摘要
主蒸汽流量的准确测量对大型汽轮机的正常运行至关重要,传统测量方法工作量大且测量精度不高。为了提高主蒸汽流量的测量精度,以某600MW大型汽轮机为研究对象,提出一种基于长短期记忆神经网络(LSTM)的汽轮机主蒸汽流量测量模型,同时分析了调节级后压力的非正常波动对模型的影响。研究结果表明:LSTM模型能够实现主蒸汽流量的精确测量,平均百分比绝对误差(MAPE)为0.799%、均方根误差(RMSE)为15.132;调节级后压力在10%和1.2MPa范围内波动时,MAPE不大于1%、RMSE不大于20。LSTM模型具有较高的测量精度与较好的稳定性,研究结果对汽轮机主蒸汽流量测量具有一定参考价值。
The accurate measurement of main steam flow is very important for the normal operation of large steam turbine.The traditional measurement method has a large workload and low measurement accuracy.In view of the above problems,a main steam flow measurement model of a 600MW large steam turbine based on LSTM is proposed.Meanwhile,the influence of abnormal pressure fluctuation after the regulating stage on the model is analyzed.The results show that the LSTM model can accurately measure the main steam flow.The MAPE is 0.799%and the RMSE is 15.132;When the pressure after the regulating stage fluctuates in the range of 10%and 1.2MPa,MAPE is not more than 1%,RMSE is not more than 20.The model has high measurement accuracy and good stability,and the research results have a certain reference value for the main steam flow measurement of steam turbine.
作者
李浩
李路江
宁大鑫
韩旭
LI Hao;LI Lu-jiang;NING Da-xin;HAN Xu(State Grid Hebei Energy Technology Service Co.,Ltd.,Hebei Shijiazhuang 050000,China;Hebei Province Key Laboratory of Low-Carbon High-Efficiency Power Generation Technology,North China Electric Power University,Hebei Baoding 071003,China)
出处
《机械设计与制造》
北大核心
2023年第11期35-39,共5页
Machinery Design & Manufacture
基金
国网河北能源技术服务有限公司科技项目(TSS2020-18)
河北省自然科学基金项目(E2020502001)。
关键词
汽轮机
主蒸汽流量
LSTM
测量精度
偏差
Steam Turbine
Main Steam Flow Rate
LSTM
Measurement Accuracy
Deviation
