Neural Network Prediction Research Articles

Prediction of moisture content ratio of emulsified asphalt chip seal based on machine learning and electrical parameters

The moisture content ratio (MCR) of the emulsified asphalt chip seal can determine its curing degree. However, the MCR of emulsified asphalt chip seal is difficult to measure on actual projects, and there is a lack of a method to assess its MCR. The objective of this study is to establish a prediction methodology for the MCR of emulsified asphalt chip seal based on machine learning and electrical parameters. Features such as electrical parameters and MCR of emulsified asphalt chip seal at different times were measured experimentally. The importance of the features was evaluated using a Random Forest (RF) model. A Back Propagation Neural Network (BPNN) prediction model was established using the important features. The weights and biases of the BPNN were optimized and initialized using the Improved Particle Swarm Optimization (IMPSO) algorithm. As a result, the RF-IMPSO-BPNN emulsified asphalt chip seal MCR prediction model was developed. This model was compared with five other models. The results show that compared to the RF-PSO-BPNN model, the improved RF-IMPSO-BPNN model can improve the ability of the neural network to find the global optimal solution. Compared to the other four machine learning models, the RF-IMPSO-BPNN model can achieve higher prediction accuracy while reducing the human and material resources of the various devices to collect part of the data. In addition, the emulsified asphalt chip seal cures at low MCR. The model predictions are more accurate at low MCR. Therefore, this study developed the RF-IMPSO-BPNN emulsified asphalt chip seal MCR prediction model, which can use fewer features to achieve higher accuracy and provide a rapid and non-destructive idea for judging its curing.

Trend Prediction of Vibration Signals for Pumped-Storage Units Based on BA-VMD and LSTM

Under “dual-carbon” goals and rapid renewable energy growth, increasing start-stop frequency poses new challenges to safe operations of pumped-storage power plant equipment. Ensuring equipment safety and predictive maintenance under complex conditions urgently requires vibration warnings and trend forecasting for pumped-storage units. In this study, the measured vibration-signal characteristics of pumped-storage units in a strong background-noise environment are obtained using a noise-reduction method that integrates BA-VMD and wavelet thresholding. We monitored the vibration-signal data of hydroelectric units over a long period of time, and the measured vibration-signal characteristics of pumped-storage units in a strong background-noise environment are accurately obtained using a noise-reduction method that integrates BA-VMD and wavelet thresholding. In this paper, a BP neural network prediction model, a support vector machine (SVM) prediction model, a convolutional neural network (CNN) prediction model, and a long short-term memory network (LSTM) prediction model are used to predict the trend of vibration signals of the pumped-storage unit under different operating conditions. The model prediction effect is analyzed by using the different error evaluation functions, and the prediction results are compared with the predicted results of the four different methods. By comparing the prediction effects of the four different methods, it is concluded that LSTM has higher prediction accuracy and can predict the vibration trends of hydropower units more accurately.

Generalization of neural network models for complex network dynamics

Differential equations are a ubiquitous tool to study dynamics, ranging from physical systems to complex systems, where a large number of agents interact through a graph. Data-driven approximations of differential equations present a promising alternative to traditional methods for uncovering a model of dynamical systems, especially in complex systems that lack explicit first principles. A recently employed machine learning tool for studying dynamics is neural networks, which can be used for solution finding or discovery of differential equations. However, deploying deep learning models in unfamiliar settings-such as predicting dynamics in unobserved state space regions or on novel graphs-can lead to spurious results. Focusing on complex systems whose dynamics are described with a system of first-order differential equations coupled through a graph, we study generalization of neural network predictions in settings where statistical properties of test data and training data are different. We find that neural networks can accurately predict dynamics beyond the immediate training setting within the domain of the training data. To identify when a model is unable to generalize to novel settings, we propose a statistical significance test.

Trajectory error compensation method for grinding robots based on kinematic calibration and joint variable prediction

Trajectory accuracy, a crucial metric in assessing the dynamic performance of grinding robots, is influenced by the uncertain movement of the tool center point, directly impacting the surface quality of processed workpieces. This article introduces an innovative method for compensating trajectory errors. Initially, a strategy for error compensation is derived using differential kinematics theory. Subsequently, a robot kinematic calibration method utilizing ring particle swarm optimization (RPSO) is proposed to address static errors in the grinding robot. Simultaneously, a method for predicting robot joint variables based on a dual-channel feedforward neural network (DCFNN) is designed to mitigate dynamic errors. Finally, a simulation platform is developed to validate the proposed method. Simulation analysis using extensive data demonstrates an 89.3% improvement in absolute position accuracy and a 74.2% reduction in error fluctuation range, outperforming sparrow search algorithm (SSA), improved mayfly algorithm (IMA), multi-representation integrated predictive neural network (MRIPNN), etc. Algorithmic comparison reveals that kinematic calibration significantly reduces the average trajectory error, while joint variable prediction notably minimizes error fluctuation. Validation through trajectory straightness testing and a 3D printing propeller grinding experiment achieves a trajectory straightness of 0.2425 mm. Implementing this method enables achieving 86.1% surface machining allowance within tolerance, making it an optimal solution for grinding robots.

Calibration of discrete meta-parameters of bamboo flour based on magnitude analysis and BP neural network.

In the research and development of technology and equipment for bamboo products deep processing, such as filling, drying, and medicinal use of bamboo flour (BF), the poor compaction and fluidity of BF materials entails the need for accurate discrete element model (DEM) and BF parameters to provide a reference for the simulation of BF processing operationsand the development of related equipment. The average particle size of the 5 types of BFs ranges from 0.136 mm to 0.293 mm, and the small particle size of BF particles causes to the number of BF particles in bamboo processing equipment to reach tens of millions or even billions. When conventional methods are used for simulation, ordinary computers cannot provide the required computing power. To address the aforementioned challenges, this paper proposes a calibration method for the discrete element contact parameters of BFs based on dimensional analysis and a back propagation (BP) neural network. Using particle scaling theory and dimensional analysis methods, the average particle size of the BF was increased to 1 mm, and the main discrete element contact parameters of the five types of BF to be tested were used as input layers. The injection method and sidewall collapse method were used to obtain the angle of repose (AR) as the output layer. Fifty groups were randomly selected using MATLAB for EDEM simulation, and the simulation results were trained using the BP neural network algorithm; an ideal neural network model was obtained, the discrete element parameters of different BFs were predicted, and physical experiments were performed to verify two types of AR and mold hole compression under calibrated parameters. The relative error between the simulated AR obtained through calibration parameters and the physical experimental values is less than 2.3%. Through BF parameter validity verification, the simulated maximum compression displacement and compression ratio after stabilization were 34.81 mm and 0.477, which were close to the actual experimental results of 34.77 mm and 0.461, respectively, verifying the accuracy of the neural network prediction model. The research results provide a reference for the simulation of BF processing operations and the development of related equipment.

Dynamics of finite width Kernel and prediction fluctuations in mean field neural networks*

Abstract We analyze the dynamics of finite width effects in wide but finite feature learning neural networks. Starting from a dynamical mean field theory description of infinite width deep neural network kernel and prediction dynamics, we provide a characterization of the O ( 1 / width ) fluctuations of the dynamical mean field theory order parameters over random initializations of the network weights. Our results, while perturbative in width, unlike prior analyses, are non-perturbative in the strength of feature learning. We find that once the mean field/µP parameterization is adopted, the leading finite size effect on the dynamics is to introduce initialization variance in the predictions and feature kernels of the networks. In the lazy limit of network training, all kernels are random but static in time and the prediction variance has a universal form. However, in the rich, feature learning regime, the fluctuations of the kernels and predictions are dynamically coupled with a variance that can be computed self-consistently. In two layer networks, we show how feature learning can dynamically reduce the variance of the final tangent kernel and final network predictions. We also show how initialization variance can slow down online learning in wide but finite networks. In deeper networks, kernel variance can dramatically accumulate through subsequent layers at large feature learning strengths, but feature learning continues to improve the signal-to-noise ratio of the feature kernels. In discrete time, we demonstrate that large learning rate phenomena such as edge of stability effects can be well captured by infinite width dynamics and that initialization variance can decrease dynamically. For convolutional neural networks trained on CIFAR-10, we empirically find significant corrections to both the bias and variance of network dynamics due to finite width.

Rolled Thickness Prediction for Titanium/Steel‐Clad Plates Based on Combined Method of Theoretical and Neural Network

To address the issue of low precision in rolled thickness prediction for titanium/steel‐clad plates, theoretical calculation and back propagation (BP) neural network prediction models are established based on the slab method and the rolled measured data, respectively. Furthermore, to solve the model's black‐box problem while also improving thickness prediction accuracy, three combined prediction models are built using the principles of compensating multiplicative error, additive error, and average error based on the model's advantages and features for thickness prediction. During the modeling phase, 1925 groups of manufacturing data for titanium/steel‐clad plate rolling are selected and normalized to form the data set. The trial‐and‐error method is used to determine the optimal network structure for the BP neural network. The model's performance is assessed by calculating the mean square error, mean absolute error, and coefficient of determination (R2) and comparing the relative errors in prediction outcomes across different models. In the results, it is demonstrated that the average error combined model (Combined Model III) performs better and predicts more accurately.

Precision and uncertainty in natural gas calorific value estimation: advanced combinatorial predictive models for complex pipeline systems

Abstract In recent years, natural gas pipelines have been characterized by multiple intake points, the mixing of different gas sources, and significant variations in gas quality. The existing volumetric measurement system is no longer suitable for the development and operation of natural gas pipelines in China. To ensure accurate and equitable implementation of natural gas measurement, there is a gradual shift towards energy measurement. However, due to numerous measurement interfaces, installing chromatographs at all measurement stations would lead to substantial investment costs. Therefore, predicting the calorific value of natural gas is a key technology for energy measurement. In view of the issues existing in the current methods of obtaining natural gas calorific value, different prediction models of natural gas calorific value are constructed. According to the influencing factors of natural gas calorific value, the uncertainty evaluation is carried out, and the accuracy of different calorific value prediction models is analyzed by case data. The maximum error of regression equation (RE), response surface analysis (RSA), BP neural network and genetic algorithm (GA) prediction model is 2.29%, 0.43%, 0.59% and 0.45% respectively. However, the combined calorific value (CCV)prediction model is 0.41%. The results indicate that the predicted value calculated by the CCV prediction model is closer to the actual value and has higher prediction performance, which provides a new method for the calorific value prediction and measurement management of natural gas.

Examining the personal growth of college teacher educators through the lens of human development ecology: An approach utilizing artificial neural networks (ANNs) modeling

College teacher educators play a crucial role in improving teaching quality and have a significant impact on educational development. They must engage in a process of self-growth to achieve personal development, considering the interaction between individuals and their environment. This process involves deep exploration of their knowledge, skills, and attitudes, as well as an understanding of the broader institutional and cultural contexts. Through self-reflection and continuous learning, college teacher educators can foster personal growth by acquiring new teaching strategies, developing effective communication and interpersonal skills, and cultivating a reflective and adaptive teaching practice. Collaboration with colleagues, engagement with professional communities, and involvement in educational research and innovation are also essential aspects of personal growth. Creating supportive environments that facilitate the continuous development of college teacher educators is crucial for enhanced teaching quality and educational advancement. In this study, an artificial neural network (ANN) was employed to estimate the impact of changes in knowledge and skill development and reflection and continuous learning on the acquisition of new educational strategies, development of communication and interpersonal skills, and reflective and adaptive practice. The ANN's predictions indicate that increasing reflection and continuous learning has a more positive impact on acquiring new educational strategies compared to increasing knowledge and skill development alone. Additionally, when both knowledge and skill development and reflection and continuous learning are combined, it results in substantial growth in the development of communication and interpersonal skills and reflective and adaptive practice. The accuracy of the neural network's predictions was evaluated using linear regression, showing an acceptable level of error when compared to the experimental results.

A Novel Neural Network Prediction Model for Mineral Content Based on the Data of Elemental Logging and Well Logging

The mineral content of shale plays an important role in the study of reservoir characteristics. Obtaining high-precision mineral content data from a simple and straightforward method is helpful for the study of reservoir properties. Due to the complex and varied elemental combination of minerals and the lithological diversity in different formations, there are limitations in the traditional statistical methods for predicting mineral content. In this study, a multi-source fusion dataset is constructed first using a detailed workflow to process multi-source data from elemental logging and conventional well logging from the Funing shale formations in China. A hybrid neural network model consisting of a back-propagation artificial neural network (BP-ANN) and gated recurrent unit (GRU) structure, with a custom constraint loss function, is developed to predict the mineral contents. The results show that the predictive performance of this hybrid multi-source model performs statistically significantly better than that of the BP-ANN and GRU models using single source data. In addition, the accuracy of the hybrid model is further improved by using the custom constraint loss function. This method provides high-resolution distributions of mineral content compared to the X-ray powder diffraction testing method. It will help enhance the understanding of the formation.

Training machine learning interatomic potentials for accurate phonon properties

Abstract One of the major challenges in the development of universal machine learning interatomic potentials is accurately reproducing phonon properties. This issue appears to arise from the limitations of available datasets rather than the models themselves. To address this, we develop an extensive dataset of phonon calculations using density-functional perturbation theory. We then show how this dataset can be used to train neural-network force fields, by implementing the training and the prediction of force constants in periodic crystals. This approach improves the quality of phonon properties prediction while reducing the number of structures needed for neural network training. We demonstrate the efficiency of this method using two examples of ternary phase diagrams: Ti—Nb—Ta and Li—B—C. In both cases, neural network predictions for the energy and forces show a considerable improvement, while phonon properties are predicted with high precision for all structures across the entire phase diagrams.

Integrated optimization control strategy for vehicle longitudinal and lateral assisted driving

In order to coordinate the control objectives of the longitudinal and lateral auxiliary driving system of the vehicle and the vehicle dynamics control requirements, this paper takes the four-wheel independent drive electric vehicle as the control object, and combines deep learning and model predictive control to establish the vehicle generalized control force prediction-dynamic optimization distribution control strategy. The upper-level algorithm calculates the generalized control force that meets the vehicle control requirements according to the reference objectives of the longitudinal and lateral auxiliary driving strategies; the middle-level module includes the vehicle expected generalized force neural network prediction model and the active set numerical solution algorithm; the lower level calculates the tire slip rate and sideslip angle according to the expected tire force, realizes the tire force control requirements by controlling the motor torque and the braking system, and adjusts the vehicle movement direction through the steering system. This paper uses simulation tests and hardware-in-the-loop tests to verify the algorithm. The results show that the established algorithm can simultaneously meet the longitudinal and lateral control requirements of the vehicle, with a lateral path following error of 0.0117 m and a vehicle speed error of 0.59 km/h.

Prediction of carbon dioxide levels in the European Alps based on machine learning algorithms

With the rapid development of world industrialization, the continuous increase of global carbon dioxide emissions has led to the gradual deterioration of the ecological environment and the obvious aggravation of the greenhouse effect. In this paper, the carbon dioxide content in the air over the European Alps is taken as the research object, and the RNN and LSTM neural network prediction models are respectively used to compare and predict it. The results show that the fitting effect of LSTM is better than that of RNN, and the fitting effect of the prediction model will also improve with the increase of iteration times and sample size. Since carbon monoxide and methane in the air will cause changes in carbon dioxide content, this paper adds the two factors into the LSTM prediction model as influencing factors, and the goodness-of-fit can reach 0.95, which is higher than the prediction results when there are only one influencing factor or no influencing factor. In order to reduce the running time, XGBoost, LightGBM and random forest algorithms are respectively used in this paper and Bayesian optimization algorithm is used to predict the carbon dioxide content. The results show that the prediction effect is slightly lower than LSTM. Therefore, this paper takes the above three algorithms as the base model. Linear regression experiments are carried out for the meta-model's Stacking fusion algorithm. The goodness-of-fit can reach 0.92, which significantly improves the prediction effect compared with the base model. Finally, the sensitivity analysis of the Stacking fusion model is carried out in this paper. The experimental results show that the model has strong stability.

Multi-branch deep learning neural network prediction model for the development of angular biosensors based on sEMG.

Human gait motion intention recognition is very important for the lower extremity exoskeleton robot to accurately synchronize and respond to the user's natural motion. And motion intention recognition is generally performed through sEMG. Deep learning neural networks perform well in dealing with high-dimensional data and nonlinear relationships such as sEMG, but different deep learning neural networks have their own advantages in dealing with different types of data. Therefore, a multi-branch deep learning neural network, which enables different neural networks to process different feature items, could achieve more accurate and efficient motion intention recognition. The purpose of this study is to 1) Establish a multi-branch deep learning neural network model to achieve accurate gait recognition and effective estimation of joint angles. 2) Quantify the performance of the multi-branch deep learning neural network model in gait recognition and joint angle prediction using sEMG. This study involved the collection of sEMG and plantar pressure data during walking in human subjects. Firstly, the collected signals are filtered and denoised to ensure the quality and reliability of the data. Calculate the time domain features and the frequency domain features to capture the key information of gait. Then, using the sensitivity difference of different structural neural networks to different feature data, a multi-branch deep learning neural network model is developed, in which the extracted features are used as the input of the model. The output of the model includes gait cycle and joint angle, so as to realize the accurate recognition of human gait and the effective estimation of joint angle. The results show that the proposed method has high accuracy in identifying human gait and estimating joint angles. The multi-branch neural network model successfully integrates time-domain and frequency-domain features and provides reliable prediction of gait cycle and joint angle. The highest accuracy of gait recognition is 95.42%, the lowest is 90.11%, and the average is 92.16%. The average error of joint angle estimation is 3.19. This study designed a human walking gait recognition and joint angle prediction model to achieve accurate human lower limb motion intention recognition.The model can be integrated into the sEMG sensor to design a angular biosensors, which can predict the human joint angle in real time.

A Comparison of PSO-based Neural Network Predictions of F2 Ionospheric Layer Critical Frequency in Malaysia with IRI-2016

This paper discussed the prediction of the equatorial ionospheric F2 layer critical frequencies, foF2, using a backpropagation neural network (BPNN) model in conjunction with a particle swarm optimization (PSO) algorithm, named as the BPNN–PSO model, for various solar and geomagnetic activities, and compared it with the IRI-2016 model. The critical frequency data were taken from an ionosonde located at Universiti Tun Hussein Onn Malaysia in Johor, Malaysia (1.86° N, 103.80° E). The model’s outputs were analyzed using root-mean-square error (RMSE). The BPNN–PSO model outperformed the IRI-2016 model during low, medium, and high solar activity. The BPNN–PSO model had the lowest RMSE of 0.20 MHz and performed best during periods of high solar activity. This is much better than the RMSE for IRI-2016 model, which was 2.95 MHz. In addition, compared with the IRI-2016 model, the BPNN–PSO model made accurate predictions during both quiet and geomagnetic storm conditions. The BPNN–PSO model had the lowest RMSE of 0.54 MHz during an intense storm, while the IRI-2016 model had the RMSE of 2.84 MHz during this storm.

A method for optimizing different geometric shields of D-T neutron generators by combining BP neural network and Analytic Hierarchy Process

In this paper, an optimization of shielding structures with different geometries is established for the D-T neutron generator system by combining Back Propagation (BP) neural network and Analytic Hierarchy Process (AHP). The D-T neutron generator (Model NG-9) used in the system was developed independently by Northeast Normal University. After investigating the rule of shielding performance among spherical, cylindrical and cubic geometries, the spherical shield is selected for BP neural network prediction to determine the total dose rate through it. Information about spherical multilayer-shielding structures and properties calculated by MCNP code is used to train the neural network. The predicted result serves as a parameter of the evaluation function, which provides a comprehensive assessment of the dose rate penetrated the shield, the shielding mass, and the shielding volume. Together with AHP, the weight factors are determined for all the optimization objectives to construct the evaluation function. By comparing its values, the optimal shielding structures for spherical, cylindrical and cubic materials are found. Against MCNP simulated values, the total dose rates’ errors of the optimal shielding structures for the sphere, cylinder, and cube are 1.72 %, -4.94 %, and -5.17 %, respectively. This result demonstrates that the combination of BP neural network and AHP is more effective in addressing multi-objective optimization problems related to the design of radiation shielding for various geometries.

Fully Interpretable Deep Learning Model Using IR Thermal Images for Possible Breast Cancer Cases.

Breast cancer remains a global health problem requiring effective diagnostic methods for early detection, in order to achieve the World Health Organization's ultimate goal of breast self-examination. A literature review indicates the urgency of improving diagnostic methods and identifies thermography as a promising, cost-effective, non-invasive, adjunctive, and complementary detection method. This research explores the potential of using machine learning techniques, specifically Bayesian networks combined with convolutional neural networks, to improve possible breast cancer diagnosis at early stages. Explainable artificial intelligence aims to clarify the reasoning behind any output of artificial neural network-based models. The proposed integration adds interpretability of the diagnosis, which is particularly significant for a medical diagnosis. We constructed two diagnostic expert models: Model A and Model B. In this research, Model A, combining thermal images after the explainable artificial intelligence process together with medical records, achieved an accuracy of 84.07%, while model B, which also includes a convolutional neural network prediction, achieved an accuracy of 90.93%. These results demonstrate the potential of explainable artificial intelligence to improve possible breast cancer diagnosis, with very high accuracy.

Artificial neural network models for forecasting the extracted yield of essential oils from Curcuma longa L. by hydro-distillation

Abstract Turmeric is commonly acknowledged in traditional medical practices for its strong healing properties. In the present work, hydro-distillation was employed to extract essential oils from turmeric powder. The response surface methodology (RSM) was applied to investigate the effects of various parameters, including evaporation rate (0.23, 0.5, 0.8, 0.97, 1.36, 2 ml/min), solid/liquid ratio (4:100, 6:100, 8:100, 1:10, 11:100 g/ml), and extraction duration (13–250 min) on the yield of essential oils. The central composite design (CCD) proved to be an effective tool for evaluating the extraction yield of essential oils. A three-layer artificial neural network (ANN) was utilized to develop the extraction model, employing the Levenberg–Marquardt (LM) optimization algorithm. The neural network’s input layer comprised the solid/liquid ratio, evaporation rate, and extraction time, while the output layer indicated the yield of essential oil extraction. The most appropriate model included a hidden layer with 16 neurons, achieving R 2 and MSE values of 0.9989 and 0.0013, respectively. This investigation indicates that an artificial neural network prediction model serves as an effective method for estimating essential oil yield.

A Spatial Interpolation Method for Meteorological Data Based on a Hybrid Kriging and Machine Learning Approach

ABSTRACTConventional spatial interpolation methods for meteorological data are usually based on linear interpolation. However, with the improvements in the temporal and spatial resolution of observational data, local neighbouring stations are susceptible to the influence of underlying surface changes and high terrain gradients. Moreover, for interpolation at a single time point, the inability to extract continuous change information effectively from adjacent times limits the interpolation performance. In this paper, an improved hybrid deep learning‐kriging method is proposed that combines a graph neural networks (GNNs) prediction model with the kriging interpolation algorithm. The GNNs considers dynamic changes over time and combines spatial and temporal information to estimate (interpolate) meteorological data at target weather stations using reanalysis data as input. The experimental results show that the hybrid method exhibits good performance in interpolating station data in complex terrain areas and under uneven surface conditions. The interpolation effectiveness of this method is markedly improved compared to that of traditional kriging methods. Moreover, when applied to station‐to‐grid interpolation, the hybrid method still provides better interpolation results than those of kriging methods. Therefore, this research provides a new method and perspective for meteorological data interpolation.

Neural Networks for Prediction of 3D Printing Parameters for Reducing Particulate Matter Emissions and Enhancing Sustainability

Neural Networks for Prediction of 3D Printing Parameters for Reducing Particulate Matter Emissions and Enhancing Sustainability