Operation State Monitor And Current Sensor Fault Tolerant Control Of DFIG

Wind energy is still a great potential renewable energy. After a period of rapidgrowth, the global wind power industry has entered a transition period of steadygrowth. After a period of explosive growth, China’s wind power industry growthrate gradually slowed down and stabilized. At the same time, some new growthareas of wind power appeared. Offshore wind power development will be a newgrowth point, at wind power industry. Europe, especially Britain start early angrown fast in the field of offshore wind power. China has only just begun in thefield of offshore wind power. Thus, there is a huge development potential at field ofoffshore wind power in China. At present, China’s wind power industrydevelopment patterns changes from the scale growth gradually to high-quality,high-efficiency technology growth. Therefore, wind power control system researchhas been put forward higher requirements. Facing the lack of core technology ofwind power control system, targeting to DFIG wind power generation system,thispaper study on three aspects, namely the rotor current control strategy, onlineidentification of motor parameters and the stator current sensor fault tolerantcontrol.DFIG has been the mainstream type of wind power industry. It is also the mostmature wind power generator. DFIG has always been the focus and hot pots ofresearch. DFIG rotor current is the controller’s direct control object. Its controlperformance directly determines the performance of the system. The basic DFIGdynamic mathematical model has been established and analyzed. In order toimproving the performance of the traditional deadbeat control in DFIG dynamic andsteady state, an improved deadbeat control strategy has been proposed in this paper.Different improved deadbeat control strategy has been proposed, before and afterconnecting to the grid, because of structural mutations and different requirements.Online identification of motor parameters has great significance for long-termstable operation of wind power generation system. Model reference adaptive control(MRAC) theory was applied to DFIG parameters online identification. A stable andconvergent adaptive algorithm has been designed based on DFIG dynamic modeland Popov super-stability theories. According to coupling problems of Doubly-fedmotor parameters, Gradually-update method was applied in model referenceadaptive parameter identification. Online identification of DFIG’s parameters hasbeen achieved in Matlab/simulink simulation and physical experiments.Stator current observer (SCO) and rotor current observer (RCO) based onMRAS has been established in this paper. Taking advantage of input error of observer, the stator current sensor failure can be detected immediately.Common-mode mutation of input error caused by soft-failure can be eliminated bysubtracting input error of SCO and RCO. That can prevent incorrect operation ofsensor fault detection. The current sensor’s failure measurements can be replaced byobservations of SCO. The sensor fault tolerant control can be accomplished. Bymeans of analyzing observations and measurements with or without failure, the faultsensor position can be accurately located.