Proposed System Identification Approach Research Articles

Online non-affine nonlinear system identification based on state-space neuro-fuzzy models

This paper proposes a new general recurrent state-space neuro-fuzzy model structure. Three topologies are under assessment, including the state-input recurrent neuro-fuzzy system, the series-parallel recurrent neuro-fuzzy system and the parallel recurrent neuro-fuzzy system. Moreover, the underlying generalised state-space Takagi–Sugeno system is proven to be a universal approximator, and some stability conditions derived for this system. The online training is carried out based on a constrained unscented Kalman filter, where weights, membership functions and consequents are recursively updated. Results from experiments on a benchmark MIMO system demonstrate the applicability and flexibility of the proposed system identification approach.

An efficient numerical and experimental system identification approach for a flexible manipulator

Purpose – The purpose of this paper is to propose an efficient and straightforward approach for system identification of a rotating single link flexible manipulator (RSLFM). Also, the achieved results are experimentally validated through identification procedure. Design/methodology/approach – The proposed system identification approach is applied to a RSLFM with a tip mass. At first, the dynamic model of the system is derived using Lagrange method. Then, an efficient method is developed for identification of such a system. This method facilitates the nonlinear complicated identification problem of the RSLFM to a simplified root finding problem. Findings – The main advantage of the developed method is to convert a complicated system identification process to a simple nonlinear equation solution. This approach uses small-size input/output data set and requires a short-time interval of data acquisition, which gives important advantages in lower computational load and lower execution time. The investigated approach is studied on experimental system identification of a single link flexible manipulator. To demonstrate this fact, the developed method is successfully applied in identification of two other mechanical systems; the inverted pendulum on a cart and the ball and beam apparatus. Originality/value – In this work, the proposed identification approach has been originally applied to a RSLFM and two other mechanical examples. All obtained identification results show the performance and applicability of the developed method clearly. This approach is not restricted in using state space or transfer function. It has significant superiority in comparison with other known approaches including autoregressive with exogenous input (ARX) and Box-Jenkins (BJ).

A new approach for online T-S fuzzy identification and model predictive control of nonlinear systems

This paper proposes a new unsupervised fuzzy clustering algorithm (NUFCA) to construct a novel online evolving Takagi–Sugeno (T-S) fuzzy model identification method and an adaptive predictive process control methodology. The proposed system identification approach consists of two main steps: antecedent T-S fuzzy model parameters identification and consequent parameters identification. The NUFCA combines the K-nearest neighbour and fuzzy C-means methods into a fuzzy modelling method for partitioning of the input–output data and identifying the antecedent parameters of the fuzzy system; then the recursive least squares method is exploited to obtain initialization type consequent parameters and to construct a method for on-line fuzzy model identification. The integration of the proposed adaptive identification method with the generalized predictive control results in an effective adaptive predictive fuzzy control methodology. For better demonstration of the robustness and efficiency of the proposed methodology, it is applied to the identification of a model for the estimation of the flour concentration in the effluent of a real-world wastewater treatment plant (WWTP); and to control a simulated continuous stirred tank reactor (CSTR), and a real experimental setup composed of two coupled DC motors. The results show that the developed evolving T-S fuzzy model methodology can identify nonlinear systems satisfactorily and can be successfully used for a prediction model of the process for the generalized predictive controller. It is also shown that the algorithm is robust to changes in the initial parameters, and to unexpected disturbances.

A new design method based on firefly algorithm for IIR system identification problem

In this paper a population based evolutionary optimization methodology called firefly algorithm (FFA) is applied for the optimization of system coefficients of the infinite impulse response (IIR) system identification problem. FFA is inspired by the flash pattern and characteristics of fireflies. In FFA technique, behaviour of flashing firefly towards its competent mate is structured. In this algorithm attractiveness depends on brightness of light and a bright firefly feels more attraction for the brighter one. For this optimization problem, brightness varies inversely proportional to the error fitness value, so the position of the brightest firefly gives the optimum result corresponding to the least error fitness in multidimensional search space. Incorporation of different control parameters in basic movement equation results in balancing of exploration and exploitation of search space. The proposed FFA based system identification approach has alleviated from inherent drawbacks of premature convergence and stagnation, unlike genetic algorithm (GA), particle swarm optimization (PSO) and differential evolution (DE). The simulation results obtained for some well known benchmark examples justify the efficacy of the proposed system identification approach using FFA over GA, PSO and DE in terms of convergence speed, identifying plant coefficients and mean square error (MSE) fitness values produced for both same order and reduced order models of adaptive IIR filters.

IIR system identification using differential evolution with wavelet mutation

In this paper an improved version of Differential Evolution (DE) technique called Differential Evolution with Wavelet Mutation (DEWM) is applied to the infinite impulse response (IIR) system identification problem. Instead of fixed value of scaling factor in standard DE, an iteration dependent scaling factor governed by the wavelet function during the mutation process is adopted in the proposed technique. This modification in the mutation process ensures not only the faster searching in the multidimensional search space but also the solution produced is very close to the global optimal solution. Apart from this, the proposed technique DEWM has alleviated from inherent drawbacks of premature convergence and stagnation, unlike Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The simulation results obtained for some well known benchmark examples justify the efficacy of the proposed system identification approach using DEWM over GA, PSO and DE in terms of convergence speed, plant coefficients and mean square error (MSE) values produced for both the same order and reduced order models of adaptive IIR filters.

A novel social emotional optimisation algorithm for IIR system identification problem

In this paper an evolutionary optimisation methodology based on social emotional optimisation algorithm (SEOA) is applied to the infinite impulse response (IIR) system identification problem. In SEOA methodology, behaviour of human beings for achieving higher social status in society is structured. In this virtual world, the individual with the highest rank in society gives the optimal solution in multidimensional search space. Earning the highest social status by means of cooperation and competition with others not only results in better exploration and exploitation of problem space but also ensures faster convergence to optimal solution. The proposed SEOA based system identification approach has resolved the inherent drawbacks of premature convergence and stagnation, unlike genetic algorithm (GA), particle swarm optimisation (PSO) and differential evolution (DE). The simulation results obtained for some well known benchmark examples justify the efficacy of the proposed system identification approach using SEOA over GA, PSO and DE in terms of convergence speed, unknown plant coefficients and mean square error (MSE) values produced for both the same order and reduced order models of adaptive IIR filters.

Collective Animal Behaviour Based Optimization Algorithm for IIR System Identification Problem

In this paper a novel optimization technique which is developed on mimicking the collective animal behaviour (CAB) is applied to the infinite impulse response (IIR) system identification problem. Functionality of CAB is governed by occupying the best position of an animal according to its dominance in the group. Enrichment of CAB with the features of randomness, stochastic and heuristic search nature has made the algorithm a suitable tool for finding the global optimal solution. The proposed CAB has alleviated from the defects of premature convergence and stagnation, shown by real coded genetic algorithm (RGA), particle swarm optimization (PSO) and differential evolution (DE) in the present system identification problem. The simulation results obtained for some well known benchmark examples justify the efficacy of the proposed system identification approach using CAB over RGA, PSO and DE in terms of convergence speed, unknown plant coefficients and mean square error (MSE) values produced for IIR system models of both the same order and reduced order.

Craziness based particle swarm optimization algorithm for IIR system identification problem

In this paper a variant of particle swarm optimization (PSO), called craziness based particle swarm optimization (CRPSO) technique is applied to the infinite impulse response (IIR) system identification problem. A modified version of PSO, called CRPSO adopts a number of random variables for having better and faster exploration and exploitation in multidimensional search space. Incorporation of craziness factor in the basic velocity expression of PSO not only brings diversity in particles but also ensures convergence to optimal solution. The proposed CRPSO based system identification approach has alleviated from the inherent drawbacks of premature convergence and stagnation, unlike real coded genetic algorithm (RGA), particle swarm optimization (PSO) and differential evolution (DE). The simulation results obtained for some well known benchmark examples justify the efficacy of the proposed system identification approach using CRPSO over RGA, PSO and DE in terms of convergence speed, unknown plant coefficients and mean square error (MSE) values produced for both the same order and reduced order models of adaptive IIR filters.

Relative Transfer Function Identification Using Speech Signals

An important component of a multichannel hands-free communication system is the identification of the relative transfer function between sensors in response to a desired source signal. In this paper, a robust system identification approach adapted to speech signals is proposed. A weighted least-squares optimization criterion is introduced, which considers the uncertainty of the desired signal presence in the observed signals. An asymptotically unbiased estimate for the system's transfer function is derived, and a corresponding recursive online implementation is presented. We show that compared to a competing nonstationarity-based method, a smaller error variance is achieved and generally shorter observation intervals are required. Furthermore, in the case of a time-varying system, faster convergence and higher reliability of the system identification are obtained by using the proposed method than by using the nonstationarity-based method. Evaluation of the proposed system identification approach is performed under various noise conditions, including simulated stationary and nonstationary white Gaussian noise, and car interior noise in real pseudo-stationary and nonstationary environments. The experimental results confirm the advantages of proposed approach.