Search

Brushless DC (BLDC) motors are among the most widely used electrical motors. Design of a BLDC motor is the most fundamental problem when dealing with the BLDC motors. This thesis presents an intelligent decision support system that can be used to design BLDC motors. A hybrid approach, that includes an object oriented paradigm using frames and procedural attachments together with a rule based mechanism, is used to build the knowledge base of the proposed architecture. The design strategy is implemented using a rule-based successive iterative method. An evolutionary fuzzy system was used to derive the modification rules of the system. The antecedent and consequent of each fuzzy modification rule was encoded as the individual of an evolutionary system. The evolutionary system evolves the set of modification rules to find a set of optimized rules. The proposed system developed design which had superior efficiency, weight and motor constant compared to design developed using the conventional design method.

This paper proposes a probabilistic sensor model for the optimization of sensor placement. Traditional schemes rely on simple sensor behaviour and environmental factors. The consequences of these oversimplifications are unrealistic simulation of sensor performance and, thus, suboptimal sensor placement. In this paper, we develop a novel probabilistic sensing model for sensors with line-of-sight-based coverage (e.g., cameras) to tackle the sensor placement problem for these sensors. The probabilistic sensing model consists of membership functions for sensing range and sensing angle, which takes into consideration sensing capacity probability as well as critical environmental factors such as terrain topography. We then implement several optimization schemes for sensor placement optimization, including simulated annealing, limited-memory Broyden-Fletcher-Goldfarb-Shanno method, and covariance matrix adaptation evolution strategy.

This paper proposes the use of a kernel density estimation to measure similarities between trajectories. The similarities are then used to predict the future locations of a target. For a given environment with a history of previous target trajectories, the goal is to establish a probabilistic framework to predict the future trajectory of currently observed targets based on their recent moves. Instead of clustering trajectories into groups, we calculate the similarity between a given test trajectory and the set of all past trajectories in a dataset. Next, we use a weighted mechanism for prediction, that can be used in target tracking and collision avoidance applications. The proposed method is compared with two other commonly used similarity models (PCA and LCSS) over a dataset of simulated trajectories, and two datasets of real observations. Results show that the proposed method significantly outperforms the existing models for those datasets and experimental settings.

Wireless Sensor Networks (WSN) have been studied intensively for various applications such as monitoring and surveillance. Sensor deployment is an essential part of WSN, because it affects both the cost and capability of the sensor network. However, most deployment schemes proposed so far have been based on over-simplified assumptions, where results may be far from optimal in practice. Our proposal aims at automating and optimizing sensor deployment based on realistic topographic information, and is thus different from previous work in two ways: 1) it takes into account the 3D nature of the environment ; 2) it allows the use of anisotropic sensors. Based on the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), the proposed approach shows good potential for tackling diverse problems in the WSN domain. Preliminary results are given for a mountainous area of North Carolina where coverage is maximized.

An evolutionary system for derivation of fuzzy classification rules is presented. This system uses two populations: one of fuzzy classification rules, and one of membership function definitions. A constrained-syntax genetic programming evolves the first population and a mutation-based evolutionary algorithm evolves the second population. These two populations co-evolve to better classify the underlying dataset. Unlike other approaches that use fuzzification of continuous attributes of the dataset for discovering fuzzy classification rules, the system presented here fuzzifies the relational operators ldquogreater thanrdquo and ldquoless thanrdquo using evolutionary methods. For testing our system, the system is applied to the Iris dataset. Our experimental results show that our system outperforms previous evolutionary and non-evolutionary systems on accuracy of classification and derivation of interrelation between the attributes of the Iris dataset. The resulting fuzzy rules of the system can be directly used in knowledge-based systems.