M. R. Mosavi, A. Rashidinia,
Volume 13, Issue 3 (9-2017)
Abstract
Differential Global Positioning System (DGPS) provides differential corrections for a GPS receiver in order to improve the navigation solution accuracy. DGPS position signals are accurate, but very slow updates. Improving DGPS corrections prediction accuracy has received considerable attention in past decades. In this research work, the Neural Network (NN) based on the Gaussian Radial Basis Function (RBF) has been developed. In many previous works all parameter of RBF NN are optimizing by evolutionary algorithm such as Particle Swarm Optimization (PSO), but in our approach shape parameter and centers of RBF NN are calculated in better way, in addition, search space for PSO algorithm will be reduced which cause more accurate and faster approach. The obtained results show that RMS has been reduced about 0.13 meter. Moreover, results are tabulated in the tables which verify the accuracy and faster convergence nature of our approach in both on-line and off-line training methods.
Azzedine Khati,
Volume 20, Issue 3 (9-2024)
Abstract
In this research paper, a multivariable prediction control method based on direct vector control is applied to command the active power and reactive power of a doubly-fed induction generator used into a wind turbine system. To obtain high energy performance, the space vector modulation inverter based on fuzzy logic technique (fuzzy space vector modulation) is used to reduce stator currents harmonics and active power and reactive power ripples. Also the direct vector control model of the doubly-fed induction generator is required to ensure a decoupled control. Then its classic proportional integral regulators are replaced by the multivariable prediction controller in order to adjust the active and reactive power. So, in this work, we implement a new method of control for the doubly-fed induction generator energy. This method is carried out for the first time by combining the MPC strategy with artificial intelligence represented by Fuzzy SVM-based converter in order to overcome the drawbacks of other controllers used in renewable energies. The given simulation results using Matlab software show a good performance of the used strategy, particularly with regard to the quality of the energy supplied.
Sahbasadat Rajamand, Abdulhamid Zahedi,
Volume 22, Issue 3 (9-2026)
Abstract
Noise parameters in many target tracking projects are assumed as known factors which is a main challenge because of uncertainty in measurement and state-model noise. Thus, many papers are focused on the accurate estimation of noise statistics. This paper is concentrated on this subject where it is tried to present three simple efficient methods in this regard. Estimation using n-step prediction, applying Kalman filter covariance and using Gamma distribution for noise parameters are the main concepts of the three proposed methods. Simulation results show the efficiency of all methods compared to other methods in the literature where the Gamma-distribution-based method is the most efficient work among other suggested ones in term of estimation error.
Elahe Parham, Mohamad Feshki, Alireza Fallahi, Hamid Soltanian-Zadeh,
Volume 22, Issue 3 (9-2026)
Abstract
The discovery of relationships between brain connectivity and human intelligence is of great interest. In this study, we identify structural connections correlated with human intelligence and investigate the predictability of intelligence from brain structural connectivity. The study uses data from 137 healthy subjects from the Human Connectome Project (HCP, 1200 Subjects Release). Structural connectivity was estimated using tractography derived from diffusion tensor imaging (DTI) data. A connectivity matrix was constructed using the mean fractional anisotropy (FA) of white-matter pathways between 116 regions defined by the AAL atlas. Global graph measures and correlation analysis were applied to identify connections relevant to predicting fluid intelligence (Gf) and crystallized intelligence (Gc). For prediction, three regression models of linear regression, support vector regression (SVR), and multi-layer perceptron (MLP) were employed. Most connections associated with Gf or Gc were located in the right hemisphere. Connections originating from prefrontal, right temporal, limbic, and right occipital regions were related to Gf, whereas connections originating from prefrontal, temporal, and left parietal regions were related to Gc. Among the models, SVR showed superior performance, achieving R² values of 0.45 and 0.52 for Gf and Gc, respectively. No significant relationships were observed between global graph measures and Gf or Gc scores. These findings demonstrate that DTI-based structural connectivity can be used to predict both fluid and crystallized intelligence, with fine-grained regional definitions enabling more specific connectivity patterns than in previous studies.
