基于支持向量机的主汽温趋势预测

Trend Forecast of Main Steam Temperature Based on SVM

主汽温是火电机组热力系统中的重要参数,其大迟延、大惯性、时变性等特性使得主汽温的控制难以达到理想效果,准确预测主汽温趋势对改善其控制效果具有重要意义。通过采集现场运行数据,利用灰色关联分析确定主汽温的主要影响因素,再利用支持向量机对主汽温进行回归预测,预测结果与实际对象有较高的相关度,对主汽温调节、参数优化及机组运行有指导意义。

Main steam temperature is an important parameter of the thermodynamic system in thermal power units, whose characteristics such as large delay, large inertia, time-varying make it difficult to get an ideal control effect. It is very significant to predict the trend of the main steam temperature accurately to improve its control effectiveness. In this paper, based on the data collected on site, the main factors affecting the main steam temperature are detected by grey relational analysis, and then the main steam temperature is predicted by the support vector machine（SVM）. The results show that the predicted results are highly correlated with the actual objects, possessing guiding significance for the main steam temperature regulation, parameter optimization and units operation.