Model Structure Research Articles

Improving Transparency of Decision Models Through the Application of Decision Analytic Models with Omitted Objects Displayed (DAMWOOD).

The coronavirus disease 2019 (COVID-19) pandemic has increased public awareness of the influence of epidemiological and economic decision models on public policy decisions. Alongside this is an increased scrutiny on the development, analysis, reporting and utilisation of decision models for public policy making. Therefore, it is important that model developers can clearly explain and justify to all stakeholders what is included and excluded from a model developed to support decision-making, to both improve transparency and trust in decision-making. Our aim is to provide tools for improving communication between modellers and decision-makers, leading to improved transparency in decision-making. To do so, we extend the recently described directed acyclic graphs with omitted objects displayed (DAGWOOD) approach from Haber et al. (Ann Epidemiol 68:64-71, 2022) to decision analytic models, giving the decision analytic models with omitted objects displayed (DAMWOOD) approach. DAMWOOD is a framework for the identification of objects omitted from a decision model, as well as for consideration of the effects of omissions on model outcomes. Objects omitted from a decision model are classed as either an exclusion (known and unknown confounders), misdirection (alternative model pathways) or structure (e.g. model type, methods for estimating relationships between objects). DAMWOOD requires model developers to use explicit statements and provide illustration of included and omitted objects, supporting communication with model users and stakeholders, allowing them to provide input and feedback to modellers about which objects to include or omit in a model. In developing DAMWOOD, we considered two challenges we encountered in modelling for pandemic policy response. First, the scope of the decision problem is not always made sufficiently explicit by decision-makers, requiring modellers to intuit which policy options should be considered, and/or which outcomes should be considered in their evaluation. Second, there is rarely sufficient transparency to ensure stakeholders can see what is included in models and why. This limits stakeholders' ability to advocate to decision-makers for the prioritisation of specific outcomes and challenge the model results. To illustrate the application of DAMWOOD, we apply it to a previously published COVID-19 vaccine allocation optimisation model. The DAMWOOD diagrams illustrate the ways in which it is possible to improve the communication of model assumptions. The diagrams make explicit which outcomes are omitted and provide information on the expected impact of the omissions on model results. We discuss the usefulness of DAMWOOD for framing the decision problem, communicating the model structure and results and engaging with those making and affected by the decisions the model is developed to inform.

Estrogenic Responsiveness of Brown Trout Primary Hepatocyte Spheroids to Environmental Levels of 17α-Ethinylestradiol

Three-dimensional (3D) fish hepatocyte cultures are promising alternative models for replicating in vivo data. Few studies have attempted to characterise the structure and function of fish 3D liver models and illustrate their applicability. This study aimed to further characterise a previously established spheroid model obtained from juvenile brown trout (Salmo trutta) primary hepatocytes under estrogenic stimulation. The spheroids were exposed for six days to environmentally relevant concentrations of 17α-ethinylestradiol—EE2 (1–100 ng/L). The mRNA levels of peroxisome (catalase—Cat and urate oxidase—Uox), lipid metabolism (acyl-CoA long chain synthetase 1—Acsl1, apolipoprotein AI—ApoAI, and fatty acid binding protein 1—Fabp1), and estrogen-related (estrogen receptor α—ERα, estrogen receptor β—ERβ, vitellogenin A—VtgA, zona pellucida glycoprotein 2.5—ZP2.5, and zona pellucida glycoprotein 3a.2—ZP3a.2) target genes were evaluated by quantitative real-time polymerase chain reaction. Immunohistochemistry was used to assess Vtg and ZP protein expressions. At the highest EE2 concentration, VtgA and ZP2.5 genes were significantly upregulated. The remaining target genes were not significantly altered by EE2. Vtg and ZP immunostaining was consistently increased in spheroids exposed to 50 and 100 ng/L of EE2, whereas lower EE2 levels resulted in a weaker signal. EE2 did not induce significant changes in the spheroids’ viability and morphological parameters. This study identified EE2 effects at environmentally relevant doses in trout liver spheroids, indicating its usefulness as a proxy for in vivo impacts of xenoestrogens.

Coordinate system setting for post-machining of impeller shape by wire arc DED and evaluation of processing efficiency

Wire arc additive manufacturing (WAAM) is a direct energy deposition (DED) process that uses arc welding. It is a method of stacking beads made by melting metal wires with an arc heat source generated by a short-circuit current. Compared to other metal additive manufacturing methods, this process can be used to quickly produce large and complex-shaped metal parts. However, due to the multi-bead stacking method, the surface is highly curved and the dimensional errors are large; therefore, post-processing of the surface by cutting is required. Impellers, which are widely used in various industries, have complex shapes and high material consumption during cutting; therefore, the WAAM process can improve the manufacturing efficiency. In this study, a manufacturing process for an impeller with a diameter of 160 mm was developed by using the WAAM process. A 6-bladed fan-type impeller used for high-pressure fluid delivery was similarly modeled, and the product was additively manufactured using an Inconel 625 alloy wire. Manufacturing conditions that ensure productivity and quality or the product were determined through experimentation. Considering the post-processing of the WAAM-fabricated structure, the robot and tool paths of the impeller model were designed, and the error in the process coordinate system caused by attaching and detaching the workpiece between the two processes was reduced. Through the post-processing of the WAAM-fabricated structure, the production efficiency and process reliability were verified when the conventional manufacturing method and WAAM process were applied.

Identification of heart rate dynamics during treadmill and cycle ergometer exercise: the role of model zeros and dead time

Background The response of heart rate to changes in exercise intensity is comprised of several dynamic modes with differing magnitudes and temporal characteristics. Investigations of empirical identification of dynamic models of heart rate showed that second-order models gave substantially and significantly better model fidelity compared to the first order case. In the present work, we aimed to reanalyse data from previous studies to more closely consider the effect of including a zero and a pure delay in the model. Methods This is a retrospective analysis of 22 treadmill (TM) and 54 cycle ergometer (CE) data sets from a total of 38 healthy participants. A linear, time-invariant plant model structure with up to two poles, a zero and a dead time is considered. Empirical estimation of the free parameters was performed using least-squares optimisation. The primary outcome measure is model fit, which is a normalised root-mean-square model error. Results A model comprising parallel connection of two first-order transfer functions, one with a dead time and one without, was found to give the highest fit (56.7 % for TM, 54.3 % for CE), whereby the non-delayed component appeared to merely capture initial transients in the data and the part with dead time likely represented the true dynamic response of heart rate to the excitation. In comparison, a simple first-order model without dead time gave substantially lower fit than the parallel model (50.2 % for TM, 47.9 % for CE). Conclusions This preliminary analysis points to a linear first-order system with dead time as being an appropriate model for heart rate response to exercise using treadmill and cycle ergometer modalities. In order to avoid biased estimates, it is vitally important that, prior to parameter estimation and validation, careful attention is paid to data preprocessing in order to eliminate transients and trends.

Neural network analysis in time series forecasting

Objectives. To build neural network models of time series (LSTM, GRU, RNN) and compare the results of forecasting with their mutual help and the results of standard models (ARIMA, ETS), in order to ascertain in which cases a certain group of models should be used.Methods. The paper provides a review of neural network models and considers the structure of RNN, LSTM, and GRU models. They are used for modeling time series in Russian macroeconomic statistics. The quality of model adjustment to the data and the quality of forecasts are compared experimentally. Neural network and standard models can be used both for the entire series and for its parts (trend and seasonality). When building a forecast for several time intervals in the future, two approaches are considered: building a forecast for the entire interval at once, and step-by-step forecasting. In this way there are several combinations of models that can be used for forecasting. These approaches are analyzed in the computational experiment.Results. Several experiments have been conducted in which standard (ARIMA, ETS, LOESS) and neural network models (LSTM, GRU, RNN) are built and compared in terms of proximity of the forecast to the series data in the test period.Conclusions. In the case of seasonal time series, models based on neural networks surpassed the standard ARIMA and ETS models in terms of forecast accuracy for the test period. The single-step forecast is computationally less efficient than the integral forecast for the entire target period. However, it is not possible to accurately indicate which approach is the best in terms of quality for a given series. Combined models (neural networks for trend, ARIMA for seasonality) almost always give good results. When forecasting a non-seasonal heteroskedastic series of share price, the standard approaches (LOESS method and ETS model) showed the best results.

Health Technology Assessment Reports for Non-Oncology Medications in Canada from 2018 to 2022: Methodological Critiques on Manufacturers' Submissions and a Comparison between Manufacturer and Canadian Agency for Drugs and Technologies in Health (CADTH) Analyses.

Identifying key differences between manufacturers' submitted analysis and economic reanalysis by the Canadian Agency for Drugs and Technologies in Health (CADTH) is an important step toward understanding reimbursement recommendations. We compared economic values reported in manufacturers' analysis with the CADTH reanalysis and also assessed methodological critiques. Two reviewers extracted data from the clinical and economic reports in publicly available CADTH reports from 2018 to 2022. We used the Wilcoxon rank-sum test to assess the difference between mean economic values, and the Chi-square test to assess the association between the CADTH critique final recommendations. Of the total submissions, 99.4% included effectiveness critiques, 88.8% included model structure critiques, 69.1% included utility score critiques, and 78.7% included cost critiques. The median incremental cost-utility ratio (ICUR) in the manufacturers' analyses was $138,658/quality-adjusted life-year (QALY), 2.5-fold lower than the CADTH's reanalysis at $380,251/QALY (p<0.001). The median CADTH reanalysis for 3-year budget impact analysis (BIA) was $4,575,102, which was 27% higher than the manufacturers' submitted 3-year BIA (p<0.001). CADTH requested a price reduction for 95% of all submissions, and the median price reduction request was 63.5%. In 2021 and 2022, the willingness-to-pay threshold identified in CADTH reports remained constant at $50,000 per QALY gained for all medications. There was high frequency of CADTH critiques on manufacturers' submissions in all four aspects of economic submissions: effectiveness, cost, utility score and structure. We observed a higher median incremental cost and lower median incremental QALYs in the CADTH reanalysis compared with the manufacturers' submissions. The resulting higher ICUR in the CADTH reanalysis often leads to a recommendation that the manufacturer needs to reduce its price. The 3-year budget impact was higher in the CADTH reanalyses compared with manufacturers' submissions.

Updating strategy of safe operation control model for dense medium coal preparation process based on Bayesian network and incremental learning

AbstractThe effectiveness of control decisions provided by the safe operation control model for the dense medium coal preparation process may decline due to its inability to adapt to changing working conditions. To address this issue, this paper investigates a safe operation control model update strategy based on Bayesian network and incremental learning. This strategy can update the model structure and parameters according to different conditions, ensuring the effectiveness of the updated model. Considering that the old model has effective information to explain the new working conditions, the Bayesian network structure update learning method based on incremental learning is proposed. This method retains the components of the old model that can describe the joint probability distribution of the sampled data under the new working conditions while updating the remaining structure. This approach improves the efficiency of model updating. The simulation results show that the updated model obtained by the proposed method can effectively deal with new abnormal conditions.

Research on the Parametric Generation Method of Model with Multi-Information for the Wooden Architectural Heritage

ABSTRACT There are many traditional wooden buildings with complex structural systems, which leads to tedious modelling processes, and challenges to achieve multidisciplinary collaboration between architecture and structure. This study establish a multi-information model using information and parameterization methods. First, this study identifies typical traditional Chuan-Dou style wood frames in Jingdezhen as an example and uses the parameterization method to quickly generate building information models based on the composition rules of traditional wooden buildings using the Revit-Dynamo platform. Second, by embedding mechanical parameters in the model components, the building and structure information are integrated into a multidimensional integrated model. Finally, a method for applying the generated multi-information model in the project is discussed. This building and structure multi-information model verifies the rationality of the component structure in the building model in real time, thereby optimizing the design scheme and achieving the full-scale recording of the detailed statistics of the building and structural information of the components. These research results have significantly improved the modelling speed of traditional wooden buildings, simplified the tedious process from architectural design to structural calculation, and provided an important research foundation for the intelligent and digital protection of architectural heritage.

Hybrid Modeling for On-Line Fermentation Optimization and Scale-Up: A Review

Modeling is a crucial tool in the biomanufacturing industry, namely in fermentation processes. This work discusses both mechanistic and data-driven models, each with unique benefits and application potential. It discusses semi-parametric hybrid modeling, a growing field that combines these two types of models for more accurate and easy result extrapolation. The characteristics and structure of such hybrid models will be examined. Moreover, its versatility will be highlighted, showing its usefulness in various stages of process development, including real-time monitoring and optimization. Scale-up remains one of the most relevant topics in fermentation processes, as it is important to have reproducible critical quality attributes, such as titer and yield, on larger scales. Furthermore, the process still relies on empirical correlations and iterative optimization. For these reasons, it is important to improve scale-up predictions, through e.g., the use of digital tools. Perspectives will be presented on the potential that hybrid modeling has by predicting performance across different process scales. This could provide more efficient and reliable biomanufacturing processes that require less resource consumption through experimentation.

Impact of Structural Differences on the Modeled Cost-Effectiveness of Noninvasive Prenatal Testing.

Noninvasive prenatal testing (NIPT) was developed to improve the accuracy of prenatal screening to detect chromosomal abnormalities. Published economic analyses have yielded different incremental cost-effective ratios (ICERs), leading to conclusions of NIPT being dominant, cost-effective, and cost-ineffective. These analyses have used different model structures, and the extent to which these structural variations have contributed to differences in ICERs is unclear. To assess the impact of different model structures on the cost-effectiveness of NIPT for the detection of trisomy 21 (T21; Down syndrome). A systematic review identified economic models comparing NIPT to conventional screening. The key variations in identified model structures were the number of health states and modeling approach. New models with different structures were developed in TreeAge and populated with consistent parameters to enable a comparison of the impact of selected structural variations on results. The review identified 34 economic models. Based on these findings, demonstration models were developed: 1) a decision tree with 3 health states, 2) a decision tree with 5 health states, 3) a microsimulation with 3 health states, and 4) a microsimulation with 5 health states. The base-case ICER from each model was 1) USD$34,474 (2023)/quality-adjusted life-year (QALY), 2) USD$14,990 (2023)/QALY, (3) USD$54,983 (2023)/QALY, and (4) NIPT was dominated. Model-structuring choices can have a large impact on the ICER and conclusions regarding cost-effectiveness, which may inadvertently affect policy decisions to support or not support funding for NIPT. The use of reference models could improve international consistency in health policy decision making for prenatal screening. NIPT is a clinical area in which a variety of modeling approaches have been published, with wide variation in reported cost-effectiveness.This study shows that when broader contextual factors are held constant, varying the model structure yields results that range from NIPT being less effective and more expensive than conventional screening (i.e., NIPT was dominated) through to NIPT being more effective and more expensive than conventional screening with an ICER of USD$54,983 (2023)/QALY.Model-structuring choices may inadvertently affect policy decisions to support or not support funding of NIPT. Reference models could improve international consistency in health policy decision making for prenatal screening.

Predicting seismic sustainability for a complex CHESST interaction by AHP using LWST

Underground and over-ground structures are affected by the external forces (tsunami, earthquake, storm, etc.) that are directly related to the environment and change in climate. For this reason, a complex CHESST (Climate-Human-Environment-Structure-Soil-Tunnel) interaction model is proposed in this research which is categorized based on the structure types (building, bridge, elevated, road/rail). The best CHESST interaction model is selected based on the AHP (Analytic Hierarchy Process) by using LWST (Logical Weightage Selection Technique). The LWST depends on the proposed modified logical scale. The logical weightage matrices are prepared for the three criteria (Environment and Climate, Social, and Economic), three sub-criteria under each criteria, and goal (selection of best CHESST interaction model). The weightage selection technique is validated based on the acceptable rate of decision change sensitivity. The elevated (alternative of the CHESST interaction model) shows the maximum value of seismic sustainability index (SSI) among all alternatives based on the AHP results. Also, the maximum fluctuation rate of strain is found for the elevated structure of the CHESST interaction model from the shake table test results. The most seismic sustainable CHESST interaction model is elevated expressway according to the SSI because of showing highest priority of 0.368. However, there is a scope to enhance this research in future by considering more alternatives, criteria, sub-criteria, etc.

A modeller’s fingerprint on hydrodynamic decision support modelling

Model results can have far-reaching societal implications, requiring fit-for-purpose models. However, model output is resulting from a particular path chosen with each modelling decision. We interviewed fourteen modellers in the Dutch water management sector in order to study how decision support hydrodynamic modellers make modelling decisions. An inductive-content analysis was performed. We identified eight motivation-categories. Individual and team considerations mostly motivate modelling decisions. We identified patterns between the motivation-categories and their occurrence across modelling steps. Modelling decisions during model implementation were found to be more in the modeller’s direct sphere of influence, while decisions concerning model structure and data selection more outside of it. So, even though modellers can leave their fingerprint, their sphere of influence and thus their fingerprint’s clarity is bounded by institutionalised predefined decisions. Thus, models and their results are shaped within a broader sphere than the modeller’s alone, requiring a broader consideration of organisations and standards.

A computationally efficient [formula omitted]-spectral form for partial dispersion analyses within the wave finite element framework

This paper addresses the computation of frequency-dependent dispersion curves (i.e., k(ω)) and wave modes within the framework of the Wave Finite Element Method (WFEM) and in the context of high-dimensional periodic unit cell models. Numerous applications, ranging from phononics to vibroacoustics, now rely on dispersion analyses or wave expansion over a subset of eigensolutions –complex wavenumbers and Bloch waves– resulting from the resolution of an eigenvalue problem with a T-palindromic quadratic structure (T-PQEP). To exploit the structure of finite element models, various structure-preserving linearizations such as the Zhong-Williams and the (S+S−1)-transform have already been developed to achieve partial wave resolution of large T-PQEP, primarily targeting the dominating (least decaying) waves. In this paper we derive an alternative linearization of the T-PQEP for the k(ω) problem, which leads to enhanced targeting of the eigenvalues around the unit circle and reduces the inaccuracies induced by root multiplicity. A specific form of the problem is then proposed as an optimal compromise between ease of implementation, numerical stability, convergence and accuracy enhancement. The performance of our proposed linearization is compared against existing ones across various iterative eigensolvers, since the generalized eigenvalue problems involve complex non-hermitian matrices, which are not extensively included in eigensolvers. Results indicate that the proposed linearization should be favoured for the WFEM, as it provides numerical enhancements in dispersion and wave vectors computation for large eigenvalue problems, as well as for further wave expansion applications.

Physico−mathematical model of the voltage−current characteristics of light-emitting diodes with quantum wells based on the Sah−Noyce−Shockley recombination mechanism

Herein, a physical and mathematical model of the voltage−current characteristics of a p−n heterostructure with quantum wells (QWs) is prepared using the Sah−Noyce−Shockley (SNS) recombination mechanism to show the SNS recombination rate of the correction function of the distribution of QWs in the space charge region of diode configuration. A comparison of the model voltage−current characteristics (VCCs) with the experimental ones reveals their adequacy. The technological parameters of the structure of the VCC model are determined experimentally using a nondestructive capacitive approach for determining the impurity distribution profile in the active region of the diode structure with a profile depth resolution of up to 10 Å. The correction function in the expression of the recombination rate shows the possibility of determining the derivative of the VCCs of structures with QWs with a nonideality factor of up to 4.

Chronological DC transmission expansion planning considering new energy and load uncertainties

Abstract The integration of renewable energy into the grid and the expansion of new loads have significantly increased the uncertainty of the power system, which has brought many new challenges to power system planning. Hence, it is necessary to establish a transmission expansion planning model that adapts to future development to improve the economic benefits of power grid planning. This paper proposes a chronological DC transmission expansion planning method for new energy sources, such as wind and solar energy, considering the time characteristics of source loads. First, the method establishes a chronological DC transmission expansion planning model by simulating long-term grid operations. Among them, the objective function aims to minimize the sum of investment costs and power generation operating costs related costs. Secondly, to simulate unit operation, a loop optimization modeling method is used to optimize the model structure. Finally, simulation analysis is conducted using the improved Garver’s 6 bus system under random planning scenarios as a basis for comparison. By comparing various scenarios, it is concluded that the proposed chronological DC transmission expansion planning can provide a more economical solution than non-chronological planning. This helps ensure that the system operates cost-effectively under uncertain source and load scenarios while enhancing the integration capabilities of wind and solar power generation.

MPE-YOLO: enhanced small target detection in aerial imaging

Aerial image target detection is essential for urban planning, traffic monitoring, and disaster assessment. However, existing detection algorithms struggle with small target recognition and accuracy in complex environments. To address this issue, this paper proposes an improved model based on YOLOv8, named MPE-YOLO. Initially, a multilevel feature integrator (MFI) module is employed to enhance the representation of small target features, which meticulously moderates information loss during the feature fusion process. For the backbone network of the model, a perception enhancement convolution (PEC) module is introduced to replace traditional convolutional layers, thereby expanding the network’s fine-grained feature processing capability. Furthermore, an enhanced scope-C2f (ES-C2f) module is designed, utilizing channel expansion and stacking of multiscale convolutional kernels to enhance the network’s ability to capture small target details. After a series of experiments on the VisDrone, RSOD, and AI-TOD datasets, the model has not only demonstrated superior performance in aerial image detection tasks compared to existing advanced algorithms but also achieved a lightweight model structure. The experimental results demonstrate the potential of MPE-YOLO in enhancing the accuracy and operational efficiency of aerial target detection. Code will be available online (https://github.com/zhanderen/MPE-YOLO).

Linear-quadratic Tobit regression model with a change point due to a covariate threshold

This paper considers a linear-quadratic Tobit regression model, which is developed for modelling the mixture structure with a line segment and a quadratic segment intersecting at an unknown change point. Due to the smoothness of such model structure, the regression coefficients and the change point can be obtained by directly maximising the likelihood function. The proposed estimator has rapid convergence rate and high estimation accuracy, without other computation burden. The asymptotic properties for the proposed estimator are derived by using empirical process theory. A sup-likelihood ratio test procedure is developed for testing the existence of a change point, and its limiting distributions are derived. The good finite sample performances of the proposed estimator are illustrated by numerical studies and empirical applications to the MGUS and GDP data.

A randomized clinical trial testing a health literacy intervention to reduce disparities in access to care among Justice-Impacted Adults (JIA)

BackgroundLow health literacy is costly and observed among justice-impacted adults (JIA), a group that often faces numerous barriers in accessing healthcare and a disproportionate burden of illness. Health literacy interventions for JIA are critically needed to improve healthcare access and related outcomes.MethodsThis manuscript describes the protocol for a longitudinal mixed-methods randomized clinical trial that assesses the effectiveness of a coach-guided health literacy intervention on JIA’s healthcare access. The intervention was previously piloted with justice impacted adults. We will recruit 300 JIA ages 18 + in San Diego, California. Participants will be randomized 1:1 to the Treatment Group (i.e., coach-guided intervention providing 12 sessions of individualized health coaching and service navigation over 6 months) or the Control Group (i.e., self-study of the health coaching program, and brief service navigation support). We will quantitatively assess JIA’s healthcare access defined as: use of healthcare, health insurance status, and regular source of care at 6-months as the primary outcomes. Participants will also be surveyed at 12-months. Statistical analyses will incorporate the intent-to-treat (ITT) principle and we will estimate mixed-effects logistic regression for the primary outcomes. We will also conduct qualitative interviews at 6 and 12-months with 40 purposively sampled participants, stratified by study arm, who reported healthcare access barriers at baseline. Interviews will explore participants’ satisfaction with the intervention, healthcare attitudes, self-efficacy for and barriers to healthcare access over time, perceived contribution of the intervention to health and well-being, and diffusion of intervention-related information within participants’ social networks. We will conduct deductive thematic analyses of qualitative data.DiscussionLow health literacy among JIA is a foundational challenge requiring tailored intervention strategies. Findings from this trial may inform policies and the structure of service delivery models to build health literacy among JIA in institutional and community settings throughout the United States and elsewhere.Trial registrationThis study is registered with the United States’ ClinicalTrials.gov registry under protocol # 161,903.

Fermionic UV models for neutral triple gauge boson vertices

Searches for anomalous neutral triple gauge boson couplings (NTGCs) provide important tests for the gauge structure of the standard model. In SMEFT (“standard model effective field theory”) NTGCs appear only at the level of dimension-8 operators. While the phenomenology of these operators has been discussed extensively in the literature, renormalizable UV models that can generate these operators are scarce. In this work, we study a variety of extensions of the SM with heavy fermions and calculate their matching to d = 8 NTGC operators. We point out that the complete matching of UV models requires four different CP-conserving d = 8 operators and that the single CPC d = 8 operator, most commonly used by the experimental collaborations, does not describe all possible NTGC form factors. Despite stringent experimental constraints on NTGCs, limits on the scale of UV models are relatively weak, because their contributions are doubly suppressed (being d = 8 and 1-loop). We suggest a series of benchmark UV scenarios suitable for interpreting searches for NTGCs in the upcoming LHC runs, obtain their current limits and provide estimates for the expected sensitivity of the high-luminosity LHC.

A method for predicting methane production from anaerobic digestion of kitchen waste under small sample conditions.

Anaerobic digestion (AD) has the great potential to treat organic waste and achieve remarkable results effectively. However, it is very tough to establish an accurate mechanistic model for this process. Data-driven modeling technology has opened a new door to solving this problem. While when the sample set is small, traditional data-driven modeling methods are often powerless. In this paper, an effective method is proposed for data-driven high-precision modeling in small sample scenarios. A time series generative adversarial network (TimeGAN) is first utilized to augment the original high-quality small-sample data collected during the AD methane production. A novel hybrid kernel extreme learning machine (HKELM) is then designed to form a better structure of the data-driven model, whose regularization coefficient C0 is optimized by the sparrow search algorithm (SSA). Finally, this semi-finished model (SSA-HKELM) is trained by the augmented data to form the final mathematical model (TimeGAN-SSA-HKELM) for the AD methane generation process. Comparative experiments of the methane daily production prediction error haveverified the effectiveness of the method, which can be extended to other similar small sample data-driven modeling scenarios.