Fu-Sheng Cheng

Nonconvex Economic Dispatch by Integrated Artificial Intelligence

This paper presents a new algorithm by integrating evolutionary programming (EP), tabu search (TS), and quadratic programming (QP) methods to solve the nonconvex economic dispatch problem (NED). A hybrid EP and TS were used for quality control and the Fletchers quadratic programming technique was used for solving. EP and TS determine the segment of a cost curve used, which is piecewise quadratic natured. Operation constraints are modeled as linear equality or inequality equations, resulting in a typical QP problem. Fletchers QP was chosen to enhance the performance. The fitness function is constructed from priorities without penalty terms. Numerical results show that the proposed method is more effective than other previously developed evolutionary computation algorithms.

Detection and Classification of Multiple Power-Quality Disturbances With Wavelet Multiclass SVM

This paper presents an integrated model for recognizing power-quality disturbances (PQD) using a novel wavelet multiclass support vector machine (WMSVM). The so-called support vector machine (SVM) is an effective classification tool. It is deemed to process binary classification problems. This paper combined linear SVM and the disturbances-versus-normal approach to form the multiclass SVM which is capable of processing multiple classification problems. Various disturbance events were tested for WMSVM and the wavelet-based multilayer-perceptron neural network was used for comparison. A simplified network architecture and shortened processing time can be seen for WMSVM.

Classification of Multiple Power Quality Disturbances Using Support Vector Machine and One-versus-One Approach

This paper presents a classifier for recognizing power quality disturbances (PQD) problem. The so called support vector machine (SVM) is an effective classification tool, but it can only process binary classification problems. This paper integrated SVM and the one-versus-one (OVO) approach to form the OVO-based SVM (OSVM) which can process the multiple classification problem such as PQD. Using the proposed methodology can reduce a great quantity of the training data, less memory space and computing time are required. With IEEE 14-bus power system, seven power quality disturbing events were tested and compared with artificial neural network (ANN). The simulation results were conducted to show the shortened processing time and effectiveness of the proposed approach.

Operation strategy of cogeneration systems under environmental constraints

This paper studies the economical operation of cogeneration systems under emission constraints. It attempts to control the production of atmospheric emissions such as NO/sub x/ and SO/sub x/ caused by the operation of fossil-fueled thermal generation. The objective of this paper includes fuel cost, population cost and tie-line energy cost, subject to fuel mix ratios in a boiler, operational constraints and emission condition. Evolutionary programming was adopted to solve this problem. The steam and fuel ratios of boilers and generation output will be obtained by considering the time-of-use (TOU) dispatch between cogeneration systems and utility companies. With the introduction of optimization methods, one can explore the potential for operational changes in unit commitment and dispatching to achieve minimal cost while complying with environmental standards.

A hybrid programming for distribution reconfiguration of dc microgrid

This paper presents a hybrid programming (HP) technique to solve the dc microgrid reconfiguration problem for loss reduction and service restoration. By using the proposed algorithm, a more efficient network configuration can be obtained to reduce loss. The problem is optimized in a stochastic searching manner similar to that of the evolutionary programming (EP). The initial population is determined by opening the switches with the lowest current in every mesh derived in the optimal power flow (OPF) with all switches closed. To avoid prematurity, HP technique was applied to adjust the number of mutant elements adaptively. Tabu Lists with heuristic rules were employed in the searching process to enhance performance. Simulation results show that the proposed algorithm has advantages than the earlier developed algorithms. The optimization strategy is general and can be used to solve other power system optimization problems as well.

MRAS-based sensorless wind energy control for wind generation system using RFNN

This paper presents an analysis of a high-performance model reference adaptive system (MRAS) observer for the sensorless control of a induction generator (IG). The sensorless control is based on a model reference adaptive system observer for estimating the rotational speed. The proposed output maximization control is achieved without mechanical sensors such as wind speed or position sensor, and the new control system will deliver maximum electric power with light weight, high efficiency, and high reliability. The concept has been developed and analyzed using a turbine directly driven IG. The estimation of the rotor speed is on the basis of the MRAS control theory. A sensorless vector-control strategy for an IG operating in a grid-connected variable speed wind energy conversion system is presented.

Application of Fuzzy Neural Network Sliding Mode Controller for Wind Driven Induction Generator System

An induction generator (IG) speed drive with the application of a sliding mode controller and a proposed fuzzy neural network (FNN) controller is introduced in this paper. Grid connected wind energy conversion system (WECS) present interesting control demands, due to the intrinsic nonlinear characteristic of wind mills and electric generators. The FNN torque compensation is feedforward to increase the robustness of the wind driven induction generator system. A multivariable controller is designed to drive the turbine speed to extract maximum power from the wind and adjust to the power regulation. Moreover, a sliding mode speed controller is designed based on an integral-proportional (IP) sliding surface. When sliding mode occurs on the sliding surface, the control system acts as a robust state feedback system.

An improved evolutionary programming approach for distribution loss reduction by feeder switching

This paper proposed an improved evolutionary programming (IEP) technique to minimize distribution feeder losses. In the minimization process, switches were indicated by pre-assigned integers. In this paper, the authors focused on the minimization of the real power loss, as well as on voltage and current constraints subjected to the radial network structure. Many tests were conducted to show its effectiveness.

Design of an adaptive intelligent control scheme for switched reluctance wind generator

Purpose This paper aims to present the analysis, design and implementation of functional link-based recurrent fuzzy neural network (FLRFNN) for the control of variable-speed switched reluctance generator (SRG). Design/methodology/approach The node connecting weights of the FLRFNN are trained online by back-propagation (BP) algorithms. The proposed estimator requires less processing time than traditional methods and can be fully implemented using a low-cost digital signal processor (DSP) with MATLAB toolboxes. The DSP-based hybrid sensor presented in this paper can be applied to a wind energy-conversion system where the SRG is used as a variable-speed generator. The current transducer is used to monitor the energized current and proximity sensors for rotor salient. Findings The authors have found that optimal based on FLRFNN with Grey controller can resolve the regulation of the system with uncertainty model and unknown disturbances. This technique can maintain the system stability and reach the desired performance even with parameter uncertainties. Originality/value This design will improve the performance of SRG to operate more smoothly. This application is currently being studied because the SRG has well-known advantages such as robustness, low manufacturing cost and good size-to-power ratio. Performance of the proposed controller can offer better stability characteristics. Finally, the SRG has a very good efficiency in the whole operating range.