A system with fast torque response is very beneficial in applications where direct torque control (DTC) is highly desirable. The principle of DTC is to control machine torque and flux directly by selecting the switching states of inverter according to the errors of torque and flux; however, the response of direct torque control is slower during start-up and during change in load. Fuzzy control with grey prediction strategy is proposed for the implementation of DTC to improve its slow response. Moreover, it uses less sensors than the conventional method. Experimental implementation of the fuzzy logic controller and the grey predictor is carried out to verify the behavior of the controller. The controller is implemented with a single board computer that uses a TMS320C30 DSP.

This paper reports a new dynamic modeling method for switched reluctance motor (SRM) which can account for neighboring phase coupling effects. The dynamic model is developed using a few sets of finite element method (FEM)-computed phase flux linkage/current/position data under the condition of simultaneous excitation of two neighboring phases. The combination of piecewise Hermite cubic spline and Fourier series techniques is used to represent each FEM-calculated flux linkage data set. The methods to implement and to verify the dynamic model using simulation techniques are presented. The model is ready to be incorporated with various converter topologies and control strategies. It can provide valuable information for motor design evaluation and its controller design evaluation. The modeling method has been implemented using Saber circuit simulator for a 4-phase SRM driven by a split DC-link inverter. Dynamic simulation results have been obtained for two different SRM winding configurations. The significance of its neighboring phase coupling effects is evaluated by comparing the simulation results of a model including the effects with the simulation results of a degraded model neglecting the effects.

Using wind turbine generators (WTG) as an alternative supply in a rural distribution system will affect both the mean values and the probability distributions of system reliability indices due to the fluctuating characteristics of wind speed. This paper investigates the impacts on the reliability indices and their probability distributions of using WTG as an alternative supply. A time sequential simulation technique is used to simulate the time-varying nature of wind speed and the random failure of system elements. The probability distributions of basic load point and system reliability indices for a modified rural distribution system are presented. The effects of WTG parameters on the index probability distributions are discussed.

Turn-to-turn fault detection is based on the principle that symmetrical (unfaulted) motors powered by symmetrical multiphase voltage sources will have no negative sequence currents flowing in the leads. A turn-to-turn fault will break that symmetry and give rise to a negative sequence current which may then be used as a measure of fault severity or to initiate protective action, such as a circuit breaker trip. A new way of looking at the effects of turn faults has been developed that improves sensitivity andspeedwhile reducing the probability of misdetection, taking into account voltage balance, load or voltage variation, and instrument errors. The method has been implemented on a PC and tested, in real time, with controllable faults in a small motor. Reliable detection of one shorted turn out of 648 turns per phase (in a Y-connected motor) was demonstrated with the fault indicator becoming fully developed in two cycles of line frequency after initiation of the fault.

The technique of calculating the air gap torque of an induction motor utilizing only the terminal voltages and currents has been known for some time; however implementation of this technique in a commercial instrument that must compete in accuracy with electromechanical instruments raises a number of issues that must be resolved. Because the derivation of the theorem relies on assumptions of symmetry in the motor, questions have been raised as to its accuracy in real machines. Also, the effects of saturation and core losses usually are ignored but are always present, especially in '' commodity''-type motors. There also are issues associated with the way data are processed, especially in the presence of strong harmonics. Finally, there is the question of how the system can be put into the field using only the information available to the user on the motor name plate. To clarify these issues, an experiment was set up using a PC equipped with standard A/D boards and a common 3/4 HP induction motor. The algorithm was implemented in C code on the PC. Upon the conclusion of these tests, the code was ported to a switchboard instrument and tested using a 15 HP motor. Results in both platforms were good, and the switchboard implementation has been adopted into the standard motor protection relay product.

A current controller composed of a current regulator associated to a parameter adaptation scheme is proposed. The current regulator considers the inherent coupled structure of the induction machine. The adaptive block is coupled to this regulator in order to carry out a high performance current control with low sensitivity to parameter variations. It is shown that this controller can also be used to estimate the machine saturation characteristic and the rotor time constant as part of an indirect vector control autocommissioning procedure. A discrete time approach is utilized for the design considering microprocessor implementation.

This paper presents a new technique to solving the economic dispatch problem based on a reduced order model of the original system. Loads are first modeled by their appropriate voltage-dependent models as load admittances. Load admittances are then added to the bus admittance matrix and their respective buses are eliminated. The obtained model is a reduced model of the original system. The admittance matrix is of the same order as the number of voltagecontrolled buses in the system. The variables of the reduced model are the voltage-controlled buses’ angles and active power generations. Newton Raphson method is used to calculate the angles and active generations of the reduced model while minimizing the operational cost. Load bus voltages and angles of the original system are then calculated by a direct method and load admittances are modified. The process is repeated until convergence is achieved. The simulation is carried out on IEEE 118 bus test system. A comparison between the new approach and the penalty factors method has been made. It is shown that operational cost is improved and solution time is significantly reduced when compared to the penalty factors method of economic dispatch.

In this paper an improved neural network application for short-term load forecasting purposes is presented. To speed up the learning process on one side, and not to jeopardize the stability performance of the learning process on the other side, the adaptive approach to the learning-rate parameter has been employed. Also, instead of learning overall load characteristics, the preprocessing of input data has been designed with the idea to learn only load demand behavior that is important for a certain period. The proposed neural network has shown good performance, even in the case of the incomplete data temperature set and at high irregularities in weekly load data.

This paper presents the Load Increase Index, which is a new computationally efficient loadability limit index derived from the continuation power flow method. The continuation power flow is a well-known procedure for identifying the maximum loading point of a static power system network. The proposed index is a byproduct of the continuation power flow, and may be obtained with little extra computational effort while yielding considerable information about the distance to the critical loading point. This paper also presents the mathematical relationship of the proposed index to two other commonly used voltage stability indices: the smallest singular value of the loadflow Jacobian and the determinant of the loadflow Jacobian. The proposed index has been verified on numerous test systems of various sizes.

Network reconfiguration for load balancing is the determination of switching-options for a particular set of loads on a distribution system that improves load balancing the most and is performed altering the topological structure of distribution feeders by changing open/closed status of sectionalizing and tie-switches. By transferring loads from the heavily loaded feeders to the lightly loaded ones, network reconfiguration can balance feeder loads and alleviate overload conditions of a network. In this paper a novel method is proposed by formulating an algorithm to reconfigure distribution networks for load balancing. Systematic logical approach is used to determine the feasible switching combinations, select the status of the switches, and find the best combination of switches for maximum improvement in load balancing. The algorithm has been developed in two stages. In the first stage, a limited number of switching combinations is formed by trinary logic principle. A simultaneous switching approach is applied logically to find the best combination by selecting the appropriate switching options, connecting the switches accordingly, and assigning the loads for the newly formed combinations. In the next stage, the search is extended to investigate whether any other switching combination gives rise to maximum improvement in load balancing, compared to the improvement obtained in the first stage. This is carried out by considering one after another, the branches next to the open-branches of the best configuration, and the switching configuration that gives the maximum improvement in load balancing is identified. The proposed method has been tested on a 69-bus test system, and the test results indicate that the algorithm proposed is able to determine the appropriate switching-options of the optimal (or near optimal) configuration with less computation. This paper proposes a scheme to implement the simultaneous switching algorithm for real-time on-line control of three-phase distribution networks, with symmetrical loading per phase.