To explore the response of yield, crop growth as well as water and N use efficiency to different fertigation regimes, three years field experiment were carried out. The factors in the experiment comprised five N levels providing 0, 60, 100, 140 and 180 kg N ha(-1), and one real time model (Daisy model)-based N fertilization method, as well as three irrigation levels. Results showed that prolonged N fertigation increased yield consistently but N fertigation should be carried out early so that shoot growth is optimized and prolonged N have sufficient time to take effect on tuber yield accumulation. In all three years, the highest fresh yield and marketable yield was obtained when 180 kg N ha was applied. Compared to the 180 kg N ha(-1) treatment, realtime model-based N fertilization resulted in comparable yield (not different) in all seasons. Nevertheless, model based N fertilization resulted in yield that was significantly lower than the predicted maximum yield across the seasons, implying the necessity to adjust the parameters related to critical N concentration of Daisy model. The economic optimal were 142, 194 and 195 kg N ha(-1) in 2013, 2014 and 2015, respectively. Deficit irrigation did not cause reduced soil water content, ratio vegetation index (RVI) and yield, but irrigation efficiency was increased significantly in 2013.