Efficient utilization of a residential photovoltaic (PV) array with grid connection is difficult due to power fluctuation and geographical dispersion. Reliable energy management and control system are required for overcoming these obstacles. This study provides a new residential energy management system (REMS) based on the convolution neural network (CNN) including PV array environment. The CNN is used in the estimation of the nonlinear relationship between the residence PV array power and meteorological datasets. REMS has three main stages for the energy management such as forecasting, scheduling, and real functioning. A short term forecasting strategy has been performed in the forecasting stage based on the PV power and the residential load. A coordinated scheduling has been utilized for minimizing the functioning cost. A real-time predictive strategy has been used in the actual functioning stage to minimize the difference between the actual and scheduled power consumption of the building. The proposed approach has been evaluated based on real-time power and meteorological data sets.

由于功率波动和地理位置分散，很难有效利用带有电网连接的住宅光伏（PV）阵列。需要可靠的能源管理和控制系统来克服这些障碍。这项研究提供了一种基于卷积神经网络（CNN）的新型住宅能源管理系统（REMS），其中包括光伏阵列环境。CNN用于估算驻地光伏阵列功率与气象数据集之间的非线性关系。REMS具有三个主要阶段的能源管理，例如预测，计划和实际运行。在预测阶段已根据光伏发电量和住宅负荷执行了短期预测策略。已利用协调调度来最小化运行成本。在实际运行阶段已使用了实时预测策略，以最大程度地减少建筑物的实际能耗与计划能耗之间的差异。已经基于实时功率和气象数据集对提出的方法进行了评估。