基于遗传算法和最小二乘支持向量机可靠性分配*

为了提高系统可靠性的精确快速分配，采用支持向量机对系统可靠性进行建模，采用逆向思维对系统可靠性进行分配;为了提高求解速度和鲁棒性，用最小二乘法对支持向量机进行算法优化,并用遗传算法对最小二乘支持向量机进行参数优化;为了提高分配精度，用三角模糊数进行模糊处理;最后针对某系统的可靠性，采用遗传算法优化和模糊处理的最小二乘支持向量机进行分配，并与神经网络和普通遗传算法优化的最小二乘支持向量机进行对比。结果表明,用遗传算法优化和模糊数处理的最小二乘支持向量机具有分配精度高，泛化能力强等优点。

Reliability allocation based on genetic algorithm and LS-SVM

For improving precise and rapid system reliability allocation, made the model by support vector machines, used the reliability by reverse thinking. In order to improve the solution speed and robustness, allocated least square method to optimization. At the same time used genetic algorithmfor parameter optimization in least squares support vector machines. Used the triangle fuzzy number for improving distribution accuracy. At last allocated some system reliability by using least squares support vector machines which was optimized by genetic algorithm and triangle fuzzy number. Compared with neural network, the results show the least squares support vector machines which was optimized by genetic algorithm and triangle fuzzy number has the advantages include such as high accuracy and strong generalization ability.