Application of machine learning in the process of commander decision support in the military fuel distribution system

Forecasting of groundwater levels is very useful for planning integrated management of groundwater and surface water resources in a basin. In the present study, artificial neural network models have been developed for groundwater level forecasting in a river island of tropical humid region, eastern India. ANN modeling was carried out to predict groundwater levels 1 week ahead at 18 sites over the study area. The inputs to the ANN models consisted of weekly rainfall, pan evaporation, river stage, water level in the drain, pumping rate and groundwater level in the previous week, which led to 40 input nodes and 18 output nodes. Three different ANN training algorithms, viz., gradient descent with momentum and adaptive learning rate backpropagation (GDX) algorithm, Levenberg–Marquardt (LM) algorithm and Bayesian regularization (BR) algorithm were employed and their performance was evaluated. As the neural network became very large with 40 input nodes and 18 output nodes, the LM and BR algorithms took too much time to complete a single iteration. Consequently, the study area was divided into three clusters and the performance evaluation of the three ANN training algorithms was done separately for all the clusters. The performance of all the three ANN training algorithms in predicting groundwater levels over the study area was found to be almost equally good. However, the performance of the BR algorithm was found slightly superior to that of the GDX and LM algorithms. The ANN model trained with BR algorithm was further used for predicting groundwater levels 2, 3 and 4 weeks ahead in the tubewells of one cluster using the same inputs. It was found that though the accuracy of predicted groundwater levels generally decreases with an increase in the lead time, the predicted groundwater levels are reasonable for the larger lead times as well.

The increasing penetration of renewable energy sources and the variability of demand in microgrids have intensified the need for accurate short-term load forecasting (STLF). This study develops a multivariable STLF framework using the Bayesian Regularization Algorithm (BRA) and validates its performance against a Bidirectional Long Short-Term Memory (Bi-LSTM) network. Temperature, humidity, and feeder current were employed as input variables, while electrical load (MW) served as the output. Both models were implemented in MATLAB R2018a with a 120-minute forecasting horizon. Performance evaluation was conducted using regression analysis and statistical metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the coefficient of determination (R²). The results show that the BRA model achieved superior regression accuracy (R² ≈ 0.997), demonstrating robustness and generalization under noisy conditions. By contrast, Bi-LSTM attained R² ≈ 0.991 with lower error magnitudes (MAE = 0.0293, RMSE = 0.0412), reflecting its strength in capturing sequential dependencies. A comparative review with existing literature, including confirms the competitiveness of Bi-LSTM while highlighting BRA’s resilience against noisy fluctuations typical of Nigerian microgrid environments. This study contributes to microgrid energy management by providing practical insights into selecting forecasting models, emphasizing that BRA is well-suited for noisy, small datasets, while Bi-LSTM excels in sequence-sensitive forecasting. The findings also suggest the potential for hybrid BRA–Bi-LSTM architectures to enhance forecasting accuracy in real-world microgrid applications.

State of health estimation for lithium-ion battery based on the coupling-loop nonlinear autoregressive with exogenous inputs neural network

Headland-bay beach planform and tidal range: A neural network model

In this research paper, a nonlinear autoregressive with exogenous input (NARX) model of the nonlinear system based on neural network and time series analysis is proposed to deal with the one-month forecast of the produced power from photovoltaic modules (PVM). The PVM is a monocrystalline cell with a rated production of 175 watts that is placed at Heliopolis University, Bilbéis city, Egypt. The NARX model is considered powerful enough to emulate the nonlinear dynamic state-space model. It is extensively performed to resolve a variety of problems and is mainly important in complex process control. Moreover, the NARX method is selected because of its quick learning and completion times, as well as high appropriateness, and is distinguished by advantageous dynamics and interference resistance. The neural network (NN) is trained and optimized with three algorithms, the Levenberg–Marquardt Algorithm (NARX-LMA), the Bayesian Regularization Algorithm (NARX-BRA) and the Scaled Conjugate Gradient Algorithm (NARX-SCGA), to attain the best performance. The forecasted results using the NARX method based on the three algorithms are compared with experimentally measured data. The NARX-LMA, NARX-BRA and NARX-SCGA models are validated using statistical criteria. In general, weather conditions have a significant impact on the execution and quality of the results.

This paper presents a flexible pentagonal shape Ultra-Wide Band (UWB) antenna design using Artificial Neural Network (ANN) for WLAN, 5G, and WiMAX applications. The pentagonal patch is placed on top of flexible polyimide substrate and simulated using the well-known 3-D electromagnetic (EM) simulator HFSS, v.18.1. Due to large computing cluster required by the EM simulator to solve the design under consideration in addition to the time consumed, ANN is used to synthesize the design and reduce the cost and time consumed to analyze the aforementioned structure. Neural Network with 1 hidden layer of 10 neurons based on Bayesian Regularization algorithm is presented. An error of less 5% is produced during the learning, validation, and testing processes. Neural network is a good candidate to represent the pentagonal shape antenna used for UWB applications.

Development of an energy cost prediction model for a VRF heating system

Recently, Artificial Neural Network (ANN) has been developed as an alternative to classical statistical methods in animal production. The methods can do classification or prediction by analyzing the information in the data set with the help of the neural network without requiring any preconditions (for example, distribution of data, non-linear data, highly correlated variables, etc.). In this context, we hypothesized that ANN, which is not only used in large and complex data sets but also estimates better in small data sets compared to classical statistical methods. The ability of ANN of Bayesian Regularization (BR-ANN) or Levenberg-Marquardt (LM-ANN) algorithms and Multiple Regression (MR) model to predict meat quality traits were assessed in a comparative study. The multilayer ANN algorithms obtained prediction data of meat quality measurements from pre-slaughter information using 1, 2, 4, 6 and 8 neurons in the hidden layer applied 10 times for each model. The performance of the methods was assessed according to the coefficient of determination (R2) criteria, root mean squared error (RMSE) and residual prediction deviation (RPD). The comparison of the findings of BR-ANN and ML-ANN algorithms showed a similar ability to predict meat quality traits (error of prediction and R2 values between 0.32-2.72 and 0.19-0.49, respectively). However, MR model predictions had lower performance than ANN algorithms, resulting in a wider error of prediction interval (0.4-3.44) and low R2 (0.16-0.44). The RPD for meat quality traits was fair for BR-ANN and LM-ANN but was poor for MR. Based on our results, the ANN algorithms produced more reasonable prediction values than the MR model. ANN algorithms can be used as an acceptable alternative method for simple physical measurements of meat quality. ANN algorithms can be used reliably in small data sets. Key words: Bayesian networks, cattle, meat science

Intelligent and automatic systems based on arrays of non-specific-response chemical sensors were recently developed in our laboratory. For multidetermination applications, the normal choice is an array of potentiometric sensors to generate the signal, and an artificial neural network (ANN) correctly trained to obtain the calibration model. As a great amount of information is required for the proper modelling, we proposed its automated generation by using the sequential injection analysis (SIA) technique. First signals used were steady-state: the equilibrium signal after a step-change in concentration. We have now adapted our procedures to record the transient response corresponding to a sample step. The novelty in this approach is therefore the use of the dynamic components of the signal in order to better discriminate or differentiate a sample. In the developed electronic tongue systems, detection is carried out by using a sensor array formed by five potentiometric sensors based on PVC membranes. For the developed application we employed two different chloride-selective sensors, two nitrate-selective sensors and one generic response sensor. As the amount of raw data (fivefold recordings corresponding to the five sensors) is excessive for an ANN, some feature extraction step prior to the modelling was needed. In order to attain substantial data reduction and noise filtering, the data obtained were fitted with orthonormal Legendre polynomials. In this case, a third-degree Legendre polynomial was shown to be sufficient to fit the data. The coefficients of these polynomials were the input information fed into the ANN used to model the concentrations of the determined species (Cl-, NO3- and HCO3-). Best results were obtained by using a backpropagation neural network trained with the Bayesian regularisation algorithm; the net had a single hidden layer containing three neurons with the tansig transfer function. The results obtained from the time-dependent response were compared with those obtained from steady-state conditions, showing the former superior performance. Finally, the method was applied for determining anions in synthetic samples and real water samples, where a satisfactory comparison was also achieved.

The annual sediment load of a river is generally determined either from direct measurements of the sediment load throughout the year or from any of the many sediment transport equations that are available today. Due to lack of a long-term sediment concentration data, sediment rating curves and flux estimation are the most widely applied. This paper has investigated the abilities of statistical models to improve the accuracy of streamflow–suspended sediment relationships in daily and annual suspended sediment estimation. In this study, a comparison was made between suspended sediment rating curves and artificial neural networks (ANNs) for the El Kebir catchment. Daily water discharge and daily suspended sediment data from the gauging station of Ain Assel, were used as inputs and targets in the models which were based on the cascade-forward and feed-forward back-propagation using Levenberg–Marquardt and Bayesian regularization algorithms. The model results have shown that the ANN models have the highest efficiency to reproduce the daily sediment load and the global annual sediment yields. Our estimation based on the available data indicated that the areas along the El Kebir River have experienced high sediment fluxes that could have obvious impacts on the sediment trapping and siltation in the Mexa reservoir.

Pumpkin antioxidants have been found to benefit diabetics. This current study was attempted to optimize slow freezing treatment for a pumpkin to obtain maximum antioxidant gain using response surface methodology (RSM) and Bayesian regularized neural network (BRANN) approaches. A central composite design was used to generate the freezing experiment and to examine response change as a function of temperature and freezing time. Feedforward neural networks with a 2-15-1 structure were developed and trained using the Bayesian regularization algorithm. The results showed that the freezing data were well fitted to quadratic models generating R2 for total phenolic compounds (TPC), flavonoid of 0.850 and 0.857 respectively. The RSM optimized freezing of -20oC for 9 hrs were well confirmed to produce an increase in TPC and flavonoid by 54.44% and 60.4% respectively. The BRANN performances were found to be similar to that of RSM. While overfitting was mitigated during the supervised training, the BRANN model served excellent predictive and confirmatory tool for the optimization. In conclusion, slow freezing at -20oC for 9 hrs significantly increases TPC and flavonoid of pumpkin. This novel process may be adopted to provide healthier pumpkins food products for targeted consumers.

Application of the wavelet transform coupled with artificial neural networks for quantification purposes in a voltammetric electronic tongue

Constitutive modelling of the TiB2–B4C composite by experiments, simulation and neutral network

In our globally connected world, threats from various aspects are going at an alarming rate. These are controlling with different security systems such as metal detector, closed circuit cameras, and scanning systems. All these aids are meant to recognize and identify the explosives and others weapons. Here, it is more important to identify the particular person or persons, who were planning to distract the society or particular event. This paper is aimed to design that to control the threats by identifying the suspected people by a simple face recognition technique using simple PC or laptop with the help of scientific software MATLAB and its neural network tool box. In general, all major events are fully securitized with well-developed protection systems but only problem with non-major and small events, where security systems are matter of financial issues. So, militants and other destroyers are taking advantage of these situations and creating a panic and terror situations. This paper is also designed like that a PC or laptop with camera can be a face recognition system to identify the suspected peoples and most wanted criminal. By recognizing the people, we can mostly avoid the threats from these people and dangerous situations. Neural network is a science that has been extensively applied to numerous pattern recognition problems such as character recognition, object recognition, and face recognition, where this paper has programmed for face recognition with the back-propagation algorithm and simulated with the software MATLAB and its neural network tool box. Here, the back propagation plays the central operation role to get the key features were extracted from the picture for training the network. Since the major role of the project is mainly focusing on the training of the neural network, already extracted key features of the person’s image from the database were taken for training the back-propagation network. Here, we have taken 7 input units, 6 hidden units, and 4 output units contained back-propagation network. The output unit, 4 output units, generates the 4-bit output which gives the person identity.

In this paper we present a classifier that separates the classes “target in wind turbine clutter” from “no target in wind turbine clutter” using a group of feedforward neural networks. The training, test and validating data sets are generated using an analytical model of the wind turbine. The focus of this paper is on the question whether or not neural networks can serve as classifier for the mentioned scenario without explicit clutter and noise estimation and thresholding. For this reason, we choose a simplified signal and radar model and generate data sets based on randomly selected parameter configurations. The simulations are carried out in Matlab using Matlab’s neural network toolbox.