Intelligent transformer station fault diagnosing method under IEC61850 standard

Abstract

The invention discloses an intelligent transformer station fault diagnosing method under the IEC61850 standard. The implementation steps of the invention comprise: obtaining the static state information of the intelligent transformer station through the function configuration model for the intelligent transformer station; obtaining the dynamic information of the intelligent transformer station through the messages of the intelligent transformer station; inputting the obtained static state information and the dynamic information to a well-trained fault diagnosing model based on a deep-learning network; outputting the fault diagnosing result for the intelligent transformer station; detecting the fault diagnosing result; if the fault diagnosing result shows that faults occur to the intelligent transformer station, then, according to the fault type classification, carrying out fault wave recording; and outputting control information to restore the fault state of the intelligent transformer station. The method of the invention has a strong learning ability, can achieve high and accurate fault detection, has a good anti-noise performance and can detect fault at a high pace, therefore, making it especially suitable for solving non-linear problems and greatly advantageous in mode identification and classification.

Description

technical field [0001] The invention relates to a fault diagnosis and evaluation technology of an intelligent substation, in particular to an intelligent substation fault diagnosis method under the IEC61850 standard. Background technique [0002] The scale of the modern power system is getting bigger and bigger, the voltage level of the transmission line is getting higher and higher, the length is getting longer and longer, and the environment in which the transmission line is located is also more complicated, so the possibility and frequency of failures are also inevitably increased greatly. . Therefore, the research on power fault detection algorithm has very important theoretical significance and broad application prospects. In addition, in order to adapt to the development requirements of digital substations and future substation automation systems, implementing the IEC61850 communication standard in the fault diagnosis system can greatly improve the interoperability be...

AI Technical Summary

Problems solved by technology

There is no longer a one-to-one correspondence between the network physical topology and the input and output of functional information and signals. Traditional fault diagnosis and evaluation methods are mostly based on fault detection and analysis of secondary electrical circuits, which cannot be applied to smart substations. Substation fault detection and analysis are difficult to carry out

At the same time, most of the existing substation fault diagnosis methods only use the operating status and alarm information of the primary equipment in the substation itself, or use the circuit breaker action and fault reports to perform fault diagnosis, and give the probability of equipment failure.

The changes in the secondary system structure of smart substations and the increased automation and intelligence requirements for management and maintenance methods have made traditional fault diagnosis and evaluation methods unable to meet the needs of substation intelligent operation in terms of diagnosis depth and diagnosis methods.

At present, there are many fault diagnosis algorithms, but there are still two problems: 1) Although the common algorithm is simple in calculation, its robustness is poor and it is easily affected by noise; 2) The intelligent algorithm has poor adaptability and lacks self-learning ability

Images