Fiber laser micromachining is found extensive applications at industrial level because it is cheap and simple to use.Due to its high strength and low conductivity titanium is difficult to machine with conventional met...Fiber laser micromachining is found extensive applications at industrial level because it is cheap and simple to use.Due to its high strength and low conductivity titanium is difficult to machine with conventional methods.In this investigation,micro holes were fabricated using a 30 W fiber laser on 2 mm thickα-titanium(Grade 2)and the process parameters were optimized through response surface methodology(RSM)and teaching learning-based optimization(TLBO)approach.Experimental runs were designed as per rotatable central composite design(RCCD).Material removal rate(MRR),hole circularity(HC),deviation in diameter(DEV)and heat affected zone(HAZ)were selected as output.A third-order polynomial prediction model was established using RSM.Analysis of variance(ANOVA)suggested that the developed model was 93.5%accurate.The impact of input factors on responses were studied by 3D surface plots.RSM desirability indicates that optimum micro drilling conditions are scan speed 275.43 mm/s,frequency 24.61 kHz,power 36.23%and number of passes 49.75.TLBO indicates that optimum micro drilling conditions are scan speed 100 mm/s,frequency 20 kHz,power 20%and number of passes 50.Comparison between RSM and TLBO suggested that TLBO provided better optimization results.Surface morphology of the fabricated micro holes were analyzed with scanning electron microscopy(SEM).展开更多
提出了一种基于协同进化教与学优化(Co-evolutionary Teaching-and-Learning based Optimization,CTLBO)算法的二维最大熵多阈值分割方法。首先,给出了二维熵多阈值分割的最优化模型。然后,针对教与学优化(Teaching-and-Learning based ...提出了一种基于协同进化教与学优化(Co-evolutionary Teaching-and-Learning based Optimization,CTLBO)算法的二维最大熵多阈值分割方法。首先,给出了二维熵多阈值分割的最优化模型。然后,针对教与学优化(Teaching-and-Learning based Optimization,TLBO)算法存在的早熟收敛和停滞问题,提出了一种CTLBO算法,并将该算法应用于二维熵多阈值分割最优化模型的求解。该算法将整个班级分为多个子班级,每个子班级的学员同时向所有子班级的老师学习,从而提高种群多样性。此外,每隔一定的代数,各子班级的老师组成新的班级进行信息交流,从而提高收敛速度。最后,应用仿真实验对所提方法的有效性和可行性进行了验证。实验结果表明:与基于传统TLBO算法及其相关改进算法、粒子群算法的图像分割方法相比,所提方法具有更好的优化能力和分割性能。展开更多
文摘Fiber laser micromachining is found extensive applications at industrial level because it is cheap and simple to use.Due to its high strength and low conductivity titanium is difficult to machine with conventional methods.In this investigation,micro holes were fabricated using a 30 W fiber laser on 2 mm thickα-titanium(Grade 2)and the process parameters were optimized through response surface methodology(RSM)and teaching learning-based optimization(TLBO)approach.Experimental runs were designed as per rotatable central composite design(RCCD).Material removal rate(MRR),hole circularity(HC),deviation in diameter(DEV)and heat affected zone(HAZ)were selected as output.A third-order polynomial prediction model was established using RSM.Analysis of variance(ANOVA)suggested that the developed model was 93.5%accurate.The impact of input factors on responses were studied by 3D surface plots.RSM desirability indicates that optimum micro drilling conditions are scan speed 275.43 mm/s,frequency 24.61 kHz,power 36.23%and number of passes 49.75.TLBO indicates that optimum micro drilling conditions are scan speed 100 mm/s,frequency 20 kHz,power 20%and number of passes 50.Comparison between RSM and TLBO suggested that TLBO provided better optimization results.Surface morphology of the fabricated micro holes were analyzed with scanning electron microscopy(SEM).
文摘提出了一种基于协同进化教与学优化(Co-evolutionary Teaching-and-Learning based Optimization,CTLBO)算法的二维最大熵多阈值分割方法。首先,给出了二维熵多阈值分割的最优化模型。然后,针对教与学优化(Teaching-and-Learning based Optimization,TLBO)算法存在的早熟收敛和停滞问题,提出了一种CTLBO算法,并将该算法应用于二维熵多阈值分割最优化模型的求解。该算法将整个班级分为多个子班级,每个子班级的学员同时向所有子班级的老师学习,从而提高种群多样性。此外,每隔一定的代数,各子班级的老师组成新的班级进行信息交流,从而提高收敛速度。最后,应用仿真实验对所提方法的有效性和可行性进行了验证。实验结果表明:与基于传统TLBO算法及其相关改进算法、粒子群算法的图像分割方法相比,所提方法具有更好的优化能力和分割性能。
