Miroslav B. Milovanović

Quasi-Sliding Mode Control With Orthogonal Endocrine Neural Network-Based Estimator Applied in Anti-Lock Braking System

This paper presents a new control method for nonlinear discrete-time systems, described by an input-output model which is based on a combination of quasi-sliding mode and neural networks. First, an input-output discrete-time quasi-sliding mode control with inserted digital integrator, which additionally reduces chattering, is described. Due to the presence of various nonlinearities and uncertainties, the model of the controlled object cannot be described adequately enough. These imperfections in modeling cause a modeling error, resulting in rather poor system performances. In order to increase the steady-state accuracy, an estimated value of the modeling error in the next sampling period is implemented into the control law. For this purpose, we propose two improved structures of the neural networks by implementing the generalized quasi-orthogonal functions of Legendre type. These functions have already been proven as an effective tool for the signal approximation, as well as for modeling, identification, analysis, synthesis, and simulation of dynamical systems. Finally, the proposed method is verified through digital simulations and real-time experiments on an anti-lock braking system as a representative of the considered class of mechatronic systems, in a laboratory environment. A detailed analysis of the obtained results confirms the effectiveness of the proposed approach in terms of better steady-state performances.

Modeling of Dynamic Systems Using Orthogonal Endocrine Adaptive Neuro-Fuzzy Inference Systems

This paper presents a new method for designing adaptive neuro-fuzzy inference systems (ANFIS). Improvements are made by introducing specially developed orthogonal functions into the very structure of ANFIS, specifically, into the layer that imitates Sugeno stile defuzzification. These functions are specially tailored for analysis and synthesis of dynamic systems and they also contain an adaptive measure of the variability of the systems operating in a real environment, which can be implemented inside the ANFIS as hormonal effect.

Adaptive PID control based on orthogonal endocrine neural networks

A new intelligent hybrid structure used for online tuning of a PID controller is proposed in this paper. The structure is based on two adaptive neural networks, both with built-in Chebyshev orthogonal polynomials. First substructure network is a regular orthogonal neural network with implemented artificial endocrine factor (OENN), in the form of environmental stimuli, to its weights. It is used for approximation of control signals and for processing system deviation/disturbance signals which are introduced in the form of environmental stimuli. The output values of OENN are used to calculate artificial environmental stimuli (AES), which represent required adaptation measure of a second network-orthogonal endocrine adaptive neuro-fuzzy inference system (OEANFIS). OEANFIS is used to process control, output and error signals of a system and to generate adjustable values of proportional, derivative, and integral parameters, used for online tuning of a PID controller. The developed structure is experimentally tested on a laboratory model of the 3D crane system in terms of analysing tracking performances and deviation signals (error signals) of a payload. OENN-OEANFIS performances are compared with traditional PID and 6 intelligent PID type controllers. Tracking performance comparisons (in transient and steady-state period) showed that the proposed adaptive controller possesses performances within the range of other tested controllers. The main contribution of OENN-OEANFIS structure is significant minimization of deviation signals (17%-79%) compared to other controllers. It is recommended to exploit it when dealing with a highly nonlinear system which operates in the presence of undesirable disturbances.

Application of neural networks with orthogonal activation functions in control of dynamical systems

In this article, we present a new method for the synthesis of almost and quasi-orthogonal polynomials of arbitrary order. Filters designed on the bases of these functions are generators of generalised quasi-orthogonal signals for which we derived and presented necessary mathematical background. Based on theoretical results, we designed and practically implemented generalised first-order (k = 1) quasi-orthogonal filter and proved its quasi-orthogonality via performed experiments. Designed filters can be applied in many scientific areas. In this article, generated functions were successfully implemented in Nonlinear Auto Regressive eXogenous (NARX) neural network as activation functions. One practical application of the designed orthogonal neural network is demonstrated through the example of control of the complex technical non-linear system – laboratory magnetic levitation system. Obtained results were compared with neural networks with standard activation functions and orthogonal functions of trigonometric shape. The proposed network demonstrated superiority over existing solutions in the sense of system performances.

Wood resource management using an endocrine NARX neural network

Planning and forecasting wood resources implies a challenging analysis, which has a direct impact on planning human resources, production timeline, as well as stock management of wooden assortments, which requires a complex data analysis taking into account all inputs that define the yield of wooden material. This paper includes an analysis of monthly time series data from 1991 to 2015 which can be characterized as long time dependence data. In recent years, artificial neural networks have become a popular tool for time dependence data treatment. Therefore, a prediction of monthly requirements of treated wood is performed by developing a new type of neural network in this research. The nonlinear autoregressive model with exogenous inputs (NARX) is used as a foundation of a new network. NARX is a type of recurrent neural network which is a very effective tool for approximation of any nonlinear function, especially ones which could occur during a nonlinear time sequence prediction. The main contribution of this paper is the introduction of an artificial endocrine factor inside the standard NARX structure. The developed ENARX model provides an extra sensitivity of the network to environmental conditions and external disturbances, as well as its improved adaptive capability. The proposed network shows better forecasting performances compared to the default NARX network, thus establishing itself as an excellent prediction tool in the field of wood science, engineering and technology.

ORTHOGONAL HYBRID-FUZZY CONTROLLERS

The main idea of this paper is to present a possibility of application of hybrid-fuzzy controllers in control systems theory. In this paper, we have described a new method оf using orthogonal functions in control of dynamical systems. These functions generate genarilzed quasi-orthogonal filter, which are used in the concluding phase of the fuzzy controllers. Proposed hybrid-fuzzy controllers of Takagi-Sugeno type has been applied to a DC servo drive system and performed experiments have verified efficiency and improvements of a new control method.