摘要
针对解决高频特征的非侵入式负荷监测方法硬件价值高,不利于推广,而低频特征难以识别多电器的同时投切,且对相似功率特征的电器不易识别的问题,本文提出了一种基于电器功耗类型特征的负荷分解方法,利用能够标识电器的功耗类型特征,建立负荷分解改进模型,使用遗传算法进行求解,进而识别用电器的运行状态,获取其运行功率。实验表明,该方法能有效识别用电器的运行状态及功耗模式,且能处理多用电器具有相似功率特征或同时投切的情况;通过性能对比,证明提出模型较传统PQ优化模型在负荷分解性能提高的同时,节省了算法计算时间;对不同采样间隔的数据负荷分解结果证明了该模型对数据采样频率的鲁棒性。该模型适用于处理智能电表获取的数据,适用于大规模推广。
Currently non-intrusive load monitoring technology with features extracted from high frequently sampling data has high hardware cost and is not suitable for large-scale application.However,the low-frequency features are difficult to identify when multiple electrical appliances are switched on at the same time,and electrical appliances with similar power features are not easily recognized.To solve these problems,this paper proposes a load disaggregation model based on features of power consumption patterns.By establishing a new mathematical optimization model for load decomposition and calculating it using genetic algorithm,operating states of the appliance are identified.Experiments show that the model can effectively identify the operating states and consumption patterns of the electrical appliances,and can deal with the situation that multiple appliances have similar power characteristics or simultaneous switching.Through the performance comparison,it is proved that the proposed model is more efficient than the traditional PQ optimization model in load decomposition performance.At the same time,the calculation time of the algorithm is saved,and load decomposition results for different sampling intervals prove the robustness of the model to the data sampling frequency.The model is suitable for processing data acquired by smart meters without extra measurement hardware installed.
作者
王慧娟
邢艺兰
杨鹏
李想
Wang Huijuan;Xing Yilan;Yang Peng;Li Xiang(School of Computer Science and Engineering,North China Institute of Aerospace Engineering,Langfang 065000,Chi-na)
出处
《北华航天工业学院学报》
CAS
2021年第5期5-8,47,共5页
Journal of North China Institute of Aerospace Engineering
基金
河北省高等学校科学科学技术研究项目重点课题(ZD2020161)
北华航天工业学院博士启动基金课题(BKY-2020-42)。
关键词
非侵入式
负荷分解
功耗类型
数学优化
Non-intrusive
load decomposition
power consumption type
mathematical optimization
