Model Input Research Articles

Innovating bio-lubricant design: utilizing Mordred descriptors for model development and thermal stability prediction of novel compounds

Purpose Developing bio-lubricants for high-temperature applications, such as engine operations, requires reliable thermal stability assessment. This study aims to create a predictive tool to evaluate thermal stability using existing organic compounds data. The model predicts the onset temperature (T onset) of bio-lubricant candidates derived from epoxidation with various amine nucleophiles, enhancing initial selection efficiency and reducing reliance on traditional trial-and-error methods. Design/methodology/approach Using the Python library Mordred, molecular descriptors were calculated from the SMILES structures of 126 compounds with known T onsets. These descriptors were inputs for machine learning models predicting the thermal stability of bio-lubricants. The models were evaluated on training and test data sets and validated with novel synthesized compounds. Findings The predictive model showed strong performance on the test data set, with an R² of 0.93 and a root mean square error of 17.77. In external validation, the model estimated the thermal stability of a novel bio-lubricant compound (MA-EPO) at 290.7°C, closely matching the actual T onset of approximately 300°C. Originality/value This research introduces a method for predicting the thermal stability of bio-lubricants using machine learning and open-source libraries. This approach significantly advances the field by improving the efficiency of bio-lubricant development for high-temperature applications. It provides a cost-effective and time-saving alternative to traditional experimental methods, serving as a valuable resource for researchers and developers in the bio-lubricant industry.

First 2D electron density measurements using multi-delay coherence imaging spectroscopy in the MAST-U Super-X divertor

Abstract 2D profiles of electron density and neutral temperature are inferred from multi-delay coherence imaging spectroscopy data of divertor plasmas using a non-linear inversion technique. The inference is based on imaging the spectral line-broadening of Balmer lines and can differentiate between the Doppler and Stark broadening components by measuring the fringe contrast at multiple interferometric delays simultaneously. The model has been applied to images generated from simulated density profiles to evaluate its performance. Typical mean absolute errors of 30% are achieved, which are consistent with Monte Carlo uncertainty propagation accounting for noise, uncertainties in the calibrations, and in the model inputs. The analysis has been tested on experimental data from the MAST-U Super-X divertor, where it infers typical electron densities of 2–3 1019 m−3 and neutral temperatures of 0–2 eV during beam-heated L-mode discharges. The results are shown to be in reasonable agreement with the other available diagnostics.

Diagnosis of Wind Turbine Yaw System Based on Self-Attention–Long Short-Term Memory (LSTM)

Addressing the challenges and significant risks associated with diagnosing faults in wind turbine yaw systems, along with the typically low diagnostic accuracy, this study introduces a Long Short-Term Memory (LSTM) neural network augmented by a self-attention mechanism (SAM) as a novel fault diagnosis technique for wind turbine yaw systems. The method integrates the automatic weighting capability of the self-attention mechanism on input features with the advantage of LSTM in processing time series data, thereby effectively capturing key information and long-term dependencies in the operating data of the yawing system. This combination enhances the accuracy of fault feature extraction to more accurately identify various types of fault modes within the yawing system. Six types of feature parameters are extracted from the raw data collected by the SCADA (Supervisory Control And Data Acquisition) system of the wind turbine and are utilized as inputs for the diagnostic model. These parameters are then fed into the self-attention–LSTM neural network model to diagnose the health status of the yaw system, including yaw bearing damage, yaw gearbox failure, yaw motor failure, and sensor failure. The experimental results demonstrate that the accuracy of LSTM fault diagnosis, when enhanced with the self-attention mechanism, can reach 98.67% with an appropriate amount of training samples, verifying its significant advantages in terms of accuracy and stability of fault diagnosis. The proposed fault diagnosis method exhibits a better model fitting effect, strong generalization ability, and high accuracy compared to other methods, providing robust support for the reliable operation and maintenance of wind turbines.

Toward Machine Learning Electrospray Ionization Sensitivity Prediction for Semiquantitative Lipidomics in Stem Cells.

Specificity, sensitivity, and high metabolite coverage make mass spectrometry (MS) one of the most valuable tools in metabolomics and lipidomics. However, translation of metabolomics MS methods to multiyear studies conducted across multiple batches is limited by variability in electrospray ionization response, making batch-to-batch comparisons challenging. This limitation creates an artificial divide between nontargeted discovery work that is broad in scope but limited in terms of absolute quantitation ability and targeted work that is highly accurate but limited in scope due to the need for matched isotopically labeled standards. These issues are often observed in stem cell studies using metabolomic and lipidomic MS approaches, where patient recruitment can be a years-long process and samples become available in discrete batches every few months. To bridge this gap, we developed a machine learning model that predicts electrospray ionization sensitivity for lipid classes that have shown correlation with stem cell potency. Molecular descriptors derived from these lipids' chemical structures are used as model input to predict electrospray response, enabling quantitation by MS with moderate accuracy (semiquantitation). Model performance was evaluated via internal and external validation using cultured cells from various stem cell donors, achieving global percent errors of 40% and 20% for positive and negative electrospray ion modes, respectively. Although this accuracy is typically insufficient for traditional targeted lipidomics experiments, it is sufficient for semiquantitative estimation of lipid marker concentrations across batches without the need for specific chemical standards that many times are unavailable. Furthermore, the precision for model-predicted concentrations was 16.9% for the positive mode and 7.5% for the negative mode, indicating promise for data harmonization across batches. The set of molecular descriptors used by the models described here was able to yield higher accuracy than those previously published in the literature, showing high promise toward semiquantitative lipidomics.

Development of a Microsimulation Model to Project the Future Prevalence of Childhood Cancer in Ontario, Canada.

Estimates of the future prevalence of childhood cancer are informative for health system planning but are underutilized. We describe the development of a pediatric oncology microsimulation model for prevalence (POSIM-Prev) and illustrate its application to produce projections of incidence, survival, and limited-duration prevalence of childhood cancer in Ontario, Canada, until 2040. POSIM-Prev is a population-based, open-cohort, discrete-time microsimulation model. The model population was updated annually from 1970 to 2040 to account for births, deaths, net migration, and incident cases of childhood cancer. Prevalent individuals were followed until death, emigration, or the last year of simulation. Median population-based outcomes with 95% credible intervals (CrIs) were generated using Monte Carlo simulation. The methodology to derive model inputs included generalized additive modeling of cancer incidence, parametric survival modeling, and stochastic population forecasting. Individual-level data from provincial cancer registries for years 1970 to 2019 informed cancer-related model inputs and internal validation. The number of children (aged 0-14 y) diagnosed with cancer in Ontario is projected to rise from 414 (95% CrI: 353-486) in 2020 to 561 (95% CrI: 481-653) in 2039. The 5-y overall survival rate for 2030-2034 is estimated to reach 90% (95% CrI: 88%-92%). By 2040, 24,088 (95% CrI: 22,764-25,648) individuals with a history of childhood cancer (diagnosed in Ontario or elsewhere) are projected to reside in the province. The model accurately reproduced historical trends in incidence, survival, and prevalence when validated. The rising incidence and prevalence of childhood cancer will create increased demand for both acute cancer care and long-term follow-up services in Ontario. The POSIM-Prev model can be used to support long-range health system planning and future health technology assessments in jurisdictions that have access to similar model inputs. This article describes the development of a population-based, discrete-time microsimulation model that can simulate incident and prevalent cases of childhood cancer in Ontario, Canada, until 2040.Use of an open cohort framework allowed for estimation of the potential impact of net migration on childhood cancer prevalence.In addition to supporting long-term health system planning, this model can be used in future health technology assessments, by providing a demographic profile of incident and prevalent cases for model conceptualization and budget impact purposes.This modeling framework is adaptable to other jurisdictions and disease areas where individual-level data for incidence and survival are available.

Patient- and fraction-specific magnetic resonance volume reconstruction from orthogonal images with generative adversarial networks.

Although deep learning (DL) methods for reconstructing 3D magnetic resonance (MR) volumes from 2D MR images yield promising results, they require large amounts of training data to perform effectively. To overcome this challenge, fine-tuning-a transfer learning technique particularly effective for small datasets-presents a robust solution for developing personalized DL models. A 2D to 3D conditional generative adversarial network (GAN) model with a patient- and fraction-specific fine-tuning workflow was developed to reconstruct synthetic 3D MR volumes using orthogonal 2D MR images for online dose adaptation. A total of 2473 3D MR volumes were collected from 43 patients. The training and test datasets were separated into 34 and 9 patients, respectively. All patients underwent MR-guided adaptive radiotherapy using the same imaging protocol. The population data contained 2047 3D MR volumes from the training dataset. Population data were used to train the population-based GAN model. For each fraction of the remaining patients, the population model was fine-tuned with the 3D MR volumes acquired before beam irradiation of the fraction, named the fine-tuned model. The performance of the fine-tuned model was tested using the 3D MR volume acquired immediately after the beam delivery of the fraction. The model's input was a pair of axial and sagittal MR images at the isocenter level, and the output was a 3D MR volume. Model performance was evaluated using the structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), root mean square error (RMSE), and mean absolute error (MAE). Moreover, the prostate, bladder, and rectum in the predicted MR images were manually segmented. To assess geometric accuracy, the 2D Dice Similarity Coefficient (DSC) and 2D Hausdorff Distance (HD) were calculated. A total of 84 3D MR volumes were included in the performance testing. The mean±standard deviation (SD) of SSIM, PSNR, RMSE, and MAE were 0.64±0.10, 93.9±1.5dB, 0.050±0.009, and 0.036±0.007 for the population model and 0.72±0.09, 96.2±1.8dB, 0.041±0.007, and 0.028±0.006 for the fine-tuned model, respectively. The image quality of the fine-tuned model was significantly better than that of the population model (p<0.05). The mean±SD of DSC and HD of the population model were 0.79±0.08 and 1.70±2.35mm for prostate, 0.81±0.10 and 2.75±1.53mm for bladder, and 0.72±0.08 and 1.93±0.59mm for rectum. Contrarily, the mean±SD of DSC and HD of the fine-tuned model were 0.83±0.06 and 1.29±0.77mm for prostate, 0.85±0.07 and 2.16±1.09mm for bladder, and 0.77±0.08 and 1.57±0.52mm for rectum. The geometric accuracy of the fine-tuned model was significantly improved than that of the population model (p<0.05). By employing a patient- and fraction-specific fine-tuning approach, the GAN model demonstrated promising accuracy despite limited data availability.

Machine Learning-Based Reconstruction and Prediction of Groundwater Time Series in the Allertal, Germany

State-of-the-art hydrogeological investigations use transient calibrated numerical flow and transport models for multiple scenario analyses. However, the transient calibration of numerical flow and transport models still requires consistent long-term groundwater time series, which are often not available or contain data gaps, thus reducing the robustness and confidence of the numerical model. This study presents a data-driven approach for the reconstruction and prediction of gaps in a discontinuous groundwater level time series at a monitoring station in the Allertal (Saxony-Anhalt, Germany). Deep Learning and classical machine learning (ML) approaches (artificial neural networks (TensorFlow, PyTorch), the ensemble method (Random Forest), boosting method (eXtreme gradient boosting (XGBoost)), and Multiple Linear Regression) are used. Precipitation and groundwater level time series from two neighboring monitoring stations serve as input data for the prediction and reconstruction. A comparative analysis shows that the input data from one measuring station enable the reconstruction and prediction of the missing groundwater levels with good to satisfactory accuracy. Due to a higher correlation between this station and the station to be predicted, its input data lead to better adapted models than those of the second station. If the time series of the second station are used as model inputs, the results show slightly lower correlations for training, testing and, prediction. All machine learning models show a similar qualitative behavior with lower fluctuations during the hydrological summer months. The successfully reconstructed and predicted time series can be used for transient calibration of numerical flow and transport models in the Allertal (e.g., for the overlying rocks of the Morsleben Nuclear Waste Repository). This could lead to greater acceptance, reliability, and confidence in further numerical studies, potentially addressing the influence of the overburden acting as a barrier to radioactive substances.

Developing a Neural Network Based Microwave Sensing System for Accurate Salinity Prediction in Water

Abstract High and low salinity levels play a crucial role in the vitality of organisms and affect natural ecosystems, agricultural yields and human health. To mitigate the risks associated with high blood pressure and cardiovascular diseases, the World Health Organization (WHO) advocates reducing salt consumption among adults, suggesting an intake of no more than 5 g daily. In this study, a non-invasive microwave (MW) sensing approach, that is augmented by deep neural network (DNN) models is proposed to predict salinity levels. The MW detection measurement system, including a Horn antenna, has been developed to evaluate the salt content in bottled spring waters (BSWs). The system with DNN model provides a novel solution for real-time water quality monitoring. The input and output dataset for DNN model were generated using four different BSWs, each with a salt content ranging from 0 to 32 g and increased by 1 g. The developed DNN model, designed with six fully connected layers, uses reflection coefficients (RCs) as input dataset to predict salt content in grams accurately. The accuracy performance of the DNN model in various bandwidths was evaluated by dividing the 1–13 GHz range into 78 different bands and the lowest error rate was found to be in the 1–8 GHz bandwidth (2.18%). Furthermore, each BSW was measured five times, and the performance of the model was evaluated according to the number of measurements. In three or more measurements, the model demonstrated notable improvement(15.3%) in predicting salt content.

Evaluating the Impacts of Parameter Uncertainty in a Practical Transportation Demand Model

The inherent uncertainty in travel forecasting models—arising from potential and unknown errors in input data, parameter estimation, or model formulation—is receiving increasing attention from both the scholarly and practicing communities. In this research, we investigate the variance in forecasted traffic volumes resulting from varying the mode and destination choice parameters in an advanced trip-based travel demand model. Using Latin hypercube sampling to construct several hundred combinations of parameters across the plausible parameter space, we introduce substantial changes to implied travel impedances and modal utilities, on the order of a 10 percent variation. However, the aggregate effects of these changes on forecasted traffic volumes are small, with a variation of approximately 1 percent on high-volume facilities. It is likely that in this example—and perhaps in others—the network assignment places constraints on the possible volume solutions and limits the practical impacts of parameter uncertainty. Nevertheless, parameter uncertainty may not be the largest contributor to error in practical travel forecasts. Further research should examine the robustness of this finding across other less constrained networks and within activity-based travel model frameworks.

A Simple, Scalable Large Deformation Solid Mechanics Implementation in the MOOSE Framework

This paper describes a large deformation solid mechanics solver implemented as part of the freely-available and open-source MOOSE finite element simulation framework. The paper documents the choices made in developing the solid mechanics framework and describes novel formulations for the gradient operator and constitutive modeling framework made to simplify implementations of different coordinate systems, stabilized gradient operators, and different constitutive model inputs and outputs. In the process, the paper describes a new formulation that casts objective integration of the Cauchy stress as a linear transformation of the small stress rate. Finally, the paper presents key implementation details and examines the parallel efficiency of the solid mechanics solver implemented in MOOSE. The implementation retains good weak scaling efficiency beyond 1,000 parallel processes. The papers includes a discussion of the factors limiting the parallel efficiency of implicit, large deformation solid mechanics codes on current high-performance computers, with the main current limitation being the scalability of the algebraic multigrid methods used to solve the linearized equilibrium equations.

Dynamics and hazards of pyroclastic avalanches at Etna volcano (Italy)

We present a multidisciplinary research aimed at quantifying the conditional probabilities for hazards associated with pyroclastic avalanches at Etna, which combines physical and numerical modeling of granular avalanches and probabilistic analysis. Pyroclastic avalanches are modeled using the depth-averaged model IMEX-SfloW2D, which is able to simulate the transient propagation and emplacement of granular flows generated by the collapse of a prescribed volume of granular material. Preliminary sensitivity analysis allowed us to identify the main controlling parameters of the dynamics, i.e. the total avalanche mass, the initial position of the collapsing granular mass (and the associated terrain morphology), the initial avalanche velocity, and the two rheological parameters which determine the mechanical properties of the flow. While the first two parameters can be considered as “scenario parameters” in the definition of the hazards, the initial velocity and the rheological parameters need to be calibrated. We therefore adopted a methodology for the statistical calibration of the physical model parameters based on field observations. We used data from the pyroclastic avalanche that occurred on February 10, 2022 at Etna, for which we had an accurate mapping of the deposit and some estimates of the total mass and the initial volume. We then run a preliminary ensemble of numerical simulations, with fixed initial volume and position, to calibrate the other input parameters. Based on the accuracy of the matching of the simulated and observed deposits (measured by the Jaccard Index), we extracted from the simulation ensemble a subsample of equally probable combinations of initial velocities and rheological parameters. We then built an ensemble of model input parameters, with varying (i) avalanche volumes, (ii) initial positions, (iii) velocity, and (iv) rheological coefficients. The initial volume range was chosen within the range of observed pyroclastic avalanches at Etna (i.e., between 0.1 and 3 × 106 m3), using a prescribed probability distribution extracted from the literature data. The initial positions have been chosen on the flanks of the South East Crater of Etna, with homogeneous spatial distribution. The initial velocity and the rheological coefficients were chosen from the subsample created with the calibration. Finally, a semi-automatic procedure (digital workflow) running the Monte Carlo simulation allowed us to produce the first probabilistic map of pyroclastic avalanche invasion at Etna. Such a map, conditional to the occurrence of a pyroclastic avalanche event, can be used to identify the hazardous areas of the volcano and to plan mitigation measures.

IFD-BiC: A Class-incremental Continual Learning Method for Diesel Engine Fault Diagnosis

Abstract In recent years, deep learning-based fault diagnosis methods for diesel engines have been developed assuming that a comprehensive dataset is available for one-time learning. However, in reality, it is impractical to obtain a complete dataset all at once. Instead, new data is gradually acquired over time. To address this issue, this paper proposes an incremental fault diagnosis method with bias correction (IFD-BiC). When receiving incremental fault data, the IFD-BiC model incorporates new classifier nodes and updates network parameters using a combination of distillation loss and cross-entropy. During the update process, old samples saved in the exemplar set are replayed, and a bias correction layer is used to correct the prediction bias caused by the imbalance between new and old samples. This enables the model to learn new fault tasks without experiencing catastrophic forgetting. To ensure the applicability of the method to various operating conditions of diesel engines, cyclic angular vibration with consistent length at different speeds is used as the model input. Through experimental validation across a pair of distinct diesel engine models, the proposed method achieves highly accurate fault diagnosis on stream format training data. Moreover, an evaluation against leading incremental learning techniques affirms the proposed method’s advantage. &amp;#xD;

Integrated neuronetwork modeling of EEG for diagnostic disorders of brain activity

The article discusses the structured multistage modeling of EEG by means of applied mathematics to customize the input parameter space for neural network prediction. Also, the approaches and models are analyzed for their accuracy in determining the relevant signal features and adaptability to real data. The activity of potentials during brain activity is a biological process that depends on many factors and hides a space of parameters, the search for which and their definition can open us up to a new perspective on the nature and activity of the human brain. A rational way of obtaining data on brain activity is a non-invasive electroencephalogram, which registers the potential difference on the electrodes relative to the base. For further processing of the received data, it is necessary to remove noise and artifacts from them. In this work, a stan-dard algorithm of frequency filtering and filtering from noise caused by the power grid is used. After processing the data, an overview of mathematical models is offered, which with a certain degree of accuracy try to simulate the behavior of the signal or the peak moments of certain features. Added to this is the use of the LSTM model to predict the further behavior of the signal with the preliminary introduction of chaos into the model due to the modified activation function (Gaussian noise) and the input modeling of the weights.

Cost-effectiveness of NALIRIFOX compared to other first-line treatments for metastatic pancreatic cancer.

733 Background: NALIRIFOX was recently approved for first-line treatment of metastatic pancreatic cancer, where combination chemotherapy remains the main treatment modality. We conducted a cost-effectiveness analysis of NALIRIFOX compared to both FOLFIRINOX and gemcitabine/protein-bound paclitaxel (GP). Methods: A partitioned survival model with three health states (progression-free, post-progression and death) was constructed to model the disease course. Using published data from NAPOLI-3, parametric survival curves were generated for NALIRIFOX, from which survival curves for FOLFIRINOX and GP were extrapolated using pooled efficacy data reported by a network meta-analysis of clinical trials. Model inputs were retrieved from the literature or other publicly available resources. Specifically, drug acquisition costs were from average wholesale prices in the Red Book while administration, disease monitoring and supportive care costs were from the CMS Physician Fee Schedule. Utility values to estimate quality-adjusted life years (QALY) were from the CALGB80303 trial. Adverse effects that were considered included anemia, neutropenia, febrile neutropenia, thrombocytopenia, diarrhea, peripheral neuropathy, nausea and vomiting, and fatigue. Adverse effect rates were based on data pooled from clinical trials while costs and disutility were obtained from a secondary analysis of claims data and a clinician survey, respectively. The time horizon was 5 years and utilized a third-party payer perspective. Both costs ($USD 2024) and outcomes were discounted at 3% annually. A probabilistic sensitivity analysis was conducted by varying model parameters based on 95% confidence intervals if known or 10% in both directions otherwise. Results: The overall cost associated with NALIRIFOX was $129,314, followed by GP ($104,593) and FOLFIRINOX ($41,528). Among the 3 regimens, NALIRIFOX incurred the highest treatment-related cost ($105,533) but the lowest adverse effects management cost ($16,541). On average, patients on NALIRIFOX gained a total of 1.14 life years (LY) or after adjusting for utility, 0.86 QALY. FOLFIRINOX led to 1.09 LY or 0.82 QALY while GP resulted in 0.99 LY or 0.75 QALY. Due to its higher cost and lower survival benefit, GP was dominated by FOLFIRINOX. Compared to FOLFIRINOX, the incremental cost-effectiveness ratio (ICER) of NALIRIFOX was $1,952,062/QALY. In our sensitivity analysis, the probability of NALIRIFOX being cost-effective was 0% at all conventional ICER thresholds up to $200,000/QALY. Conclusions: Due to the high drug acquisition cost, the ICER of NALIRIFOX far exceeds currently acceptable thresholds for economic value and NALIRIFOX is considered not cost-effective. This is despite that in our model among first line treatments for metastatic pancreatic cancer, NALIRIFOX was associated with reduced costs for managing adverse effects.

Dapagliflozin for the treatment of heart failure with reduced ejection fraction in Brazil: a cost-effectiveness analysis.

Heart failure, a complex clinical syndrome with high morbidity and mortality, has become a significant burden on public health. Recently, a new class of antidiabetic agents-the sodium-glucose cotransporter 2 (SGLT2) inhibitors-was associated with a significant reduction on mortality and hospitalization in HF with reduced ejection fraction (HFrEF) when added to standard pharmacological treatment. Considering the lack of data on its cost-effectiveness, the present study aims to estimate the incremental cost-effectiveness ratio of add-on dapagliflozin treatment for HFrEF from the Brazilian public healthcare system perspective. We built a Markov model to estimate the clinical outcomes and costs of 1,000 hypothetical subjects with established HFrEF in a lifetime horizon. The model inputs were based on the Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure (DAPA-HF) trial and local data. The main outcome was the incremental cost-effectiveness ratio (ICER) per quality-adjusted life year (QALY) gained. Deterministic and probabilistic sensitivity analyses, as well as scenario analyses, were performed. The addition of dapagliflozin to standard care treatment in 1,000 HFrEF patients yielded an expected value of 366.99 additional QALYs at an incremental cost of US$ 1,517,878.49, resulting in an ICER of US$ 4,136.08 per QALY gained, being a cost-effective strategy considering the Brazilian official cost-effectiveness threshold (US$ 8,000/QALY). In probabilistic sensitivity analyses, 96.60% of the simulations were also cost-effective. In the scenario analyses, results were similar for individuals with and without diabetes. Dapagliflozin is likely to be cost-effective when added to standard HFrEF therapy in Brazil. This study was supported by the National Institute of Science and Technology for Health Technology Assessment (Instituto de Avaliação de Tecnologias em Saúde-IATS).

Reconstruction of the solid–liquid two-phase flow field in the pipeline based on limited pipeline wall information

Pipeline hydraulic transportation is the primary method for transporting deep-sea mineral resources and fossil fuels. Pipeline blockage often causes excessive pressure in the pipeline, leading to pipeline breakage or even cargo leakage, which severely impacts transportation safety and can easily trigger secondary disasters. Therefore, clarifying the global flow field within pipelines, such as particle distribution, is crucial for monitoring and controlling pipeline systems. This study uses a limited number of easily measurable pipeline wall sensor pressure values as inputs of deep learning models for flow field reconstruction, with the global flow field of solid–liquid two-phase flow in the three-dimensional pipeline as the output. Three model frameworks from existing studies are summarized, and their reconstruction effects are compared. Based on this, a new framework is proposed. It expands the low-dimensional sensor pressure values to the same size as the global flow field using a pseudo-decoder and then processes them through an autoencoder. The results indicate that the new framework achieves further accuracy improvements compared to the previous three frameworks, with R2 and mean squared error reaching 0.933 and 5.13 ×10−4, respectively. Additionally, the effects of the skip connection configuration of the model, dataset size, and model learning rate, as well as the number and arrangement of pressure sensors on reconstruction accuracy, are investigated. Finally, the transferability of the model is demonstrated by reconstructing the pressure and fluid velocity fields of the pipeline two-phase flow.

Exploring building performance evaluation methods to estimate energy savings following field implementation

Research on energy conservation measures and advanced control strategies is becoming increasingly present to improve building energy performance and may reach the stage of field testing in existing buildings. Therefore, it becomes critical to properly assess the benefits following on-site implementation. Different approaches have been conducted in the past to predict building energy performance and evaluate energy savings. However, it remains not clear how the calculation methods and their underlying assumptions eventually influence the savings estimates. This paper aims to compare the energy savings obtained following the implementation of data-driven measures in an existing building and estimated using three different approaches that are commonly found in the literature. The intricacies of these baseline model-free and baseline model-based methods were investigated, and various aspects were explored such as weather normalization, baseline modelling time-step (monthly, daily, hourly, sub-hourly), input sets (including weather, time-index variables, room air temperatures, heating system conditions), and techniques (linear, multi-linear, gaussian process regression). Results show that savings estimates can significantly vary from one performance evaluation method to another, and from one model to another, highlighting the importance of properly selecting both the evaluation method and the baseline model. Baseline model-based approaches were found more appropriate than the other methods, and daily linear regressions against outdoor air temperature and hourly gaussian process regression using weather and time-index variables were found suitable for building performance evaluation.

Using the CIPP Model to elicit perceptions of health professions faculty and students about virtual learning

BackgroundThe outbreak of the Coronavirus epidemic has caused a huge crisis initiating fundamental changes in education since this crisis has turned face-to-face education into virtual training. This questionnaire-based study employed the comprehensive CIPP Model (Context, Input, Process, Product) to obtain the perspectives of both faculty members and students from six different faculties at one university in Iran concerning implementation of virtual learning in during COVID 19.MethodologyThe participants in this cross-sectional study were 522 students and 38 members of the faculty in six different faculties who were selected via stratified random sampling. The research tool was a validated and reliable researcher-made questionnaire developed based on the context, input, process, and product (CIPP) evaluation model. The survey included a scale for each of the four CIPP components, with scales comprised of 9 to 12 questions. The data were analyzed through SPSS 23 using descriptive and inferential (Mann–Whitney, Kruskal–Wallis, and Spearman correlation) methods.ResultsThe scale items for each of the CIPP components that elicited the highest levels of agreement by both professors and students were as follows based on a five point scale where higher scores indicated higher levels of respondent agreement: Context: Topics presented in the virtual training are determined according to the course plan (3.63), and virtual education reduces the teacher's control over class (3.56); Input: Designated hours are suitable for virtual learning classes (3.29); Process: Professors have less commitment and responsibility in providing virtual courses (3.48); and Product: Student participation in virtual classes is low (3.78), and virtual learning saves time (3.67).” For both students and faculty, the mean scores for the context, input, process, and product scales all averaged near the mid-point of the scale, 3.00. No significant difference was observed between professors and students except for the input construct. Students responded significantly differently according to their age and educational level on the product construct, and significantly differently according to their faculty and marital status on each of the CIPP constructs.ConclusionThis study has highlighted various issues related to virtual education, and the opinions of professors and students regarding changes to the online educational program should be strategically considered. The present findings can facilitate decision-making and policy-making at the macro level, enabling officials to plan appropriately, take professional measures, and decide whether to continue, cease, or revise educational goals, inputs, processes, and products.

Physics-guided degradation trajectory modeling for remaining useful life prediction of rolling bearings

Remaining useful life (RUL) prediction has great significance in reducing operating costs and enhancing the maintainability and safety of rolling bearings. Recently, significant progress has been achieved in this field by leveraging deep learning approaches. However, advanced deep learning methods suffer from a black-box nature that makes them lack interpretability. Moreover, the prediction results may sometimes not follow laws of physics. To tackle this drawback, a physics-guided degradation trajectory modeling method is proposed for RUL prediction of rolling bearings, where physical knowledge is embedded in input preparation, model construction, and output specification. Specifically, phase space reconstruction is leveraged to construct the physics-guided inputs, transforming the degradation prediction of rolling bearings into the variation estimation of phase space trajectories. The degradation trajectory prediction strategy is derived from the physics-guided inputs and the decreased value of RUL is predicted through a compact one-dimensional convolutional neural network with the nonnegative bounded function. Simulation studies are performed with a theoretical model of defective rolling bearings to illustrate the effectiveness of the constructed physics-guided inputs. In addition, comparative analyses with advanced RUL prediction approaches under unseen working conditions and across different machines are conducted. The results demonstrate that the proposed method outperforms other approaches in both accuracy and robustness, showing its superior ability in RUL prediction of rolling bearings.

Rapid urbanization decelerates regional net anthropogenic phosphorus input: Evidence from the Chengdu-Chongqing urban agglomeration in China.

Mitigating the pressure of regional phosphorus (P) inputs driven by human activities is essential for the prevention and control of non-point source pollution as well as for effective environmental management. This study emploied the net anthropogenic phosphorus input (NAPI) model and the coupling coordination degree model (CCDM) to quantitatively analyze the spatiotemporal evolution of phosphorus inputs and urbanization levels in the Chengdu-Chongqing urban agglomeration (CCUA) in Southwest China from 2011 to 2022. By integrating urbanization, socioeconomic and land use data, we identified key driving factors and specific indicators influencing changes in regional phosphorus inputs and their components. Compared to 2011, urbanization levels increased by 16.60% in 2022, while regional phosphorus inputs decreased by 21.52%. P input from human food and livestock feed was the main component of NAPI during the study period (54.27%-61.28%), followed by fertilizer application (37.46%-44.62%). The shift in industrial structure towards the tertiary sector (explanation rate 20.8%) and construction land expansion (explanation rate 17.7%) were identified as key drivers of phosphorus input changes. Additionally, urbanization has a more significant direct standardized path coefficient on P inputs (-7.30) than socioeconomic (0.44) and land use factors (0.37) based on Partial Least Squares Path Modeling (PLS-PM), suggesting that regional phosphorus inputs decrease with increasing urbanization levels. These findings provide new insights for the formulation of effective phosphorus management strategies in rapidly urbanizing areas, and are conducive to the sustainable management of urbanization progress and ecological environment.