Maximum power point tracking(MPPT) techniques are used to maintain photovoltaic modules operating points at the local maximum power points under non-uniform irradiance conditions(NUIC). For global maximum power point ...Maximum power point tracking(MPPT) techniques are used to maintain photovoltaic modules operating points at the local maximum power points under non-uniform irradiance conditions(NUIC). For global maximum power point tracking(GMPPT) within an appropriate period, a hybrid artificial fish swarm algorithm(HAFSA) is proposed in this paper, which was developed using particle swarm optimization(PSO) to reformulate AFSA and improve its principal parameters. Simulation results show that under NUIC, compared with PSO and AFSA, the proposed algorithm has better performance with respect to convergence speed and convergence accuracy. Under NUIC, the average convergence times for 1000 simulation experiments completed with PSO, AFSA, and HAFSA are 0.4830 s, 0.4003 s and 0.3152 s respectively, and the average tracking time of the HAFSA algorithm is reduced by 34.74% and 21.26% compared with PSO and AFSA, respectively. The convergence times of the velocity inertia ω relative constant and linear decrement method decreased by 35.48% and 8.19%, the convergence time of the Visual relative constant mode decreased by 10.16%, and the convergence time of the Step relative constant mode decreased by 17.88%. The proposed GMPPT algorithm is simulated in MATLAB, and the algorithm tracks GMPP with excellent efficiency and fast speed.展开更多
针对电源组合故障预测的需求,提出了一种基于人工鱼群算法(Artificial Fish Swarm Algorithm,AFSA)优化灰色神经网络的故障预测方法;文中首先对AFSA算法和灰色神经网络进行了介绍;然后在此基础上提出了基于AFSA优化灰色神经网络的故障...针对电源组合故障预测的需求,提出了一种基于人工鱼群算法(Artificial Fish Swarm Algorithm,AFSA)优化灰色神经网络的故障预测方法;文中首先对AFSA算法和灰色神经网络进行了介绍;然后在此基础上提出了基于AFSA优化灰色神经网络的故障预测模型,并给出了AFSA优化灰色神经网络参数的算法步骤;最后对制导雷达波束控制系统中的某电源组合进行了故障预测,预测结果表明该预测方法误差较小,达到了预期效果。展开更多
为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改...为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改进人工鱼群算法在迭代中的收敛速度和种群多样性.其次,将自适应全局-局部记忆机制引入到标准AFSA算法的觅食行为中,以进一步提高勘探能力.最后,增加了基于平均适应度的行为选择机制,以提供更合理的行为选择,减少算法的复杂性.通过使用CloudSim平台进行实验验证,分别测试在不同任务规模下IAFSA的算法效能.实验结果表明,改进人工鱼群算法在降低系统任务最大完工时间和成本上均表现出了显著的优势.展开更多
基金supported by National Natural Science Foundation of China (No.61501106)Science and Technology Foundation of Jilin Province (No. 20180101039JC and JJKH20170102KJ)
文摘Maximum power point tracking(MPPT) techniques are used to maintain photovoltaic modules operating points at the local maximum power points under non-uniform irradiance conditions(NUIC). For global maximum power point tracking(GMPPT) within an appropriate period, a hybrid artificial fish swarm algorithm(HAFSA) is proposed in this paper, which was developed using particle swarm optimization(PSO) to reformulate AFSA and improve its principal parameters. Simulation results show that under NUIC, compared with PSO and AFSA, the proposed algorithm has better performance with respect to convergence speed and convergence accuracy. Under NUIC, the average convergence times for 1000 simulation experiments completed with PSO, AFSA, and HAFSA are 0.4830 s, 0.4003 s and 0.3152 s respectively, and the average tracking time of the HAFSA algorithm is reduced by 34.74% and 21.26% compared with PSO and AFSA, respectively. The convergence times of the velocity inertia ω relative constant and linear decrement method decreased by 35.48% and 8.19%, the convergence time of the Visual relative constant mode decreased by 10.16%, and the convergence time of the Step relative constant mode decreased by 17.88%. The proposed GMPPT algorithm is simulated in MATLAB, and the algorithm tracks GMPP with excellent efficiency and fast speed.
文摘针对电源组合故障预测的需求,提出了一种基于人工鱼群算法(Artificial Fish Swarm Algorithm,AFSA)优化灰色神经网络的故障预测方法;文中首先对AFSA算法和灰色神经网络进行了介绍;然后在此基础上提出了基于AFSA优化灰色神经网络的故障预测模型,并给出了AFSA优化灰色神经网络参数的算法步骤;最后对制导雷达波束控制系统中的某电源组合进行了故障预测,预测结果表明该预测方法误差较小,达到了预期效果。
文摘为提高云计算任务调度的效率,减少系统执行任务的最大完工时间以及成本,本文提出一种改进的人工鱼群任务调度算法(improved artificial fish swarm algorithm,IAFSA).首先,将反向学习策略应用于种群初始化和鱼群的行为选择中,以提高改进人工鱼群算法在迭代中的收敛速度和种群多样性.其次,将自适应全局-局部记忆机制引入到标准AFSA算法的觅食行为中,以进一步提高勘探能力.最后,增加了基于平均适应度的行为选择机制,以提供更合理的行为选择,减少算法的复杂性.通过使用CloudSim平台进行实验验证,分别测试在不同任务规模下IAFSA的算法效能.实验结果表明,改进人工鱼群算法在降低系统任务最大完工时间和成本上均表现出了显著的优势.
