---

A non-linear finite element model could be a useful tool in the development of a method of predicting soil pressure-sinkage behaviour, and can be used to investigate and analyze soil compaction. This study was undertaken to emphasize that the finite element method (FEM) is a proper technique to model soil pressure-sinkage behaviour. For this purpose, the finite element method was used to model soil pressure-sinkage behaviour and a two-dimensional finite element program was developed to perform the required numerical calculations. This program was written in FORTRAN. The soil material was considered as an elastoplastic material and the Mohr-Coulomb elastoplastic material model was adopted with the flow rule of associated plasticity. In order to deal with material non-linearity, incremental method was adopted to gradually load the soil and a total Lagran-gian formulation was used to allow for the geometric non-linear behaviour in this study. The FEM model was verified against previously developed models for one circular footing problem and one strip footing problem and the finite element program was used to pre-dict the pressure-sinkage behaviour of the footing surfaces. Statistical analysis of the veri-fication confirmed the validity of the finite element model and demonstrated the potential use of the FEM in predicting soil pressure-sinkage behaviour. However, experimental verification of the model is necessary before the method can be recommended for exten-sive use.

---

This paper introduces a novel approach to improve performance of speech recognition systems using a combination of features obtained from speech reconstructed phase space (RPS) and frequency domain analysis. By choosing an appropriate value for the dimension, reconstructed phase space is assured to be topologically equivalent to the dynamics of the speech production system, and could therefore include information that may be absent in analysis approaches based on linear methods. Also, complicated systems such as speech production system can present cyclic and oscillatory patterns and Poincare sections could be used as an effective tool in analysis of such trajectories. In this research, a statistical modeling approach based on Gaussian Mixture models (GMM) was applied to the Poincare sections of speech RPS. The final feature set is obtained from a feature selection stage omong parameters of GMM model and the usual Mel Frequency Cepstral coefficients (MFCC). An HMM-based speech recognition system and the TIMIT speech database are used to evaluate performance of the proposed feature extraction system for isolated and continuous speech recognition. Experiments represent about 5.7% absolute isolated phoneme recognition accuracy improvement in isolated phoneme recognition performance. The new approach is shown to be a viable and effective alternative to traditional feature extraction methods, particularly for signals such as speech with strong nonlinear characteristics.

---

The attenuation of mechanical load is one of the most effective approaches in wind turbine components cost reduction, and improving the control system reduces mechanical loads with minimum effort. In modern wind turbines, electrically-excited synchronous generators are mostly applied in direct-drive structure. In current research, generator field voltage along with the blade pitch angle is employed for tower load reduction in a novel multivariable-adaptive control structure. The controller is designed based on the extracted model with aerodynamic, vibratory and electrical interactions. The centralized multivariable structure is chosen to simultaneously reduce rotor speed fluctuations and tower vibrations. Since the nonlinear wind turbine model is complex, the controller is designed via optimization process. The nonlinear aerodynamic behavior of blades influences the closed-loop performance in different operating condition; therefore controller is adapted to the condition by employing gain-scheduling method. The effects of signal noise, digital control and higher-order dynamics of electrical system might defect the closed-loop stability. The designed controller is implemented on a wind turbine simulator which includes the before-mentioned effects. By comparing the performance of the multivariable adaptive controller with a two input-one output multivariable controller, it is proven that the mechanical loads acting on tower have been greatly decreased.