning monitoring optimization based on the transformer fusion indicators. The transformer fusion indicators are used to comprehensively quantify the fusion effect of the transformer aging monitoring data; S3, obtain the transformer warning optimization indicators after performing the transformer warning monitoring optimization, and judge whether to perform dynamic priority scheduling of the transformer warning monitoring resources based on the transformer warning optimization indicators. The transformer warning optimization indicators are used to comprehensively quantify the compliance degree of the transformer warning monitoring optimization.

[0081] In this embodiment, the aging monitoring data of the transformer is collected by sensors, thereby realizing the evaluation of the acquisition quality of the aging monitoring data of the transformer and the fusion of the transformer warning data, and judging whether to perform the optimization of the transformer warning monitoring based on the fusion effect of the transformer warning data, and further realizing the more timely warning monitoring of the intelligent transformer aging performance during the multi-sensor interaction process.

[0082] In summary, the embodiment of the present application evaluates the acquisition quality of the collected aging monitoring data of the transformer to judge whether to perform data fusion, then evaluates the effect of the transformer warning data fusion and judges whether to perform the optimization of the transformer warning monitoring, and finally obtains the transformer warning optimization index after performing the optimization of the transformer warning monitoring to judge whether to perform the dynamic priority scheduling of the transformer warning monitoring resources, thereby realizing the evaluation of the transformer warning data fusion and the monitoring optimization, and further realizing the improvement of the real-time performance of the intelligent transformer aging performance warning, effectively solving the problem that the prior art has the problem of untimely warning monitoring of the aging performance of the intelligent transformer based on multi-sensor interaction.

[0083] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0084] The present invention is described with reference to the flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0085] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions in the process Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in one or more processes and / or boxes. Figure 1 One process or more processes and / or boxes Figure 1 The steps of the functions specified in one box or more boxes.

[0087] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed as including the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. Intelligent electrical equipment early warning monitoring system based on multi-sensor interaction, characterized in that: include: Data collection compliance judgment module, data fusion effect judgment module and monitoring optimization compliance judgment module; The data collection compliance judgment module is used to evaluate the collection quality of the collected transformer aging monitoring data to obtain the transformer collection index, and judge whether to perform data fusion based on the relationship between the transformer collection index and the preset warning collection threshold range. The transformer collection index is used to quantitatively evaluate the compliance degree of the quality of the transformer aging monitoring data collected by the sensor. The method for obtaining the transformer collection index is as follows: ； In the formula, Indicates transformer acquisition indicators, Indicates the type number of the monitoring sensor. , Indicates the total number of monitoring sensor types. Indicates the average transformer monitoring data corresponding to the nth monitoring sensor within the preset time interval, Indicates the transformer monitoring reference value corresponding to the nth monitoring sensor, Indicates the transformer monitoring reference deviation corresponding to the nth monitoring sensor, Indicates the total amount of transformer monitoring data collected within the preset time interval. Indicates the amount of transformer monitoring loss data within the preset time interval. It represents the transformer acquisition redundancy correlation coefficient corresponding to the nth monitoring sensor and the jth monitoring sensor within the preset time interval, , represents the reference acquisition redundancy correlation coefficient, Indicates transformer monitoring acquisition correction factor; The data fusion effect judgment module is used to evaluate the effect of transformer early warning data fusion after data fusion is performed to obtain a transformer fusion index, and judge whether to perform transformer early warning monitoring optimization based on the relationship between the transformer fusion index and the preset early warning fusion threshold range. The transformer fusion index is used to comprehensively quantify the fusion effect of transformer aging monitoring data. The method for obtaining the transformer fusion index is as follows: ； In the formula, represents the transformer fusion index, Indicates the number of times data is fused within the preset time interval. , Indicates the total number of data fusions within the preset time interval. represents the transformer collection index of the rth data fusion of transformer aging monitoring data, Indicates the reference optimal transformer acquisition index, represents the amount of early warning monitoring fusion data of the rth data fusion of transformer aging monitoring data, Indicates the total amount of data collected for early warning monitoring within the preset time interval. represents the reference maximum fusion time, represents the early warning monitoring fusion time of the rth data fusion of transformer aging monitoring data, represents the predicted performance aging value of the rth data fusion of transformer aging monitoring data, represents the reference predicted performance aging value, Indicates the fusion signal-to-noise ratio of early warning monitoring at a preset time interval, represents the reference monitoring signal-to-noise ratio; The monitoring optimization compliance judgment module is used to obtain the transformer early warning optimization index after the transformer early warning monitoring optimization is performed, and judge whether to perform the dynamic priority scheduling of transformer early warning monitoring resources based on the relationship between the transformer early warning optimization index and the preset early warning optimization threshold range. The transformer early warning optimization index is used to comprehensively quantify the compliance degree of the transformer early warning monitoring optimization. The method for obtaining the transformer early warning optimization index is as follows: ； In the formula, Represents the transformer early warning optimization index, Indicates the transformer collection index to be optimized, represents the optimized sampling frequency, represents the reference sampling frequency, Indicates the total amount of data collected for early warning monitoring is optimized. Indicates the amount of data lost by optimizing the collection warning monitoring. represents the transformer fusion index to be optimized, It refers to the optimal transformer early warning fusion index, It represents the average prediction performance aging value of the optimized early warning monitoring fusion, Indicates the reference early warning monitoring predicted performance aging value.

2. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 1, characterized in that: The specific steps of evaluating the collection quality of the collected transformer aging monitoring data to obtain the transformer collection index are as follows: Performing statistical analysis on the transformer aging monitoring data collected by the monitoring sensor within a preset time interval to obtain the corresponding transformer monitoring average data; Acquire transformer collection and evaluation data within a preset time interval, wherein the transformer collection and evaluation data includes a total amount of transformer collection and monitoring data, an amount of transformer collection and monitoring lost data, and a transformer collection redundancy correlation coefficient; Obtain transformer acquisition indicators based on transformer monitoring average data, transformer acquisition evaluation data, and transformer acquisition related reference data obtained from a preset database; The monitoring sensors include current sensors, voltage sensors, temperature sensors, vibration sensors and gas sensors; The transformer monitoring average data includes transformer monitoring average temperature data, transformer monitoring average current data, transformer monitoring average voltage data, transformer average vibration amplitude data and transformer average methane concentration data; The transformer acquisition related reference data includes a transformer monitoring reference value, a transformer monitoring reference deviation, a reference acquisition redundancy correlation coefficient and a transformer monitoring acquisition correction factor.

3. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 1, characterized in that: The specific process of judging whether to perform data fusion based on the relationship between the transformer collection index and the preset warning collection threshold range is as follows: Compare the transformer collection index with a preset warning collection threshold range obtained from a preset database; If the transformer collection index is within the preset warning collection threshold range, data fusion is performed, and the fusion effect of the transformer warning monitoring data is evaluated to obtain the transformer fusion index; If the transformer collection index exceeds the preset warning collection threshold range, data fusion will not be performed, and the preset personnel will be reminded to re-collect the transformer aging monitoring data.

4. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 3, characterized in that: The specific steps of evaluating the fusion effect of transformer early warning monitoring data to obtain transformer fusion indicators are as follows: Obtain transformer fusion evaluation data within a preset time interval after executing data fusion; The transformer fusion evaluation data includes transformer collection indicators, early warning monitoring fusion data volume, total collection early warning monitoring data volume, early warning monitoring fusion time, predicted performance aging value and early warning monitoring fusion signal-to-noise ratio; The transformer fusion evaluation data is combined with the transformer fusion reference data obtained from a preset database to obtain a transformer fusion index; The transformer fusion reference data includes a reference optimal transformer acquisition index, a reference maximum fusion time, a reference predicted performance aging value, and a reference monitoring signal-to-noise ratio.

5. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 1, characterized in that: The specific process of judging whether to perform transformer early warning monitoring optimization based on the relationship between the transformer fusion index and the preset early warning fusion threshold range is as follows: If the transformer fusion index is within the preset warning fusion threshold, the transformer warning monitoring optimization will not be performed, and the transformer aging monitoring data will be optimized and uploaded to the cloud storage; The optimization process includes transformer aging monitoring data anomaly detection, transformer aging monitoring data feature extraction and transformer aging monitoring data compression processing; If the transformer fusion index exceeds the preset early warning fusion threshold range, the transformer early warning monitoring optimization is executed, and the transformer early warning optimization index after the transformer early warning monitoring optimization is executed is obtained.

6. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 5, characterized in that: The specific process of transformer early warning monitoring optimization is as follows: A1, optimize the monitoring data collection of the transformer. When the transformer fusion index after the monitoring data collection optimization is within the preset early warning fusion threshold range, stop the transformer early warning monitoring optimization, otherwise execute A2; A2, perform data fusion optimization on the transformer, and determine that the transformer fusion index after data fusion optimization is within the preset warning fusion threshold range, then stop the transformer early warning monitoring optimization, otherwise execute A3; A3, optimize the early warning monitoring architecture of the transformer, and obtain the transformer early warning optimization index after executing the transformer early warning monitoring optimization.

7. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 5, characterized in that: The specific steps of obtaining the transformer early warning optimization index after performing transformer early warning monitoring optimization are as follows: Obtain the transformer indicators to be optimized and transformer optimization evaluation data after performing transformer early warning monitoring optimization; The transformer indicators to be optimized include transformer acquisition indicators to be optimized and transformer fusion indicators to be optimized; The transformer optimization evaluation data includes optimizing the sampling frequency, optimizing the total amount of data collected for early warning monitoring, optimizing the amount of lost data collected for early warning monitoring, and optimizing the average predicted performance aging value of early warning monitoring fusion; Obtaining transformer early warning optimization indicators according to transformer indicators to be optimized, transformer optimization evaluation data and transformer optimization reference data obtained from a preset database; The transformer optimization reference data includes a reference sampling frequency, a reference optimal transformer early warning fusion index, and a reference early warning monitoring prediction performance aging value.

8. The intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in claim 1, characterized in that: The specific process of judging whether to perform dynamic priority scheduling of transformer early warning monitoring resources based on the relationship between the transformer early warning optimization index and the preset early warning optimization threshold range is as follows: If the transformer early warning optimization index is within the preset early warning optimization interval, the dynamic priority scheduling of transformer early warning monitoring resources will not be performed, and the transformer fusion index will be continuously monitored to see if it is within the preset early warning fusion threshold range; If the transformer early warning optimization index exceeds the preset early warning optimization interval, the dynamic priority scheduling of transformer early warning monitoring resources is executed; The dynamic priority scheduling of transformer early warning monitoring resources includes algorithm priority scheduling and monitoring resource scheduling.

9. A method for applying to the intelligent electrical equipment early warning monitoring system based on multi-sensor interaction as claimed in any one of claims 1 to 8, characterized in that: The following steps are involved: S1, evaluating the collection quality of the collected transformer aging monitoring data to obtain a transformer collection index, and judging whether to perform data fusion based on the transformer collection index, wherein the transformer collection index is used to quantitatively evaluate the degree of compliance of the quality of the transformer aging monitoring data collected by the sensor; S2, after executing data fusion, evaluating the effect of transformer early warning data fusion to obtain a transformer fusion index, and judging whether to perform transformer early warning monitoring optimization based on the transformer fusion index, wherein the transformer fusion index is used to comprehensively quantify the fusion effect of transformer aging monitoring data; S3, obtaining the transformer early warning optimization index after executing the transformer early warning monitoring optimization, and judging whether to execute the dynamic priority scheduling of transformer early warning monitoring resources based on the transformer early warning optimization index, wherein the transformer early warning optimization index is used to comprehensively quantify the compliance degree of the transformer early warning monitoring optimization.

Images