基于GA-BP的水力机组振动预测研究

振动是影响水力机组安全稳定运行的重要影响因素,为此基于GA-BP算法建立了机组振动预测模型,统筹水力机组运行过程中受机械、水力和电磁等因素的关联性影响,结合广东省某水电站~#1机组状态监测数据进行分析验证。结果表明,与传统BP算法相比,在对~#1机组主轴、上机架振动预测中,GA-BP算法训练迭代步数分别从26步减少至6步和63步减少至26步,预测平均相对误差分别从10.18%减小至4.62%和11.42%减小至4.92%,模型的预测性能获得显著提高,为保障机组安全运行提供了重要的技术支撑。

Research on Vibration Prediction of Hydraulic Unit Based on GA-BP

Vibration is an important factor affecting the safe and stable operation of hydraulic units. Based on the GA-BP algorithm, this paper establishes the vibration prediction model of the unit. Taking into account the influence of mechanical, hydraulic and electromagnetic factors during the operation of the unit, #1 unit state of a hydropower station in Guangdong Province is analyzes using the monitoring data. Compared with the traditional BP algorithm, the results show that the GA-BP algorithm is used to predict the vibration of the main shaft and upper frame of #1 unit, and the training iteration is reduced from 26 steps to 6 steps, and from 63 steps to 26 steps, respectively; Meanwhile, the average relative error of prediction is reduced from 10.18% to 4.62%, and from 11.42% to 4.92%, respectively. The prediction performance of the model is significantly improved, which can provide important technical support for the safe operation of the unit.