基于熵增理论的优化算法研究及其应用

针对传统机械优化算法所存在的问题提出了一种基于熵增理论的优化算法。该算法将系统的演化过程和优化问题在解空间的寻优过程相对应,将系统的构形参数与优化问题的设计变量相对应,以系统的平衡态作为优化设计问题的最优点,以系统熵最大或系统能量最小作为系统收敛的准则。在算法研究过程中,运用模拟退火算法这种可获得全局最优的抽样算法作为系统演化的机制,可以保证优化问题的全局最优性。运用该算法对数学测试函数和内燃机配气机构凸轮的轮廓曲线进行了设计,所得结果表明了该算法的有效性。

Study on the Optimal Algorithm and its Application Based on the Theory of Entropy

This paper presented a new approach of optimal algorithm based on theory of entropy increase. This method takes into account some inter-related pairs, and these include the evolution and optimization of system with that of solving process, the structure of system with that of designing variants. The system's steady status was set as the optimal solution, and maximum entropy or minimum energy of the system as the algorithm of system convergence. Utilization of the simulated annealing algorithm realized the global optimal result. The obtained results supported the validity of the algorithm.