Abstract: The coal consumption has time-varying and nonlinear characteristics. In order to improve the prediction accuracy of coal consumption, a coal consumption prediction model based on Catfish Particle Swarm algorithm and Least Squares Support Vector Machine（CEPSO-LSSVM） is proposed. LSSVM parameter is encoded into the position of the particle, and minimum of the cross validation error of network training set is taken as optimal target, and then the parameters of LSSVM are obtained by the exchange information among particles, and “catfish effect” is introduced to keep the diversity of particle swarm to overcome the local optimum of the traditional particle swarm optimization algorithm, and coal consumption prediction model is built according to the optimum parameters, and the simulation test is carried out on actual coal consumption data. The results show that, compared with other prediction models, the proposed model can get better parameters, and coal consumption prediction accuracy can be improved. It is more suitable for complex coal consumption prediction.

Key words: coal consumption, Least Squares Support Vector Machine（LSSVM）, Particle Swarm Optimization（PSO） algorithm, catfish effect

摘要： 针对煤炭消费量的时变性、非平稳性特点，为了提高煤炭消费量预测精度，提出了一种鲶鱼粒子群算法优化最小二乘支持向量机（LSSVM）的煤炭消费量预测模型（CEPSO-LSSVM）。将LSSVM参数编码成粒子位置串，并根据煤炭消费量训练集的交叉验证误差最小作为参数优化目标，通过粒子间信息交流找到最优LSSVM参数，并引入“鲶鱼效应”，保持粒子群的多样性，克服传统粒子群算法的局部最优，根据最优参数建立煤炭消费量预测模型，并采用实际煤炭消费量数据进行仿真测试。结果表明，相对于其他预测模型，CEPSO-LSSVM可以获得更优的LSSVM参数，提高了煤炭消费量预测精度，更加适用于复杂非线性的煤炭消费量预测。

关键词: 煤炭消费量, 最小二乘支持向量机, 粒子群优化算法, 鲶鱼效应