Abstract
Solving linear inequalities is widely used in various fields, and it plays a more and more crucial role in practical engineering applications. However, the existing recurrent neural network models to solve this problem only achieve global convergence without any noise. To overcome this disadvantage, in this article, we propose a novel integral design scheme for finding the robust solution of time-varying matrix inequalities. The core idea of this model is to add an integral term in the construction of the error function to make the model have error memory, so as to eliminate static difference. Meanwhile, appropriate activation functions (AFs) are used in the integral term noise-tolerance zeroing neural network (ITNTZNN) model, which can make error function accomplish finite-time convergence. The noise tolerance property of the ITNTZNN model is proved by theoretical analysis, and the upper limit of convergence time is obtained. Numerical simulative results ulteriorly verify the finite-time and noise-tolerant properties of the ITNTZNN model.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
