摘要
以某电厂600 MW机组直接空冷凝汽器为对象,使用BP神经网络建模方法,以风机功率、环境温度和负荷为输入,凝汽器背压作为输出建立直接空冷凝汽器背压预测模型。通过对预测数据和实际数据进行对比,其误差基本满足现场凝汽器背压的控制精度。最后举例说明了预测模型在提供凝汽器背压参考值和最好经济背压的风机功率优化这两方面的应用,具有较好的工程实用性。
Taking the 600 MW unit direct air-cooling condenser in a certain factory as the object, adopting the modeling method of BP neural network, taking the blower power, environment temperature and load as inputs, and the value of the condenser back-pressure as output, the direct air-cooling condenser back-pressure prediction model is built. According to the comparison of the actual data and the predicted data, the error meets the control accuracy of field condenser back-pressure. Finally an example is given to illustrate the application of the prediction model in two aspects ： providing condenser back-pressure reference value and optimizing the best economical back-pressure of the blower power. The resutls also show that the proposed method has good practical value.
出处
《电力科学与工程》
2014年第5期67-70,共4页
Electric Power Science and Engineering
关键词
直接空冷凝汽器
背压
BP神经网络
预测
direct air-cooling condenser
back-pressure
BP neural network
prediction
