Research On Energy Optimal Scheduling Strategy Of Smart Home Equipments Based On Photovoltaic Power

In recent years,the development trend of smart home technology is to use renewable energy for home energy supply,and combine smart technology with green energy technology.At the same time,due to the development of smart grid,dynamic electricity price policy is gradually applied to residential users,which provides a new vision to improve the structure of residence energy consumption and save energy.The research is based on the dual power supply,the one power supply is municipal smart grid power and the other is photovoltaic power generation system,for this,research the smart home devices’ energy optimal scheduling strategy based on photovoltaic power generation.(1)Research on the energy consumption of smart home equipment.Determine the composition,equipment type and energy consumption form of intelligent home equipment system,analyze the dynamic price mechanism,select time-sharing price and real-time price as the price information according to the characteristics of residential power consumption of residents,and finally select 14 kinds of energy consumption equipment.According to the function characteristics of equipment and user requirements,the energy consumption model of various equipment is established,It lays a foundation for optimizing the scheduling of smart home equipment.(2)Research on photovoltaic power generation prediction model based on smart home,propose a VMD-HS-BP photovoltaic power prediction model.Firstly,the input variables with higher influence factors are selected by correlation analysis method,and then the photovoltaic power is decomposed by using the variational mode decomposition(VMD)technology to solve the problem of poor prediction accuracy caused by the nonlinear characteristics of the original data.Secondly,use harmony search algorithm(HS)to optimize BP neural network.Finally,use the improved network to predict the decomposition of photovoltaic power,and the final value is to sum the decomposition.The results show that,compared with the BP model and the HS-BP model,VMD-HS-BP model’s the RMSPE increases by 55.29% and 23.20% respectively,and the MAPE increases by 52.60% and22.76%;The prediction speed increased by 30.22% and 14.50%,the prediction accuracy and the speed of VMD-HS-BP model are obviously improved,which is great for photovoltaic power prediction.(3)Optimal scheduling strategy for energy consumption of smart home equipment based on dynamic electricity price.Firstly,establish a multi-objective optimal scheduling model,which takes the minimization of power cost,the maximization of renewable energy utilization and the maximization of user satisfaction as optimization objectives,and the physical characteristics and power balance of equipment as constraints.Then,use NSGA-II algorithm to solve the model based on the above multi-objective optimal dispatching model and photovoltaic power prediction.The simulation results show that NSGA-II can achieve good results in solving the day ahead optimal dispatching problem.Under the condition of time of use price and real-time price,the power cost after dispatching is reduced by 67.80% and 61.16% respectively,The utilization rate of renewable energy can reach 41.8% and 39.0%,at the same time,it can meet the user’s satisfaction with electricity consumption and realize the efficient and economic operation of smart home devices..This research is of great significance for the construction of smart home and the promotion and application of green energy.