Data-driven Modeling Approach Research Articles

Use of Deep-Learning-Accelerated Gradient Approximation for Reservoir Geological Parameter Estimation

The estimation of space-varying geological parameters is often not computationally affordable for high-dimensional subsurface reservoir modeling systems. The adjoint method is generally regarded as an efficient approach for obtaining analytical gradient and, thus, proceeding with the gradient-based iteration algorithm; however, the infeasible memory requirement and computational demands strictly prohibit its generic implementation, especially for high-dimensional problems. The autoregressive neural network (aNN) model, as a nonlinear surrogate approximation, has gradually received increasing popularity due to significant reduction of computational cost, but one prominent limitation is that the generic application of aNN to large-scale reservoir models inevitably poses challenges in the training procedure, which remains unresolved. To address this issue, model-order reduction could be a promising strategy, which enables us to train the neural network in a very efficient manner. A very popular projection-based linear reduction method, i.e., propel orthogonal decomposition (POD), is adopted to achieve dimensionality reduction. This paper presents an architecture of a projection-based autoregressive neural network that efficiently derives an easy-to-use adjoint model by the use of an auto-differentiation module inside the popular deep learning frameworks. This hybrid neural network proxy, referred to as POD-aNN, is capable of speeding up derivation of reduced-order adjoint models. The performance of POD-aNN is validated through a synthetic 2D subsurface transport model. The use of POD-aNN significantly reduces the computation cost while the accuracy remains. In addition, our proposed POD-aNN can easily obtain multiple posterior realizations for uncertainty evaluation. The developed POD-aNN emulator is a data-driven approach for reduced-order modeling of nonlinear dynamic systems and, thus, should be a very efficient modeling tool to address many engineering applications related to intensive simulation-based optimization.

Childhood maltreatment and adult aggression: The moderating role of neurocognitive ability and substance use

BackgroundAchieving positive intervention outcomes for aggressive behaviour in adulthood is challenging. This difficulty is enhanced by the complex presentations of those engaging in such behaviours and the impact this has on their engagement with interventions. ObjectiveThis study assessed the cumulative impact of childhood maltreatment, substance use, and neurocognitive ability (working memory, cognitive flexibility, decision making, response inhibition, and cognitive control) on aggressive behaviour in adulthood. Participants and settingAdult participants (N = 179; 69 % female) recruited from the community, and clinical and forensic services, aged 18 to 81 (M = 40.5, SD = 15.9). MethodsParticipants completed an online Qualtrics survey and remote neurocognitive testing via Inquisit. ResultsWe implemented an a priori approach, assessing three-way interactions between childhood maltreatment, substance use, neurocognitive ability, and aggression. No three-way interactions were significant. We then utilised a data-driven modelling approach, using automatic linear forward stepwise modelling to identify the most important variables for predicting aggression. Four were significant: physical maltreatment (b = 0.053, p < .001), drug use risk level (b = 0.015, p < .001), poorer response inhibition (b = 0.001, p = .016), and the interaction between poorer response inhibition and physical maltreatment (b = 0.205, p = .017). ConclusionsPhysical maltreatment in childhood, drug use risk level, and response inhibition impact significantly on adult aggression, indicating a need for early intervention for children who have experienced maltreatment. Consideration should be given to how maltreatment in childhood may impact on ability to engage with interventions as an adult, particularly response inhibition difficulties that may hinder skill implementation.

Modeling Enzyme Kinetics: Current Challenges and Future Perspectives for Biocatalysis.

Biocatalysis is becoming a data science. High-throughput experimentation generates a rapidly increasing stream of biocatalytic data, which is the raw material for mechanistic and novel data-driven modeling approaches for the predictive design of improved biocatalysts and novel bioprocesses. The holistic and molecular understanding of enzymatic reaction systems will enable us to identify and overcome kinetic bottlenecks and shift the thermodynamics of a reaction. The full characterization and modeling of reaction systems is a community effort; therefore, published methods and results should be findable, accessible, interoperable, and reusable (FAIR), which is achieved by developing standardized data exchange formats, by a complete and reproducible documentation of experimentation, by collaborative platforms for developing sustainable software and for analyzing data, and by repositories for publishing results together with raw data. The FAIRification of biocatalysis is a prerequisite to developing highly automated laboratory infrastructures that improve the reproducibility of scientific results and reduce the time and costs required to develop novel synthesis routes.

GA-SmaAt-GNet: Generative adversarial small attention GNet for extreme precipitation nowcasting

In recent years, data-driven modeling approaches have gained significant attention across various meteorological applications, particularly in weather forecasting. However, these methods often face challenges in handling extreme weather conditions. In response, we present the GA-SmaAt-GNet model, a novel generative adversarial framework for extreme precipitation nowcasting. This model features a unique SmaAt-GNet generator, an extension of the successful SmaAt-UNet architecture, capable of integrating precipitation masks (binarized precipitation maps) to enhance predictive accuracy. Additionally, GA-SmaAt-GNet incorporates an attention-augmented discriminator inspired by the Pix2Pix architecture. This innovative framework paves the way for generative precipitation nowcasting using multiple data sources. We evaluate the performance of SmaAt-GNet and GA-SmaAt-GNet using real-life precipitation data from The Netherlands, revealing notable improvements in overall performance and for extreme precipitation events compared to other models. Specifically, our proposed architecture demonstrates its main performance gain in summer and autumn, when precipitation intensity is typically at its peak. Furthermore, we conduct uncertainty analysis on the GA-SmaAt-GNet model and the precipitation dataset, providing insights into its predictive capabilities. Finally, we employ Grad-CAM to offer visual explanations of our model’s predictions, generating activation heatmaps that highlight areas of input activation throughout the network.

Aerodynamic Performance Prediction for Wide-Incidence Turbines Using Graph Neural Network Models Driven by Small-Scale Experimental Data

Abstract To rapidly and accurately predict turbine rotor blade losses within a wide range of incidences (−50–30 deg), graph neural networks (GNNs) are utilized to predict the aerodynamic parameters of two-dimensional turbine blades based on a small-scale experimental dataset. By comparing the backpropagation neural network (BPnn) model and computational fluid dynamics (CFD) results, it is demonstrated that GNNs with appropriately designed graph structures can accurately and quickly predict high-fidelity aerodynamic parameters based on limited experimental data. Unlike traditional data-driven modeling approaches, two innovative methods for improving blade profiles into graph structures are proposed. Relatively few input features are used to comprehensively and effectively represent the turbine blade profile by applying blade profile features in the GNN. The research findings indicate that due to the graph structure, which divides the turbine blade profile into five nodes based on five key points, coupled with high-fidelity experimental data and the unique weight updating mechanism of the graph attention network (GAT) model, the GAT-5 model exhibits the best performance among the studied models. Additionally, when assessing unknown validation blade profiles, the GAT-5 model maintains an absolute error below 6% at an incidence angle of 30 deg compared to the experimental results.

The use of ensembles trees machine learning method in estimating damping of granular materials

Granular Materials (GM) employed within the mechanism of particle dampers attenuate the vibration energy due to their interparticle and particle-wall interactions. Estimating their damping effect using the analytical equivalent single mass approach overlooked the particles' individual losses that are built into the total damping. Alternatively, the numerical techniques (e.g., Discrete Element) are time-inefficient and computationally demanding. Therefore, this study explores the implementation of Machine learning (ML) algorithms to estimate the damping effect of GM. The ML model in this study will rely on a Data-Driven Modeling approach (DDM) incorporating the ensemble tress nonlinear regressor method. The models' training and testing data were obtained from an experimental setup of an acrylic beam internally integrated with Stainless Steel (SS) and glass spheres undergoing low excitation amplitude (RMS &lt;1N). The main aim was to map between seven input features (e.g. filling ratio) and one targeted output: the beam's damped frequency response. Three ensemble trees' algorithms were used to create the DDM; Decision Tree, Random Forest, and XGBoost. The hyperparameter combination based on the Gridsearch CV function increases the prediction accuracy of each model. The developed ML models provided high accuracy (86-93%) in predicting the damping effect of the granular materials spheres.

Childhood adversity and adolescent mental health: Examining cumulative and specificity effects across contexts and developmental timing.

Associations between adversity and youth psychopathology likely vary based on the types and timing of experiences. Major theories suggest that the impact of childhood adversity may either be cumulative in type (the more types of adversity, the worse outcomes) or in timing (the longer exposure, the worse outcomes) or, alternatively, specific concerning the type (e.g., parenting, home, neighborhood) or the timing of adversity (e.g., specific developmental periods). In a longitudinal sample from the Future of Families and Wellbeing Study (N = 4,210), we evaluated these competing hypotheses using a data-driven structured life-course modeling approach using risk factors examined at child age 1 (infancy), 3 (toddlerhood), 5 (early childhood), and 9 (middle childhood). Results showed that exposures to more types of adversity for longer durations (i.e., cumulative in both type and timing) best predicted youth psychopathology. Adversities that occurred at age 9 were better predictors of youth psychopathology as compared to those experienced earlier, except for neglect, which was predictive of internalizing symptoms when experienced at age 3. Throughout childhood (across ages 1-9), aside from the accumulation of all adversities, parental stressand low collective efficacy were the strongest predictors of internalizing symptoms, whereas psychological aggression was predictive of externalizing symptoms.

An advanced high dimensional model representation approach for internal combustion engine modeling and optimization

This work introduces an efficient data-driven approach for engine performance modeling and optimization. A highly accurate and robust high-dimensional model representation (HDMR) surrogate model is developed based on the dataset generated from a coupled KIVA4-CHEMKIN Ⅱ software. The HDMR model is integrated with a multi-objective optimization algorithm, enabling the automatic identification of optimal combustion strategies. Results suggest that the constructed HDMR model is able to accurately predict the engine performance and exhaust gas emissions with negligible CPU costs. Moreover, the model effectively identifies optimal engine operating conditions, leading to enhanced fuel efficiency.

Exploring Potential Human Health Risks Linked to Heavy Metal(Loid)s in Dietary Fishes: Utilizing Data-Driven and Computational Modelling Approaches.

There are significant concerns about the risks to human health posed by metal(loid) contamination in dietary fishes in Bangladesh. Therefore, this study aimed to evaluate heavy metal(loid) contamination in fish and their associated health risks using published data from 2000 to 2022. Additionally, the safe limit of fish consumption was estimated using the target hazard quotient (THQ) followed by computational modelling and artificial neural networks (ANN). Results showed that freshwater and herbivorous fishes pose the least non-cancer risks, whereas saltwater and carnivorous fishes pose the highest non-cancer risks to Bangladeshi consumers. However, freshwater and omnivorous fish consumption pose the highest cancer risks compared to all studied metal(loid)s. In particular, among the heavy metal(loid)s, As, Cr, Hg, and Ni pose significant cancer and non-cancer risks to Bangladeshi consumers. On the contrary, the ANN and Decision tree regression (DTR) characterized the dataset, simulation model, or data testing condition, reaching 94.7% accuracy and allowing us to measure the safest fish consumption limit. The herbivorous fishes are less contaminated and allow greater consumption (175.09gday-1). Contrarily, the allowable intake rates of carnivorous and omnivorous fishes are 153.05 and 168.63gday-1, respectively. Besides, the safe consumption rate of saltwater fishes was 156.51gday-1, which was lower than freshwater (180.59gday-1) and euryhaline fishes (182.17gday-1). Therefore, this study will assist fish consumers in selecting less contaminated fish, ensuring safe consumption levels, and ultimately reducing health risks associated with metal(loid) contamination in fish.

A data-driven underground gas storage production system string failure prediction model for time-varying reliability analysis

In underground gas storage (UGS) systems, the production tubing string system undergoes complex thermo-mechanical loading conditions, leading to fatigue damage accumulation and potential failure risks. To accurately assess the fatigue reliability and mitigate risks, it is crucial to develop a data-driven physics-informed uncertainty modeling approach that captures the coupled thermo-mechanical effects and accounts for various uncertainties. This study proposes a data-driven physics-informed modeling method based on thermo-mechanical coupling for fatigue reliability assessment and risk mitigation in critical UGS systems. A quantitative analysis method is introduced, which relies on small-scale multi-stage loading experiments and combines the stochastic degradation process with time-series reliability theory. To analyze the fatigue life of joints under complex loading conditions, a modified S-N curve model is developed. This model takes into account the influences of the dimension coefficient, stress concentration factors, surface finish coefficients, and thermal load. It considers the uncertainties associated with different temperature loads and surface finish on fatigue life. By integrating physics-informed modeling, uncertainty quantification, and fatigue damage accumulation analysis, this approach provides a comprehensive framework for fatigue reliability assessment and risk mitigation in critical UGS systems. It enables informed decision-making for safe and reliable operations, as well as optimized maintenance strategies, ultimately enhancing the overall system performance and mitigating potential failures.

Informatics-enhanced prediction of failure strength in skeletal muscle tissue

Skeletal muscles are susceptible to injury during daily activities and competitive sports. Reliable prediction of tissue failure strength is a major challenge due to the large uncertainty in the mechanical behavior of skeletal muscle tissue. The present study reveals a strong correlation between histological characteristics and skeletal muscle failure strength by means of mechanical and histological experiments. We propose a data-driven hybrid modeling approach that enables an effective integration of data science and informatics tools to capture tissue failure strength. Uncertainty in the tissue failure strength is propagated into the posterior information of reduced model parameters via the Bayesian inference framework and parameter space compression. A histological enhancement to the softening hyperelasticity model is made by linking a quantified tissue-scale histological characteristic and stiff model parameters using artificial neural networks. The model is applied to skeletal muscle tissue from different species and sites to assess its predictive capabilities for physiological differences. The results show that the approach can achieve reliable predictions of skeletal muscle tissue failure strength. The proposed approach can be extended to different scales to enrich the understanding of structure–property linkages for biomaterials.

Output power prediction of stratospheric airship solar array based on surrogate model under global wind field

Stratospheric airships are lighter-than-air vehicles capable of continuous flying for months. The energy balance of the airship is the key to long-duration flights. The stratospheric airship is entirely powered by the solar array. It is necessary to accurately predict the output power of the array for any flight state. Because of the uneven solar radiation received by the solar array, the traditional model based on components has a slow simulation speed. In this study, a data-driven surrogate modeling approach for prediction the output power of the solar array is proposed. The surrogate model is trained using the samples obtained from the high-accuracy simulation model. By using the input parameter preprocessor, the accuracy of the surrogate model in predicting the output power of the solar array is improved to 98.65%. In addition, the predictive speed of the surrogate model is ten million times faster than the traditional simulation model. Finally, the surrogate model is used to predict the energy balance of stratospheric airships flying throughout the year under actual global wind fields.

Predictive Models for Aggregate Available Capacity Prediction in Vehicle-to-Grid Applications

The integration of vehicle-to-grid (V2G) technology into smart energy management systems represents a significant advancement in the field of energy suppliers for Industry 4.0. V2G systems enable a bidirectional flow of energy between electric vehicles and the power grid and can provide ancillary services to the grid, such as peak shaving, load balancing, and emergency power supply during power outages, grid faults, or periods of high demand. In this context, reliable prediction of the availability of V2G as an energy source in the grid is fundamental in order to optimize both grid stability and economic returns. This requires both an accurate modeling framework that includes the integration and pre-processing of readily accessible data and a prediction phase over different time horizons for the provision of different time-scale ancillary services. In this research, we propose and compare two data-driven predictive modeling approaches to demonstrate their suitability for dealing with quasi-periodic time series, including those dealing with mobility data, meteorological and calendrical information, and renewable energy generation. These approaches utilize publicly available vehicle tracking data within the floating car data paradigm, information about meteorological conditions, and fuzzy weekend and holiday information to predict the available aggregate capacity with high precision over different time horizons. Two data-driven predictive modeling approaches are then applied to the selected data, and the performance is compared. The first approach is Hankel dynamic mode decomposition with control (HDMDc), a linear state-space representation technique, and the second is long short-term memory (LSTM), a deep learning method based on recurrent nonlinear neural networks. In particular, HDMDc performs well on predictions up to a time horizon of 4 h, demonstrating its effectiveness in capturing global dynamics over an entire year of data, including weekends, holidays, and different meteorological conditions. This capability, along with its state-space representation, enables the extraction of relationships among exogenous inputs and target variables. Consequently, HDMDc is applicable to V2G integration in complex environments such as smart grids, which include various energy suppliers, renewable energy sources, buildings, and mobility data.

Risk-aware scheduling and dispatch of flexibility events in buildings

Residential and commercial buildings, equipped with systems such as heat pumps (HPs), hot water tanks, or stationary energy storage, have a large potential to offer their consumption flexibility as grid services. In this work, we leverage this flexibility to react to consumption requests related to maximizing self-consumption and reducing peak loads. We employ a data-driven virtual storage modeling approach for flexibility prediction in the form of flexibility envelopes for individual buildings. The risk-awareness of this prediction is inherited by the proposed scheduling algorithm. A Mixed-integer Linear Program (MILP) is formulated to schedule the activation of a pool of buildings in order to best respond to an external aggregated consumption request. This aggregated request is then dispatched to the active individual buildings, based on the previously determined schedule. The effectiveness of the approach is demonstrated by coordinating up to 500 simulated buildings using the Energym Python library and observing about 1.5 times peak power reduction in comparison with a baseline approach while maintaining comfort more robustly. We demonstrate the scalability of the approach by solving problems with 2000 buildings in about 21 s, with solving times being approximately linear in the number of considered assets.

Data-driven modelling of coastal storm erosion for real-time forecasting at a wave-dominated embayed beach

Emergency managers have an increasing need for tools to enhance preparedness to extreme coastal storms and support disaster risk reduction measures. With the emergence of Early Warning Systems (EWSs) for coastal storm hazards, a fundamental challenge is the accurate prediction of sandy beach erosion at lead times of days to weeks corresponding to an approaching storm event. This work presents a data-driven modelling approach to predict storm-driven beach erosion (shoreline change) using a large dataset of 276 individual storm events at Narrabeen-Collaroy Beach, SE Australia. Correlation analysis between individual storm characteristics and shoreline response at three locations along the embayment with varying exposure to the prevailing waves indicates that cumulative storm wave energy is the dominant driver of storm erosion at this site. This is followed by the pre-storm beach width, storm wave direction and to a minimal extent, storm wave period and water levels. A multi-linear regression model of storm erosion is developed and found to accurately predict shoreline change due to individual storm events (RMSE = 3.7 m–6.4 m). This work highlights the value of high-frequency shoreline data for storm erosion forecasting and provides a framework for real-time forecasting applications.

Long-range hydrological drought forecasting using multi-year cycles in the North Atlantic Oscillation

With global temperatures, populations and ecological stressors expected to rise, hydrological droughts are projected to have progressively severe economic and environmental impacts. As a result, hydrological drought forecasting systems have become increasingly important water resource management tools for mitigating these impacts. However, high frequency behaviours in meteorological or atmospheric conditions often limit the lead times of hydrological drought forecasts to seasonal timescales, either through poorer performance of multi-year meteorological forecasts or the lack of multi-year lags in atmosphere-hydrology systems. By contrast, low frequency behaviours in regionally important teleconnection systems (such as the North Atlantic Oscillation, NAO) offer a novel way to forecast hydrological drought at longer lead times. This paper shows that, by using a data-driven modelling approach, long-term behaviours within the NAO can be skilful predictors of hydrological drought conditions at a four-year forecasting horizon. Multi-year semi-periodic patterns in the NAO were used to forecast regional groundwater drought coverage in the UK (proportion of groundwater boreholes in drought), with the greatest forecast performance achieved for longer duration droughts, and for hydrogeological regions with longer response times. Model errors vary from 14 % (proportion of boreholes, (MAE)) in flashy hydrological regions or short droughts (<3 months), to 2 % for longer duration droughts (>8 months). Model fits of r2 up to 0.8 were produced between simulated and recorded regional drought coverage. As such our results show that teleconnection indices can be a skilful predictor of hydrological drought dynamics at multi-year timescales, opening new opportunities for long-lead groundwater drought forecasts to be integrated within existing drought management strategies in Europe and beyond.

Dynamic Hybrid Models With Active Sampling and Adaptive Selection of Double-Domain Features for the Tuning of Microwave Cavity Filters.

Microwave cavity filters are essential electromechanical coupling devices in communication systems. Structural-parameter tuning by experienced operators improves the filter performance but is demanding and time-consuming. The automatic tuning method has received extensive research attentions using data-driven modeling approaches. However, two main issues affect the accuracy and efficiency of the model construction: 1) features of tuning processes, as model inputs, have limited adaptability and extraction accuracy to different resonant states and 2) models require plentiful training data and the training process is time-consuming. Thus, dynamic hybrid models are developed in this study with self-selected inputs, self-organized samples, and a self-learning structure. First, spatial features are extracted to flexibly depict the tuning characteristic, and double-domain (spatial or circuital) features are selected adaptively to accommodate distinct resonance states. Second, a trustworthiness-curiosity-driven active sampling method is exploited to attain fewer and better-training data. Third, an improved glsms broad learning system acrlong BLS is developed using new modules of incremental node calculation and weight pruning, characterized by more lightweight and flexible structures. The proposed method is effective and flexible demonstrated by simulations and experiments, and the tuning task of microwave cavity filters is fulfilled in a more accurate and efficient manner.

Numerical simulation-based targeting of the Au mineralization within the Hongbu Altered Rock type gold deposit, Eastern Shandong Peninsula, China

Altered rock type deposits are significant gold deposits widely distributed throughout the world, having important economic and industrial value. The Eastern Shandong Peninsula is the third largest gold mining camps in the world with known gold reserves of more than 5300 tons. The majority of the gold deposits here are located in three main orefields, namely the northwest Jiaodong (Laizhou-Zhaoyuan), the Qixia-Penglai-Fushan, and the Muping-Rushan orefields. The Hongbu deposit is a medium-sized altered rock type deposit within the northwest Jiaodong orefield. With decades of mining, the Hongbu deposit faces the dilemma of resource depletion and loss of undigitized data. Therefore, prospecting and exploration have to focus on deep and peripheral areas of known Au mineralization. This study built a three-dimensional model of the Hongbu deposit and sliced it to obtain nine cross-sections to conduct multi-field coupled numerical simulation (including heat conduction, fluid migration, chemical reaction, and material migration). Locations with relatively developed Au mineralization (the top 15 % within the concentration range of Au) are defined as having metallogenic potential, which are highlighted and projected to the geological map of the Hongbu district. The results also show that the spatial distributions of Au mineralization within these nine sections are different, indicating that the formation of Au mineralization is controlled by the morphology of the Hexi fault (other simulation parameters of these nine sections are the same) and the pyrite sericite cataclasite. This research proves that numerical simulation methods is effective in mineral exploration, especially for those areas with insufficient data to use data-driven prospectivity modeling approaches. Future development into both mathematical geosciences and computational science will provide more detailed answers for remaining hot issues related to mineral exploration.

Minimal model identification of drum brake squeal via SINDy

The industrial standard in the design and development process of NVH(Noise Vibration Harshness) characteristic of brakes is the application of Finite Element(FE) models with a high number of degrees of freedom in the range of one or several millions. Nevertheless, parallel experimental investigations are still indispensable. On the other hand, minimal models with, due to the inclusion of the self-excitation process, at least two degrees of freedom are well known to be capable to explain qualitatively phenomena as instability of the desired non-vibrating solution or limit cycle oscillation but are in general very inaccurate in predicting the dynamics of a specific real brake. This is because the underlying physical assumptions are already too restrictive and model parameters (especially those referring to nonlinearities) are widely unknown. To overcome this problem, the data-driven modeling approach SINDy(Sparse Identification of Nonlinear Dynamics) is applied to identify appropriate nonlinear functions for a brake squeal minimal model. A problem thereby is the limited database. It turns out that the naive implementation of the method yielding the lowest possible residuum does not necessarily provide physically meaningful models and results, respectively. Instead, a constrained model that incorporates physical knowledge is used to robustly identify parameters and reproduce realistic dynamic behavior. Thereby, several appropriate models with coexisting limit cycles and stationary equilibrium are identified. In particular, it was found that the angular position of the brake drum has a significant influence on the model parameters and therefore must be taken into account in a model with long-term validity.

An efficient data-driven approach for modeling and optimization of reactivity-controlled compression ignition engine fueled with polyoxymethylene dimethyl ethers and hydrogen

This paper presents an efficient data-driven approach for optimizing the utilization of PODEn and hydrogen in reactivity-controlled compression ignition (RCCI) engines. A novel high dimensional model representation (HDMR) technique is adopted to construct a highly accurate surrogate model for the RCCI engine based on the dataset generated by a detailed CFD model that has been calibrated against experimental data. The HDMR model is then coupled with Non-dominated Sorting Genetic Algorithm III (NSGA III), a multi-objective genetic algorithm, to search for the optimal operating conditions where the RCCI engine fueled by PODEn and H2 achieves the highest combustion performance. Results suggest that the constructed HDMR model is highly accurate, being able to predict the engine performance within milliseconds. The coupled HDMR-NSGA III approach successfully identifies the optimal engine operating condition, which significantly improves the combustion performance and reduces pollutant emissions compared with the base condition.