Abstract: MSMA sensor is a new type of sensor that converts mechanical force signal into induced voltage signal. In order to better realize the application of MSMA sensor, it is necessary to model it. Due to the nonlinear relationship between the input and output of the sensor, it will increase the complexity of modeling, and there are many unknown parameters to be identified. The mathematical model of MSMA sensor is established by mechanism method combined with electromagnetic theory, and the unknown parameters of MSMA sensor mathematical model are identified by improved particle swarm optimization algorithm. In the identification process, the adjustment coefficient is introduced to dynamically adjust the weight and maximum speed of the particles, speed up the convergence of the algorithm, improve the social learning factor, and enhance the global search ability of the algorithm. Finally, the induced voltage calculated by the sensor mathematical model is compared with the induced voltage obtained by the test, which verifies the correctness of the model, and also shows that the algorithm can effectively solve the identification problem of the MSMA sensor mathematical model.