针对传统遗传算法(GA)和粒子群算法(PSO)在优化计算过程中存在的优缺点,研究了二者结合的混合算法(GA-PSO),用其优化基于Hermite多项式的投影寻踪回归模型,解决了模型中最佳投影方向的优化计算问题,并运用该模型进行需水量预测。实例研究表明,基于混合算法优化的投影寻踪回归模型可靠,且预测精度较高,能有效地解决需水预测中存在的数据非线性、高维数等实际难点,在需水预测中有一定的适用性。 Aiming at the advantages and disadvantages of traditional genetic algorithm (GA) and particle swarm optimization (PSO) in the process of optimization calculation, a hybrid algorithm (GA-PSO) combining the two is studied. Projection pursuit regression model based on Hermite polynomials , Which solves the optimization problem of the best projection direction in the model and uses the model to forecast water demand. The case study shows that projection pursuit model based on hybrid algorithm is reliable and predictive accuracy is high, which can effectively solve the practical difficulties of data nonlinearity and high dimension in water demand prediction. Applicability.