Downsizing the battery capacity requirement of photovoltaic/hydrogen systems by adjusting the asymmetric time series using improved prediction — Based power management strategy

This paper aims at downsizing the necessary capacity of battery used in a system including photovoltaic (PV) source and hydrogen system which composes of fuel cell (FC) and water electrolyzer (EL) to supply to the load. The size of battery mainly depends on the power management strategy (PMS). A PMS based on the prediction of the difference between the intermittent PV power and the load using trend — AR model and Kalman filter (KF) was proposed. FC and EL, which have slow dynamic responses, will generate or consume this trend — prediction power while battery, with its fast transient characteristics, will cover the fluctuation error between prediction and actual power. Simulation results show that the proposed PMS can decrease the required battery capacity in comparison with no — prediction PMS. However, when the fluctuation distribution is asymmetric or when the efficiency of charging/discharging process of battery is taken into account, the evolution of the energy in battery may increase or decrease continuously, consequently requiring higher capacity or stopping the operation of battery when it is fully charged or deeply discharged. Therefore, the PMS was improved by combination of KF with the control of the state of charge (SOC) of battery to adjust the battery power in such a way that the power alternates between charge and discharge meanwhile the allowed maximum power changing rates of FC and EL are secured. By adjusting the symmetry of battery power distribution, the required capacity of battery could be smaller. Simulation results indicates that the proposed improved PMS could be applied to the systems for estimating the battery capacity and it shows better effectiveness in decreasing the required capacity of battery than the trend — prediction PMS.

[1]  Masumi Ishikawa,et al.  Prediction of time series by a structural learning of neural networks , 1996, Fuzzy Sets Syst..

[2]  Hui Liu,et al.  Comparison of two new ARIMA-ANN and ARIMA-Kalman hybrid methods for wind speed prediction , 2012 .

[3]  José L. Bernal-Agustín,et al.  Hourly energy management for grid-connected wind–hydrogen systems , 2008 .

[4]  Heather Cruickshank,et al.  The role for low carbon electrification technologies in poverty reduction and climate change strategies: A focus on renewable energy mini-grids with case studies in Nepal, Peru and Kenya , 2012 .

[5]  P. Seferlis,et al.  Power management strategies for a stand-alone power system using renewable energy sources and hydrogen storage , 2009 .

[6]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[7]  Jing Shi,et al.  On comparing three artificial neural networks for wind speed forecasting , 2010 .

[8]  Prabodh Bajpai,et al.  Hybrid renewable energy systems for power generation in stand-alone applications: A review , 2012 .

[9]  Masayoshi Ishida,et al.  Power Control Method Using Kalman Filter Prediction for Stable Operation of PV/FC/LiB Hybrid Power System Based on Experimental Dynamic Characteristics , 2015 .

[10]  Prodromos Daoutidis,et al.  Modeling and Control of a Renewable Hybrid Energy System With Hydrogen Storage , 2014, IEEE Transactions on Control Systems Technology.

[11]  Alagan Anpalagan,et al.  Optimization classification, algorithms and tools for renewable energy: A review , 2014 .

[12]  Takataro Hamajima,et al.  Optimization of SMES Compensation Capacity for Stochastic Power Using a Kalman Filter , 2010 .