Transformer oil chromatographic data cleaning method based on machine learning and neural network

Abstract

The invention discloses a method for cleaning transformer oil chromatography data, which is used for cleaning transformer oil chromatography online data through a neural network machine learning method, and comprises the following steps of: S1, acquiring historical transformer oil chromatography online monitoring data and transformer oil chromatography offline monitoring data, and integrating and preprocessing the historical transformer oil chromatography online monitoring data and transformer oil chromatography offline monitoring data; s2, further processing the preprocessed data into a training set; and S3, extracting gas concentration data in the training set, inputting the gas concentration data into an LSTM neural network for training, and enabling the neural network to find a model capable of converting online monitoring data to be close to offline monitoring data. And S4, inputting to-be-cleaned transformer oil chromatography online monitoring data into the trained model to obtain cleaned data. According to the method, the online monitoring data of the transformer oil chromatography can be learned and cleaned through the neural network machine, so that the method is fast and convenient, and meanwhile, the accuracy is kept at a relatively high level.

Description

technical field [0001] The invention belongs to the technical field of online monitoring, and in particular relates to a method for cleaning chromatographic data of transformer oil. Background technique [0002] A transformer is a complex device that plays a very important role in power facilities such as power plants and substations. Due to its complex structure and long-term resistance to high voltage and electric power, transformers are likely to have various hidden dangers. Transformer oil chromatographic analysis is an important test method, which can detect latent faults such as moisture, local overheating, and low-energy discharge inside the transformer by analyzing various gas components in the transformer insulating oil and their changes. At present, there are mainly two methods of online monitoring and offline detection for chromatographic analysis of transformer oil. Offline detection has high accuracy, but requires cumbersome sampling and testing, and the real-...

AI Technical Summary

Problems solved by technology

Off-line detection has high accuracy, but requires cumbersome sampling and testing, and the real-time performance is poor; online monitoring and sampling are convenient, but due to the difficulty in calibration of monitoring equipment, the test data after long-term work is often not accurate enough and can only reflect a general trend

Images