JERAJournal of Electronic Research and Application2208-35022208-3510Bio-Byword Scientific Publishing Pty. Ltd.10.26689/jera.v10i1.13988ArticleDynamic Memristor-Inspired Zeroing Neural Network with Hybrid Activation for Finite-Time Convergence and Remote Sensing Image Fusionhttps://artdesignp.com/journal/JERA/10/1/10.26689/jera.v10i1.13988HeJiaqi,LiMu20261012026-02-27Solving time-varying nonlinear equations in real time is a significant challenge in modern computing. Dynamic Memristor-Inspired Zeroing Neural Networks (DMZNN) have shown strong performance in this field, but their convergence speed and robustness heavily rely on the design of the activation function. This paper proposes a novel hybrid activation function inspired by the nonlinear characteristics of memristors. By integrating cubic and sublinear terms, the proposed function facilitates multi-stage error decay, effectively addressing the slow convergence and poor noise resistance of traditional activation functions. Theoretical analysis shows that the DMZNN model, built upon this activation function, can converge in finite time. Robustness under parameter perturbations and additive noise is rigorously proven using Lyapunov theory. Simulation results demonstrate that the convergence speed of the DMZNN model is obviously faster than that of traditional ZNN models when solving second-order, third-order, and fourth-order time-varying nonlinear equations. Additionally, in the application of remote sensing image fusion, DMZNN outperforms traditional gradient-based methods in both fusion quality and processing speed, demonstrating its practical effectiveness and superiority in real-world applications.

Jie J, 2021, An Improved Finite Time Convergence Recurrent Neural Network with Application to Time-Varying Linear Complex Matrix Equation Solution. Neural Processing Letters, 53(1): 777–786.

Wu W, Zhang Y, Tan N, 2024, Adaptive ZNN Model and Solvers for Tackling Temporally Variant Quadratic Program with Applications. IEEE Transactions on Industrial Informatics, 20(11): 13015–13025.

Chen J, Pan Y, Zhang Y, 2024, ZNN Continuous Model and Discrete Algorithm for Temporally Variant Optimization with Nonlinear Equation Constraints via Novel TD Formula. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 54(7): 3994–4004.

Tang C, Ding Y, 2025, A New Parameter-Changing Integral ZNN Model with Nonlinear Activation for Solving Inequality Constraint Time-Varying Quadratic Programming. IEEE Access, 13: 52874–52888

Zhang Y, Wang L, Zhong G, 2025, Design and Analysis of a Variable-Parameter Noise-Tolerant ZNN for Solving Time-Variant Nonlinear Equations and Applications. Applied Intelligence, 55(6): 460.

Fu Z, Zhang Y, Li W, 2024, Solving Future Nonlinear Equation System via ZNN and Novel General ILR3S Formula with Multitype Manipulator Applications. IEEE Transactions on Industrial Electronics, 71(10): 12623–12633.

Liao B, Xu J, Hua C, et al., 2025, Predefined-Time ZNN Model with Noise Reduction for Solving Quadratic Programming and Its Application to Binary Assignment Problem in Logistics. The Journal of Supercomputing, 81(12): 1228.

Jie J, Zhu J, Gong J, et al., 2022, Novel Activation Functions-Based ZNN Models for Fixed-Time Solving Dynamic Sylvester Equation. Neural Computing and Applications, 34(17): 14297–14315.

Xiao L, Li X, Cao P, et al., 2023, A Dynamic-Varying Parameter Enhanced ZNN Model for Solving Time-Varying Complex-Valued Tensor Inversion with Its Application to Image Encryption. IEEE Transactions on Neural Networks and Learning Systems, 35(10): 13681–13690.

Li J, Zhu J, Li C, et al., 2022, CGTF: Convolution-Guided Transformer for Infrared and Visible Image Fusion. IEEE Transactions on Instrumentation and Measurement, 71: 1–14.