摘要
在火电厂燃煤锅炉运行过程中,受热面的积灰是降低锅炉运行效率和安全性的一个重要原因。对此,各研究者根据建立固定的预测模型制定吹灰策略,认为是及时吹灰而忽略了吹灰需要一定的准备时间。针对上述问题,不仅在数据预处理时,采用时间序列随机选取的方法反映不同工况下的灰污沉积厚度,而且提出了基于清洁因子(CF)的Elman时序神经网络动态预测模型。为了构造合理的网络模型,采用试凑法得到最优隐含层节点数。最终,通过正态概率密度(PDF)曲线验证了该模型预测的精准性,并对比分析了在不同预测起始点滚动预测的结果。经实例仿真,结果显示Elman网络模型的预测结果与实际监测数据的吻合度较高。从而为下一步的吹灰优化研究奠定了坚实的基础。
During the operation of coal-fired boilers in thermal power plants, the accumulation of ash on the heated surface is an important cause to reduce the efficiency and safety of the boiler. In this regard, each researcher developed a soot blowing strategy based on the establishment of a fixed forecasting model, believing that soot blowing in time and ignoring the need for certain preparation time soot blowing. In order to solve the above problems, this paper not only uses the method of random selection of time series to reflect the thickness of ash deposition under different conditions, but also proposes a dynamic prediction model of Elman time series neural network based on cleanliness factor ( CF). In order to construct a reasonable network model, the optimal number of hidden layer nodes is obtained by trial and error method. Finally, the prediction accuracy of the model is verified by the normal probability density curve (PDF), and the results of rolling prediction at different starting points are compared and analyzed. The simulation results show that the prediction results of Elman network model are in good agreement with the actual monitoring data. It lays a solid foundation for further research on soot blowing optimization.
作者
贾志琴
史元浩
梁建宇
李登耀
Jia Zhiqin;Shi Yuanhao;Liang Jianyu;Li Dengyao(School of Electrical and Control Engineering,North University of China,Taiyuan 030000,China;Datang Baoding Thermal Power Plant,Baoding 071052,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2019年第9期50-56,共7页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61533013)
山西省重点研发计划(NO.201703D111011)
山西省自然科学基金(201801D121159)
山西省青年自然科学基金(201801D221208)
中北大学校基金(2016032,2017025)资助项目
