This paper presents a novel approach to simulate a Knowledge Based System for diagnosis of Breast Cancer using Soft Computing tools like Artificial Neural Networks (ANNs) and Neuro Fuzzy Systems. The feed-forward neur...
详细信息
ISBN:
(纸本)9781424429271
This paper presents a novel approach to simulate a Knowledge Based System for diagnosis of Breast Cancer using Soft Computing tools like Artificial Neural Networks (ANNs) and Neuro Fuzzy Systems. The feed-forward neural network has been trained using three ANN algorithms, the backpropagationalgorithm (BPA), the Radial Basis Function (RBF) Networks and the Learning Vector Quantization (LVQ) Networks;and also by Adaptive Neuro Fuzzy Inference System (ANFIS). The simulator has been developed using MATLAB and performance is compared by considering the metrics like accuracy of diagnosis, training time, number of neurons, number of epochs etc. The simulation results show that this Knowledge Based Approach can be effectively used for early detection of Breast Cancer to help oncologists to enhance the survival rates significantly.
This paper presents a method to discriminate a temporary fault from a permanent one in an extra high voltage (EHV) transmission line so that improper reclosing of the line onto a fault is avoided. The fault identifica...
详细信息
ISBN:
(纸本)9780769535210
This paper presents a method to discriminate a temporary fault from a permanent one in an extra high voltage (EHV) transmission line so that improper reclosing of the line onto a fault is avoided. The fault identification prior to reclosing is based on optimized artificial neural network associated with standard Error back-propagation, Levenberg Marquardt algorithm and Resilient back-propagation training algorithms together with Taguchi's Method. The algorithms are developed using MATLAB software. A range of faults are simulated on EHV modeled transmission line using SimPowerSytems, and the spectra of the fault data are analyzed using fast Fourier transform to extract features of each type of fault. For both training and testing purposes, the neural network is fed with the normalized energies of the DC component, the fundamental and the first four harmonics of the faulted voltages. The developed algorithm is effectively trained, verified and validated with a set of training, dedicated testing and validation data respectively.
The last decade witnessed a significant increase in net private capital inflows in China. Some of them are short-term capital flows, which are typically considered to be highly volatile. For effectively forecasting th...
详细信息
ISBN:
(纸本)9780769539270
The last decade witnessed a significant increase in net private capital inflows in China. Some of them are short-term capital flows, which are typically considered to be highly volatile. For effectively forecasting the short-term capital flows, a three-layered neural feedforward network was employed in this paper. In light of the weakness of the conventional back-propagation algorithm, the Levenberg-Marquardt algorithm was used to train the neural network. The simulation results indicate that the predictive model can be used to carry out the prediction of short-term capital flow.
General Purpose Processors (GPPs) and ASICs have traditionally been the common means for building and implementing Artificial Neural Network's (ANNs). However Such computing paradigms suffer from the constant need...
详细信息
ISBN:
(纸本)0769524567
General Purpose Processors (GPPs) and ASICs have traditionally been the common means for building and implementing Artificial Neural Network's (ANNs). However Such computing paradigms suffer from the constant need of establishing a trade-off between flexibility and performance Due to the technological advance in the development of progammable logic devices, Field Programmable Gate Arrays (FPGAs) have become attractive for realizing ANNs. FPGAs have shown to exhibit excellent flexibility in terms of reprogramming the same hardware and at the same time achieving good performance by enabling parallel computation. In this paper various implementations of ANNs on FPGAs are investigated and compared. The research described in this paper proposes three partially parallel architectures and a fully parallel architecture to realize the back- propagationalgorithm on an FPGA. The proposed designs are coded in Handel-C and functionally, verified by synthesizing them on a Virtex2000e FPGA chip. The partially parallel architectures and the fully parallel architecture are found to be 2.25 and 4 times faster than the software implementation rcspectively for different benchmarks.
As a successively and locally plastic deformation process, backward ball spinning is applied for the purpose of manufacturing thin-walled tubular part with longitudinal inner ribs. Obtaining the desired inner ribs is ...
详细信息
As a successively and locally plastic deformation process, backward ball spinning is applied for the purpose of manufacturing thin-walled tubular part with longitudinal inner ribs. Obtaining the desired inner ribs is one of the most critical tasks in backward ball spinning of thin-walled tubular part with longitudinal inner ribs and the formability of inner ribs depends greatly on the process parameters, such as ball diameter, feed ratio, wall thickness reduction and wall thickness of tubular blank. As a nonlinear dynamics system simulating structure and function of biological neural network in the human brain, back-propagation artificial neural network (BPANN) is used in backward ball spinning of thin-walled tubular part with longitudinal inner ribs. The attractiveness of BPANN comes from its remarkable information processing characteristics pertinent mainly to nonlinearity, adaptability, high parallelism, learning capability, fault and noise tolerance so that it can be more efficient in solving complex and nonlinear optimization problems in backward ball spinning of thin-walled tubular parts with longitudinal inner ribs. Not only can BPANN successfully predict the formability of the inner ribs, but it can simulate the influences of the process parameters on the height of inner ribs as well. In the end, the process parameters are matched so rationally that the desired spun parts can be obtained. (c) 2007 Elsevier B.V. All rights reserved.
A method based on neural network with back-propagation algorithm (BPA) and Adaptive Smoothing Errors (ASE), and a Genetic algorithm (GA) employing a new concept named Adaptive Relaxation (GAAR) is presented in this pa...
详细信息
A method based on neural network with back-propagation algorithm (BPA) and Adaptive Smoothing Errors (ASE), and a Genetic algorithm (GA) employing a new concept named Adaptive Relaxation (GAAR) is presented in this paper to construct learning system that can find an Adaptive Mesh points (AM) in fluid problems. AM based on reallocation scheme is implemented on different types of two steps channels by using a three layer neural network with GA. Results of numerical experiments using Finite Element Method (FEM) are discussed. Such discussion is intended to validate the process and to demonstrate the performance of the proposed learning system on three types of two steps channels. It appears that training is fast enough and accurate due to the optimal values of weights by using a few numbers of patterns. Results confirm that the presented neural network with the proposed GA consistently finds better solutions than the conventional neural network.
Hindcasting of wave parameters is necessary for many applications in coastal and offshore engineering and is generally made with the help of sophisticated numerical models. This paper presents alternative hindcast mod...
详细信息
Hindcasting of wave parameters is necessary for many applications in coastal and offshore engineering and is generally made with the help of sophisticated numerical models. This paper presents alternative hindcast models based on Artificial Neural Networks (ANNs), Fuzzy Inference System (FIS) and Adaptive-Network-based Fuzzy Inference System (ANFIS). The data set used in this study comprises wave and wind data gathered from deep water location in Lake Ontario. Wind speed, wind direction, fetch length and wind duration were used as input variables, while significant wave height, peak spectral period and mean wave direction were the Output parameters. Different topologies of ANNs were considered to predict the wave parameters and the relative importance of input parameters were determined. Finally, the results of ANNs-based models, FIS- and ANFIS-based models were compared. Results indicated that error statistics of soft computing models were similar, while ANFIS models were marginally more accurate than FIS and ANNs models. (C) 2008 Elsevier Ltd. All rights reserved.
The performance of a biological Fe2+ oxidizing fluidized bed reactor (FBR) was modeled by a popular neural network-back-propagation algorithm over a period of 220 days at 37 degrees C under different operational condi...
详细信息
The performance of a biological Fe2+ oxidizing fluidized bed reactor (FBR) was modeled by a popular neural network-back-propagation algorithm over a period of 220 days at 37 degrees C under different operational conditions. A method is proposed for modeling Fe3+ production in FBR and thereby managing the regeneration of Fe3+ for heap leaching application, based on an artificial neural network-back-propagation algorithm. Depending on output value, relevant control strategies and actions are activated, and Fe3+ production in FBR was considered as a critical output parameter. The modeling of effluent Fe3+ concentration was very successful, and an excellent match was obtained between the measured and the predicted concentrations.
Product unit neural networks with exponential weights (PUNNs) can provide more powerful internal representation capability than traditional feed-forward neural networks. In this paper, a convergence result of the back...
详细信息
Product unit neural networks with exponential weights (PUNNs) can provide more powerful internal representation capability than traditional feed-forward neural networks. In this paper, a convergence result of the back-propagation (BP) algorithm for training PUNNs is presented. The monotonicity of the error function in the training iteration process is also guaranteed. A numerical example is given to support the theoretical findings. (c) 2008 Elsevier B.V. All rights reserved.
With the development of our country's electricity market, power system load forecasting, especially short-term load forecasting (STLF), has been get more and more important function in the reliable, security, and ...
详细信息
ISBN:
(纸本)9781424421138
With the development of our country's electricity market, power system load forecasting, especially short-term load forecasting (STLF), has been get more and more important function in the reliable, security, and economical operation of power system. The paper has introduced the principle of artificial neural networks and how we can use it to forecast electric power load. In this paper, a three-layer feed-forward back-propagation (BP) network is applied for short-term load forecasting. In order to improve the accuracy of short-term load forecasting, the data were detached into three groups: workday data, weekend data and festival data which were used to be trained for grouping forecasting models. The proposed networks are trained with 2-year (2002-2003) actual data of Nanchang city and are tested for target years (2004) including workday model, weekend model and festival model. Very reasonable results have been obtained for sample data.
暂无评论