多目标优化函数定义 Multi-Objective optimization

A. Multi-Objective optimization

A. 多目标优化

 Fig. 17  From: An integrated multi-objective optimization method with application to train crashworthiness design

图17：一种集成的多目标优化方法，应用于训练耐撞性设计

Assuming that there are M objective functions, the problem can be mathematically formulated as follows:

假设有 M 个目标函数，则该问题可以数学地表述如下：

 where x is a vector of N variables, which can be both integer and continuous, defined in the search space Σ. The problem is generally subjected to equality and inequality constraints, which can be expressed, respectively, as:
其中 x 是 N 个变量的向量，可以是整数，也可以是连续变量，在搜索空间 Σ 中定义。该问题通常受到相等和不等式约束的影响，它们可以分别表示为：

        For multi-objective optimization, the problem is thus to simultaneously minimize the components of a vector of functions (f=(f1, …, fM)T), which is subject to several constraints. The problem is typically not uniquely solvable, but a set of equally efficient alternative solutions is possible. These solutions, among all the found ones, constitute the Pareto-optimal front, when they are plotted in the objective space. When comparing individuals i and j, with objective vector functions fi and fj we affirm that i dominates j in the Pareto sense if and only if:
        因此，对于多目标优化，问题在于同时最小化函数 (f=(f1, …, fM)T) 向量的分量，该向量受多个约束。该问题通常不是唯一可以解决的，但一组同样有效的替代解决方案是可能的。在所有找到的解中，当它们在目标空间中绘制时，它们构成了帕累托最优前沿。当使用客观向量函数 fi 比较个体 i 和 j 时， fj 我们确认 i 在帕累托意义上支配 j，当且仅当： 

B. Hypervolume Metric