So the improvement , such as the self - adaptive fitness function combined with the penalty function methods , self - adaptive crossover probability and the bp operator enlightened by neutral network , especially the bp network to improve the local optimal capacity were used 其次结合惩罚函数法对适应度进行了优化。采用自适应交叉概率。受神经网络中bp算法的启示,构造bp算子,提高小种群遗传算法的局部搜索寻优能力。
According to the work condition of one 1450 five - stand tandem cold mill , reduction distribution as decision variable , the penalty function was established , in which rolling pressure distribution ratio was as objective function and the constraint condition of reduction distribution was as penalty term 根据某1450五机架冷连轧机生产工况,以压下量分配为自变量,以轧制力成比例分配为目标函数,将压下量分配的约束条件作为惩罚项,建立惩罚函数。
For the mathematics model of differential / algebraic equation system , the penalty function is introduced in numerical computation , and runge - kutta method and newton - raphson method are used respectively in solving the ordinary differential equations and nonlinear equation systems . the process and the results are carried out under corba environment 本论文采用近年发展的惩罚函数方法进行数值积分,对于常微分方程,采用runge - kutta方法进行数值积分,对于代数方程,采用newton - raphson迭代方法求解,并在corba环境下实现。
On the basis of analyzing limitation of using penalty function method dealing with constraints , an amending method based on the knowlege regulating strategy is suggested to amende the mapping relationship of infeasible constraints in decoding , thus making the regulated individuals map into the space to obtain the most promising optimal solution 在分析惩罚函数法对约束处理效率较低的情况下,提出了一种基于知识调整策略的修正法,对解码中不满足约束的映射关系进行修正,使调整后个体映射到最有希望获得最优解的空间中。
By introducing the concepts of constraints measure and feasible degree , the extent to which the decision variables subject to the inequality constraints and equality constraints is described and then the self - adaptive penalty function is constructed , whose values are adaptively changed with the constraints feasible degree 该策略通过引入约束可行测度、可行度等概念来描述决策变量服从于不等式约束和等式约束的程度,并以此构造处理约束条件的自适应惩罚函数,惩罚值随着约束可行度的变化而动态自适应地改变。
In order to make the best of both merits in local and global planning , chaos algorithm and classical algorithm are mixed together . to ensure the systematic real - time responsibility , a novel algorithm named conversely decentralized autonomous control is put forward on motion planning when considering orientation 为了统筹兼顾局部规划和全局规划的各自优点,将混沌算法与经典算法(如惩罚函数法以及基于梯度的各种算法)混合使用;为了保证控制系统的实时性,对有姿态要求的运动规划提出了一种新的算法? ?逆向自律分散控制。