In general, neural networks cannot match nonlinear systems exactly, neuro identiÞer has to include robust modiÞcation in order to guarantee Lyapunov stability. In this paper input-to-state stability approach is applied to access robust training algorithms of discrete-time neural networks. We conclude that the gradient descent law and the backpropagation-like algorithm for the weights adjustment...

At present quality of software systems is a major issue, still well defined criteria to measure it needs to be established. The object-oriented (OO) systems, which is different from procedural paradigm requires valid and effective metrics to assess quality of the software. There is considerable research interest in developing and applying sophisticated techniques to construct models for estimat...

for Digital Neuro-Computers Paolo Ienne LAMI & MANTRA, Swiss Federal Institute of Technology, CH-1015 Lausanne In Proceedings of the International Joint Conference on Neural Networks, volume II, pages 1987{90, Nagoya, Japan, October 1993. Abstract As the eld of neural-networks matures toward real-world applications, a strong need for hardware systems to e ciently compute larger network arises. ...

The knowledge about the seasonal rainfall in some Brazilian regions is essential for agriculture and the adequate management of water resources. For this purpose, linear and nonlinear models are commonly used for seasonal rainfall prediction, while some of them are based on Artificial Neural Networks, demonstrating great potential as shown in literature. According to this tendency, this work pr...

Aircraft burst fault is uncertainty and ambiguity. Considering QAR data as the research object, the fault diagnosis system based on the T-S fuzzy neural network combined with aircraft maintenance processes is built. First, the system designs the network performance oversight function to improve genetic neural network program. Then the fuzzy logic is used to deal with fuzzy rules, which can dete...

The paper is continuing a work for a velocity controller synthesis by neuro-fuzzy techniques in a vector control with an asynchronous motor structure. The input – output data prepared in the previous paper are applied to a program in order to train a neural network able to generate a file for the speed fuzzy controller. The fuzzy solution is analyzed comparatively with the standard structure th...

In its first part, this contribution reviews shortly the application of neural network methods to medical problems and characterizes its advantages and problems in the context of the medical background. Successful application examples show that human diagnostic capabilities are significantly worse than the neural diagnostic systems. Then, paradigm of neural networks is shortly introduced and th...

Abstract. Inverse problems related to the interaction of radiation with participating media have been attracting the attention of various researchers for many years due to the relevant applications not only in engineering, but also in astrophysics, physical oceanography (hydrologic optics), remote sensing and atmosphere/hydrosphere optics among many others. In this work the determination the si...

The main contribution of the present paper is to train efficient neural networks for seismic design of double layer grids subject to multiple-earthquake loading. As the seismic analysis and design of such large scale structures require high computational efforts, employing neural network techniques substantially decreases the computational burden. Square-on-square double layer grids with the va...

simulation of groundwater fluctuations plays a crucial role in management of watersheds and water demand balancing. recently, wavelet analysis has been used widely in time series decomposition and coupling with neural networks for hydrological modeling. in this paper, the ability of the wavelet-dynamic artificial neural networks (w-ann) model was applied in forecasting one-month-ahead of ground...