Search

This study explores strategies for long‐term reservoir simulations by combining generic rule‐based reservoir management model (RMM) and machine learning (ML) models for two major multipurpose reservoirs — Allatoona Lake and Lake Sidney Lanier in the southeastern United States. First, a standalone RMM is developed to simulate daily release and storage during Water Year 1981–2015. Next, using Long‐Short Term Memory (LSTM) as the ML technique, a standalone LSTM model is trained based on reservoir inflow and meteorological observations to simulate reservoir release and estimate reservoir storage through water balance calculation. Three hybrid modeling strategies are developed, one using RMM output as an additional LSTM input (H1), another using LSTM as the initial release estimate in RMM (H2), and the third combining the first two strategies (H3). The Nash–Sutcliffe efficiency (NSE) for release (NSE‐r), storage (NSE‐s), and their mean (NSE‐avg) are used for model evaluation. Overall, H1 improves NSE‐r to 0.65 and 0.54 for Allatoona and Lanier, respectively, compared to standalone RMM (0.44 and 0.21); however, its storage trajectory did not produce a physically feasible solution, similar to LSTM. H2 and especially H3 show that they can retain the best features from RMM and LSTM, with H3 NSE‐avg being 0.695 and 0.55 for Allatoona and Lanier outperforming RMM (0.615 and 0.29). The findings suggest a robust simulation capacity for large‐scale water management in future studies. [ABSTRACT FROM AUTHOR]

This paper introduces an advanced Short‐term Nodal Load Forecasting (STNLF) method that forecasts nodal load profiles for the next day in power systems, based on the combined use of three machine learning techniques. Least Absolute Shrinkage and Selection Operator (LASSO) is employed to reduce the number of features for a single nodal load forecasting. Principal Component Analysis (PCA) is used to capture the features of historical loads in low‐dimensional space compared to the original high‐dimensional load space where features are barely possible to depict. Bayesian Ridge Regression (BRR) is utilized to decide the parameters of the prediction model from a statistics perspective. Tests based on modified PJM load data demonstrate the effectiveness of the proposed STNLF method compared to the state‐of‐the‐art General Regression Neural Network (GRNN) method. Moreover, the reliability of the day‐ahead Unit Commitment (UC) solution is shown to have been improved, based on the forecasted load data using the proposed STNLF method. [ABSTRACT FROM AUTHOR]

Hydrologic predictions at rural watersheds are important but also challenging due to data shortage. Long short-term memory (LSTM) networks are a promising machine learning approach and have demonstrated good performance in streamflow predictions. However, due to its data-hungry nature, most LSTM applications focus on well-monitored catchments with abundant and high-quality observations. In this work, we investigate predictive capabilities of LSTM in poorly monitored watersheds with short observation records. To address three main challenges of LSTM applications in data-scarce locations, i.e., overfitting, uncertainty quantification (UQ), and out-of-distribution prediction, we evaluate different regularization techniques to prevent overfitting, apply a Bayesian LSTM for UQ, and introduce a physics-informed hybrid LSTM to enhance out-of-distribution prediction. Through case studies in two diverse sets of catchments with and without snow influence, we demonstrate that 1) when hydrologic variability in the prediction period is similar to the calibration period, LSTM models can reasonably predict daily streamflow with Nash–Sutcliffe efficiency above 0.8, even with only 2 years of calibration data; 2) when the hydrologic variability in the prediction and calibration periods is dramatically different, LSTM alone does not predict well, but the hybrid model can improve the out-of-distribution prediction with acceptable generalization accuracy; 3) L2 norm penalty and dropout can mitigate overfitting, and Bayesian and hybrid LSTM have no overfitting; and 4) Bayesian LSTM provides useful uncertainty information to improve prediction understanding and credibility. These insights have vital implications for streamflow simulation in watersheds where data quality and availability are a critical issue. [ABSTRACT FROM AUTHOR]

Although there is increased interest in utilizing machine learning (ML) to support drug development, technical hurdles associated with complex algorithms have limited widespread adoption. In response, we have developed Pharm‐AutoML, an open‐source Python package that enables users to automate the construction of ML models and predict clinical outcomes, especially in the context of pharmacological interventions. In particular, our approach streamlines tedious steps within the ML workflow, including data preprocessing, model tuning, model selection, results analysis, and model interpretation. Moreover, our open‐source package helps to identify the most predictive ML pipeline among defined search spaces by selecting the best data preprocessing strategy and tuning the ML model hyperparameters. The package currently supports multiclass classification tasks, and additional functions are being developed. Using a set of five publicly available biomedical datasets, we show that our Pharm‐AutoML outperforms other ML frameworks, including H2O with default settings, by demonstrating improved predictive accuracy of classification. We further illustrate how model interpretation methods can be utilized to help explain the fine‐tuned ML pipeline to end users. Pharm‐AutoML provides both novice and expert users improved clinical predictions and scientific insights. [ABSTRACT FROM AUTHOR]

Diversified ion-selective separation applications have dramatically incentivized the exploitation and performance modulation of highly ion-selective nanofiltration (NF) membranes. However, the ion selectivity of NF membranes is synergistically governed by multi-scale factors of membrane structural parameters and operational parameters, with the intrinsic ion-selective mechanism still ambiguous. Herein, we proposed an ensemble machine learning (ML) method to decouple key factors affecting the ionic selectivity of polyamide NF membranes. Membrane structural parameters and operational parameters were typically extracted as input variables and linked to mono−/divalent ion selectivity by model training based on Random Forest and XGBoost algorithms. The feature importance assessment indicated the critical role of membrane structure parameters on ion selectivity, wherein pore radius dominated the mono−/divalent anionic selectivity while zeta potential for mono−/divalent cationic selectivity. Partial dependence analyses further depicted intensive insights regarding the influence of membrane structural parameters on ion selectivity. Moreover, stochastic dataset splitting measurements demonstrated the accurate predictive capability of the model simultaneously possessing excellent stability and reliability. We anticipated that the implementation of ensemble ML in explicating the intricate ion-selective mechanism created platforms for understanding the structure-membrane performance correlation and orientally manufacturing highly ion-selective NF membranes. [Display omitted] • Ensemble machine learning was implemented to decouple the key features affecting ion selectivities of polyamide membranes. • RF and XGBoost algorithms were adopted to learn from relevant variables and evaluate permutation features importance. • Mono-/divalent anionic selectivity was dominated by pore size while zeta potential for mono-/divalent cationic selectivity. [ABSTRACT FROM AUTHOR]

Nanofiltration (NF) plays an increasingly central role in water/salts separation, which puts forward tailored requirements on NF membranes in a variety of application scenarios. However, the ambiguous separation mechanisms of NF including membrane structure parameters and operating conditions hinder the rational design of versatile NF membranes. Herein, machine learning was used to explore the correlation between membrane structure parameters and operating conditions with water/salts selectivity, and reveal the importance of the diverse features based on literature data. Two structural features of polyamide NF membrane (pore radius and zeta potential) and two operating parameters (pressure and feed concentration) were typically extracted and associated with water/salts selectivity. Random Forest and XGBoost models were employed to learn from relevant variables and assess their importance. The results showed that membrane structure parameters attached more importance to water/salts selectivity than that of operating conditions accompanying the variable influence for different typed salts, where symmetrical salts were mainly governed by size sieving while Donnan exclusion for asymmetrical salts. Structure-performance relationships between pore radius, zeta potential and diverse water/salts selectivity were established using partial dependence plot analysis. It is anticipated that the constructed comprehensive insights can be further leveraged to tackle performance modulation and oriented design of multi-scenarios NF membranes. [Display omitted] • Machine learning was used to understand NF separation mechanisms. • Random Forest and XGBoost models were employed to learn from relevant variables and assess features importance. • Symmetrical salts were mainly governed by size sieving while Donnan exclusion for asymmetrical salts. • The integrated management of pore size and zeta potential for the oriented design of versatile NF membranes was highlighted. [ABSTRACT FROM AUTHOR]

Neutron scattering is a powerful but expensive technique to study materials and discover new matter. Advanced detector technology has significantly improved the efficiency of neutron experiments, increasing the complexity of neutron data reduction and analysis. Machine learning (ML) brings new directions for neutron diffraction data reduction and experiment operation. This work presents an ML‐assisted data reduction and analysis method for precise recognition of Bragg peaks and the corresponding regions of interest; it can then automatically screen and align a measured crystal using the recognized peaks, and subsequently plan and optimize the data collection with user‐provided information and uncertainty quantification values of detected peaks. This method shows robust performance in different complex sample environments and enables automated single‐crystal neutron diffraction. [ABSTRACT FROM AUTHOR]

[Display omitted] • The design of novel single-phase refractory high entropy alloys via machine learning. • Extended dataset and optimal input features for machine learning algorithms. • Machine learning model validation by mechanical alloying experiment. Herein, we proposed a strategy to design single-phase refractory high entropy alloys (RHEAs) with the assistance of machine learning algorithms. Based on an extensive dataset (1807 entries) built in this work, we applied multiple machine learning algorithms to train the dataset. After the blind test, we found that the Gradient boosting (GB) model can distinguish the single-phase-solid solution and non-single-phase-solid solution alloys with a test accuracy of 96.41%. Given the GB model, we predicted over 100 equiatomic oxidation-resistance RHEAs from the composition space of eight metallic elements. After that, we synthesized ten of these predicted single-phase RHEAs by mechanical alloying. The XRD patterns show that all of them are single-phase BCC solid solution. The experimental results agree well with the prediction results, indicating the excellent performance of the machine learning model in single-phase RHEAs prediction. With the aid of the machine learning method, single-phase oxidation-resistant RHEAs were successfully designed. Our work presents a novel strategy with outstanding performance and evident effectiveness on the accelerated discovery of novel metallic materials used for extreme environments. [ABSTRACT FROM AUTHOR]

Objective: This study aimed to evaluate the diagnostic performance of machine learning (ML)–based ultrasound (US) radiomics models for risk stratification of gallbladder (GB) masses. Methods: We prospectively examined 640 pathologically confirmed GB masses obtained from 640 patients between August 2019 and October 2022 at four institutions. Radiomics features were extracted from grayscale US images and germane features were selected. Subsequently, 11 ML algorithms were separately used with the selected features to construct optimum US radiomics models for risk stratification of the GB masses. Furthermore, we compared the diagnostic performance of these models with the conventional US and contrast-enhanced US (CEUS) models. Results: The optimal XGBoost-based US radiomics model for discriminating neoplastic from non-neoplastic GB lesions showed higher diagnostic performance in terms of areas under the curves (AUCs) than the conventional US model (0.822–0.853 vs. 0.642–0.706, p < 0.05) and potentially decreased unnecessary cholecystectomy rate in a speculative comparison with performing cholecystectomy for lesions sized over 10 mm (2.7–13.8% vs. 53.6–64.9%, p < 0.05) in the validation and test sets. The AUCs of the XGBoost-based US radiomics model for discriminating carcinomas from benign GB lesions were higher than the conventional US model (0.904–0.979 vs. 0.706–0.766, p < 0.05). The XGBoost-US radiomics model performed better than the CEUS model in discriminating GB carcinomas (AUC: 0.995 vs. 0.902, p = 0.011). Conclusions: The proposed ML-based US radiomics models possess the potential capacity for risk stratification of GB masses and may reduce the unnecessary cholecystectomy rate and use of CEUS. Clinical relevance statement: The machine learning-based ultrasound radiomics models have potential for risk stratification of gallbladder masses and may potentially reduce unnecessary cholecystectomies. Key Points: • The XGBoost-based US radiomics models are useful for the risk stratification of GB masses. • The XGBoost-based US radiomics model is superior to the conventional US model for discriminating neoplastic from non-neoplastic GB lesions and may potentially decrease unnecessary cholecystectomy rate for lesions sized over 10 mm in comparison with the current consensus guideline. • The XGBoost-based US radiomics model could overmatch CEUS model in discriminating GB carcinomas from benign GB lesions. [ABSTRACT FROM AUTHOR]

We propose an explainable machine learning (ML) model with uncertainty quantification (UQ) to improve multi-step reservoir inflow forecasting. Traditional ML methods have challenges in forecasting inflows multiple days ahead, and lack explainability and UQ. To address these limitations, we introduce an encoder–decoder long short-term memory (ED-LSTM) network for multi-step forecasting, employ the SHapley Additive exPlanation (SHAP) technique for understanding the influence of hydrometeorological factors on inflow prediction, and develop a novel UQ method for prediction trustworthiness. We apply these methods to forecast 7-day inflow in snow-dominant and rain-driven reservoirs. The results demonstrate the effectiveness of the ED-LSTM model, with high forecasting accuracy for short lead times. Our UQ method provides reliable uncertainty estimates, covering 90% of data with a 90% confidence level. The SHAP analysis reveals the importance of historical inflow and precipitation as influential factors. These findings and methods may support reservoir operators in optimizing water resources management decisions. • Propose an ED-LSTM network to improve multi-step reservoir inflow forecasting. • Develop an uncertainty quantification method to quantify the predictive uncertainty. • Apply SHAP to quantify the effect of hydrometeorological drivers on inflow prediction. • The proposed framework can be applied to hydrologically diverse reservoirs. [ABSTRACT FROM AUTHOR]

The parameterization of key photosynthesis parameters is one of the key uncertain sources in modeling ecosystem gross primary productivity (GPP). Solar‐induced chlorophyll fluorescence (SIF) offers a good proxy for GPP since it marks the actual process of photosynthesis; while machine learning (ML) provides a robust approach to model the GPP‐SIF relationship. Here, we trained the boosted regressing tree (BRT) and the Random Forest ML models with Greenhouse Gases Observing Satellite SIF data and in situ GPP observations from 49 eddy covariance towers. These trained ML GPP‐SIF models were fed into the Energy Exascale Earth System Model (E3SM) Land Model (ELM) to generate ELM‐simulated global SIF estimates, which were then benchmarked against satellite SIF observations with a surrogate modeling approach. Our results indicated good modeling performance of the ML‐based GPP‐SIF relationship. The ELM model when fed with the ML GPP‐SIF models also can well predict the spatial‐temporal variations in SIF. We also found high model accuracy for the surrogate modeling. Model parameter sensitivity analysis suggested that the fraction of leaf nitrogen in RuBisCO (flnr) is the most sensitive parameter to the SIF; other sensitive parameters include the Ball‐Berry stomatal conductance slope (mbbopt) and the vcmax entropy (vcmaxse). The posterior uncertainty in simulated GPP was greatly reduced after benchmarking, and the model produced improved spatial patterns of mean GPP relative to FLUXCOM GPP. Our integrated approach provides a new avenue for improving land models and using remote‐sensing SIF, which can be further improved in the future with more ground‐ and satellite‐based observations. Plain Language Summary: Model estimation of photosynthesis product, that is, gross primary productivity (GPP), is a challenging but vital task. One of the keys is to find better values for key parameters. This parameter searching process requires good proxies for GPP that can be widely available across space and time, good statistical methods to relate proxies to GPP and to make best estimations that reduce the gaps between modeled results and observations. Here, we designed a new method that use solar‐induced chlorophyll fluorescence (SIF, a good proxy for photosynthesis) as a key input, and employ machine learning (a robust way to relate SIF and GPP) and surrogate modeling (a good method for finding the best parameters), to improve the photosynthesis parameterization in the Energy Exascale Earth System Model (E3SM) Land Model (ELM), a state‐of‐the‐art terrestrial biosphere model. Our results demonstrate that this new integrated approach has great potential for improving the parameterization of key photosynthesis parameters in land models. Key Points: We built a unique method to improve gross primary productivity (GPP) modeling in land modelsThis method integrates solar‐induced chlorophyll fluorescence observations, machine learning, and surrogate modelingThe method reduced posterior uncertainties in simulated GPP and improved the modeling of its spatial patterns [ABSTRACT FROM AUTHOR]

The application of logistic regression (LR) and Cox Proportional Hazard (CoxPH) models are well‐established for evaluating exposure–response (E–R) relationship in large molecule oncology drugs. However, applying machine learning (ML) models on evaluating E–R relationships has not been widely explored. We developed a workflow to train regularized LR/CoxPH and tree‐based XGboost (XGB) models, and derive the odds ratios for best overall response and hazard ratios for overall survival, across exposure quantiles to evaluate the E–R relationship using clinical trial datasets. The E–R conclusions between LR/CoxPH and XGB models are overall consistent, and largely aligned with historical pharmacometric analyses findings. Overall, applying this interpretable ML workflow provides a promising alternative method to assess E–R relationships for impacting key dosing decisions in drug development. [ABSTRACT FROM AUTHOR]