Neural Networks in System Identification

Abstract

Abstract. Neural Networks are non-linear black-box model structures, to be used with conventional parameter estimation methods. They have good general approximation capabilities for reasonable non-linear systems. When estimating the parameters in these structures, there is also good adaptability to concentrate on those parameters that have the most importance for the particular data set.

Artificial Neural Networks for Pollution Forecast

“…www.intechopen.com where the vector of regressors x  includes information available up to the time step n. A number of delayed values of the time series up to time step n can be used together with additional data from other measures (non linear autoregressive with exogenous inputs model, NARX; Sjöberg et al, 1994). Such values may also be filtered (e.g., using a FIR filter).…”

Intelligent predictive control of nonlinear processes using neural networks

“…To reduce the degrees of freedom and because it is advantageous in many control system designs, it is common to consider model structures that are natural extensions of well-known linear model structures like ARX, AIWAX, and OE in that a similar regression vector is considered [9]: 0 In the NARX structure the regressors are past inputs and outputs (d>O denotes the time delay)…”

“…Neural networks have on many occasions been presented as an excellent generic model structure for identification of nonlinear systems [9]. Utilization of this ability in the design of control systems has thus evolved into being one of the main classes of applications within the field of intelligent control.…”

Facility Power Usage Prediction with Artificial Neural Networks

“…A main principle in identification is to "try simple things first." The idea is to start with the simplest model which has a possibility to describe the system and continue to more complex ones if the simple model does not provide reliable results in the validation stage [18]. When a more complex model is investigated, the results with the simpler model give some guidelines how the structural parameters should be chosen in the new model.…”