论文部分内容阅读
能量解析在分解综合负荷及提高设备的能量效率方面起到重要作用。当前,能量解析方法主要存在较低准确性和效率问题。论文提出一种基于低频监控数据的多输出极限学习的能源解析方法。该方法的特征映射函数可一次随机生成且无需调整其参数,与支持向量机方法相比,其优化目标函数具有较少的优化约束条件且更易实现。用实际记录的房屋能量数据进行仿真,仿真结果表明:与支持向量机相比,本文方法具有更高的训练速度和分类精度、更少的计算时间和更强的泛化能力。
Energy analysis plays an important role in breaking down the overall load and improving the energy efficiency of the plant. At present, the energy analysis method mainly has the problem of low accuracy and efficiency. This paper presents a method for energy analysis of multi-output limit learning based on low-frequency monitoring data. Compared with the SVM method, the proposed method has fewer optimization constraints and is easier to implement than the SVM method. The simulation results show that compared with SVM, the proposed method has higher training speed and classification accuracy, less computation time and more extensive generalization ability.