由於太陽光電系統易受照度及溫度等環境因素影響而造成瞬間能量變化大，若大量併入市電系統，將影響電網的可靠度與穩定度。因此，本論文提出以非對稱歸屬函數之機率模糊類神經網路為架構之智慧型控制應用於磷酸鋰鐵電池儲能系統，目的為減緩太陽光電輸出至電網的功率波動。於此控制策略中，太陽光電實際輸出功率與平滑後輸出功率之間的差值將由電池儲能系統提供。論文中將詳細介紹非對稱歸屬函數之機率模糊類神經網路之架構與線上學習法則，並證明其收斂性。除此之外，在電池能量管理方面利用庫倫積分法實現電池電量狀態的估測，以避免電池過度充放電。根據再生能源導入電網實功率波動之規範，本研究所提之非對稱歸屬函數之機率模糊類神經網路明顯減緩太陽光電輸出功率波動，以提高電網之可靠度與穩定度。此外，與其他平滑控制方法相比，本論文透過非對稱歸屬函數之機率模糊類神經網路之控制實現既符合規範並且使所需電池容量最小化之目的。最後，利用模擬與實驗結果驗證所提功率平滑控制應用於電池儲能系統在不同照度變化情況下之成效。;As a major problem for integrating photovoltaic (PV) power plant to the grid, power fluctuations lead to poor power quality. A possible solution for regulating the intermittent output power of a PV power plant is to integrate a battery energy storage system (BESS). Therefore, an intelligent PV power smoothing control using probabilistic fuzzy neural network with asymmetric membership function (PFNN-AMF) is proposed to mitigate the fluctuation of PV output power directly fed to the grid. Moreover, the network structure of the PFNN-AMF and its online learning algorithms are described in detail. In addition, the state of charge (SOC) estimation using Coulomb counting method is adopted in the energy management of battery. According to the grid active power fluctuation limits set in this study, the proposed method is capable of mitigating the fluctuation of PV output power to improve reliability and stability of the grid. Furthermore, comparing to the other smoothing methods, a minimum energy capacity of the BESS with a small fluctuation of the grid power can be achieved by the PV power smoothing control using PFNN-AMF. Finally, the experimental results of various PV variation sceneries are realized to validate the effectiveness of the proposed intelligent PV power smoothing control.
