Device measurement data processing method, system and terminal based on deep neural network

Abstract

The invention discloses an equipment measurement data processing method, system and terminal based on a deep neural network, and relates to the technical field of power station equipment data processing. The key points of the technical solution are: a recognition model established based on a bidirectional long-short-term memory neural network and a conditional random field Entity recognition is performed on the measurement data of the target device, and a short text sequence represented by a character vector and a word vector after labeling is obtained; the short text sequence is expanded and input into a convolutional neural network, and the short text is obtained by learning the deep features in the short text According to the deep semantics of the short text, the clustering equipment measurement data is obtained after clustering processing according to the deep semantics of the short text; the mapping relationship between the historical equipment measurement data and the standard code is obtained through pre-built training model training, and the clustering equipment After the measurement data is input into the training model, combined with the mapping relationship, the new measurement data predicts the coding label. The invention can perform unified and standardized automatic encoding processing on different devices.

Description

technical field [0001] The present invention relates to the technical field of power station equipment data processing, more specifically, it relates to a deep neural network-based equipment measurement data processing method, system and terminal. Background technique [0002] The process of power station safety monitoring involves many different types of sensor equipment and operating equipment, and there are certain differences in the management of each power station, which makes data sharing difficult. At present, the definition of equipment measurement data in the core basic platforms of each station, such as monitoring systems and condition monitoring systems, only considers the implementation of their respective systems. At that time, there was no unified definition standard for equipment measurement data. The staff of each station are highly subjective to the equipment measurement data, resulting in simple text language expression, less vocabulary, and various descrip...

AI Technical Summary

Problems solved by technology

The staff of each station are highly subjective to the equipment measurement data, resulting in simple text language expression, less vocabulary, and various descriptions. Compared with PPIS data rules, the words are relatively irregular, resulting in different implementation standards and efforts. It is difficult to integrate the measurement data of equipment at each station

Images