Achieved Prediction Accuracy Research Articles

Long-Term or Short-Term? Prediction of Ship Detention Duration Based on Machine Learning

The prevalence of ship deficiencies continues to be a significant issue. Data from the Tokyo Memorandum of Understanding reveals that ship detentions in 2023 surged by more than 80% compared with the previous year. The significant number of detained ships not only disrupts ships’ daily operations but also strains port resources, leading to increased additional costs. In light of this issue, predicting the duration of ship detention becomes crucial, as accurate predictions can assist port managers in resource allocation and provide shipping companies with critical information for operational planning. This study is the first to predict ship detention duration, specifically distinguishing between long-term and short-term detained ships. Initially, key deficiency types influencing the ship detention duration were identified using an improved entropy weight–grey relational analysis. Subsequently, in consideration of the imbalance in data distribution between long-term and short-term detentions, a random forest model capable of handling imbalanced data was applied to classify these two types. The study found that fire safety, propulsion and auxiliary machinery, and pollution prevention are the three most critical deficiency types impacting detention duration; and the random forest model sampled and processed from the data level possessed the best model performance, achieving prediction accuracies of 0.71, 0.72, and 0.85 for bulk carriers, containers, and oil tankers, respectively. This research offers a comprehensive analysis of ship detention duration, making a significant contribution to both the theoretical understanding and practical applications in the maritime industry. Accurately predicting ship detention duration provides valuable insights for stakeholders, enabling them to anticipate potential detention scenarios and thus supporting shipping companies in effective fleet management while assisting port authorities in the optimal allocation of berth resources.

A thermofluid temperature modeling, prediction of closed-loop cooling system of roll grinder based on circulation differential equations

Roll grinders are essential in high-precision roll manufacturing. However, lubricant temperature fluctuations decrease the precision of grinders. The complicated dynamic heat transfer within the closed-loop cooling systems of roll grinders poses significant challenges to accurately modeling their thermal states. This study proposed a novel model based on circulation differential equations for the thermofluid temperatures in roll grinders. Based on thermodynamic principles, differential equations of each component within the closed-loop cooling system were derived. To improve model accuracy, a thermal resistance approach was utilized, and the thermal interaction between lubricant and hydrodynamic bearings under different speeds and the heat convection between hydraulic hoses and air are considered. To validate the proposed model, a series of experiments across a range of spindle speeds were carried out. The results showed the model could achieve prediction accuracy less than 5.27%. Subsequently, the influence of system design parameters on cooling performance was analyzed. The proposed temperature model better comprehends the thermodynamic interaction within closed-loop cooling systems featuring hydrodynamic bearings. The limitations and potential application in other industrial areas were discussed.

Hybrid GRU–Random Forest Model for Accurate Atmospheric Duct Detection with Incomplete Sounding Data

Atmospheric data forecasting traditionally relies on physical models, which simulate atmospheric motion and change by solving atmospheric dynamics, thermodynamics, and radiative transfer processes. However, numerical models often involve significant computational demands and time constraints. In this study, we analyze the performance of Gated Recurrent Units (GRU) and Long Short-Term Memory networks (LSTM) using over two decades of sounding data from the Xisha Island Observatory in the South China Sea. We propose a hybrid model that combines GRU and Random Forest (RF) in series, which predicts the presence of atmospheric ducts from limited data. The results demonstrate that GRU achieves prediction accuracy comparable to LSTM with 10% to 20% shorter running times. The prediction accuracy of the GRU-RF model reaches 0.92. This model effectively predicts the presence of atmospheric ducts in certain height regions, even with low data accuracy or missing data, highlighting its potential for improving efficiency in atmospheric forecasting.

Enhancing prediction accuracy of grain yield in wheat lines adapted to the southeastern United States through multivariate and multi-environment genomic prediction models incorporating spectral and thermal information.

Enhancing predictive modeling accuracy in wheat (Triticum aestivum) breeding through the integration of high-throughput phenotyping (HTP) data with genomic information is crucial for maximizing genetic gain. In this study, spanning four locations in the southeastern United States over 3 years, models to predict grain yield (GY) were investigated through different cross-validation approaches. The results demonstrate the superiority of multivariate comprehensive models that incorporate both genomic and HTP data, particularly in accurately predicting GY across diverse locations and years. These HTP-incorporating models achieve prediction accuracies ranging from 0.59 to 0.68, compared to 0.40-0.54 for genomic-only models when tested under different prediction scenarios both across years and locations. The comprehensive models exhibit superior generalization to new environments and achieve the highest accuracy when trained on diverse datasets. Predictive accuracy improves as models incorporate data from multiple years, highlighting the importance of considering temporal dynamics in modeling approaches. The study reveals that multivariate prediction outperformed genomic prediction methods in predicting lines across years and locations. The percentage of top 25% lines selected based on multivariate prediction was higher compared to genomic-only models, indicated by higher specificity, which is the proportion of correctly identified top-yielding lines that matched the observed top 25% performance across different sites and years. Additionally, the study addresses the prediction of untested locations based on other locations within the same year and in new years at previously tested locations. Findings show the comprehensive models effectively extrapolate to new environments, highlighting their potential for guiding breeding strategies.

Comprehensive quality assessment of 296 sweetpotato core germplasm in China: A quantitative and qualitative analysis

The potential for improving sweetpotato quality remains underutilized due to a lack of comprehensive quality data on germplasm resources. This study evaluated 296 core germplasms, revealing significant phenotypic diversity across 24 quality traits in both stem tips and roots. Landraces had higher sugar content in roots, while wild relatives showed increased total flavonoid and phenol contents. Accessions with red-orange flesh were rich in sugars and carotenoids, whereas those with purple flesh had higher dry matter, flavonoids, and phenols. The accessions were classified into three clusters: high sugars and carotenoids, high phenolic compounds, and high starch. A comprehensive quality scoring model identified SP286 and SP192 as superior for stem tips and roots, respectively. Near-infrared spectroscopy, combined with a random forest algorithm, enabled rapid screening of superior germplasm, achieving prediction accuracies of 97 % for stem tips and 98 % for roots. These findings offer valuable resources and high-throughput models for enhancing sweetpotato quality.

Sertraline as a Multi-Target Modulator of AChE, COX-2, BACE-1, and GSK-3β: Computational and In Vivo Studies.

Alzheimer's disease (AD) is a neurodegenerative disorder associated with the dysregulation of several key enzymes, including acetylcholinesterase (AChE), cyclooxygenase-2 (COX-2), glycogen synthase kinase 3β (GSK-3β), β-site amyloid precursor protein cleaving enzyme 1 (BACE-1), and caspase-3. In this study, machine learning algorithms such as Random Forest (RF), Gradient Boost (GB), and Extreme Gradient Boost (XGB) were employed to screen US-FDA approved drugs from the ZINC15 database to identify potential dual inhibitors of COX-2 and AChE. The models were trained using molecules obtained from the ChEMBL database, with 5039 molecules for AChE and 3689 molecules for COX-2. Specifically, 1248 and 3791 molecules were classified as active and inactive for AChE, respectively, while 858 and 2831 molecules were classified as active and inactive for COX-2. The three machine learning models achieved prediction accuracies ranging from 92% to 95% for both AChE and COX-2. Virtual screening of US-FDA drugs from the ZINC15 database identified sertraline (SETL) as a potential dual inhibitor of AChE and COX-2. Further docking studies of SETL in the active sites of AChE and COX-2, as well as BACE-1, GSK-3β, and caspase-3, revealed strong binding affinities for all five proteins. In vivo validation was conducted using a lipopolysaccharide (LPS)-induced rat model pretreated with SETL for 30 days. The results demonstrated a significant decrease in the levels of AChE (p < 0.001), BACE-1 (p < 0.01), GSK-3β (p < 0.05), and COX-2 (p < 0.05). Additionally, the downstream effects were evaluated, showing significant decreases in the apoptosis marker caspase-3 (p < 0.05) and the oxidative stress marker malondialdehyde (MDA) (p < 0.001), indicating that SETL is clinically localized in its effectiveness, mitigating both enzymatic activity and the associated pathological changes of cognitive impairment and AD.

Facilitated the discovery of new γ/γ′ Co-based superalloys by combining first-principles and machine learning

Superalloys are indispensable materials for the fabrication of high-temperature components in aircraft engines. The discovery of a novel class of γ/γ′ Co-Al-W alloys has ignited a surge of interest in Co-based superalloys, with the aspiration to transcend the inherent constraints of their Ni-based counterparts. However, the conventional methodologies utilized in the design and advancement of new γ/γ′ Co-based superalloys are frequently characterized by their laborious and resource-intensive nature. In this study, we employed a coupled Density Functional Theory (DFT) and machine learning (ML) approach to predict and analyze the stability of the crucial γ′ phase, which is instrumental in expediting the discovery of γ/γ′ Co-based alloys. A dataset comprised of thousands of reliable formation (Hf) and decomposition (Hd) energies was obtained through high-throughput DFT calculations. Through regression model selection and feature engineering, our trained Random Forest (RF) model achieved prediction accuracies of 98.07% for Hf and 97.05% for Hd. Utilizing the well-trained RF model, we predicted the energies of over 150,000 ternary and quaternary γ′ phases within the Co-Ni-Fe-Cr-Al-W-Ti-Ta-V-Mo-Nb system. The energy analyses revealed that the presence of Ni, Nb, Ta, Ti, and V significantly reduced the Hf and the Hd of γ′, while Mo and W deteriorate the stability by increasing both energy values. Interestingly, although Al reduces the Hf, it increases Hd, thereby adversely affecting the stability of γ′. Applying domain-specific screening based on our knowledge, we identified 1049 out of >150,000 compositions likely to form stable γ′ phases, predominantly distributed across 11 Al-containing systems and 25 Al-free systems. Combining the analysis of CALPHAD method, we experimentally synthesized two new Co-based alloys with γ/γ′ dual-phase microstructures, corroborating the reliability of our theoretical prediction model.

Linguistic models of abusive language

ABSTRACT Abusive language and hate speech that incites violence is a growing problem for governments trying to manage social media posts on the internet. Such language incites civil disorder by angering target individuals and groups, and tends to reinforce the identity and cohesiveness of those who use it. Abusive language is an intelligence target because it can be a leading indicator of violence. Black box predictors can achieve high accuracy but they reveal nothing about the structures, both social and technical, that underlie abusive language. We build a set of abusive language predictors leveraging both social constructs such as otherness, and linguistic properties. A stacking predictor then determines the most significant component predictors. We achieve prediction accuracy of 96.5%, with an 88% accuracy for abusive documents. This is an improvement of more than 16% points in detecting abusive documents compared to a popular empirically derived predictors, and provides insights into the mechanics of abusive language.

Generation Method for HVAC Systems Design Schemes in Office Buildings Based on Deep Graph Generative Models

The design process of heating, ventilation, and air conditioning (HVAC) systems is complex and time consuming due to the need to follow design codes. Since the design standards are not fixed, the final outcome often depends on the designer’s experience. The development of building information modeling (BIM) technology has made information throughout the building lifecycle more integrated. BIM-based forward design is now widely used, providing a data foundation for combining HVAC system design with machine learning. This paper proposes an unsupervised learning method based on deep graph generative models to uncover hidden design patterns and optimization strategies from the design results. We trained and validated four deep graph generative models—GAE, GNF, GAN, and diffusion—using HVAC system terminal pipeline layout data. Accuracy and precision metrics were used to compare the generated designs with automated forward design solutions, assessing the models’ ability to capture both local variations and broader changes in design logic. A graph-neural-network-based evaluation method was employed to measure the models’ capacity to detect changes. The results indicate that all four models achieved prediction accuracies exceeding 90% and precision rates above 75%. The models effectively captured both local modifications made by designers and global design changes, showing greater sensitivity to global layout adjustments than to local updates. When comparing the results generated by deep graph generative models and the actual design, it is obvious that the accuracy of the predictions varies significantly due to the complexity of the test buildings.

Machine learning predictions of spatter behavior in LPBF additive manufacturing

The adoption of additive manufacturing (AM) technologies, particularly Laser Powder Bed Fusion (LPBF), has been rapidly increasing in industries requiring high precision and complex geometries. Despite its advantages, LPBF faces challenges related to defects that affect material quality, with spatter formation being a significant concern. Spatters – tiny particles ejected during the printing process – can adversely affect the final product’s integrity by altering surface roughness and contributing to defects. This study introduces a comprehensive approach to predict the ejection velocity and direction of spatter particles using a suite of machine learning (ML) algorithms, including Random Forest, Gaussian Process Regression, Support Vector Machine, Regularized Linear Regressions, Gradient Boosting Trees, and Neural Networks. Our analysis reveals that the Neural Network model outperforms others, achieving prediction accuracies of 97.58% for spatter velocity and 88.22% for ejection direction, thus offering a substantial improvement in understanding and controlling spatter-related defects in LPBF processes. The practical implications of these predictions are profound, enabling manufacturers to adjust AM parameters in real time to minimize defects and enhance product quality. This study not only fills a gap in the current literature by providing a detailed comparative analysis of multiple ML algorithms for spatter ejection prediction but also paves the way for future research into real-time monitoring and control systems in AM.

Dual-agent intelligent fire detection method for large commercial spaces based on numerical databases and artificial intelligence

Fires in commercial buildings can lead to significant casualties and property damage. Thus, rapidly and accurately identifying the source of a fire and its intensity is crucial for guiding rescue operations. This study aims to explore the use of distributed fiber optic temperature sensing systems, combined with deep learning algorithms, to predict key information about fires in commercial buildings, including the location of the fire source, the intensity of fire development, and the critical planar distribution of carbon monoxide. Based on the Fire Dynamic Simulation (FDS) data, a dual-agent model was developed, and with a sensor sampling interval of 3 s, 30 s of historical data achieved prediction accuracies of 94.23 % for fire intensity and 99.28 % for fire location. Upon reducing the intervals for fire location, the accuracy reached 93.56 %. Additionally, the Random Forest algorithm for mapping carbon monoxide distribution demonstrated an overall Mean Absolute Percentage Error (MAPE) of 5.87 %. Within the 0–150 ppm range, the MAPE was 2.46 %, far below the error levels of existing CO sensors. The database configurations and the sensors' spatiotemporal arrangement were also optimized to rapid and reliable fire prediction. This research demonstrates the feasibility of using artificial intelligence for critical fire scene information detection.

Integrating Petrophysical, Hydrofracture, and Historical Production Data with Self-Attention-Based Deep Learning for Shale Oil Production Prediction

Summary As the energy industry increasingly turns to unconventional shale reservoirs to meet global demands, the development of advanced predictive models for shale oil production has become imperative. The inherent complexity of shale formations, coupled with the intricacies of hydraulic fracturing, poses significant challenges to efficient resource extraction. Our study leverages a substantial data set from the Ordos Basin to develop an advanced predictive model, integrating 18 parameters that blend static petrophysical attributes and dynamic factors, including hydraulic fracturing parameters and real-time pump pressure data. This holistic approach enables our self-attention (SA) model to accurately forecast future production rates by processing the complex interplay between reservoir characteristics and operational inputs. In testing across three wells, the model achieved average accuracies of 99.28% for daily oil production (DOP) and 99.25% for daily liquid production (DLP) over 20 days, surpassing traditional long short-term memory (LSTM) and gated recurrent unit (GRU) models, proving its efficacy in fractured well production forecasting. Furthermore, using the initial 30 days of production data as input, the model demonstrated its capability to predict DOP and DLP over a one-year period, achieving prediction accuracies of 96.2% for DOP and 99.6% for DLP rates. Our model’s profound implications for the shale industry include establishing a quantifiable link between key factors and production forecasts, guiding the optimization of controllable aspects, and serving as a decision-support tool for more efficient and cost-effective oil recovery.

Robust identification key predictors of short- and long-term weight status in children and adolescents by machine learning.

Early identification of high-risk individuals for weight problems in children and adolescents is crucial for implementing timely preventive measures. While machine learning (ML) techniques have shown promise in addressing this complex challenge with high-dimensional data, feature selection is vital for identifying the key predictors that can facilitate effective and targeted interventions. This study aims to utilize feature selection process to identify a robust and minimal set of predictors that can aid in the early prediction of short- and long-term weight problems in children and adolescents. We utilized demographic, physical, and psychological wellbeing predictors to model weight status (normal, underweight, overweight, and obese) for 1-, 3-, and 5-year periods. To select the most influential features, we employed four feature selection methods: (1) Chi-Square test; (2) Information Gain; (3) Random Forest; (4) eXtreme Gradient Boosting (XGBoost) with six ML approaches. The stability of the feature selection methods was assessed by Jaccard's index, Spearman's rank correlation and Pearson's correlation. Model evaluation was performed by various accuracy metrics. With 3,862,820 million student-visits were included in this population-based study, the mean age of 11.6 (SD = 3.64) for the training set and 10.8 years (SD = 3.50) for the temporal test set. From the initial set of 38 predictors, we identified 6, 9, and 13 features for 1-, 3-, and 5-year predictions, respectively, by the best performed feature selection method of Chi-Square test in XGBoost models. These feature sets demonstrated excellent stability and achieved prediction accuracies of 0.82, 0.73, and 0.70; macro-AUCs of 0.94, 0.86, and 0.83; micro-AUCs of 0.96, 0.93, and 0.92 for different prediction windows, respectively. Weight, height, sex, total score of self-esteem, and age were consistently the most influential predictors across all prediction windows. Additionally, several psychological and social wellbeing predictors showed relatively high importance in long-term weight status prediction. We demonstrate the potential of ML in identifying key predictors of weight status in children and adolescents. While traditional anthropometric measures remain important, psychological and social wellbeing factors also emerge as crucial predictors, potentially informing targeted interventions to address childhood and adolescence weight problems.

Constructed encoded data based coded distributed DNN training for edge computing scenario

Deep learning in unmanned aerial vehicle (UAV) deployment encounters two problems: 1, One UAV may fail to store and execute large Deep Neural Network (DNN) model. 2, One UAV fails to accomplish real time services. One possible solution is the group of UAVs (nodes) collectively generate swarm intelligence in the form of edge computing scenario, namely in the distributed computing (DC) mode with clever arrangement, say Coded Distributed Computing (CDC). In CDC systems, the redundant computation introduced by linear coding can compensate stragglers. However, since linear property cannot pass the nonlinear activation function in Deep Neural Network (DNN) training, coding/decoding for CDC need to be applied layer by layer, which slows down the training. To avoid layer-by-layer coding/decoding, we propose a novel DNN training scheme based on constructing encoded data. This construction process lies before the training process (can be done before training without any impact on training efficiency). Based on both the original data and the newly constructed encoded data, the training phase can take advantage of the (n,k) property (Wait for the first k returned data) and hence improve the training speed. The training process does not require encoding/decoding operations, and hence significantly improves the training speed. Experimental results show that the training scheme based on constructed encoded data can achieve prediction accuracy approximating that of the centralized one and significantly reduce the latency compared to the layer-by-layer linear encoding and decoding scheme.

Identifying freshness of various chilled pork cuts using rapid imaging analysis.

Determining the freshness of chilled pork is of paramount importance to consumers worldwide. Established freshness indicators such as total viable count, total volatile basic nitrogen and pH are destructive and time-consuming. Color change in chilled pork is also associated with freshness. However, traditional detection methods using handheld colorimeters are expensive, inconvenient and prone to limitations in accuracy. Substantial progress has been made in methods for pork preservation and freshness evaluation. However, traditional methods often necessitate expensive equipment or specialized expertise, restricting their accessibility to general consumers and small-scale traders. Therefore, developing a user-friendly, rapid and economical method is of particular importance. This study conducted image analysis of photographs captured by smartphone cameras of chilled pork stored at 4 °C for 7 days. The analysis tracked color changes, which were then used to develop predictive models for freshness indicators. Compared to handheld colorimeters, smartphone image analysis demonstrated superior stability and accuracy in color data acquisition. Machine learning regression models, particularly the random forest and decision tree models, achieved prediction accuracies of more than 80% and 90%, respectively. Our study provides a feasible and practical non-destructive approach to determining the freshness of chilled pork. © 2024 Society of Chemical Industry.

Analysis and Optimization of Multi-Physical Field Coupling in Boom Flow Channel of Excavator Multiway Valves

The multiway valve is the core control element of the hydraulic system in construction machinery, such as excavators. Its complex internal structure, especially the flow channels, significantly impacts the machine’s efficiency and reliability. This study focuses on the boom flow channel of excavator multiway valves and establishes a multi-physical field coupling simulation model. We propose six key flow channel structural parameters and analyze changes in the valve’s flow field, temperature field, and structural field using orthogonal test simulation data. The range analysis method identifies the primary and secondary influences of structural parameters on pressure loss, temperature, stress, and strain. A multi-objective optimization model was developed using a neural network and the Non-dominated Sorting Genetic Algorithm II(NSGA-II), with pressure loss and maximum stress as the optimization objectives. The Pareto front solution set for key flow channel parameters was calculated. The optimization results showed a 9.0% reduction in pressure loss and a 40.7% reduction in maximum stress. A test bench verified the simulation model, achieving prediction accuracies of 94.8% for pressure loss in the inlet area and 92.3% in the return area. This method can provide a reference for the optimal design of the dynamic characteristics of high-pressure multiway valves.

Streamlined Deep Learning Models for Move Prediction in Go-Game

Due to the complexity of search space and move evaluation, the game of Go has been a long-standing challenge for artificial intelligence (AI) to achieve a high level of proficiency. It was not until DeepMind proposed the deep neural network and tree search algorithm AlphaGo in 2014 that an efficient learning algorithm was developed, marking a significant milestone in AI technology. In light of the key technologies in AI Computer Go, this work examines move prediction across different Go rankings and sophisticatedly develops two deep learning models by combining and extending the feature extraction methods of AlphaGo. Specifically, effective modules for neural networks are proposed to guide learning through complicated Go situations based on the Inception module in GoogLeNet and the Convolutional Block Attention Module (CBAM). Subsequently, the two models are combined by ensemble learning to improve generalization, and these streamlined models significantly reduce the number of model parameters to the scale of one hundred thousand. Experimental results show that our models achieve prediction accuracies of 46.9% and 50.8% on two different Go datasets, outperforming conventional models by significant margins. This work not only advances AI development in the Go-game but also offers an innovative approach to related studies.

Predicting Knee Joint Contact Force Peaks During Gait Using a Video Camera or Wearable Sensors

PurposeEstimating loading of the knee joint may be helpful in managing degenerative joint diseases. Contemporary methods to estimate loading involve calculating knee joint contact forces using musculoskeletal modeling and simulation from motion capture (MOCAP) data, which must be collected in a specialized environment and analyzed by a trained expert. To make the estimation of knee joint loading more accessible, simple input predictors should be used for predicting knee joint loading using artificial neural networks.MethodsWe trained feedforward artificial neural networks (ANNs) to predict knee joint loading peaks from the mass, height, age, sex, walking speed, and knee flexion angle (KFA) of subjects using their existing MOCAP data. We also collected an independent MOCAP dataset while recording walking with a video camera (VC) and inertial measurement units (IMUs). We quantified the prediction accuracy of the ANNs using walking speed and KFA estimates from (1) MOCAP data, (2) VC data, and (3) IMU data separately (i.e., we quantified three sets of prediction accuracy metrics).ResultsUsing portable modalities, we achieved prediction accuracies between 0.13 and 0.37 root mean square error normalized to the mean of the musculoskeletal analysis-based reference values. The correlation between the predicted and reference loading peaks varied between 0.65 and 0.91. This was comparable to the prediction accuracies obtained when obtaining predictors from motion capture data.DiscussionThe prediction results show that both VCs and IMUs can be used to estimate predictors that can be used in estimating knee joint loading outside the motion laboratory. Future studies should investigate the usability of the methods in an out-of-laboratory setting.

Nonintrusive and Effective Volume Reconstruction Model of Swimming Sturgeon Based on RGB-D Sensor.

The sturgeon is an important commercial aquaculture species in China. The measurement of sturgeon mass plays a remarkable role in aquaculture management. Furthermore, the measurement of sturgeon mass serves as a key phenotype, offering crucial information for enhancing growth traits through genetic improvement. Until now, the measurement of sturgeon mass is usually conducted by manual sampling, which is work intensive and time consuming for farmers and invasive and stressful for the fish. Therefore, a noninvasive volume reconstruction model for estimating the mass of swimming sturgeon based on RGB-D sensor was proposed in this paper. The volume of individual sturgeon was reconstructed by integrating the thickness of the upper surface of the sturgeon, where the difference in depth between the surface and the bottom was used as the thickness measurement. To verify feasibility, three experimental groups were conducted, achieving prediction accuracies of 0.897, 0.861, and 0.883, which indicated that the method can obtain the reliable, accurate mass of the sturgeon. The strategy requires no special hardware or intensive calculation, and it provides a key to uncovering noncontact, high-throughput, and highly sensitive mass evaluation of sturgeon while holding potential for evaluating the mass of other cultured fishes.

Manifold learning-based reduced-order model for full speed flow field

Reduced-order models (ROMs) can effectively balance the accuracy and efficiency of computational fluid dynamics (CFD). The nonlinear flow field characteristics cannot be captured accurately by traditional ROMs, such as proper orthogonal decomposition (POD). Combining isometric mapping (ISOMAP) and local linear embedding (LLE), a novel manifold learning method (ISOMAP-Local LLE) is proposed, performing global and accurate reconstruction of the nonlinear flow field. First, the nonlinear dimensionality reduction is derived by considering the global isometric characteristic and the local reconstruction relationship simultaneously. Then, the local interpolation model is constructed to improve the interpolation accuracy introduced by the global interpolation model. Finally, the flow field reconstruction is accomplished based on the inverse mapping. Furthermore, the criteria of hyperparameters have been established to achieve high-precision prediction. Several examples covering full speed flow field are carried out to demonstrate the accuracy and efficiency of ISOMAP-Local LLE. The proposed manifold learning-based ROM achieves prediction accuracy that surpasses that of other ROMs (such as POD, ISOMAP, LLE, etc.), resulting in significant time cost savings for CFD simulations.