研究目的
Investigating the optimal control of energy flow in a home energy management system (HEMS) including the uncertainty introduced by electric vehicles (EVs).
研究成果
The stochastic optimization framework (SOFW) and the proposed stochastic optimization model can effectively manage the uncertainties of EV plug-in times and SoC, providing optimal control for ESS and EV charging within a HEMS. The preliminary results are promising, showing feasibility in meeting local energy management targets.
研究不足
The approach is limited to use scenarios where simple assumptions on daily EV use hold true. Higher-state resolution in dynamic programming and inclusion of more uncertainty factors are areas for future optimization.
1：Experimental Design and Method Selection:
The study proposes a stochastic optimization framework (SOFW) using dynamic programming to solve a stochastic optimization model. The framework is designed to manage energy storage systems (ESS) and EV charging stations within a HEMS, considering the uncertainties of EV plug-in times and state of charge (SoC).
2：Sample Selection and Data Sources:
The use case involves a residential house in Italy with photovoltaics (PV), ESS, and an EV. Data for EV plug-in times and SoC are modeled using a Markovian process and Monte-Carlo simulation based on historical data.
3：List of Experimental Equipment and Materials:
The scenario includes rooftop PV panels with 7.6 kW nominal electric power, an ESS with 9.6 kWh energy capacity and a 6.4 kW inverter, and an EV with 30 kWh capacity charged with a 3 kW EV Charging Supply Equipment.
4：6 kW nominal electric power, an ESS with 6 kWh energy capacity and a 4 kW inverter, and an EV with 30 kWh capacity charged with a 3 kW EV Charging Supply Equipment.
Experimental Procedures and Operational Workflow:
4. Experimental Procedures and Operational Workflow: The SOFW continuously generates optimal control decisions for ESS and EV charging based on real-time data and predictions. The optimization horizon is set to 24 hours with one-hour time resolution.
5：Data Analysis Methods:
The performance of the optimization approach is evaluated based on its ability to cope with EV plug-in time and SoC uncertainty, as well as local energy management indicators.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
