Two-layer Back-propagation Neural Network Research Articles

Assessment of climate change impact on inflows to Amandara headwork using HEC-HMS and ANNs

AbstractThis research has assessed the impact of climate change on temperature, precipitation, and inflows to the Amandara headwork in Pakistan. Trend Analysis using the Mann–Kendall test and Innovative Trend Analysis has been performed. Rainfall-runoff modeling is executed using the Hydrological Engineering Centre-Hydrological Modeling System (HEC-HMS) and Artificial Neural Networks including Feed Forward Neural Network, Conjugate Gradient, Two-layer Backpropagation Neural Network, and Broyden Fletcher-Goldfarb-Shanno. Mean daily hydro-meteorological data (1992 to 2023) was utilized for this study in which 70% was employed for calibration while the remaining 30% was used for validation of the model. Two GCMs namely CSIROMk3-6–0 and HadGEM2-ES with four Representative Concentration Pathways; RCP 2.6, 4.5, 6.0, and 8.5, were employed for future forecasting of temperature and precipitation. This future predicted data was then used to forecast flows up to 2050 by HEC-HMS. The performance of the models was assessed using correlation coefficient (R), Root Mean Square Error, Mean Bias Error, and Nash Sutcliffe Efficiency. Significant patterns in the runoff and temperature with no trend in precipitation were found. GCMs showed an increase in the range of 3–9 °C in temperature, 300 to 500 mm in precipitation, and 45 to 54% in peak flows.

New Face Recognition System Based on DCT Pyramid and Backpropagation Neural Network

Face recognition has emerged as a prominent biometric identification technique with applications ranging from security to human-computer interaction. This paper proposes a new face recognition system by appropriately combining techniques for improved accuracy. Specifically, it incorporates a discrete cosine transform (DCT) pyramid for feature extraction, statistical measures for dimensionality reduction of the features, and a two-layer backpropagation neural network for classification. The DCT pyramid is used to effectively capture both low- and high-frequency information from face images to improve the ability of the system to recognise faces accurately. Meanwhile, the introduction of statistical measures for dimensionality reduction helps in decreasing the computational complexity and provides better discrimination, leading to more efficient processing. Moreover, the two-layer neural network introduced, which plays a vital role in efficiently handling complex patterns, further enhances the recognition capabilities of the system. As a result of these advancements, the system achieves an outstanding 99 % recognition rate on the Olivetti Research Laboratory (ORL) data set, 98.88 % on YALE, and 99.16 % on AR. This performance demonstrates the robustness and potential of the proposed system for real-world applications in face recognition.

Design of marine propellers with prescribed and optimal spanwise circulation distributions based on genetic algorithms and neural network

Lifting-surface design methods are proposed for marine propellers based on genetic algorithms (GAs) and a two-layer back-propagation neural network (BPNN). For propellers with prescribed spanwise circulation distributions, the design problem is solved as GA-based optimization of camber surface geometry and pitch distribution which produce a circulation distribution to best fit the prescribed one. An in-house code based on the vortex lattice method (VLM) is employed to simulate circulation distribution and hydrodynamic performance of the propeller. Computer codes are developed in this research based on existing genetic algorithms, with a measure devised to accelerate convergence and improve the quality of solution. To optimize the spanwise circulation distribution, GAs are utilized again to explore the BPNN model established via the MATLAB toolbox and the in-house VLM code. Then the optimal circulation distribution is taken as the prescribed one for the GA-based design method mentioned above. Numerical tests are conducted to determine proper ranges of modeling parameters, such as the population size for GAs and the number of vortex lattices, and to assess the performance of the BPNN established. To numerically validate the proposed methods, viscous-flow simulation by solving the Reynolds-averaged Navier-Stokes equations is carried out for the propeller designed by using both methods mentioned above. The predicted pressure distributions over blade surfaces correlate consistently with the circulation distributions used to design the propeller, thus indicating that present design methods are effective and reasonably accurate.

Investigation of Mechanical Properties of Self Compacting Polymeric Concrete with Backpropagation Network

Acrylic polymer that is highly stable against chemicals and is a good choice when concrete is subject to chemical attack. In this study, self-compacting concrete (SCC) made using acrylic polymer, nanosilica and microsilica has been investigated. The results of experimental testing showed that the addition of microsilica and acrylic polymer decreased the tensile, compressive and bending strength of the concrete. The addition of nanosilica and an increase in polymer content increased the bending strength of concrete and decreased the tensile and compressive strengths. Because, in the laboratory, the number of samples were limited and the amount of variation was small, comprehensive results cannot be achieved. With the help of neural networks, estimating any amount within the range of the input data is possible. In this paper, in addition to the experimental results, a backpropagation neural network (BNN) was used to simulate the testing on the strength of self-compacting polymeric concrete. The results showed that the use of the normalized mean squared error, resilient backpropagation training, tangent-sigmoid and log sigmoid transfer functions and five neurons in each hidden layers in a two-layer BNN produced good results with a regression value of 0.95 and error of 0.17.

Coupled multiview autoencoders with locality sensitivity for three-dimensional human pose estimation

Estimating three-dimensional (3D) human poses from a single camera is usually implemented by searching pose candidates with image descriptors. Existing methods usually suppose that the mapping from feature space to pose space is linear, but in fact, their mapping relationship is highly nonlinear, which heavily degrades the performance of 3D pose estimation. We propose a method to recover 3D pose from a silhouette image. It is based on the multiview feature embedding (MFE) and the locality-sensitive autoencoders (LSAEs). On the one hand, we first depict the manifold regularized sparse low-rank approximation for MFE and then the input image is characterized by a fused feature descriptor. On the other hand, both the fused feature and its corresponding 3D pose are separately encoded by LSAEs. A two-layer back-propagation neural network is trained by parameter fine-tuning and then used to map the encoded 2D features to encoded 3D poses. Our LSAE ensures a good preservation of the local topology of data points. Experimental results demonstrate the effectiveness of our proposed method.

Generalization Capability of Neural Network Models for Temperature-Frequency Correlation Using Monitoring Data

The parametric approach to eliminating the temperature-caused modal variability in vibration-based structural damage detection requires a correlation model between the modal properties and environmental temperatures. This paper examines the generalization capability of neural network models, established using long-term monitoring data, for correlation between the modal frequencies and environmental temperatures. A total of 770 h modal frequency and temperature data obtained from an instrumented bridge are available for this study, which are further divided into three sets: training data, validation data, and testing data. A two-layer back-propagation neural network (BPNN) is first trained using the training data by the conventional training algorithm, in which the number of hidden nodes is optimally determined using the validation data. Then two new BPNNs are configured with the same data by applying the early stopping technique and the Bayesian regularization technique, respectively. The reproduction and prediction capabilities of the two new BPNNs are examined in respect of the training data and the unseen testing data, and compared with the performance of the baseline BPNN model. This study indicates that both the early stopping and Bayesian regularization techniques can significantly ameliorate the generalization capability of BPNN-based correlation models, and the BPNN model formulated using the early stopping technique outperforms that using the Bayesian regularization technique in both reproduction and prediction capabilities.

Quantification of Odours from Piggery Effluent Ponds using an Electronic Nose and an Artificial Neural Network

An Artificial Neural Network (ANN) and an electronic nose, AromaScan, were used to predict the piggery odour concentrations emanating from an effluent pond and to develop a confident, rapid, and cost-effective technique for odour measurement. Odour samples from five different piggery effluent ponds were analysed using the AromaScan and dynamic dilution olfactometry. The resulting sensor data were used to train the artificial neural network to correlate the responses to the odour concentrations measured by olfactometry. Effectiveness was evaluated through simulation with various pre-processing techniques and network architectures. The simulation results have shown that a two-layer back-propagation neural network, which has a tan-sigmoid transfer function in the hidden layer and a linear transfer function in the output layer, could be trained to predict piggery odour concentrations with high value of the correlation coefficient R of 0·984 under the best network performance. The results from the application of scaling and principal component analysis suggest that these techniques are necessary not only to avoid the failure of the network caused by saturation but also to enhance performance. An early stopping technique was shown to provide benefits to the network performance in terms of a decrease in computation time and overfitting. It was found that the optimal number of hidden neurons for the network was 20. Odour concentration of unknown samples were able to be predicted with significant accuracy.