| 摘 要: 粒子群优化算法(PSO)是一种群体协作优化算法,将其应用到能源管理系统的数据分析中,对于提高能源预测性能具有重要意义。文章采用轻量级梯度提升机(Light Gradient Boosting Machine,LightGBM)模型对能源数据进行详细分析,并通过PSO对LightGBM 模型进行优化,提升模型性能。通过对能源数据进行预测,并与多元线性回归(LR)、随机森林(RF)、LightGBM 和极限梯度提升(XGB)进行对比实验发现,经过PSO 优化后的LightGBM模型的表现更优异,预测效果显著优于其他模型。实验结果显示,决定系数(R2)高达91.56%,同时均方误差(MSE)明显降低,进一步验证了PSO-LightGBM模型的优越性。 |
| 关键词: 能源管理系统;数据分析;PSO;LightGBM |
|
中图分类号: TP31
文献标识码: A
|
|
| Data Analysis of Energy Management System Based on PSO-LightGBM |
|
WANG Lang, LIU Yongfei, XU Linzhang
|
(Guangdong Yue Gang Water Supply Co., Ltd., Shenzhen 518021, China)
wangl@163.com; liuyf@163.com; zhanglinxu@163.com
|
| Abstract: Particle Swarm Optimization(PSO) is a group collaborative optimization algorithm that is of great significance for improving energy prediction performance when applied to data analysis in energy management systems. In this paper, Light Gradient Boosting Machine (LightGBM) model is used to analyze the energy data in detail, and PSO is used to optimize the LightGBM model to improve the model performance. By predicting the energy data and comparing experiments with multiple Linear Regression (LR), Random Forest (RF), LightGBM and Extreme Gradient Boosting ( XGB), it is found that the PSO-optimized LightGBM model performs much better, and the prediction effect is significantly better than the other models. The experimental results show that the coefficient of determination (R2 ) is as high as 91.56% , while the mean square error (MSE) is significantly reduced, which further verifying the superiority of the PSO-LightGBM model. |
| Keywords: energy management system; data analysis; PSO; LightGBM |
