水电站计算机监控系统综合智能告警研究

故障诊断技术是保证水电机组及相关设备安全可靠运行的关键。由于电站集控中心接入监控对象设备量越来越多，故障自主诊断成为重要研究问题。本文针对水电站运行特点，提出了一种结合深度学习和规则推理的计算机监控系统综合智能告警方法。首先，介绍监控数据的前期处理流程，并以卷积神经网络（Convolutional Neural Network，CNN）为基础，构建基于深度学习的故障诊断模型；然后整合出几种较为宽泛的故障类型，利用历史样本训练 CNN 模型；最后结合基于专家经验的规则推理，完成对故障诊断结果的校核、细化及补充。算例结果表明，本文所提方法能够有效实现水电站故障自主诊断，为水电站智能化建设提供技术支撑。

Research on comprehensive intelligent alarm of computer monitoring system of hydropower station

Fault diagnosis technology is the key to ensuring the safe and reliable operation of hydropower units and related equipment. Aiming at the characteristics of hydropower station operation, this paper proposes a comprehensive intelligent warning method for computer monitoring system, combining deep learning and rule inference. First, the pre-processing process of monitoring data is introduced, and a fault diagnosis model is built, based on deep learning based on Convolutional Neural Network (CNN). Then, several broader types of faults are integrated and historical samples are used for training CNN model. Finally, by rule-based reasoning based on expert experience, the verification, refinement and supplement of fault diagnosis results are completed. The calculation example results show that the method proposed in this paper can effectively realize the autonomous diagnosis of hydropower station faults and provide technical support for the intelligent construction of power stations.