摘要
针对需要对火电厂设备健康状态监测与预警的问题,首先建立动态综合评价模型,对单台设备的健康样本数据进行归一化和标准化,对各项指标特征进行动态加权,得到动态综合指标函数。再利用神经网络对每项指标进行分析,计算预测值与实时监测值的相似度,实现在线状态评估与诊断。因火电厂系统由多级设备构成,建立多层次综合评价模型,通过每一层级对上一层级的评价,对火电厂各层级实现在线评估与诊断。对于提前预警,利用神经网络进行预测,实时监视设备状态变化,给出报警信息,提高设备运行效率。
In view of the need to monitor and warn the health status of thermal power plant equipment, the dynamic comprehensive evaluation model is first established, the health sample data of single equipment is normalized and standardized, and the dynamic comprehensive index function is obtained by dynamically weighting the characteristics of each index. Then the neural network is used to analyze each index, calculate the similarity between the predicted value and the real-time monitoring value, and realize the online status assessment and diagnosis. Because the thermal power plant system consists of multi-level equipment, a multi-level comprehensive evaluation model is established. Through the evaluation of the previous level at each level, online evaluation and diagnosis of each level of thermal power plant is realized. For early warning, the use of neural networks for forecasting, real-time monitoring of equipment status changes, give alarm information, improve the efficiency of equipment operation.
作者
赵海颐
Zhao Haiyi(School of electrical and new energy engineering,China Three Gorges University,Yichang Hubei,443002)
出处
《电子测试》
2019年第15期67-68,75,共3页
Electronic Test
关键词
火电厂
动态综合评价
多层次综合评价
神经网络
thermal power plant
Dynamic integrated evaluation
Multilevel integrated evaluation
Neural network
