Papers by Keyword: Back-Propagation

Abstract: A three-layer structure back-propagation network model based on the non-linear relationship between the size of the CaCO₃ nanocrystalline and the technological factors, such as reaction time, reaction temperature, raw material adding amount of NaCO₃ and CaCl₂, was established. Moreover, in order to accelerate the converging rate and avoid the non-converging situation, the momentum terms are introduced. Besides, the variable learning speed is adopted. At the same time, the input variables were pretreated by using the main component analysis firstly. And the results show that the improved back propagation neural networks model is very efficient for predication of the CaCO₃ nanocrystalline size.

Abstract: Based on the idea of standard back-propagation (BP) learning algorithm, an improved BP learning algorithm is presented. Three parameters are incorporated into each processing unit to enhance the output function. The improved BP learning algorithm is developed for updating the three parameters as well as the connection weights. It not only improves the learning speed, but also reduces the occurrence of local minima. Finally, the algorithm is tested on the XOR problem to verify the validity of the improved BP.

Abstract: The korshunskite samples were prepared in precipitation by the one-step reaction method at atmospheric pressure. The three-layer structure back-propagation network model based on the non-linear relationship between the amount of the korshunskite whiskers and the technological factors, such as the adding amount of raw materials NaOH, MgCl₂, MgO, and reaction temperature, is established. And the results show that the improved back propagation neural networks model is very efficient for predication of the korshunskite whiskers preparation.

Abstract: Machining processes, such as milling, are considered to be too complex to be modeled accurately by using analytical or even numeric means due to involvement of various control parameters, some of them highly vague and imprecise. Such situation calls for application of nonconventional methods for modeling the responses of interest with acceptable degree of accuracy. In this work, a computational intelligence tool, possessing quick learning ability, has been used for modeling and predicting tool’s flank wear and workpiece surface roughness in milling of cold work tool steel. Six numeric and two categorical input parameters were used in the artificial neural network model. 116 data sets were used for training the network, while 13 were used for testing. Both the responses were modeled with acceptable degree of accuracy.

Abstract: Over the past decade, the growth of the housing construction in Malaysia has been increase dramatically and the level of urbanization process in Malaysia is considered to be important in planning for low-cost housing needs. Unfortunately, there is a clear miss-match between the supply and the demand of low cost housing in Malaysia. Due to the problems faced, there have been several attempts in predicting housing demands using the artificial-neural networks (ANN) technique particularly back-propagation (BP). However, the training process of BP can result in slow convergence or even network paralysis and can easily get stuck at local minima. This paper presents a new approach to improve the training efficiency of BP algorithms to forecast low-cost housing demand in one of the states in Peninsular Malaysia. The proposed algorithm (BPM/AG) adaptively modifies the gradient based search direction by introducing the value of gain parameter in the activation function. The results show that the proposed algorithm significantly improves the learning process with more than 31% faster in term of CPU time and number of epochs as compared to the traditional approach. The proposed algorithm can forecast low-cost housing demand very well with 6.62% of MAPE value.

Abstract: Measurements of seepage are fundamental for earth dam surveillance. However, it is difficult to establish an effective and practical dam seepage prediction model due to the nonlinearity between seepage and its influencing factors. Genetic Algorithm for Levenberg-Marquardt(GA-LM), a new neural network(NN) model has been developed for predicting the seepage of an earth dam in China using 381 databases of field data (of which 366 in 2008 were used for training and 15 in 2009 for testing). Genetic algorithm(GA) is an ecological system algorithm, which was adopted to optimize the NN structure. Levenberg-Marquardt (LM) algorithm was originally designed to serve as an intermediate optimization algorithm between the Gauss-Newton(GN) method and the gradient descent algorithm, which was used to train NN. The predicted seepage values using GA-LM model are in good agreement with the field data. It is demonstrated here that the model is capable of predicting the seepage of earth dams accurately. The performance of GA-LM has been compared with that of conventional Back-Propagation(BP) algorithm and LM algorithm with trial-and-error approach. The comparison indicates that the GA-LM model can offer stronger and better performance than conventional NNs when used as a quick interpolation and extrapolation tool.

Abstract: Neural network learning algorithms are widely used in medical diagnosis, bioinformatics, intrusion detection, homeland security and other fields. The common of these applications is that all of them need to extract patterns and predict trends from a large number of complex data. In these applications, how to protect the privacy of sensitive data and personal information from disclosure is an important issue. At present, the vast majority of existing neural network learning algorithms did not consider how to protect the data privacy in the process of learning. So this paper proposes two privacy-preserving back-propagation neural network protocols applied to horizontally partitioned data and vertically partitioned data separately. The two protocols are suitable for multiple participants in a distributed environment. The results of experiments show the difference of the test error rate between the proposed two protocols and the respective non-privacy preserving versions.

Abstract: This paper develops a three-layer back-propagation artificial neural network model to analyze and predict the correlation between processing parameters and properties of the damage tolerance type titanium alloy TC21. The inputs of the ANN are working temperatures, deformation extent, deformation rate and heat treatment conditions. And the outputs are mechanical properties namely ultimate strength, yield strength, elongation, reduction of area, plane strain fracture toughness and microstructure concerned parameters such as β phase fraction, βphase grain size, substructure length and thickness. The ANN is trained with experimental data and achieves a very good performance, which has already been applied to the optimization of processing for forging of aero-parts.

Abstract: The supplier selection and evaluation is a key factor of the intelligent supplier selection & evaluation system in e-manufacturing. The model used for supplier selection is Fuzzy inference system which is introduced in the paper. The paper started with the brief introduction of the intelligent internet supplier selection & evaluation system. It concentrated to introduce the application of the fuzzy set model for supplier selection. This paper will introduce the design of the fuzzy sets model, and the evaluation results.

Abstract: The supplier evaluation is a key section of the intelligent internet supplier selection & evaluation system. The model used for supplier evaluation is Back Propagation Neural Network model which is introduced in the paper. The paper started with the brief introduction of the intelligent internet supplier selection & evaluation system. It provides a outline of the research project and then it concentrated to introduce the application of the BP NN model for supplier evaluation. The application introduced in the paper will include the design of the BP NN model, Training of the BP NN model and test results.