摘要
当前,大多数企业只关注电力设备的设计寿命,在电力设备使用后期,运维成本可能会高于电力设备的效益,从而造成资金的浪费。针对该问题,本文基于全生命周期成本(LCC)对电力设备的环境状态进行了深入分析,并评估了其综合性能。首先,选取了某大型企业的一台35 kV电压互感器的数据,使用主成分分析法(PCA)来降低数据维数。然后,使用K-Means方法来筛选数据的异常值并进行数据的聚类。接着,提出了一种LCC-PCA-KMeans-DQN模型,并使用DQN算法对该电压互感器进行了仿真实验与分析。结果表明,本文建立的基于全生命周期的LCC-PCA-KMeans-DQN模型能够科学计算出电压互感器的使用寿命范围,并通过对该设备进行使用环境分析和综合性能评估,弥补了传统方法的不足,有效降低了企业的运维成本。
Currently,most enterprises only focus on the design life of power equipment.In the later stage of power equipment use,the operation and maintenance costs may be higher than the benefits of power equipment,resulting in a waste of funds.In response to this issue,this article conducts an in-depth analysis of the environmental status of power equipment based on the life cycle cost(LcC)and evaluates its comprehensive performance.Firstly,data from a 35 kV voltage transformer of a large enterprise was selected and principal component analysis(PCA)was used to reduce the dimensionality of the data.Then,use the K-Means method to screen for outliers in the data and perform data clustering.Subsequently,an LCC-PCA KMeans DQN model was proposed,and the DQN algorithm was used to simulate and analyze the voltage transformer.The results show that the LCC-PCA KMeans DQN model based on the entire life cycle established in this article can scientifically calculate the service life range of voltage transformers,and through the analysis of the equipment s usage environment and comprehensive performance evaluation,it can make up for the shortcomings of traditional methods and effectively reduce the operation and maintenance costs of enterprises.
作者
张中印
袁洪跃
ZHANG Zhong-yin;YUAN Hong-yue(Beijing Brelai Intelligent Technology Group Co.,Ltd.,Beijing 100018;School of Electrical Engineering,Henan Mechanical and Electrical Vocational College,Zhengzhou 451192)
出处
《环境技术》
2024年第9期226-233,241,共9页
Environmental Technology
关键词
全生命周期
PCA降维
K-MEANS聚类
深度强化学习
电力设备环境分析
综合性能评估
full lifecycle
PCA dimensionality reduction
K-Means clustering
deep reinforcement learning
environmental analysis of power equipment
comprehensive performance evaluation