Bad data identification method based on coupling of proximity analysis and neural network prediction

Abstract

The invention provides a bad data detection method based on coupling of proximity analysis and neural network prediction. The method mainly comprises the steps of obtaining original data, conducting dimensionless processing on the original data, calculating the average distance between a sample and a fixed number of adjacent samples, and recording the samples with the distance values smaller than a set value as bad samples. And using the pre-screened normal sample data to learn and train to establish a BP neural network model, performing loop test on the bad samples through the neural network model, and finally obtaining a bad data set. According to the invention, by coupling the proximity analysis and the neural network model, the bad data is preliminarily judged by calculating the average distance between the sample and the adjacent sample, then the bad sample is predicted by using the neural network, and the bad sample is re-identified according to the prediction result. The method is very suitable for processing denitration system data with the characteristics of multiple parameters, large fluctuation and the like, is high in identification precision and accuracy, and can be widely applied to identification of bad data in the field of flue gas denitration.

Description

technical field [0001] The invention belongs to the field of nitrogen oxide treatment, in particular to a bad data identification method based on proximity analysis coupled neural network prediction. Background technique [0002] With the development of technologies such as the Internet of Things, machine learning, and big data analysis, researches on intelligent regulation and catalyst management of denitration systems based on big data analysis of SCR denitration systems are increasing. However, due to measurement errors, equipment failures, transmission failures and other problems in the big data collection process of the SCR denitration system, the original samples often contain some bad data. These bad data not only interfere with the control of the denitration system, but also have many adverse effects on the intelligent regulation and catalyst management based on big data analysis, limiting the popularization and application of these technologies. Therefore, how to i...

AI Technical Summary

Problems solved by technology

[0003] The traditional bad data identification methods mainly include physical discrimination method and mathematical statistics method. The physical discrimination method is based on people's objective understanding of the known data, and judges the method that the measured data deviates from the normal value due to external interference and human error. However, due to the SCR denitrification The large amount of data generated by the system, the heavy workload of manual judgment and the need for operators with rich experience make this method difficult to implement

Mathematical statistics method is to identify bad data through mathematical statistical theory. However, due to the late development of SCR denitrification technology, there are few studies on mathematical statistics of SCR denitrification system data at present, and there is a lack of guidance for suitable mathematical statistical models. sex is difficult to guarantee

Images