大型炼钢炉炉温异常检测方法研究

发布时间：2018-06-22 17:21

本文选题：炼钢炉 + 炉温　；参考：《计算机仿真》2014年12期

【摘要】：在炼钢炉内高温高压环境下,异常温度信号往往瞬间形成,在炉内气浪、异常压力鼓动等外部干扰下,温度异常信号会迅速发生非正常衰减,甚至与其它异常信号发生混淆,造成温度测量不准。传统的炉内温度异常检测方法在信号发生不可控衰减或混淆的情况下,信号分离和识别的过程存在较大困难,在外部干扰下,对温度信号的单独识别、提取效果不佳,温度检测存在较大弊端。提出一种采用约束模糊聚类算法的大型炼钢炉炉温异常信号检测方法。根据傅里叶变换方法对信号采集设备采集到的信号进行隔离处理,克服由于信号混淆带来的困难。根据约束模糊聚类算法,对隔离后的信号进行聚类,使其成为有效的识别特征,完成了异常温度的检测。实验结果表明,利用改进算法进行大型炼钢炉炉温异常检测,能够极大提高异常温度信号检测的准确率。
[Abstract]:Under the environment of high temperature and high pressure in steelmaking furnace, abnormal temperature signals often form instantly. Under external interference such as gas wave, abnormal pressure agitation and so on, abnormal temperature signals will rapidly decay and even be confused with other abnormal signals. The temperature measurement is inaccurate. Under the condition of uncontrolled attenuation or confusion of signals, the process of signal separation and recognition is difficult. Under the external disturbance, the temperature signal can be identified separately, so the extraction effect is not good. There are some disadvantages in temperature measurement. A method for detecting abnormal temperature signal of large steelmaking furnace using constrained fuzzy clustering algorithm is proposed. According to the Fourier transform method, the signals collected by signal acquisition equipment are isolated and processed to overcome the difficulties caused by signal confusion. According to the constrained fuzzy clustering algorithm, the isolated signals are clustered to make it an effective recognition feature, and the abnormal temperature detection is completed. The experimental results show that the detection accuracy of abnormal temperature signal can be greatly improved by using the improved algorithm to detect the abnormal temperature of large steelmaking furnace.
【作者单位】：辽宁科技大学软件学院;东北大学信息科学与工程学院;
【分类号】：TN911.6;TF748;TK311

【参考文献】