Neural network modeling by system dynamics methodology
03 Dec 2003-Vol. 1, pp 424-428
TL;DR: The present paper describes the modeling of neural networks by the concepts of system dynamics methodology and provides several examples of different neural networks topology.
Abstract: The present paper describes the modeling of neural networks by the concepts of system dynamics methodology. The presented models contribute to the understanding of the neural networks dynamics. The paper provides several examples of different neural networks topology. The main components of neural networks are modeled by the classical system dynamics entities.
Citations
More filters
TL;DR: It is concluded that the battery price remains a crucial explanatory variable for annual electric car sales in simulation exercises.
Abstract: This paper describes the modelling process of soft-linking two system dynamics models of the automotive ecosystem: the Powertrain Technology Transition Market Agent model and the Transport, Energy,
5 citations
Cites background from "Neural network modeling by system d..."
...The complementarity of SD with the latter was shown by Kofjac et al. (2003)....
[...]
30 May 2005
TL;DR: Because a system dynamics model is equivalent to a specially-designed artificial neural network, both of which operate under the same numerical propagation constraints, this work uses the Artificial neural network training algorithms and takes advantage of historical data to assist system dynamicsmodel construction.
Abstract: The study of system dynamics starts from model construction and simulation to understand and solve dynamical complicated problems. Traditional approaches of modeling process depend on experts' experiences and the trial-and-error procedure, so it is difficult to guarantee a useful model. Because a system dynamics model is equivalent to a specially-designed artificial neural network, both of which operate under the same numerical propagation constraints, we use the artificial neural network training algorithms and take advantage of historical data to assist system dynamics model construction. Experimental studies show that this approach is feasible.
2 citations
Cites background from "Neural network modeling by system d..."
...Kofjac [7] first demonstrated the possibility of interconnecting SD methodology and ANN modeling....
[...]
17 May 2021
TL;DR: In this article, the authors developed a conceptual model of the company's financial logistics based on the system dynamics principles to estimate the stationary trajectory financial flow and short-term and long-term gaps.
Abstract: Purpose – the purpose of the article is to develop a conceptual model of the company’s financial logistics based on the system dynamics principles. Research methodology – the article is based on the system analysis and system dynamics methods to define, classify and simulate a company financial flow. Findings – the definition of financial logistics for a business system has been defined. The authors make a classification of the company’s financial flow by the main economic activities and time series factors. Research limitations – commercial data, used for the practical implementation of the model, are confidential and cannot be disclosed. P ractical implications – the model is implemented by transformation of a system dynamic flow graph into VENSIM programs. It may estimate the stationary trajectory financial flow and short-term and long-term gaps. Originality/Value – the conceptual model of the company’s financial logistics is determined based on the system dynam-ics principles. The model includes the advantages of the financial management methods and contemporary econometric analysis instruments based on a system dynamics. https://doi.org/10.3846/cibmee.2021.630
1 citations
01 Nov 2019
TL;DR: The simulation shows that the modified cuckoo search algorithm which is applied to the actuator modeling, can effectively improve the modeling accuracy and generalization ability.
Abstract: The paper proposed a modified cuckoo search algorithm by studying the nonlinear modeling problem of electric actuator, and a neural network identification model of actuator is established. The algorithm makes the Levy flight mechanism adaptive by dynamically adjusting the step size. The proposed method is proved to be effective and feasible by simulation experiments and comparison with traditional algorithms. The simulation shows that the modified cuckoo search algorithm which is applied to the actuator modeling, can effectively improve the modeling accuracy and generalization ability.
References
More filters
01 Jan 1974
4,080 citations
01 Jan 1974
2,849 citations
Book•
01 Jul 1994TL;DR: In this chapter seven Neural Nets based on Competition, Adaptive Resonance Theory, and Backpropagation Neural Net are studied.
Abstract: 1. Introduction. 2. Simple Neural Nets for Pattern Classification. 3. Pattern Association. 4. Neural Networks Based on Competition. 5. Adaptive Resonance Theory. 6. Backpropagation Neural Net. 7. A Sampler of Other Neural Nets. Glossary. References. Index.
2,665 citations
Book•
26 Feb 1988318 citations