Transformer substation three-dimensional model data fusion method based on edge calculation

Abstract

The invention relates to a transformer substation three-dimensional model data fusion method based on edge calculation, which comprises the steps of data acquisition, preprocessing, dimension reduction, state evaluation model establishment and equipment state evaluation, and can monitor the state of transformer substation equipment in real time, accurately evaluate the health condition of the equipment and improve the reliability of the transformer substation. And the method and the device are widely applied to the technical field of transformer substations.

Description

Technical Field

[0001] The present invention relates to a substation three-dimensional model data fusion method based on edge computing, and belongs to the technical field of substations. Background Art

[0002] With the rapid development of power systems, the daily inspection and management of substations has become increasingly complex and important. However, there are still specific problems such as serious data island phenomenon, low inspection efficiency and insufficient equipment failure warning capabilities. Data unified application also faces difficulties such as data standardization, difficulty in model construction and high complexity of fusion algorithms. Therefore, the study of model-driven substation edge computing fusion technology is of great significance to improve the efficiency of substation inspection and management.

[0003] For example, the online monitoring data formats and protocols of equipment provided by different manufacturers are different, which makes it impossible to effectively interconnect data and affect the comprehensive utilization efficiency of data. Traditional manual inspection methods are time-consuming and labor-intensive, and are easily interfered by human factors, and cannot detect and deal with potential problems in a timely manner. The lack of effective data analysis methods makes it difficult to achieve a comprehensive and accurate assessment of equipment status, and thus it is impossible to predict and prevent equipment failures in advance.

[0004] Moreover, due to the wide variety of equipment and complex technical protocols, data standardization is challenging. It is not easy to build a model that accurately reflects the characteristics of the equipment, and the model needs to be updated as the equipment ages and upgrades. It is necessary to design an efficient and accurate fusion algorithm to process multi-source and heterogeneous data, taking into account factors such as real-time performance and accuracy.

[0005] Data fusion is an important concept in the field of information technology. It involves integrating data from different sources, formats and structures to provide more comprehensive and accurate information. In complex systems such as substations, data fusion is particularly critical because it can help operators better understand the overall status of the system and make more informed decisions.

[0006] In the operation of power equipment, the intelligent management process runs through the key of its complete management system, which makes the core functional modules of the power equipment system closely connected and in a unified network structure system, each playing a different role. Tests and inspections are important forms of daily inspection of the operating status of power equipment. The inspection time and methods are different. Usually, these two inspection methods cannot be used at the same time, but the power equipment status evaluation is closely linked to tests and inspections, and neither is indispensable. In other words, after the test and inspection, the relevant personnel must comprehensively analyze the subjective and objective influencing factors of all aspects, strictly follow the relevant regulations, and conduct necessary status evaluation of various types of power equipment again. This is an important guarantee for improving the effectiveness of the evaluation results of the power equipment status. As for the evaluation of the power equipment status, it has a variety of functions, such as querying the power equipment status evaluation data and reviewing the power equipment status evaluation results.

[0007] During the inspection of power equipment, relevant personnel need to evaluate the status of each component of the power equipment. According to the evaluation results of each component, the overall situation of the power equipment is clear, whether it is in a "healthy" state, and no abnormal situation occurs in any component. Only then can it be said that the power equipment is in stable operation. Once a component is not in a "normal state", for example, a warning state or a serious state, the power equipment needs to be evaluated as the most serious state. Usually, in the evaluation of the status of power equipment components, the evaluation results are based on a set of status parameters, with the status judgment as the basis. The degradation of each status parameter can be divided into multiple levels. The higher the level, the more serious the degradation of the power equipment.

[0008] With the advancement of intelligent management, although models are increasingly used in power systems, it is not easy to build models that accurately reflect the characteristics of equipment. Currently, model-driven technology is mainly used in the field of software development to improve development efficiency and code quality. With the rise of cyber-physical systems, model-driven technology has gradually been introduced into this field to simulate and predict the behavior of physical systems. In recent years, with the development of digitalization and intelligence, model-driven technology has played an increasingly important role in cyber-physical platforms, enabling more accurate monitoring and control.

[0009] Initially, cyber-physical platforms mainly appeared as data collection and monitoring systems. With the advancement of technology, the platform began to integrate more functions, such as data analysis, remote control, etc. At present, model-driven cyber-physical platforms have become an industry trend. They can not only realize real-time monitoring and analysis of data, but also optimize and analyze the system through models.

[0010] With the increasing application of processing technology, applying it to substation equipment status assessment and evaluating the status of equipment by analyzing signal characteristics of different frequencies has always been the focus of research for technical personnel in this field. Summary of the invention

[0011] The present invention discloses a substation three-dimensional model data fusion method based on edge computing. The model can monitor the status of substation equipment in real time and accurately evaluate the health status of the equipment based on edge computing, thereby improving the operating reliability of the equipment and reducing maintenance costs.

[0012] To achieve the above object, the technical solution adopted by the present invention is: a substation three-dimensional model data fusion method based on edge computing, comprising the following steps: S1: Collect data of substation equipment; Install sensors on substation equipment to collect data on voltage, current, partial discharge signals, temperature and humidity of the equipment in real time; S2: preprocessing the collected data; Preprocessing the raw data collected above; S3: Use principal component analysis method to reduce the dimension of preprocessed data; Perform principal component analysis on the preprocessed data to extract the feature vector that best represents the original data set; S4: Establish a condition assessment model based on the reduced-dimensional data combined with the historical operation data and maintenance records of the equipment; The extracted feature vectors and the historical operation data of the equipment are used to train the support vector machine model and establish a substation equipment status assessment model; S5: Evaluate the status of substation equipment using the status evaluation model and output the evaluation results; The real-time collected data is input into the trained model, and the model outputs the equipment’s health status score, failure probability, and maintenance recommendations.

[0013] Furthermore, in step S1, the collected data is effectively fused.

[0014] Furthermore, in step S2, the preprocessing also includes data standardization processing, and in the preprocessing step, effective fusion of online monitoring data is added.

[0015] Furthermore, in step S3, the method also includes data fusion and extraction of principal components.

[0016] Furthermore, in step S3, after dimensionality reduction, the state assessment model is further optimized in combination with the fusion data.

[0017] Furthermore, in step S4, the state assessment model is established based on edge computing.

[0018] Furthermore, in step S5, the evaluation results include the health status, failure probability and maintenance recommendations of the equipment.

[0019] Furthermore, the model output provides an analysis of possible equipment failure locations and causes in addition to health status scores and failure probabilities.

[0020] The beneficial effects of the present invention are as follows: the present invention can realize real-time and accurate evaluation of the status of substation equipment, timely discover potential faults and abnormalities of equipment, provide a scientific basis for the maintenance and overhaul of equipment, thereby improving the operational reliability and economic benefits of the equipment. DETAILED DESCRIPTION

[0021] The present invention is further described in detail below in conjunction with specific embodiments.

[0022] A substation three-dimensional model data fusion method based on edge computing includes the following steps: S1: Collect data of substation equipment; Install sensors on substation equipment to collect data on the equipment's voltage, current, partial discharge signal, temperature and humidity in real time, and effectively integrate the collected data.

[0023] S2: preprocessing the collected data; The raw data collected above is preprocessed; the preprocessing also includes data standardization, and in the preprocessing step, effective fusion of online monitoring data is added.

[0024] S3: Use principal component analysis method to reduce the dimension of preprocessed data; Principal component analysis was performed on the preprocessed data to extract the feature vector that best represents the original data set; after dimensionality reduction, the state assessment model was further optimized by combining the fused data.

[0025] S4: Based on the reduced-dimensional data, combined with the historical operation data and maintenance records of the equipment, a status assessment model is established; using the extracted feature vectors and the historical operation data of the equipment, the support vector machine model is trained to establish a substation equipment status assessment model; including data fusion and extraction of principal components, where the status assessment model is established based on edge computing.

[0026] S5: Evaluate the status of substation equipment using the status evaluation model and output the evaluation results; The real-time collected data is input into the trained model, and the model outputs the health status score, failure probability and maintenance recommendations of the equipment. The evaluation results include the health status, failure probability and maintenance recommendations of the equipment.

[0027] The present invention provides possible fault locations and cause analysis of equipment through model output, in addition to health status scores and fault probabilities.

[0028] Furthermore, with the collection and integration of data, data interconnection and interoperability can be achieved, which will help break down data silos and improve the efficiency of comprehensive data utilization. Provide intuitive and accurate data support to achieve real-time data processing and analysis. Comprehensively and accurately evaluate the status of equipment, issue early warning signals in a timely manner, and reduce the probability of failure. Automatically complete patrol tasks and generate reports to reduce workload and improve patrol quality.

[0029] The present invention can be deeply integrated with the information-physical platform, unify the data of physical equipment, equipment information, online monitoring data and equipment models, and adopt model data engine technology to build an information-physical platform driven by a model.

[0030] The data of online monitoring of transformers and switch cabinets in substations are integrated using computing terminals to achieve effective integration of model data and online monitoring data at the station end. On the one hand, it is used for daily inspection and maintenance at the station end, and on the other hand, it is sent to remote ends to provide support for remote command and monitoring.

[0031] By building a model of typical substation equipment (transformer, PT, CT), driven by the model, and integrating data with the substation edge computing terminal, the substation can be localized and responsive and rapid inspection can be achieved, and the model and data can be accurately unified, thereby improving the efficiency of daily inspection and management of substations.

[0032] The present invention mainly solves the problem that physical equipment, equipment information, and online monitoring data are described separately at present. Based on model drive, the equipment information, online monitoring data and other information are integrated with the equipment model by digital means, and the operating status of typical substation equipment is reproduced through the model. A cyber-physical platform based on model drive is built to provide intuitive data support for daily maintenance and inspection.

[0033] In addition, the present invention also solves the problem of inconsistent calculations of current online monitoring equipment. Based on edge computing fusion technology, the online monitoring data of multiple devices, multiple dimensions, and multiple standards are uniformly connected to the computing terminal and superimposed and integrated with the model data, providing support for unified data empowerment of digital applications.

[0034] The above are only specific application examples of the present invention and do not constitute any limitation to the protection scope of the present invention; any technical solutions formed by equivalent transformation or equivalent replacement shall fall within the protection scope of the present invention.

Claims

1. A substation three-dimensional model data fusion method based on edge computing, characterized in that: The following steps are involved: S1: Collect data of substation equipment; Install sensors on substation equipment to collect data on voltage, current, partial discharge signals, temperature and humidity of the equipment in real time; S2: preprocessing the collected data; Preprocessing the raw data collected above; S3: Use principal component analysis method to reduce the dimension of preprocessed data; Perform principal component analysis on the preprocessed data to extract the feature vector that best represents the original data set; S4: Establish a condition assessment model based on the reduced-dimensional data combined with the historical operation data and maintenance records of the equipment; The extracted feature vectors and the historical operation data of the equipment are used to train the support vector machine model and establish a substation equipment status assessment model; S5: Evaluate the status of substation equipment using the status evaluation model and output the evaluation results; The real-time collected data is input into the trained model, and the model outputs the equipment’s health status score, failure probability, and maintenance recommendations.

2. According to a method for fusion of substation three-dimensional model data based on edge computing according to claim 1, it is characterized in that: In step S1, the collected data is effectively fused.

3. According to a method for fusion of substation three-dimensional model data based on edge computing according to claim 1, it is characterized in that: In step S2, the preprocessing also includes data standardization processing, and in the preprocessing step, effective fusion of online monitoring data is added.

4. According to a method for fusion of substation three-dimensional model data based on edge computing according to claim 1, it is characterized in that: In step S3, the method further includes data fusion and extraction of principal components.

5. According to a method for fusion of substation three-dimensional model data based on edge computing according to claim 1, it is characterized in that: In step S3, after dimensionality reduction, the state assessment model is further optimized by combining the fusion data.

6. A substation three-dimensional model data fusion method based on edge computing according to claim 1, characterized in that: In step S4, the state assessment model is established based on edge computing.

7. The method for fusion of substation three-dimensional model data based on edge computing according to claim 1 is characterized in that: In step S5, the evaluation results include the health status, failure probability and maintenance recommendations of the equipment.

8. The method for fusion of substation three-dimensional model data based on edge computing according to claim 1 is characterized in that: The model output provides an analysis of possible equipment failure locations and causes in addition to health status scores and failure probabilities.