当前位置: X-MOL 学术Int. J. Numer. Meth. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Hybrid multi‐objective optimization algorithm using Taylor series model and Spider Monkey Optimization
International Journal for Numerical Methods in Engineering ( IF 2.7 ) Pub Date : 2021-01-12 , DOI: 10.1002/nme.6628
Radhika Menon 1 , Anju Kulkarni 2 , Deepak Singh 3 , Mithra Venkatesan 4
Affiliation  

Multi‐objective optimization is used for optimizing a number of objectives simultaneously. Mostly, the optimization algorithms considered the previous iterative position to find the next position updates. The main intention of this research is to design and develop a new model to solve the computational complexity, and the resource allocation problem. Based on this perspective, the Taylor series model and its predictive theory are applied to Spider Monkey Optimization (SMO), and a new optimization, named Taylor‐Spider Monkey Optimization (TaySMO) is developed. The proposed TaySMO computes the updated position of the swarm using the local leader phase and the global leader phase. However, a new position update equation is derived to enhance the searching process of the SMO. Here, multiple objectives such as, throughput, power, and fairness index are considered to solve the resource allocation problem. However, the performance of the proposed algorithm is evaluated using the conventional optimization function in terms of fitness function and convergence criteria as the mean square error (MSE) with the neural network learning is 0.3747, congestion rate of the resource allocation problem is 8.736E‐23, and MSE of the spectrum sensing is 8.74E‐23, respectively.

中文翻译:

基于泰勒级数模型和蜘蛛猴优化的混合多目标优化算法

多目标优化用于同时优化多个目标。通常,优化算法会考虑先前的迭代位置以找到下一个位置更新。这项研究的主要目的是设计和开发一种新模型,以解决计算复杂性和资源分配问题。基于此观点,将泰勒级数模型及其预测理论应用于蜘蛛猴优化(SMO),并开发了一种新的优化方法,称为泰勒-蜘蛛猴优化(TaySMO)。提议的TaySMO使用本地领导者阶段和全球领导者阶段来计算群体的更新位置。但是,导出了新的位置更新方程式以增强SMO的搜索过程。这里有多个目标,例如吞吐量,功率,考虑公平指标和公平指标来解决资源分配问题。但是,根据适合度函数和收敛标准,使用常规优化函数评估了所提出算法的性能,因为神经网络学习的均方误差(MSE)为0.3747,资源分配问题的拥塞率为8.736E- 23,频谱感测的MSE分别为8.74E-23。
更新日期:2021-01-12
down
wechat
bug