Title :
Intelligent power management (IPM) for transient recognition and control of PEM fuel cell / battery hybrid system
Author :
Karunarathne, Lakmal ; Economou, John T. ; Knowles, Kevin
Author_Institution :
Dept. of Eng. Syst. & Manage., Cranfield Univ., Swindon, UK
Abstract :
Fuel cell (FC)/ battery hybrid system power management is a decision making process which controls the power flow between each power sources. Intelligent power management (IPM) is an innovative power handling concept that can actively control the hybrid system power flow according to the load power changes and the battery state of charge variations. In addition to that, the IPM system controls the FC compressor air flow rate in order to supply the required oxygen concentration on demand. The Fuel Cell Current Limit (FCCL) controller decides the FC system operating current and the Adaptive Neuro Fuzzy Inference System (ANFIS) regulates the air flow rate by changing the compressor motor voltage. The FC system optimum compressor motor voltage which maximizes the FC system net power output is a function of the FC current. Therefore, the IPM system adapts the actual compressor motor voltage into the optimum compressor motor voltage. The ANFIS based controller back-propagates the Gaussian membership function estimation parameters such as mean (x-j i), variance (sigmaj i) and the fuzzy output (z) based on error generated by error minimization algorithm. The IPM system online adaptation process minimizes the oxygen concentration loss and maximizes the FC system net power output.
Keywords :
Gaussian processes; adaptive control; backpropagation; decision making; electric current control; fuzzy neural nets; neurocontrollers; power control; proton exchange membrane fuel cells; ANFIS; Gaussian membership function estimation parameter; PEM fuel cell; adaptive neuro fuzzy inference system; backpropagation; compressor air flow rate; compressor motor voltage; decision making process; error minimization algorithm; fuel cell current limit controller; fuel cell-battery hybrid system; intelligent power management; power flow control; Battery management systems; Control systems; Decision making; Energy management; Error correction; Fuel cells; Fuzzy control; Load flow; Power system management; Voltage; ANFIS; PEM Fuel cells; intelligent power management;
Conference_Titel :
Vehicle Power and Propulsion Conference, 2009. VPPC '09. IEEE
Conference_Location :
Dearborn, MI
Print_ISBN :
978-1-4244-2600-3
Electronic_ISBN :
978-1-4244-2601-0
DOI :
10.1109/VPPC.2009.5289738