Research Into Power Transformer Condition Assessment And Fault Warning Method Based On Edge And Cloud Collaboration

Building a cloud-side collaborative ubiquitous power Internet of Things is a strategic deployment of the State Grid Corporation of China.Power transformers are important equipment in power transmission and transformation and power supply and distribution systems,and their operating status has an important impact on the safe and stable operation of the power grid.Under the new situation,in the face of the vigorous development of the ubiquitous power Internet of Things,under the new cloud-side collaboration system,it is necessary to conduct research on im proving the level of equipment state awareness and optimizing transformer configuration management methods.This article summarizes the typical failure mechanism of large power transformers,affected by different types of stress,and analyzes the transform er in terms of insulation performance,discharge performance,overheating performance and winding mechanical performance,providing theoretical support for transformer evaluation and early warning;428 sets of fault cases were constructed,a transformer multi-dimensional data warehouse was constructed,and the case statistics and analysis were carried out to provide data support for transformer evaluation and early warning.This article analyzes and discusses various uncertainties in the process of traditional transformer evaluation.A multi-level health assessment method for power transformers is proposed from the three aspects of index criterion uncertainty,weight uncertainty and fusion uncertainty.First,select indicators based on transformer guidelines and engineering experience,and establish a multi-level model that can reflect the types and locations of transformer faults.Then,the Gaussian cloud model is used to solve the uncertainty of the index criterion boundary.Based on association rules,analytic hierarchy process and variable weights,comprehensive weights are calculated in the self-updating module,which overcomes the uncertainty of weights.The improved DSm T theory is applied to the fusion of multiple evidences,which solves the problems of high conflict and uncertainty in the fusion process.A transformer health management platform was built to realize the evaluation,early warning and real-time management of transformers.Finally,through a single case and the overall evaluation of the database,the fault type,fault location and overall state of the transformer are obtained.The evaluation accuracy rate reaches 94.82%,which verifies the practicability and effectiveness of the method.Based on the evaluation of traditional transformers,th is article is based on the ubiquitous power Internet of Things strategy and the equipment-side ubiquitous power Internet of Things top-level design formulated by the State Grid Equipment Department,and builds an online monitoring and intelligent analysis model of transformers based on the power transmission and transformation Internet of Things architecture to achieve The evaluation and early warning are intelligent and combined with the cloud platform to form a comprehensive evaluation model to form an efficient cloud-side collaboration system.It has certain guiding significance for the operation management and maintenance decision-making of the transformer under the new mode.