Ying-Tung Hsiao

Ant colony optimization for designing of PID controllers

This work presents an optimum approach to design PID controllers. The primary design goal is to obtain good load disturbance response by minimizing the integral absolute control error. At the same time, the transient response is guaranteed by minimizing the maximum overshoot, settling time, rise time of step response. This study proposes a solution algorithm based on the ant colony optimization technique to determine the parameters of the PID controller for getting a well performance for a given plant. Simulation results demonstrate that better control performance can be achieved in comparison with known methods

Ant colony optimization for best path planning

The paper presents an optimal approach to search the best path of a map considering the traffic loading conditions. The main objective of this work is to minimize the path length to get the best path planning for a given map. This study proposes a solution algorithm based on the ant colony optimization technique to search the shortest path from a desired origin to a desired destination of the map. The proposed algorithm is implemented in C++. Furthermore, the simulation program can randomly generate maps for evaluating its flexibility and performance. Simulation results demonstrate that the proposed algorithm can obtain the shortest path of a map with fast speed.

A contour based image segmentation algorithm using morphological edge detection

In this paper, a novel approach for edge-based image segmentation is proposed. Image segmentation and object extraction play an important role in supporting content-based image coding, indexing, and retrieval. However, its always a tough task to partition an object in a graph-based image. We proposed an image segmentation algorithm by integrating mathematical morphological edge detector with region growing technique. The images are first enhanced by morphological closing operations, and then detect the edge of the image by morphological dilation residue edge detector. Moreover, we deploy growing seeds into the edge image that obtained by the edge detection procedure. By cross comparing the growing result and the detected edges, the partition lines of the image are generated. In this paper, we presented the theoretical backgrounds and procedure illustrations of the proposed algorithm. Furthermore, the proposed algorithm is implemented in C++ language and evaluate on several images with promising results.

A novel optimization algorithm: space gravitational optimization

A new concept for the optimization of nonlinear functions is proposed. For most of the proposed evolutionary optimization algorithms, such as particle swarm optimization and ant colony optimization, they search the solution space by sharing known knowledge. The proposed algorithm is based on the Einsteins general theory of relativity, which we utilize the concept of gravitational field to search for the global optimal solution for a given problem. In this paper, detail procedure of the proposed algorithm is introduced. The proposed algorithm has been tested on an application that is known difficult with promising and exciting results.

Robust multiple objects tracking using image segmentation and trajectory estimation scheme in video frames

Abstract In this paper, a novel image segmentation and a robust unsupervised video objects tracking algorithm are proposed. The proposed method is able to track complete object regions in a sequence of video frames. In this work, object tracking is achieved by analysing the movement of the contours with frame by frame in the video stream. The proposed algorithm involves with three major components for analysing the shapes and motions of the object in the video frames. First, a modified mathematical morphology edge detection algorithm is utilized to extract the contour features in the video frames. Then, a contour-based image segmentation algorithm is proposed and applied to the contour features for partitioning the predetermined target objects in the video frames. Finally, a trajectory estimation scheme is developed to handle the movements of the objects in the video frames. The proposed image segmentation algorithm is capable of automatically partitioning the predetermined objects. The proposed tracking algorithm is also robust against overlapping and videos acquired by non-stationary cameras. The experimental results show that the proposed algorithm can precisely partition and track the predetermined objects in video frames.

An Adaptive PMU-based Fault Location Estimation System with a Fault-Tolerance and Load-Balancing Communication Network

Modern fault detection/location technique for an EHV/UHV transmission network usually works based on the data measured by Phaser Measurement Units (PMU). The synchronized voltage and current phasors measured by PMU are transmitted to a monitoring center for analysis. Global Positioning System (GPS) receivers are also equipped with PMU, which is called GPS-PMU, to increase the accuracy of fault detection/location by tagging all of the measured data with time stamps. Once a fault was occurred in the transmission network, the time of those measured data transmitted to the monitoring center is crucial. Therefore, a high-quality communication network is required to reduce the response time of the fault detection/location algorithm. In this study, an evolutionary routing algorithm is developed to handle guarantee the minimal data transmission delay, and also robust against faults in communication system itself. The proposed routing algorithm has been tested through two kinds of experimental simulations, and the result shows that the proposed algorithm can provide minimal transmission delay by balancing the traffic over the communication network, and while the network topology has been changed, the proposed routing algorithm can adapt to the new topology in a very short time without seriously affect the response time of the transmission network fault detection/location algorithm.

Computer network load-balancing and routing by ant colony optimization

A high efficient design of computer network is an important issue for the high transmission speed requirement of today. In computer network, the data packages have to be transmitted to the destination with a minimum delay for ensuring the quality of service guarantees. This work presents an algorithm to perform a dynamic load-balancing for transmitting the data packages with near minimum delays in the interconnection networks. The proposed algorithm is based on the ant colony optimization algorithm inspired by the simple behavior of biological ants. This work utilizes a cube topology network to evaluate the performance of the proposed algorithm. From the comparing results, the proposed algorithm can achieve good network utilization by the low rate of the bandwidth blocking.

Recognition of Multiple PQ Disturbances Using Wavelet-based Neural Networks— Part 2: Implementation and Applications

For part I see ibid., p.Z001593-8 (2005). This work proposes and implements a novel classifier integrated with the wavelet transform and dynamic structural neural network for recognizing multiple power quality disturbances in a measured waveform. The classifier has been tested under different PQ events such as with single disturbance, dual disturbances and multiple disturbances. The experimental results show that the proposed classifier can achieve high accuracy rate more than 97% under various test cases

Recognition of Multiple PQ Disturbances Using Dynamic Structure Neural Networks— Part 1: Theoretical Introduction

This work develops a new approach to recognize multiple disturbances for a power quality (PQ) event in power systems. It is usual that several different types of disturbances simultaneously exist in a PQ event. However, most of the existing methods treat a PQ event as a single type of PQ disturbance. The performance of these methods might be limited and impracticable for application in the real power systems. This work proposes a novel approach integrated the wavelet transform and dynamic structural neural network (DSNN) to identify disturbance waveforms. The proposed neural network has the capability of adapting to multiple disturbances for a PQ event. In the proposed approach, the disturbance waveforms are extracted by the wavelet transform and then fed to the DSNN for identifying the types of disturbances. The distinctive features of the proposed method are that it can estimate the amplitude of the considering event, recognize transient and steady state disturbances which are simultaneous existed in a PQ event

A novel dynamic structural neural network with neuro-regeneration and neuro-degeneration

This paper presents a novel dynamic structural neural network (DSNN) and a learning algorithm for training DSNN. The performance of a neural network system depends on several factors. In that, the architecture of a neural network plays an important role. The objective of the developing DSNN is to avoid trial-and-error process for designing a neural network system. The architecture of DSNN consists of a three-dimensional set of neurons with input/output nodes and connection weights. Designers can define the maximum connection number of each neuron. Moreover, designers can manually deploy neurons in a virtual 3D space, or randomly generate the system structure by the proposed learning algorithm. This work also develops an automatic restructuring algorithm integrated in the proposed learning algorithm to improve the system performance. Due to the novel dynamic structure of DSNN and the restructuring algorithm, the design of DSNN is fast and convenient. Furthermore, DSNN is implemented in C++ with man-machine interactive procedures and tested on many cases with promising results.