研究目的
To develop a multiobjective voltage unbalance factor (VUFMO) for estimating single-phase PV hosting capacity in 3-phase residential networks, incorporating time-varying and probabilistic behaviors of loads and PV generation.
研究成果
The developed VUFMO effectively captures time-varying and probabilistic behaviors of loads and PV generation, providing a better estimate of voltage unbalance than deterministic methods. It allows customization through weight assignment and shows that zero-sequence VUF can be significant in 3-phase 4-wire networks, especially at higher unbalance levels.
研究不足
The study assumes independence between load and solar irradiance states, uses specific test feeders which may not represent all real-world networks, and relies on historical data that might not capture all uncertainties. The ZIP model coefficients are adopted from literature without customization.
1：Experimental Design and Method Selection:
The study uses a probabilistic approach with a time-varying ZIP load model and multi-state probabilistic power flow analysis. It involves Monte Carlo simulation for random allocation of PV units and simulation on test feeders using OpenDSS in Matlab.
2：Sample Selection and Data Sources:
Historical residential load data from [18] and solar irradiance data from [19] are used to generate probability density functions for loads and PV generation over four seasons and 24 hour-segments.
3：List of Experimental Equipment and Materials:
Modified IEEE 4-bus test feeder and IEEE European low voltage test feeder are used as simulation platforms. Software includes Matlab and OpenDSS for power flow analysis.
4：Experimental Procedures and Operational Workflow:
The feeder is modified, loads and PV units are modeled probabilistically, PV penetration is varied from 10% to 60%, and voltage unbalance factors are calculated using multi-state power flow for each combined state.
5：Data Analysis Methods:
Expected values of voltage unbalance factors are computed by summing over states weighted by probabilities. Results are compared with deterministic worst-case methods.
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
