研究目的
To present an AI-based standalone PV system sizing method using differential evolution multi-objective optimization to find the optimal balance between system’s reliability and cost.
研究成果
The AI-based sizing method using DEMO was found to be significantly faster than the numerical method and capable of producing practically optimal solutions. The study demonstrates the potential of AI techniques in overcoming the limitations of conventional sizing methods for standalone photovoltaic systems.
研究不足
The study is limited by the random nature of the AI algorithm, which resulted in the exact optimal solution in only 6 out of 12 runs. The nearly optimal solutions, however, did not introduce major departure from optimal system performance.
1：Experimental Design and Method Selection
The study uses differential evolution multi-objective optimization (DEMO) to minimize two objective functions: the loss of load probability (LLP) and the life cycle cost (LCC). A numerical algorithm serves as a benchmark for the proposed method’s speed and accuracy.
2：Sample Selection and Data Sources
Hourly meteorological data for the year of 2016 were obtained from a meteorological station in Baghdad. A hypothetical hourly power load profile was used for the study.
3：List of Experimental Equipment and Materials
PV Panel (KC120-1), Battery (SKT-GFM-500), Inverter, Charge Controller, Wire, Load.
4：Experimental Procedures and Operational Workflow
The methodology involves simulating the energy flow between the system components and the load, linking components’ models together, and simulating the operation of the entire system over a year with an hourly time step.
5：Data Analysis Methods
The performance of the proposed method is validated against the numerical method's result, comparing their execution times and the accuracy of the sizing results.
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
