COMPUTATIONAL INTELLIGENCE IN ELECTRICAL ENGINEERING (INTELLIGENT SYSTEMS IN ELECTRICAL ENGINEERING)
Year:2017 | Volume:7 | Issue:4 |Papers Count:6

In this paper, an adaptive neural fuzzy controller witch optimized with genetic algorithm is introduced for performance improvement of a fixed speed wind turbine. This controller, at first, operates as a soft starter and limits the magnitude of inrush current. This object done by control of trigger angle of switches in the converter. The switching angle will be controlled by torque and speed. Then and in normal operation, with optimized control of turbine pitch angle, the fluctuations of active power and mechanical torque will be mollification. The shaft speed and error of active power are the decision variables for this purpose. In section of fuzzy logic the membership function of outputs define as the linear function of input's membership functions. In addition to proposed method, the controller set with fuzzy logic and ANFIS algorithms. The output parameters of wind turbine under all three methods, will be analyzed and compared. It is shown the proposed method has the better response.

Prediction of chaotic time series based on the phase space reconstruction theory has been applied in many research fields. In this paper, we propose an improved adaptive neuro-fuzzy inference system (ANFIS) with self-feedback and imperialist competitive learning algorithm for the application of chaotic time series prediction. Since the ANFIS is based on a feed-forward network structure, it is limited to static problems and cannot effectively cope with dynamic properties such as the time series. To surmount this trouble, we suggested an improved version of ANFIS by introducing self-feedback connections from previous outputs that model the temporal dependence. Also we suggested a new hybrid learning algorithm based on imperialist competitive algorithm (ICA) and least square estimation (LSE) to train this new ANFIS structure. This hybrid learning algorithm is free of derivation and solves the trouble of falling in local optimum in the gradient based algorithm for training the antecedent part. The proposed approach is used to model and predict the six benchmarks of chaotic time series. Analysis of the prediction results and comparisons with recent and old studies demonstrates the promising performance of the proposed approach for modeling and prediction of nonlinear and chaotic time series.

In this paper, a new method is presented to determine the optimal setting of distance relays using probabilistic modeling of the affected uncertainties. Due to model of uncertainties, the corresponding density functions of each uncertainty are presented. Then, using Monte Carlo simulation, the probability distribution of the impedance seen by the distance relay is obtained. The three zones of distance relay need to be set in a way that the relay operates correctly for an internal fault of the protection zone (sensitivity), and it does not operate for an external fault of the protection zone (selectivity). According to this, the probabilistic indices of sensitivity and selectivity are defined independently for each zone of the distance relay. In the following, different scenarios are proposed to maximize the selectivity or sensitivity indices.Finally, due to importance of selectivity in compare to sensitivity, the scenario of maximized sensitivity with perfect selectivity is proposed.According to this scenario, the problem of determining optimum setting of distance relay for each zone is defined as an optimization problem with the objective of maximizing the probability of sensitivity and with the constraint of perfect selectivity.Since the proposed formulation is nonlinear, Genetic algorithm is used to solve this problem. The proposed method is applied to the IEEE 39 bus test system and the advantages of the proposed formulation for each zone of the distance relay are presented.

This paper describes load torque estimation (LTE) issue in induction motors (IM) with uncertainty and disturbance, using dynamic sliding mode control (DSMC). In DSMC the chattering is removed due to the integrator (or low pass filter) which is placed before the input control of the plant. In the other word, the Sign function appears in derivative of input control signal and will not appear in the input control signal of the plant. However, in DSMC the augmented system (the system plus the integrator) is one dimension bigger than the actual system and then, the plant model should be completely known. To solve this problem, a new adaptive fuzzy observer (AFO) has been proposed.The advantage of the proposed approach is to have the system controlled as well as its main task i.e. LTE, which is important in practical implementation. Simulation results are presented to demonstrate the preference of the approach.

This paper presents a nonlinear controller for a Distribution Static Compensator (DSTATCOM) of microgrids incorporating the Distributed Generation (DG) units. The nonlinear control has been designed based on partial feedback linearization theory and Proportional-Integral-Derivative (PID) controllers try to adjust the voltage and trace the outputs. This paper has proposed a combination of a fuzzy system and Bees Algorithm (BA) to optimize the parameters of the PID controllers. The results confirm that the characteristics of the response of the proposed controller (i.e. settling and rise times, the maximum overshoot and the steady-state error of the voltage step response of the DSTATCOM) is significantly improved by finding a high-quality solution. The proposed hybrid tuning method for the Partial Feedback Linearizing (PFL) controller concluded a better DC voltage regulation for the capacitor within the DSTATCOM. Furthermore, in the event of fault the proposed controller tuned by the fuzzy-BA method has shown a better performance in comparison with the conventional controller or controllers tuned by Genetic Algorithm (GA) or Particle Swarm Optimization (PSO) methods on both fault duration and after clearing times.

QoS analysis in VoIP traffic is an inevitable part in multimedia networks quality improvement. Users’ acceptable VoIP quality requirement and providers’ efforts in decreasing costs and satisfying VoIP users has led researchers study in quality improvement and resource utilization field. In this article, a new approach is introduced to improve VoIP quality adaptively and decrease consumed bandwidth as far as possible. To this end, an algorithm is proposed and some intelligent modifications are carried out in sent voice packets parameters from sender side culminating in increasing received voice quality in the receiver side and decreasing consumed bandwidth.Moreover, the proposed algorithm determines types of variations and its amount in different voice quality classes. These classes are introduced using comparison among different quality types, namely, Instantaneous Quality, Perceptual Quality and Integral Quality. Inceasing VoIP network Quality, decreasing consumed bandwidth, determining appropriate quality improvement decisions in an adaptive VoIP network are accoplishments of this article.