Abstract:A deep understanding on the intrinsic mechanism is required to develop a highly specialized mechanistic model for ecosystem dynamics. However, it is usually hard to do for most of the ecological and environmental problems, because of the lack of a consistent theoretical background. Neural networks are universal and flexible models for linear and non-linear systems. This paper aimed to modeling an ecosystem using neural network models and the conventional model, and assessing their effectiveness in the dynamic simulation of ecosystem. Elman neural network model, linear neural network model, and linear ordinary differential equation were developed to simulate the dynamics of Chinese cabbage growth system recorded in the field. Matlab codes for these neural network models were given. Sensitivity analysis was conducted to detect the robustness of these models.The results showed that Elman neural network model could simulate the multivariate non-linear system at the desired accuracy. Linear neural network model may simulate such a non-linear system only in certain conditions. Conventional linear ordinary differential equation yielded divergent outputs in the dynamic simulation of the multivariate non-linear system.Sensitivity analysis indicated that the learning rate influenced the simulation performance of linear neural network model. Transfer function of the second layer in Elman neural network model would influence the simulation performance of this model, but little influence was produced when other functions were changed. Elman neural network was proven to be a robust model. Sensitivity analysis showed that the different choices of functions and parameters in neural network model would influence the performance of simulation. Sensitivity analysis is therefore powerful to detect the robustness and stability of neural network models. (c) 2006 Elsevier B.V All rights reserved.

Back-propagation neural networks for modeling complex systems

Neural Networks: Computational Models and Applications - ReadingSample

Investigation of artificial neural networks in coal blending

Neural Computing in Mechanics

Physics guided neural networks for modelling of non-linear dynamics

Neural Computing for Building Statistical Macromodels of Circuits and Systems

Application of Back Propagation Artificial Neural Networks on Dynamic Compensation of Measurement Systems

Multivariate Adaptive Regression Splines for Analysis of Geotechnical Engineering Systems

On Model-guided Neural Networks for System Identification

A Method to Model Nonlinear Systems by Neural Networks

Real-Valued Backpropagation is Unsuitable for Complex-Valued Neural Networks

Neural network modeling and dynamic behavior prediction of nonlinear dynamic systems

Neural networks modeling of electrohydraulic systems

Neural Network Modeling of Ecosystems: A Case Study on Cabbage Growth System

Robust Adaptive Control For A Class Of Complex Systems Using High-Order Neural Networks

Constructing Neural Network-Based Models for Simulating Dynamical Systems

Neural Dynamics on Complex Networks

Developing a Deep Learning Estimator to Learn Nonlinear Dynamic Systems

A Component-Oriented Reliability Model Using Back-Propagation Neural Networks

Complex Nonlinear System Modelling and Parameters Identification by Deep Neural Networks