The rank-sum test is a nonparametric method used in variety evaluation. However, the hypothesis testing of the method hasn＇t been established for multi-trait comprehensive ranking. In this paper, under null hypothesi...The rank-sum test is a nonparametric method used in variety evaluation. However, the hypothesis testing of the method hasn＇t been established for multi-trait comprehensive ranking. In this paper, under null hypothesis H0： the variety＇s ranking on each trait is random, the theoretical distribution of sum of ranks （SR） was firstly derived and further used to obtain the critical values for multi-trait comprehensive evaluation in rank-sum testing. A new C＋＋ class and its basic arithmetic were defined to deal with the miscount caused by the precision limitation of built-in data type in common statistical software under large number of varieties and traits. Finally, an application of the theoretical results was demonstrated using five starch viscosity traits of 12 glutinous maize varieties. The proposed method is so simple and convenient that it can be easily used to rank different varieties by multiple traits.展开更多
为克服单一赋权法的局限性,结合山区干线公路交通特征及交通安全评价指标的选取原则,从社会因素、驾驶因素、环境因素、管理因素和道路因素五个维度出发,选取18个综合评价指标,运用序关系分析法(Order Relation Analysis Method,G1)-指...为克服单一赋权法的局限性,结合山区干线公路交通特征及交通安全评价指标的选取原则,从社会因素、驾驶因素、环境因素、管理因素和道路因素五个维度出发,选取18个综合评价指标,运用序关系分析法(Order Relation Analysis Method,G1)-指标相关性权重确定法(Criteria Importance Through Intercriteria Correlation,CRITIC)确定各评价指标的权重,并结合折中妥协多属性决策法(VlseKriterijumska Optimizacija I Kompromisno Resenje,VIKOR)对山区干线公路交通安全进行综合评价,提出了基于G1-CRITIC-VIKOR模型的山区干线公路交通安全综合评价及比选方法。以中国西部6条山区干线公路为例进行实证研究,结果表明,G1-CRITIC-VIKOR模型的评价效果与传统的秩和比(Rank-Sum Ratio,RSR)综合评价法及加权逼近理想解排序法(Technique for Order Preference by Similarity to Ideal Solution,TOPSIS)的评价结果基本一致,且评价效果明显优于后者,具有更好的辨识性,验证了该模型的可行性和科学性。展开更多
It is evident that complex optimization problems are becoming increasingly prominent,metaheuristic algorithms have demonstrated unique advantages in solving high-dimensional,nonlinear problems.However,the traditional ...It is evident that complex optimization problems are becoming increasingly prominent,metaheuristic algorithms have demonstrated unique advantages in solving high-dimensional,nonlinear problems.However,the traditional Sparrow Search Algorithm(SSA)suffers from limited global search capability,insufficient population diversity,and slow convergence,which often leads to premature stagnation in local optima.Despite the proposal of various enhanced versions,the effective balancing of exploration and exploitation remains an unsolved challenge.To address the previously mentioned problems,this study proposes a multi-strategy collaborative improved SSA,which systematically integrates four complementary strategies:(1)the Northern Goshawk Optimization(NGO)mechanism enhances global exploration through guided prey-attacking dynamics;(2)an adaptive t-distribution mutation strategy balances the transition between exploration and exploitation via dynamic adjustment of the degrees of freedom;(3)a dual chaotic initialization method(Bernoulli and Sinusoidal maps)increases population diversity and distribution uniformity;and(4)an elite retention strategy maintains solution quality and prevents degradation during iterations.These strategies cooperate synergistically,forming a tightly coupled optimization framework that significantly improves search efficiency and robustness.Therefore,this paper names it NTSSA:A Novel Multi-Strategy Enhanced Sparrow Search Algorithm with Northern Goshawk Optimization and Adaptive t-Distribution for Global Optimization.Extensive experiments on the CEC2005 benchmark set demonstrate that NTSSA achieves theoretical optimal accuracy on unimodal functions and significantly enhances global optimum discovery for multimodal functions by 2–5 orders of magnitude.Compared with SSA,GWO,ISSA,and CSSOA,NTSSA improves solution accuracy by up to 14.3%(F8)and 99.8%(F12),while accelerating convergence by approximately 1.5–2×.The Wilcoxon rank-sum test(p<0.05)indicates that NTSSA demonstrates a statistically substantial performance advantage.Theoretical analysis demonstrates that the collaborative synergy among adaptive mutation,chaos-based diversification,and elite preservation ensures both high convergence accuracy and global stability.This work bridges a key research gap in SSA by realizing a coordinated optimization mechanism between exploration and exploitation,offering a robust and efficient solution framework for complex high-dimensional problems in intelligent computation and engineering design.展开更多
基金supported by the National Key Basic Research Program of China(2006CB101700)the Program for New Century Excellent Talents in University of Ministry of Education of China(NCET2005-05-0502)
文摘The rank-sum test is a nonparametric method used in variety evaluation. However, the hypothesis testing of the method hasn＇t been established for multi-trait comprehensive ranking. In this paper, under null hypothesis H0： the variety＇s ranking on each trait is random, the theoretical distribution of sum of ranks （SR） was firstly derived and further used to obtain the critical values for multi-trait comprehensive evaluation in rank-sum testing. A new C＋＋ class and its basic arithmetic were defined to deal with the miscount caused by the precision limitation of built-in data type in common statistical software under large number of varieties and traits. Finally, an application of the theoretical results was demonstrated using five starch viscosity traits of 12 glutinous maize varieties. The proposed method is so simple and convenient that it can be easily used to rank different varieties by multiple traits.
文摘为克服单一赋权法的局限性,结合山区干线公路交通特征及交通安全评价指标的选取原则,从社会因素、驾驶因素、环境因素、管理因素和道路因素五个维度出发,选取18个综合评价指标,运用序关系分析法(Order Relation Analysis Method,G1)-指标相关性权重确定法(Criteria Importance Through Intercriteria Correlation,CRITIC)确定各评价指标的权重,并结合折中妥协多属性决策法(VlseKriterijumska Optimizacija I Kompromisno Resenje,VIKOR)对山区干线公路交通安全进行综合评价,提出了基于G1-CRITIC-VIKOR模型的山区干线公路交通安全综合评价及比选方法。以中国西部6条山区干线公路为例进行实证研究,结果表明,G1-CRITIC-VIKOR模型的评价效果与传统的秩和比(Rank-Sum Ratio,RSR)综合评价法及加权逼近理想解排序法(Technique for Order Preference by Similarity to Ideal Solution,TOPSIS)的评价结果基本一致,且评价效果明显优于后者,具有更好的辨识性,验证了该模型的可行性和科学性。
文摘It is evident that complex optimization problems are becoming increasingly prominent,metaheuristic algorithms have demonstrated unique advantages in solving high-dimensional,nonlinear problems.However,the traditional Sparrow Search Algorithm(SSA)suffers from limited global search capability,insufficient population diversity,and slow convergence,which often leads to premature stagnation in local optima.Despite the proposal of various enhanced versions,the effective balancing of exploration and exploitation remains an unsolved challenge.To address the previously mentioned problems,this study proposes a multi-strategy collaborative improved SSA,which systematically integrates four complementary strategies:(1)the Northern Goshawk Optimization(NGO)mechanism enhances global exploration through guided prey-attacking dynamics;(2)an adaptive t-distribution mutation strategy balances the transition between exploration and exploitation via dynamic adjustment of the degrees of freedom;(3)a dual chaotic initialization method(Bernoulli and Sinusoidal maps)increases population diversity and distribution uniformity;and(4)an elite retention strategy maintains solution quality and prevents degradation during iterations.These strategies cooperate synergistically,forming a tightly coupled optimization framework that significantly improves search efficiency and robustness.Therefore,this paper names it NTSSA:A Novel Multi-Strategy Enhanced Sparrow Search Algorithm with Northern Goshawk Optimization and Adaptive t-Distribution for Global Optimization.Extensive experiments on the CEC2005 benchmark set demonstrate that NTSSA achieves theoretical optimal accuracy on unimodal functions and significantly enhances global optimum discovery for multimodal functions by 2–5 orders of magnitude.Compared with SSA,GWO,ISSA,and CSSOA,NTSSA improves solution accuracy by up to 14.3%(F8)and 99.8%(F12),while accelerating convergence by approximately 1.5–2×.The Wilcoxon rank-sum test(p<0.05)indicates that NTSSA demonstrates a statistically substantial performance advantage.Theoretical analysis demonstrates that the collaborative synergy among adaptive mutation,chaos-based diversification,and elite preservation ensures both high convergence accuracy and global stability.This work bridges a key research gap in SSA by realizing a coordinated optimization mechanism between exploration and exploitation,offering a robust and efficient solution framework for complex high-dimensional problems in intelligent computation and engineering design.
