Flame combustion stability identification method based on chaotic characteristic analysis

Abstract

A flame combustion stability identification method based on chaos characteristic analysis comprises the following steps: 1) selecting a flame video with class labels in a certain time period as a training sample, and selecting a flame video with stability to be tested in another time period as a test sample; extracting each frame of flame image of the training sample and the test sample, calculating an average gray value of the flame images, and forming a flame video illumination sequence; 2) sequentially performing high-dimensional phase space reconstruction, recurrence plot visualization, texture feature extraction and dimension reduction processing on the flame video illumination sequences of the training samples and the test samples, then selecting principal component features of the training samples and inputting the principal component features and class labels of the training samples into a classifier for training, and obtaining a trained classifier; and 3) inputting the principal component characteristics of the test sample into the trained classifier, and judging the stability of flame combustion in the test sample. According to the method, the dynamic characteristics of the flame illumination sequence are reflected by adopting the texture characteristics of the recurrence plot, so that the flame combustion stability is effectively judged.

Description

technical field [0001] The invention relates to the technical field of combustion stability detection, in particular to a flame combustion stability identification method based on chaotic characteristic analysis. Background technique [0002] Unsteady conditions during flame combustion can lead to uneven heat distribution, low efficiency and high pollutant emissions. Maintaining the stable combustion of the kiln can improve energy utilization, reduce waste gas and sewage emissions, and promote the sustainable development of the national economy. Therefore, timely and accurate detection of unstable combustion state is a problem that the industry attaches great importance to. [0003] At present, the physical measurement method is mainly used to detect the stability of the kiln flame temperature. Specifically, infrared sensors, sound pressure meters and thermocouples are used to detect the flame stability. The aforementioned components can only detect the temperature on a si...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem of inaccurate judgment of traditional flame stability, the present invention provides a flame combustion stability recognition method based on chaotic characteristic analysis

Images