According to the least squares support vector machine (LS_SVM) parameter selection has important influence on the model of performance, and conventional parameter optimization methods' effect is poor and time-consuming, this paper present a least squares support vector machine prediction model which based on particle swarm algorithm. The model based on least squares support vector machine theories, and with particle swarm algorithm to optimize the model parameters. In this paper we use the model to predict three main performance indexes of evaluating coal gasification effect of the fixed bed (gas heating value, gasification efficiency, gas production rate), through the practical data's simulation results show that the algorithm can effectively improve the prediction accuracy of the model, and the model's reliability and usability has been verified.