Residual error posterior-based abnormal value online detection and confidence degree assessment method

Abstract

The invention discloses a residual error posterior-based abnormal value online detection and confidence degree assessment method. The method comprises the steps of collecting data, establishing time series data, performing linear fitting on the time series data to obtain a linear combination formula of data at a current moment and p pieces of previous data, and predicting a data value of subsequent time; comparing the predicted data value with an actually detected data value to obtain a predicted residual error series; determining a probability density function of the predicted residual error series by adopting a KDE (Kernel Density Estimation) method; performing posterior ratio check on the predicted residual error series, and judging whether the data at the current moment is an abnormal point or not; and by taking the time series data as an input, building an SOM state model, obtaining state series and state transition probability matrixes, defining an abnormal scoring function, and outputting an abnormal score. By comparing the probability that the data is the abnormal point with the probability that the data is a normal point, the abnormal value in the pollutant discharge concentration time series data is identified online, so that the accuracy and reliability of abnormal value judgment are improved.

Description

technical field [0001] The invention relates to the field of data monitoring of pollutant emission concentration of coal-fired units, in particular to an online detection and confidence evaluation method for abnormal values ​​based on residual error posteriori. Background technique [0002] In order to effectively reduce the pollutant emission indicators of coal-fired units, my country has proposed relevant policies in recent years that the pollutant emissions of coal-fired units meet the emission indicators of gas-fired units, that is, ultra-low emissions of coal-fired units. Coal-fired units in Jiangsu Province began to vigorously carry out ultra-low emission transformation in 2014. At present, about 80% of 135MW and above units have completed ultra-low emission transformation, and realized online monitoring of pollutant-related parameters, including load conditions Parameters and chimney outlet SO 2 , NOx, soot concentration, oxygen content, temperature, humidity, pressu...

AI Technical Summary

Problems solved by technology

[0006] Most of the early outlier detection methods were based on statistical principles. It was necessary to assume the data distribution model of the sample in advance, and then use the method of hypothesis testing to judge whether the data is abnormal. However, the prior knowledge of the sample is usually lacking, and the premise of the hypothesis test is not established. , which greatly reduces the reliability of the detection method

Images