The current-limiting effect by turning off DC excitation and the speed of turning off for superconductor fault current limiter (SCFCL) are discussed. The necessity of rapid recognition of fault current is presented. Proposed SCFCL and its real-time data acquisition and control systems based on Labview and board of NI are designed and fabricated. Three important characteristics such as the value of current, changing rate of current and the lasting time of keeping high changing rate of current are extracted. According to the characteristics, off-line perceptron training in Matlab is done to find the classification surface. Also the neural cells and the method of fuzzification of the threshold current value are synthesized in FPGA of NI board with the trained weights and biases to increase the speed and veracity of pattern recognition. The experimental comparative results between complex decision method and threshold method indicates the former is more timely and veracious.