Due to the different forms, variable characteristics and various types of power loads on DSM, there are some problems in load identification using traditional methods, such as low recognition rate, difficulty in model building and difficulty in generalization. In this paper, a load identification method based on multi-parameter Hidden Markov Model is proposed, which is based on the intelligent load controller and NILM. Four load characteristic parameters are used as observation vectors of the model. Through model learning and iteration calculation, the maximum output probability and optimal state sequence of the observation sequence matching the hidden state of MPHMM model are obtained. Then the results are corrected by auxiliary discriminant algorithm to complete the final load identification. An experimental platform is built to verify the proposed method. The results show that the identification accuracy can reach more than 95% and it has good recognition effect for low power load especially. This work is supported by National Natural Science Foundation of China (No. 61403284), Henan Open Laboratory Project (No. KG2016-7), Key Scientific Research Project of Colleges and Universities in Henan (No. 18A470014), and Doctoral Fund of Henan Polytechnic University (No. B2017-20).