论文部分内容阅读
求解农业水资源优化配置模型(高维非线性优化模型),较常采用大系统分解协调原理和动态规划相结合的方法,这样减少了变量个数,便于优化求解,但协调的过程需要多次从低阶模型中返回信息,而且对于每层的寻优求解过程存在难以克服的矛盾.采用标准的粒子群优化算法则优化程度不易保证并容易陷入局部最优,优化结果对初始种群依赖性较强.因此应用免疫进化算法对标准粒子群优化算法进行改进并应用于灌区农业水资源优化配置模型的求解.算例分析表明,免疫粒子群算法为求解高维复杂的优化配置问题提供了新思路.
In order to solve the optimal allocation model of agricultural water resources (high-dimensional nonlinear optimization model), the method of large-scale system decomposition and coordination and dynamic programming is often used. This reduces the number of variables and facilitates optimization and solution. However, the coordination process needs to be repeated several times from low Order model, and there is an insurmountable contradiction for the optimal solution process of each layer.Using the standard Particle Swarm Optimization (PSO) algorithm, the optimization degree is not easy to be guaranteed and easily falls into the local optimum, and the optimization result is more dependent on the initial population. Therefore, the immune evolutionary algorithm is used to improve the standard particle swarm optimization algorithm and to solve the optimal allocation model of agricultural water resources in irrigated area.Examples show that the immune particle swarm optimization algorithm provides a new way to solve the high-dimensional complex optimization problem.