Cheng-Chien Kuo

Modified particle swarm optimization algorithm with simulated annealing behavior and its numerical verification

The hybrid algorithm that combined particle swarm optimization with simulated annealing behavior (SA-PSO) is proposed in this paper. The SA-PSO algorithm takes both of the advantages of good solution quality in simulated annealing and fast searching ability in particle swarm optimization. As stochastic optimization algorithms are sensitive to their parameters, proper procedure for parameters selection is introduced in this paper to improve solution quality. To verify the usability and effectiveness of the proposed algorithm, simulations are performed using 20 different mathematical optimization functions with different dimensions. The comparative works have also been conducted among different algorithms under the criteria of quality of the solution, the efficiency of searching for the solution and the convergence characteristics. According to the results, the SA-PSO could have higher efficiency, better quality and faster convergence speed than compared algorithms.

Partial discharge pattern recognition of power cable joints using extension method with fractal feature enhancement

Highlights? The proposed method has high recognition rate and provides fast recognition speed. ? The PD defect types can be directly identified by the degree of correlation. ? The fractal theory could extract the important features from PD 3D patterns. ? This technology can application in power station partial discharge detection. This paper proposes a new partial discharge (PD) pattern recognition using the extension method with fractal feature enhancement. First, four common defect types of XLPE power cable joints are established, and a commercial PD detector is used to measure the PD signal by inductive sensor (L-sensor). Next, the feature parameters of fractal theory (fractal dimension and lacunarity) are extracted from the 3D PD patterns. Finally, the matter-element models of the PD defect types are built. The PD defect types can be directly identified by the degree of correlation between the tested pattern and the matter-element based on the extension method. The extension method needs representative features to define the interval of the matter-element. In order to enhance the extension performance, we add fractal features that are extracted from the PD 3D patterns. To demonstrate the effectiveness of the extension method with fractal feature enhancement, the identification ability is investigated on 120 sets of field-tested PD patterns of XLPE power cable joints. Compared with the back-propagation neural network (BPNN) method, the results show that the extension method with fractal feature enhancement not only has high recognition accuracy and good tolerance when random noise is added, but that it also provides fast recognition speed.

Gas-insulated switchgear PD signal analysis based on Hilbert-Huang transform with fractal parameters enhancement

This study proposes a novel method of partial discharge (PD) electrical signal analysis based on the Hilbert-Huang transform (HHT) with fractal feature enhancement. Firstly, this study establishes four defect types of 15 kV gas-insulated switchgear (GIS) and uses a commercial high-frequency current transformer (HFCT) to measure the electrical signals caused by the PD phenomenon. Second, the authors applied HHT for the PD electrical signal process. The HHT can represent instantaneous frequency components through empirical mode decomposition (EMD) and then transform to a 3D Hilbert energy spectrum. Finally, this study extracts the fractal parameters from the 3D energy spectrum and uses a neural network (NN) for PD recognition. To demonstrate the effectiveness of the proposed method, this study uses 160 sets of field-tested PD patterns generated by GIS, and then compares the recognition rate of the signal with and without the EMD process. The result shows that the proposed method can easily separate various defect types. The method can also be employed by the construction unit to verify the GIS quality and determine the GIS insulation status.

Artificial recognition system for defective types of transformers by acoustic emission

An artificial recognition system of defective types for epoxy-resin transformers through acoustic emission (AE) from partial discharge (PD) experiment is proposed. PD detection is an efficient diagnosis method to prevent the failure of electric equipments arising from degrading insulation. However, most of the PD detection methods could be performed only at the shutdown period of equipments. By using AE, the online and real-time detection with defective types could be easily reached. Therefore, in this paper a series of high voltage tests were conducted on pre-faulty transformers to collect the AE signals for recognition system needed. The selected AE features instead of waveform are then extracted from these experimental AE signals for the input characteristic of recognition system. According to these features, effective identification of their defective types can be done using the proposed recognition system that combined particle swarm optimization with an artificial neural network. To demonstrate the effectiveness and feasibility of the proposed approach, the artificial recognition system is applied on both noisy and noiseless circumstances. The experiment showed encouraging results that even with 30% noise per discharge count, an 80% successful recognition rate can still be achieved.

Artificial identification system for transformer insulation aging

An artificial identification system to classify the insulation aging status of cast-resin transformer through current impulse method of partial discharge (PD) is proposed. The aging phenomenon of insulation materials belongs to a natural property and has strongly influences with the stability of power systems. Therefore, an effectively insulating identification technology plays an important role to enhance the system operating reliability. Since PD is a well known evidence of insulation degrading, a series of high voltage test with acceleration aging process to collect PD signals for identification system are conducted. Some selected statistical PD features instead of waveform are then extracted from these experimental PD signals as input data of the identification system. Also, an artificial neural network that combined particle swarm optimization is presented as the effectively identification tool. To demonstrate the effectiveness and feasibility of the proposed approach, the artificial identification system is applied on both noisy and noiseless circumstance. The experiment showed promising results with over 94% identification rate and even with 30% noise per discharge signal, an 85% successful identification rate can still be achieved.

A NOVEL VALIDITY INDEX FOR THE SUBTRACTIVE CLUSTERING ALGORITHM

This paper proposes a new validity index for the subtractive clustering (SC) algorithm. The subtractive clustering algorithm proposed by Chiu is an effective and simple method for identifying the cluster centers of sampling data based on the concept of a density function. The SC algorithm continually produces the cluster centers until the final potential compared with the original is less than a predefined threshold. The procedure is terminated when there are only a few data points around the most recent cluster. The choice of the threshold is an important factor affecting the clustering results: if it is too large, then too few data points will be accepted as cluster centers; if it is too small, then too many cluster centers will be generated. In this paper, a modified SC algorithm for data clustering based on a cluster validity index is proposed to obtain the optimal number of clusters. Six examples show that the proposed index achieves better performance results than other cluster validities do.

Generation Dispatch Strategy under Environment Protection Consideration

An optimal generation dispatch strategy method while considering CO2 emissions is presented in this paper. Firstly, different CO2 equivalent models for fuel generated thermal units are derived. Secondly, the Cost-CO2 emission Tradeoff Curve (CCTC) and Incremental Cost for CO2 Reduction (ICCR) curve, are introduced to assess the impact of CO2 emissions on power generation costs. Incorporating these two indexes, the single-objective and bi-objective programming methods are employed to establish a generation dispatch strategy. The salient feature of the strategy allows the decision-maker to consider both environmental and economic factors at the same time. Since generation dispatch belongs to large-scale non-linear planning problems, this study employs a particle swarm approach with novel coding scheme for generation dispatch solution, which can quickly find an optimal solution with higher probability. Finally, taking the 27-unit system from Taiwan power company as an example, simulation result of the study reveals that the proposed dispatch strategy can effectively take into consideration CO2 reduction impact on cost during off-peak, half-peak and peak load conditions to achieve environmental protection and economic purpose.

The Reactive Power and Voltage Control of Distribution Systems Using the Normalized Weighting Method

A new multi-objective formulation named normalized weighting method for the reactive power and voltage control in a distribution system is proposed. Four important objectives include reactive power, feeder loss, voltage deviation and voltage profile are all considered in this paper. These objectives are almost equaled important for electric utility companies, but they are somewhat non-commensurable with each other. In view of this, a normalized weighting method for the multi-objective problem is proposed. It can provide a set of flexible solutions by following the intention of decision makers. To increase the realism for practical use, the load and operating constraints of the system are all considered. Also, the comparative studies on the actual Tai-power system are given to demonstrate the effectiveness on the dispatch scheduling about capacitors and ULTC in a distribution system. The results have shown that the proposed method can provide a promising schedule for the distribution system.

Partial discharge measurement and pattern recognition in gas insulated switchgear

Gas insulated switchgear (GIS) has been widely used as main switching equipment in the power substation, because it has good insulation and high reliability. However, if the switchgear is with defect or beyond the server life, the ability of the insulation may be degradation. Therefore, the detecting technology and fault diagnostic method on GIS has become more and more importance. In this paper, four common defect types of 15 kV switchgears are established, and a commercial inductive sensor measures the current signal caused by the partial discharge (PD) phenomenon. Next transfer this current signal into 3D PD patterns, through these PD patterns we can find the differences characteristic among these testing models. Finally features extract and pattern recognition based on neural network (NN). This detection could determine types of failure and so assist in utility maintenance.

A reduced data set method for support vector regression

Support vector regression (SVR) has been very successful in pattern recognition, text categorization, and function approximation. The theory of SVR is based on the idea of structural risk minimization. In real application systems, data domain often suffers from noise and outliers. When there is noise and/or outliers exist in sampling data, the SVR may try to fit those improper data, and obtained systems may have the phenomenon of overfitting. In addition, the memory space for storing the kernel matrix of SVR will be increment with O(N^2), where N is the number of training data. Hence, for a large training data set, the kernel matrix cannot be saved in the memory. In this paper, a reduced support vector regression is proposed for nonlinear function approximation problems with noise and outliers. The core idea of this approach is to adopt fuzzy clustering and a robust fuzzy c-means (RFCM) algorithm to reduce the computational time of SVR and greatly mitigates the influence of data noise and outliers.