High-precision Prediction Research Articles

Modelling and optimization of well hole cleaning using artificial intelligence techniques

Ineffective hole cleaning in deviated and horizontal well drilling can lead to issues like stuck pipes, reduced rate of penetration (ROP), and drill bit damage, resulting in increased non-productive time (NPT) and operational costs. Traditional methods for assessing hole cleaning rely on experimental and empirical models that often fail to account for all influencing factors and lack real-time applicability. This study aims to improve the accuracy and practicality of hole cleaning assessment by applying Artificial Intelligence (AI) techniques, specifically Artificial Neural Networks (ANN) and Genetic Algorithms (GA), to predict downhole parameters and optimize drilling processes. These AI methods analyze the impact of key drilling parameters—such as weight on bit (WOB), ROP, rock geomechanics, drilling fluid characteristics, and rig hydraulics—on hole cleaning. Results demonstrated that AI-driven models provide high-precision predictions and enable real-time optimization, significantly reducing NPT and enhancing drilling efficiency and safety. In conclusion, AI techniques like ANN and GA offer a robust solution to improve hole cleaning, overcoming limitations of traditional methods and contributing to safer, more cost-effective drilling operations.

High Precision Prediction of Structure and Thermal Properties of Ternary Eutectic Carbonates by Machine Learning Potential for Solar Energy Application

High Precision Prediction of Structure and Thermal Properties of Ternary Eutectic Carbonates by Machine Learning Potential for Solar Energy Application

Fast Reverse Design of 4D-Printed Voxelized Composite Structures Using Deep Learning and Evolutionary Algorithm.

Designing voxelized composite structures via 4D printing involves creating voxel units with distinct material properties that transform in response to stimuli; however, optimally distributing these properties to achieve specific target shapes remains a significant challenge. This study introduces an optimization method combining deep learning (DL) and an evolutionary algorithm, focusing on a solvent-responsive hydrogel as the target material. A sequence-enhanced parallel convolutional neural network is developed and generated a dataset through finite element simulations. This DL model enables high-precision prediction of hydrogel deformation. Furthermore, a progressive evolutionary algorithm (PEA) is proposed by integrating the DL model to construct a DL-PEA framework. This framework supports rapid reverse engineering of the desired shape, and the average design time for specified target shapes is reduced to ≈3.04 s. The present findings illustrate how 4D printing of optimized hydrogel designs can effectively transform in response to environmental stimuli. This work provides a new perspective on the application of hydrogels in 4D printing and presents an efficient tool for optimizing 4D-printed voxelized compositestructures.

High-Precision prediction of curling trajectory multivariate time series using the novel CasLSTM approach

As a multivariate time series, the prediction of curling trajectories is crucial for athletes to devise game strategies. However, the wide prediction range and complex data correlations present significant challenges to this task. This paper puts forward an innovative deep learning approach, CasLSTM, by introducing integrated inter-layer memory, and establishes an encoder-predictor curling trajectory forecasting model accordingly. Additionally, tailored training techniques involving non-teacher-forcing, ExMSE loss and incremental multi-trajectory learning are devised to enhance model performance. Notably, the model demonstrates astounding accuracy, achieving sub-1cm average errors over 30m trajectories, outperforming vanilla LSTM by 41.8%. It also showcases robustness across various curling settings, with strict validation metrics on a static test set further verifying precision. Field test results reveal promising predictive capabilities for real-world scenarios as well, exhibiting applicability. The proposed technique liberates data-driven curling stone trajectory prediction from sole reliance on analytical models and tackles key challenges of long sequence forecasting. The presented technologies and insights could also generalize to prediction tasks in other remote trajectories and multivariate time series domains.

From non-affinity to high-affinity: Rapid preparation of nanobodies utilizing high-precision alphafold and structural-interaction analysis for detection of enrofloxacin in marine fish.

Traditional antibody preparation methods rely on animal immunization and experimental screening, often requiring 3-6 months, which cannot meet the urgent demands for antibodies in environmental emergencies or contamination events. Herein, a rapid antibody preparation method was proposed to transform anti-macromolecule nanobodies into high-affinity anti-small molecule (enrofloxacin) nanobodies, based on the high-precision predictive capabilities of AlphaFold, integrated with structural and interaction analysis. The high-precision prediction capability of AlphaFold ensured the accuracy of nanobody structural analysis and targeted modification. Binding structure design enabled the antigen and antibody to adopt an optimal binding conformation. Interaction analysis provided guidance for targeted modifications and served as a tool for evaluating the results. Molecular docking results revealed that the template nanobodies, which initially couldn't dock with enrofloxacin, formed 3-5 noncovalent bonds after modification. Bio-layer interferometry (BLI) assays demonstrated that the equilibrium dissociation constant (KD) of the modified nanobodies ranged from 5.56 ± 0.34 nM to 94.73 ± 4.35 nM, with half-maximal inhibitory concentration (IC50) between 231.9 and 3293.6 ng/mL. Based on the modified nanobodies and BLI platform, the detection of enrofloxacin in marine fish was achieved with a limit of detection (LOD) as low as 59.97 μg/kg. In summary, this study provided a novel perspective for the rapid nanobody preparation, showing significant potential in food safety and environmental monitoring.

Research on the Prediction Model of Coal Spontaneous Combustion Hazard Level Based on IGWO-GRNN

ABSTRACT To accurately predict the risk level of spontaneous coal combustion, we introduce an improved grey wolf optimization strategy (IGWO) to optimize the generalized regression neural network (GRNN), thereby establishing the IGWO-GRNN coal spontaneous combustion risk level prediction model. The superiority of the IGWO algorithm over the grey wolf optimization algorithm (GWO), whale optimization algorithm (WOA) and particle swarm optimization algorithm (PSO) was verified. Fifty sets of sample data of spontaneous coal combustion in the mining area were selected as the research object, and the data were randomly divided into the training set and the test set according to the ratio of 8:2. The prediction results of the IGWO-GRNN prediction model were obtained and compared with those of the GWO-GRNN, GRNN, BP (Backpropagation), and GRU (Gated Recurrent Unit) models, and the absolute value of the relative error was introduced as an evaluation index of the model prediction accuracy. The results demonstrate that IGWO outperforms GWO, WOA, and PSO in terms of optimization efficiency, with the IGWO-GRNN prediction model yielding the most accurate results. Specifically, the maximum absolute value of the relative error does not exceed 4.27%, and the average relative error is only 3.29%, indicating a substantial improvement over the other models. Therefore, this study achieves high-precision prediction under small-sample data, which provides an important reference value for the small-sample prediction problem in the mining field.

Data-Driven Approach for the Prediction of In Situ Gas Content of Deep Coalbed Methane Reservoirs Using Machine Learning: Insights from Well Logging Data.

The in situ gas content is a critical determinant of the exploitation potential and recovery of coalbed methane (CBM) resources. Deep CBM resources have enormous exploitation potential, but their intricate geological conditions hinder the acquisition of in situ gas content data. To enhance the efficiency and accuracy of acquiring in situ gas content data for deep CBM, this study integrates gray relational analysis (GRA) and the genetic algorithm (GA) into the back-propagation neural network (BPNN) model, establishing a novel prediction model for in situ gas content of deep CBM using well logging data. The results show that the multialgorithm joint model can overcome the inherent shortcomings of the BPNN. The GRA method effectively identifies the optimal input parameters for the BPNN model, the GA method optimizes the initial weights and thresholds of the BPNN, thereby enhancing the prediction accuracy and stability of the model. The mean square error (MSE) of the GRA-GA-BPNN joint model decreases by 77.60% compared with the BPNN model. Furthermore, taking the deep CBM wells in the Ningwu Basin of North China as an example, the reliability of the multialgorithm joint model was verified (4.62% average relative error only). The GRA-GA-BPNN model proposed in this study exhibits high robustness and strong generalization ability. It can achieve high-precision prediction of deep CBM in situ gas content, thereby circumventing overreliance on experimental measurements, holding significant practical application significance.

D3-ImgNet: A Framework for Molecular Properties Prediction Based on Data-Driven Electron Density Images.

Artificial intelligence technology has introduced a new research paradigm into the fields of quantum chemistry and materials science, leading to numerous studies that utilize machine learning methods to predict molecular properties. We contend that an exemplary deep learning model should not only achieve high-precision predictions of molecular properties but also incorporate guidance from physical mechanisms. Here, we propose a framework for predicting molecular properties based on data-driven electron density images, referred to as D3-ImgNet. This framework integrates group theory, density functional theory-related mechanisms, deep learning techniques, and multiobjective optimization mechanisms, embodying a methodological fusion of data analytics and system optimization. Initially, we focus on atomization energies as the primary target of our study, using the QM9 data set to demonstrate the framework's ability to predict molecular atomization energies with high accuracy and excellent exploration performance. We then further evaluate its predictive capabilities for dipole moments and forces with the QM9X data set, achieving satisfactory results. Additionally, we tested the D3-ImgNet framework on the SN2 reaction data set to demonstrate its ability to precisely predict the minimum energy paths of SN2 chemical reactions, showcasing its portability and adaptability in chemical reaction modeling. Finally, visualizations of the electronic density generated by the framework faithfully replicate the physical phenomenon of electron density transfer. We believe that this framework has the potential to accelerate property predictions and high-throughput screening of functional materials.

Information Provision from a Platform to Competing Sellers: The Role of Strategic Ambiguity

Problem definition: With the rapid growth of e-commerce, platforms are able to gather large quantities of data, which can result in high-precision predictions regarding consumer purchasing patterns and future demand. A fundamental question is whether a platform has the incentives and ability to share such nonverifiable information with its sellers. Methodology/results: We show that when information is nonverifiable, sharing the precise value of the platform’s private information by means of cheap-talk cannot result in an equilibrium. This outcome is due to the incentive of the platform to portray a more favorable market condition than that predicted, in order to encourage the sellers to raise their prices and improve market efficiency. In spite of this negative result, we demonstrate that the level of incentives misalignment between the platform and its sellers is bounded; consequently, a region-forecast information-sharing equilibrium can emerge. In this equilibrium, the support of the platform’s private information is divided into several intervals, and the platform strategically chooses to report truthfully the interval containing its private information. The structure of the partition is influenced by two main factors: the incentive of the platform to reduce market uncertainty for the sellers, and the motivation of the platform to soften competition among the sellers. Although both the sellers and the platform benefit from the ability to share some level of information, such an outcome hurts the consumers. Managerial implications: This work explains the observed practice of a platform providing nonverifiable information to its sellers via cheap-talk. The main advantage of this equilibrium is the ability to share information in a costless manner; however, the amount of information that can be shared is limited and is influenced by the level of market competition between the sellers. Funding: N. Shamir gratefully acknowledges financial support from the Israel Science Foundation [Grant 2358/22] and The Henry Crown Institute of Business Research in Israel. T. Avinadav and T. Chernonog gratefully acknowledge financial support from the Israel Science Foundation [Grant 1571/20]. Supplemental Material: The electronic companion is available at https://doi.org/10.1287/msom.2023.0262 .

K-Nearest Neighbors Approach to Analyze and Predict Air Quality in Delhi

The study considers the community of ”urban air quality improvement in modern cities” using an extensive dataset obtained from ”Air quality data of Delhi, India” for the period between 25 November 2020 and 24 January 2023. Research aims to significantly reduce air pollutants, including particulate matter, including PM2.5 and PM10, NO2, SO2, CO2, O3, and others. Different machine learning models are being used for airquality level forecasts. Among the models assessed, the Nearest Neighbors algorithm comes out on top and exhibits a very low Mean Squared Error (MSE) of 0.0002. The model’s superb precision is further supported by very low statistics in other key metrics, which confirm the Nearest Neighbors approach to forecasting the quality of air in urban zones. The Nearest Neighbors algorithm is shown to have its place in the application as a tool in the hands of researchers and decision-makers pursuing the fight against air pollution is also a signal of its efficiency and broad applicability. This modeling approach has thus the potential to first identify and later pinpoint localized empirical patterns and statistical dependencies from the data set. Its high predictive precision makes it a very valuable assistant to public health and environmental management, especially so in regions like Delhi.

A noninvasive blood glucose detection method with strong time adaptability based on fuzzy operator decision fusion and dynamic spectroscopy characteristics of PPG signals.

PPG signals are a new means of non-invasive detection of blood glucose, but there are still shortcomings of poor time adaptability and low prediction accuracy of blood glucose quantitative models. Few studies discuss prediction accuracy in the case of a large time interval span between modeling and prediction. This paper proposes an automatic optimal threshold baseline removal algorithm based on variational mode decomposition (AOT-VMD), which can adaptively eliminate high-frequency noise and baseline interference for each decomposed IMF modal component and reduce the baseline difference of PPG signals from different days. Furthermore, a fuzzy integral multi-model decision fusion algorithm based on error weight is proposed. The fuzzy integral operator is introduced to make the features with large contributions in each sub-model maintain a high-weight value in the overall prediction mechanism, which improves the prediction accuracy of blood glucose. In this paper, a self-developed portable PPG glucose meter is used to collect PPG signals, and the true blood glucose values for 8 consecutive days are collected by CGM. The proposed algorithm is used to build a model with the first day's data and predict the blood glucose values for the remaining 7 days. The experimental results show that the AOT-VMD preprocessing algorithm and the quantitative regression algorithm of the fuzzy integral multiple model decision fusion algorithm proposed in this paper perform well in measurement accuracy and time adaptability compared with the traditional methods. In addition, the proposed method requires less invasive calibration samples in the modeling stage, achieving high-precision prediction for a long period. 100% of the samples are located in areas A and B of the Clarke area in this experiment, and the algorithm has strong time generalization ability. This innovative method can promote the development of a home blood glucose noninvasive detector.

Design and application of CNN for emission detection through thermal imagery

Motorcycle exhaust emissions (EE) that do not meet regulatory standards present a significant environmental and public health issue, particularly given the rising number of motorcycles in densely populated areas. These emissions release pollutants such as carbon monoxide (CO), hydrocarbons (HC), and nitrogen oxides (NOx), which contribute to poor air quality and have adverse effects on human health. Traditional emission testing methods using gas analyzers, while commonly used, face limitations such as sensitivity to environmental fluctuations, the necessity for frequent recalibration, and an intensive testing process requiring specialized expertise. This study addresses these issues by developing an innovative method for emission detection using Convolutional Neural Networks (CNN) applied to thermal images of motorcycle exhausts. The research method involves five key stages: data acquisition, dataset formation, CNN model design and training, model testing, and validation. Thermal images were gathered from 27 motorcycles, representing various brands and engine configurations common in Indonesia, and each image set included 100 samples for both emission-compliant and non-compliant categories. By analyzing thermal patterns, the CNN model was trained to accurately detect combustion patterns indicative of emission status based on the lambda value. This approach enables the model to generalize across different motorcycle models, offering practical adaptability for widespread implementation. The results demonstrate that the CNN model delivers high predictive accuracy, precision, recall, and F1-score, making it a robust tool for assessing motorcycle emission compliance. This CNN-based approach provides a practical solution for real-time, large-scale emission monitoring and regulatory enforcement, reducing dependency on conventional methods. Its scalability and adaptability position it as a valuable advancement in emission monitoring technology, with significant potential for supporting environmental standards and improving air quality management

Construction and control parameter matching method of hydraulic pile hammer surrogate model combining transfer algorithm and knowledge embedding

Due to its complex application environment, hydraulic pile hammers are difficult to model and accurately grasp their motion laws in automation control. Therefore, to address the high-precision prediction problem of the motion law of hydraulic pile hammer systems under complex working conditions, a knowledge embedding surrogate model based on a dual neural network is designed, which can be divided into two parts: feature extraction network and prediction network. On this basis, the model introduces physics knowledge embedding technology, adds inequality constraints based on physical laws to the loss function, and introduces transfer learning algorithms to maintain high accuracy of the model under limited data conditions. The results showed that the PSM (γ = 0.5) model designed in the study had a median of -0.001 and a variance of 0.038 under undisturbed conditions, demonstrating high prediction stability. Under perturbation conditions, the median of the PSM (γ = 1.5) model was -0.003, with a variance of 0.031, indicating the highest prediction accuracy. From this, the research model has good prediction accuracy under complex working conditions, which can lay a technical foundation for predicting the motion law of hydraulic pile hammers and automatic control.

A preclinical design approach for translation of biohybrid photosensitive nanoplatform for photodynamic therapy of breast cancer.

Translating photodynamic therapy research into practice with high efficacy should follow characterization and optimization steps in an integrated process. In this way, integration of molecular-based simulations with in vitro and in vivo studies to produce more accurate models with higher prediction precision is inevitable. This study reports on the development a two-phase approach to design and synthesize, and characterize novel hybrid nano-photosensitizers to target breast cancer using molecular docking, microfluidic-based testing and animal studies. In the first phase, using pkCSM artificial intelligence tool and molecular docking, pharmacokinetic weaknesses of photoprotoporphyrin and its retention potential within albumin protein were identified. Biohybrid photosensitive nanoplatform were then synthesized using opto-microfluidics and characterized. In addition, an in-vivo method was used to qualitatively evaluate the opto-biological stability of the final optimized biohybrid nanoplatform. At the second phase, variables of photodynamic treatment were firstly optimized using design of experiment method for the optimized biohybrid photosensitive nanoplatform. Then, on-chip photodynamic studies were carried out in static and dynamic conditions. Results revealed that optimized biohybrid photosensitive nanoplatform as a nano-photosensitizer induced significant death of triple negative human cancer cells in static and dynamic cell cultures under optimized irradiation conditions. Consequently, the presented multiphase study that combined in-silico simulations with microfluidic-based synthesis and characterizations of nano-photosensitizers provided a more convergent model for development of efficacious cancer treatment for photodynamic therapy applications.

A Comparative Study of Electric Vehicles Battery State of Charge Estimation Based on Machine Learning and Real Driving Data

Electric vehicles (EVs) are rising in the automotive industry, replacing combustion engines and increasing their global market presence. These vehicles offer zero emissions during operation and more straightforward maintenance. However, for such systems that rely heavily on battery capacity, precisely determining the battery’s state of charge (SOC) presents a significant challenge due to its essential role in lithium-ion batteries. This research introduces a dual-phase testing approach, considering factors like HVAC use and road topography, and evaluating machine learning models such as linear regression, support vector regression, random forest regression, and neural networks using datasets from real-world driving conditions in European (Germany) and African (Morocco) contexts. The results validate that the proposed neural networks model does not overfit when evaluated using the dual-phase test method compared to previous studies. The neural networks consistently show high predictive precision across different scenarios within the datasets, outperforming other models by achieving the lowest mean squared error (MSE) and the highest R2 values.

A Machine Learning-Based Multi-Model Approach for Predicting Post-Deep Brain Stimulation Outcomes in Alzheimer’s disease Patients

Objectives and Background: Deep-brain stimulation (DBS) is a demonstrated rehabilitation for Alzheimer's Disease (AD), regularly resulting in an upgrade of engine work. In any case, some unfortunate incidental upshots can occur past DBS, which can deteriorate the patient's satisfaction. In this way, the medical group must painstakingly choose patients over whom to perform DBS. In the past, there have been a few endeavors to associate pre-usable information and DBS experimental results, generally centered on engine symptomatology. Objectives & Methodology: A machine learning (ML)-based strategy called a multimodel pipeline alluded to as FlowModel is designed and implemented to anticipate an enormous number of DBS medical results for AD. It is composed of a patented Artificial Neural Network (ANN) for processing medical data, using a conventional image handling technique to extract morphological biomarkers from T1 imaging, and employing a Support Vector Machine (SVM) to perform regressions. We authenticated it in 256 AD patients who underwent DBS. Results: The FlowModel demonstrated high predictive precision, achieving a correlation coefficient of as high as 0.82 and successfully predicting 63 out of 84 clinical outcome scores. Importantly, FlowModel outperformed traditional linear models, demonstrating the capability to learn effectively from preoperative data alone. Conclusion: The uniqueness of this study lies in its ability to predict multiple clinical outcomes across different modalities, independent of stimulation parameters. The introduction of this ML pipeline contributes to the precision medicine landscape by providing a robust, data-driven tool to select patients for DBS, potentially improving clinical decision-making and patient outcomes.

Study of spot distance on resistance spot welding quality: a 1DCNN-BiLSTM-Attention-based online inspection method

Abstract Resistance spot welding (RSW) is widely employed in the automotive and home appliance industries due to its high efficiency, low cost, and suitability for automation. However, traditional quality detection methods rely on destructive testing, leading to inefficiencies and resource wastage. This paper presents a novel quality inspection model for RSW that utilizes a one-dimensional convolutional neural network, bidirectional long short-term memory network, and attention mechanism (1DCNN-BiLSTM-Attention) to address the challenges of extracting temporal data under varying spot distances. The model integrates a residual linking mechanism and Kolmogorov–Arnold Networks (KAN) to enhance feature extraction and performance. Experimental results reveal that the model demonstrates strong predictive capabilities across different spot distances, with particularly notable performance at 10 mm spacing, achieving a mean absolute error (MAE) of 0.0632, a root mean square error (RMSE) of 0.0603, and an R² value of 0.7513. These findings underscore the model’s ability to provide high-precision predictions, even under conditions influenced by significant shunt effects.

Crushing Force Prediction Method of Controlled-Release Fertilizer Based on Particle Phenotype

This study proposed a method to predict the crushing force of controlled-release fertilizer granules based on their phenotypic characteristics to prevent coating damage during production, transport, and fertilization, which could affect nutrient diffusion rates. The phenotypic features, including sphericity, particle size, and texture, of three commonly used controlled-release fertilizers were obtained using machine vision, while the crushing force was measured using a universal testing machine. A principal component analysis was applied for data reduction, and the optimal parameters for the support vector machine (SVM) were selected using particle swarm optimization (PSO) combined with k-fold cross-validation. A particle swarm optimization–support vector machine (PSO-SVM) model was then developed to predict the crushing force based on fertilizer shape features. Compared with the traditional method, the innovation of this paper is that a non-destructive prediction method is proposed, which enables high-precision predictions of the crushing force by integrating multi-dimensional phenotypic features and an intelligent optimization algorithm. Comparative tests with a random forest regression, the K-nearest neighbor, a back propagation (BP) neural network, and a long short-term memory (LSTM) neural network have demonstrated that the PSO-SVM model outperforms these methods in terms of mean absolute error, root mean square error, and correlation coefficient, underscoring its effectiveness. The proportion of predictions within the −10% to +10% error range reached 0.82, 0.82, and 0.86 for the three fertilizers, confirming the high reliability and accuracy of the PSO-SVM method for non-destructive testing.

Degradation behavior judgment-enabled improved transferable squeeze aggregated excitation remaining useful life prediction network for rolling bearing

Abstract Remaining useful life prediction of rolling bearings in the various operating conditions is of significance for improving the reliability and durability of rotating machinery. However, the precision of RUL prediction is heavily affected by the significant differences in data feature distributions due to the different failure degradation behaviors of bearings and the various operating conditions. To address this issue, a novelty RUL prediction method based on degradation behavior judgment is proposed. Firstly, a method is proposed to judge the failure behavior, the deep convolutional autoencoder is used to construct the prior gradient health indicator and current health indicators of rolling bearings, and dynamic time warping is used to measure the difference between the two to judge failure degradation behavior. Then, an improved squeeze aggregated excitation network is proposed for RUL prediction. The squeeze aggregated excitation module, which embeds the residual shrinkage denoise module, is used to learn the mapping relationship between features and RUL to obtain comprehensive information. The maximum mean discrepancy is introduced into the final layer of the improved squeeze aggregated excitation network to reduce feature distribution differences. Results of the experiment show that this model has higher prediction precision and better robustness than other methods under three working conditions in the PHM2012 dataset.

Spatiotemporal prediction of solidified dendrites based on convolutional long-short-term neural network

The formation and growth of dendrites during the solidification of metals are critical factors influencing material properties. Accurate prediction of dendritic growth morphology and evolution is of great significance for materials science research and industrial applications. Traditional simulation methods, such as the phase-field method, provide high-precision results but are computationally expensive and time-consuming. Recently, with the advancement of deep learning technology, data-driven approaches have shown great potential in materials science. This paper proposes the use of Convolutional Long-Short-Term Memory Network (ConvLSTM) to capture spatiotemporal dependencies for predicting the spatiotemporal evolution of dendritic solidification in materials. By training on historical data of dendritic growth, the ConvLSTM neural network model predicts the future growth morphology of the dendritic microstructure during solidification. The model accurately forecasts the future morphology and appearance of the microstructure. Case studies were conducted using a generated dataset of dendritic solidification simulations, and the results demonstrate that the ConvLSTM model effectively captures the complex dynamic characteristics of dendritic growth and provides high-precision prediction results. This offers new insights for material design and optimization.