八元数值细胞神经网络(μ,ν)-伪概周期解的存在性OA
Existence of(μ,ν)-Pseudo Almost Periodic Solutions for Octonion-Valued Cellular Neural Networks
针对连接项含有时滞的八元数值细胞神经网络,研究(μ,ν)-伪概周期解的存在性与唯一性问题.八元数代数的乘法不满足结合性和交换性,分析其八元数值细胞神经网络的动力学行为具有重要意义.首先,提出一种不依赖于系统分解的方法,将原始细胞神经网络转化为积分方程,由于(μ,ν)-伪概周期解所构成的空间是完备的,因此建立一个适配的 Banach 空间.其次,在积分方程的基础上,通过综合运用压缩映射原理、勒贝格控制收敛定理和微分不等式技巧等方法,证明了系统存在唯一(μ,ν)-伪概周期解,得到易于验证的判别准则.最后,提出一个数值模拟验证了结果的可行性.
This paper explored the existence and uniqueness of(μ,ν)-pseudo almost periodic solutions for octonion-valued cellular neural networks with time delays in the connection terms.Since the multiplication of octonion-valued does not satisfy associativity or commutativity,analyzing the dynamic behavior of octonion-valued cellular neural net-works holds significant importance.First,a method independent of system decomposition is proposed to convert the original cellular neural network into an integral equation.Since the space composed of pseudo-almost periodic solu-tions is complete,an appropriate Banach space is established.Second,based on the integral equation,by comprehen-sively using the contraction mapping principle,Lebesgue's dominated convergence theorem,and differential inequa-lity techniques,the existence and uniqueness of the(μ,ν)-pseudo almost periodic solution of the system are proved,and easily verifiable criteria are obtained.Lastly,a numerical simulation is carried out to validate the feasibility of the results.
霍妮娜;徐玉莹
合肥大学数学与统计学院,安徽 合肥,230601合肥大学数学与统计学院,安徽 合肥,230601
数理科学
(μ,ν)-伪概周期解存在性时滞
(μ,ν)-pseudo almost periodic solutionexistencetime delay
《宿州学院学报》 2026 (3)
8-14,7
安徽省自然科学基金项目(2108085QA10)安徽省高校自然科学研究项目(2022AH051782).
评论