摘要
在信息化数字化的大环境下,实现能源管理系统的在线监测办法无形中成为了企业能源管理最基础的措施。而系统中的能源需求预测功能为企业的节能降耗提供最有力的数据支撑。为提高预测准确率,本文提出了一种改进的多维灰色模型能源预测算法,先使用指数加权方式进行源特征数据处理增加趋势平滑度,再用自适应粒子群与多维灰色模型算法组合,使预测的精度在原始数据不完整的情况下达到最高,为能源信息化水平不足提供了新的解决思路,解决了企业的系统优化,能源调度和管理的问题。将提出的改进算法在实际的企业项目中运行,结果显示:优化后的自适应粒子群多维灰色模型预测算法正确率最高可达到93.573%,比传统的预测算法精确度提高约36%左右。
In the information and digital environment, the online monitoring method of energy management system has virtually become the most basic measure of enterprise energy management. The energy demand prediction function in the system provides the most powerful data support for energy conservation and consumption reduction of enterprises. In order to improve the prediction accuracy, this paper proposes an improved multidimensional grey model energy prediction algorithm. Firstly, the exponential weighting method is used to process the source characteristic data to increase the trend smoothness, and then the adaptive particle swarm optimization and multidimensional grey model algorithm are combined to maximize the prediction accuracy when the original data is incomplete, It provides a new solution for the lack of informatization of energy data, and solves the problems of enterprise system optimization, energy scheduling and management. The results show that the accuracy of the optimized adaptive particle swarm multidimensional grey model prediction algorithm can reach 93.573%, which is about 36% higher than that of the traditional prediction algorithm.
出处
《软件工程与应用》
2021年第4期576-583,共8页
Software Engineering and Applications
