基于群智能算法的预防性维修周期优化

Optimization of Preventive Maintenance Period Based on Hybrid Swarm Intelligence

分析将蚁群优化算法应用于预防性维修周期工程寻优问题时遇到的算法参数选择困难等问题,提出将粒子群优化算法和空间划分方法引入该过程以改进原蚁群算法的寻优规则和历程.建立混合粒子群和蚁群算法的群智能优化策略:PS_ACO(Particle Swarm and Ant Colony Optimization),并将其应用于混联系统预防性维修周期优化过程中,以解决由于蚁群算法中参数选择不当和随机产生维修周期解值带来的求解精度差、寻优效率低等问题.算法的寻优结果对比分析表明:该PS_ACO算法应用于预防性维修周期优化问题,在寻优效率及寻优精度上有部分改进,且可相对削弱算法参数选择对优化结果的影响.

It was analyzed that there were some problems such as parameters value settings etc when the ant colony optimization （ACO） was applied in the PM period optimization process. And it was put forward that the particle swarm optimization （PSO） was brought into the ACO algorithm to form a new hybrid swarm optimization： PS_ACO （Particle Swarm and Ant Colony Optimization）. This new hybrid algorithm can modify the optimization rules and geographic division of ACO, and can partly solve some problems about the worse precision and inefficient optimization coming from unsuitable parameters values setting of ACO and random PM period solution. This PS_ACO algorithm was applied in the optimization process of series-parallel system PM period. The experimental data shows that： the PS_ACO can partly improve the optimization efficiency and precision, and relatively weaken the influence of parameters value settings to the optimization result.