基于集成深度随机森林算法的智能电厂设备健康评估方法

Health status assessment method of power plant equipment based on integrate deep random forest with AdaBoost

摘要:

准确地评估电厂设备健康状态,对电厂安全稳定生产、提高设备运行安全性具有十分重要的意义。针对当前电厂设备健康评估方法存在预测精度不高的问题,提出了一种基于集成深度随机森林算法的智能电厂设备健康评估方法。首先,详细介绍电厂设备健康评估系统结构,且分析了健康评估数据结构与影响因素；然后,将设备评估分为6类不同的层级,使得设备健康状态分析更方便；其次,结合深度学习与集成学习技术,提出了集成深度随机森林算法；最后,通过仿真实验分析验证了提出方法的有效性。结果表明,所提方法提高了评估模型的准确度。

Abstract:

Accurately assessing the health status of power plant equipment is of great significance to guaranteeing the safe and stable production of power plants and improving the safety of equipment operation. Aiming at the problem that the current power plant equipment health assessment method has low prediction accuracy, an intelligent power plant equipment health assessment method based on the integrated deep random forest algorithm is proposed. Firstly, the structure of the power plant equipment health assessment system is introduced in detail, and the health assessment data structure and factors influenced are analyzed. Secondly, the equipment evaluation is divided into six different levels, which makes the equipment health analysis more convenient. Then, combined with deep learning and ensemble learning technology, an integrated deep random forest algorithm is proposed. Finally, the effectiveness of the proposed method is verified by simulation experiments. The results show that the proposed method improves the accuracy of the evaluation model.