The technology of power electronics is used increasingly during recent years, and the electronic power facilities are used more and more in the power system. The non-linear electronic loads produce heavy harmonic currents and could significantly degrade the power quality. Nonlinear loads, including the un-interruptible power supply, motor control and converter, etc, are important equipment in a modern factory, however, these nonlinear loads could lead to power facility malfunction and capacitor damage. The harmonics would eventually cause severe unexpected capital loss. Identification of harmonic sources location becomes an important study for power quality. An effective tool is thus helpful for the harmonic source locating. This paper proposes a method to deal with the harmonic sources and location detection in the power system by using the artificial neural network (ANN). The non-linear loading characteristics are studied by the power flow analysis, and then the proposed methodology uses the Probabilistic Neural Networks¡]PNN¡^and wavelet-probabilistic network (WPN) for harmonic source locating. An IEEE 14-bus power system is used for study to show the effectiveness of the proposed approach.

The new approach for sensor diagnostics is presented. The approach, Enhanced Autoassociative Neural Networks (E-AANN), adds enhancement to Autoassociative Neural Networks (AANN) developed by Kramer in 1992. This enhancement allows AANN to identify faulty sensors. E-AANN uses a secondary optimization process to identify and reconstruct sensor faults. Two common types of sensor faults are investigated, drift error and shift or offset error. In the case of drift error, the sensor error occurs gradually while in the case of shift error, the sensor error occurs abruptly. EAANN catches these error types. A chiller model provided synthetic data to test the diagnostic approach under various noise level conditions. The results show that sensor faults can be detected and corrected in noisy situations with the E-AANN method described. In high noisy situations (10% to 20% noise level), E-AANN performance degrades. E-AANN performance in simple dynamic systems was also investigated. The results show that in simple dynamic situations, E-AANN identifies faulty sensors.

Annealed neural network combines the characteristics of both simulation annealing and Hopfield-Tank neural network, which are high quality solutions and fast convergence. Mobile agent planning is an important technique of information retrieval systems to provide the minimum cost of the location-aware services in mobile computing environment. By taking the time constraints of effective resources into account and the mobile agent to explore the cost optimization, we modify annealing neural network to design a new energy function and control the annealing temperature in order to deal with the dynamic temporal feature of computing environments. We not only consider the server performance and network latency when scheduling mobile agents, but also investigate the location-based constraints, such as the home site of routing sequence of the traveling mobile agent must be the start and end node. To guarantee the convergent stable state and existence of the valid solution, the energy function is reformulated into a Lyapunov function which is combined with the annealing temperature to form an activation function. The connection weights between the neurons and the activation function of state variables in the dynamic network are devised in searching for the valid solutions. Simulation of different coefficients assess the proposed model and algorithm. Furthermore, Taguchi method is used to obtain the optimal combination factors of annealing neural network. The results show that this research presents the feature of both simulated annealing and Hopfield neural network by providing fast convergence and highly quality. In addition with a larger number of sites, the experimental results demonstrate the benefits of the annealed neural network. This innovation would be applicable to improve the effectiveness of solving optimization problems.

Robot control systems evolved with genetic algorithms traditionally take the form of floating-point neural network models. This thesis proposes that digital control systems, such as quantised neural networks and logical networks, may also be used for the task of robot control. The inspiration for this is the observation that the dynamics of discrete networks may contain cyclic attractors which generate rhythmic behaviour, and that rhythmic behaviour underlies the central pattern generators which drive lowlevel motor activity in the biological world. To investigate this a series of experiments were carried out in a simulated physically realistic 3D world. The performance of evolved controllers was evaluated on two well known control tasks—pole balancing, and locomotion of evolved morphologies. The performance of evolved digital controllers was compared to evolved floating-point neural networks. The results show that the digital implementations are competitive with floating-point designs on both of the benchmark problems. In addition, the first reported evolution from scratch of a biped walker is presented, demonstrating that when all parameters are left open to evolutionary optimisation complex behaviour can result from simple components.

This thesis presents a preliminary design of a fully articulated five-fingered anthropomorphic human hand prosthesis with particular emphasis on the controller and actuator design. The proposed controller is a modified artificial neural network PID-based controller with application to the nonlinear and highly coupled dynamics of the hand prosthesis. The new solid state actuator has been designed based on electroactive polymers, which are a type of material that exhibit electromechanical behavior and a liquid metal alloy acts as the electrode. The solid state actuators reduce the overall mechanical complexity, risk failure and required maintenance of the prosthesis. / Graduate

Text classification has always been a concern in area of natural language processing, especially nowadays the data are getting massive due to the development of internet. Recurrent neural network (RNN) is one of the most popular method for natural language processing due to its recurrent architecture which give it ability to process serialized information. In the meanwhile, Convolutional neural network (CNN) has shown its ability to extract features from visual imagery. This paper combine the advantages of RNN and CNN and proposed a model called BLSTM-C for Chinese text classification. BLSTM-C begins with a Bidirectional long short-term memory (BLSTM) layer which is an special kind of RNN to get a sequence output based on the past context and the future context. Then it feed this sequence to CNN layer which is utilized to extract features from the previous sequence. We evaluate BLSTM-C model on several tasks such as sentiment classification and category classification and the result shows our model’s remarkable performance on these text tasks.

Agents being able to play board games such as Tic Tac Toe, Chess, Go and Arimaa has been, and still is, a major difficulty in Artificial Intelligence. For the mentioned board games, there is a certain amount of legal moves a player can do in a specific board state. Tic Tac Toe have in average around 4-5 legal moves, with a total amount of 255168 possible games. Both Chess, Go and Arimaa have an increased amount of possible legal moves to do, and an almost infinite amount of possible games, making it impossible to have complete knowledge of the outcome. This thesis work have created various Neural Networks, with the purpose of evaluating the likelihood of winning a game given a certain board state. An improved evaluation function would compensate for the inability of doing a deeper tree search in Arimaa, and the anticipation is to compete on equal skills against another well-performing agent (meijin) having one less search depth. The results shows great potential. From a mere one hundred games against meijin, the network manages to separate good from bad positions, and after another one hundred games able to beat meijin with equal search depth. It seems promising that by improving the training and by testing different sizes for the neural network that a neural network could win even with one less search depth. The huge branching factor of Arimaa makes such an improvement of the evaluation beneficial, even if the evaluation would be 10 000 times more slow.

Autonomous relative terrain navigation is a problem at the forefront of many space missions involving close proximity operations to any target body. With no definitive answer, there are many techniques to help cope with this issue using both passive and active sensors, but almost all require high fidelity models of the associated dynamics in the environment. Convolutional Neural Networks (CNNs) trained with images rendered from a digital terrain map (DTM) of the body’s surface can provide a way to side-step the issue of unknown or complex dynamics while still providing reliable autonomous navigation. This is achieved by directly mapping an image to a relative position to the target body. The portability of trained CNNs allows “offline” training that can yield a matured network capable of being loaded onto a spacecraft for real-time position acquisition. In this thesis the lunar surface is used as the proving ground for this optical navigation technique, but the methods used are not unique to the Moon, and are applicable in general.

Neural networks are seldom used as a modelling tool by statisticians. This is often due to the lack of knowledge in the eld of neural networks as neural networks are frequently perceived as mysterious methods that evolved from the eld of computer science. In this dissertation an attempt will be made to show that neural network methods are closely related to statistical methods. In particular we will show how a backpropagation neural network can be used for statistical applications like regression and classi cation which will include the setting up a of neural network for di erent objectives and also using a neural network for predictive inference. Through simulations we will show an e cient method to t a neural network in practical applications. A neural network will then be employed in a practical application to illustrate how to use a neural network in a regression or classi cation context. This application will also show the necessity of statistical knowledge when using a neural network as a modelling tool. / Dissertation (MSc)--University of Pretoria, 2010. / Statistics / unrestricted

Video super-resolution (VSR) aims to recover a realistic high-resolution (HR) frame from its corresponding center low-resolution (LR) frame and several neighbouring supporting frames. The neighbouring supporting LR frames can provide extra information to help recover the HR frame. However, these frames are not aligned with the center frame due to the motion of objects. Recently, many video super-resolution methods based on deep learning have been proposed with the rapid development of neural networks. Most of these methods utilize motion estimation and compensation models as preprocessing to handle spatio-temporal alignment problem. Therefore, the accuracy of these motion estimation models are critical for predicting the high-resolution frames. Inaccurate results of motion compensation models will lead to artifacts and blurs, which also will damage the recovery of high-resolution frames. We propose an effective wide activated separate 3 dimensional (3D) Convolution Neural Network (CNN) for video super-resolution to overcome the drawback of utilizing motion compensation models. Separate 3D convolution factorizes the 3D convolution into convolutions in the spatial and temporal domain, which have benefit for the optimization of spatial and temporal convolution components. Therefore, our method can capture temporal and spatial information of input frames simultaneously without additional motion evaluation and compensation model. Moreover, the experimental results demonstrated the effectiveness of the proposed wide activated separate 3D CNN.