Performance prediction of an internal-circulation membrane bioreactor based on models comparison and data features analysis

While the transportation sector is one of largest economic growth drivers for many countries, the adverse impacts of transportation on air quality are also well-noted, especially in developing countries. Carbon dioxide (CO2) emissions are one of the direct results of a transportation sector powered by burning fossil-based fuels. Detailed knowledge of CO2 emissions produced by the transportation sectors in various countries is essential for these countries to revise their future energy investments and policies. In this framework, three machine learning algorithms, ordinary least squares regression (OLS), support vector machine (SVM), and gradient boosting regression (GBR), are used to forecast transportation-based CO2 emissions. Both socioeconomic factors and transportation factors are also included as features in the study. We study the top 30 CO2 emissions-producing countries, including the Tier 1 group (the top five countries, accounting for 61% of global CO2 emissions production) and the Tier 2 group (the next 25 countries, accounting for 35% of total CO2 emissions production). We evaluate our model using four-fold cross-validation and report four frequently used statistical metrics (R2, MAE, rRMSE, and MAPE). Of the three machine learning algorithms, the GBR model with features combining socioeconomic and transportation factors (GBR_ALL) has the best performance, with an R2 value of 0.9943, rRMSE of 0.1165, and MAPE of 0.1408. We also find that both transportation features and socioeconomic features are important for transportation-based CO2 emission prediction. Transportation features are more important in modeling for 30 countries, while socioeconomic features (especially GDP and population) are more important when modeling for Tier 1 and Tier 2 countries.

The reconstruction of missing groundwater level data is of great importance in hydrogeological and environmental studies. This study provides a comprehensive and sequential approach for the reconstruction of groundwater level data near Lake Uluabat in Bursa, Turkey. This study addresses missing data reconstruction for both past and future events using the Gradient Boosting Regression (GBR) model. The reconstruction process is evaluated through model calibration metrics and changes in the statistical properties of the observed and reconstructed time series. To achieve this goal, the groundwater time series from two observational wells and lake water levels during the January 2004 to September 2019 period are used. The lake water level, the definition of the four seasons via the application of three dummy variables, and time are used as inputs in the prediction of groundwater levels in observation wells. The optimal GBR model calibration is achieved by training the dataset selected based on data gaps in the time series, while test-past and test-future datasets are used for model validation. Afterward, the GBR models are used in reconstructing the missing data both in the pre- and post-training data sets, and the performance of the models are evaluated via the Nash–Sutcliffe efficiency (NSE), Root Mean Square Percentage Error (RMSPE) and Performance Index (PI). The statistical properties of the time series including the probability distribution, maxima, minima, quartiles (Q1–Q3), standard error (SE), coefficient of variation (CV), entropy (H), and error propagation are also measured. It was concluded that GBR provides a good base for missing data reconstruction (the best performance was as high as NSE: 0.99, RMSPE: 0.36, and PI: 1.002). In particular, the standard error and the entropy of the system in one case, respectively, experienced a 53% and 35% rise, which was found to be tolerable and negligible.

Accurately estimating crop performance under various fertilizer levels in a rapid and non-destructive manner has become a vital aspect of precision agriculture technology for both economic and environmental benefits. This study aimed to estimate different sugar beet parameters, such as total chlorophyll (Chlt), chlorophyll a (Chla), chlorophyll b (Chlb), root yield (RY), sugar yield (SY), and sugar content (SC) under five nitrogen (N) levels (0, 30, 60, 90, and 120 kg N ha−1). This was achieved by using a combination of the gradient boosting regression (GBR) model with published and newly developed two- and three-band spectral indices (2D- and 3D-SRIs). The results showed that the N levels had the highest proportion of variations (80.4–92.9%) for all parameters, except for SC, which had more variation (59.9%) according to year than the N levels (37.2%). All parameters, except SC, showed a significant increase with gradually increasing N levels. Additionally, the N levels displayed linear and strong positive relationships with the chlorophyll parameters, and linear and strong negative relationships with SC, while these relationships were quadratic and strong with RY and SY. Several published and novel 3D-SRIs exhibited moderate to strong relationships (R2 = 0.65–0.89) with all parameters. The newly developed 3D-SRIs, which involve wavelengths from the visible, near-infrared, and red-edge regions, such as NDI536, 538, 534, NDI738, 750, 542, and NDI448, 734, 398, were effective in accurately estimating all parameters. Combining 2D-SRIs, 3D-SRIs, and the aggregate of all spectral indices (ASRIs) with GBR models could be a robust strategy for estimating the six observed parameters with reasonable precision. The GBR-ASF-6 SRIs and the GBR-ASF-7 SRIs models performed better in predicting Chl content and SC with R2 values of 0.99 and 0.99 (RMSE = 0.073 and 1.568) for the training dataset and R2 values of 0.65 and 0.78 (RMSE = 0.354 and 6.294) for the testing datasets, respectively. The obtained results concluded that published and newly developed 3D-SRIs, GBR based on 2D-SRIs or 3D-SRIs, and the aggregate of all ASRIs can be used in practice to accurately estimate the Chl content, production, and quality of sugar beet across a wide range of N levels under semiarid conditions.

Traditional design/optimization of metal-organic frameworks (MOFs) is time-consuming and labor-intensive. In this study, we utilize machine learning (ML) to accelerate the synthesis of MOFs. We have built a library of over 900 MOFs with different metal salts, solvent ratios, reaction durations and temperatures, and utilize zeta potentials as target variables for ML training. A total of four ML models have been used to train the collected dataset and assess their convergence performances, where Random Forest Regression (RFR) and Gradient Boosting Regression (GBR) models show strong correlation and accurate predictions. We then predicted two kinds of MOFs from RFR and GBR models. Remarkably, the experimentally data of the synthesized MOFs closely matched the predicted results, and these MOFs exhibited excellent electrocatalytic performances for oxygen evolution. This study would have general implications in the utilization of machine learning for accelerating the synthesis of MOFs for diverse applications.

Machine learning prediction of bio-polyol yields and hydroxyl values from acid-catalyzed liquefaction of lignocellulosic biomass

Understanding the solubility of drug particles in solvents has remained a big challenge in different fields. Development of advanced computational methods to predict the solubility of drugs is an important necessity due to the difficulty and time-consuming characteristics of experimental measurements. This study used hybrid machine learning (ML) models utilizing two inputs, including Pressure (X1) and Temperature (X2), to investigate the current data, and correlate the solubility of drug particles in supercritical solvent. The methods of Random Forest (RF), Extra Trees (ET), and Gradient Boosting (GB) regression models were used to build the models on the available data. RF, ET, and GB have R-squared of 0.857, 0.998, and 0.992, based on the analysis results. Additionally, in terms of MAE, they illustrated error value 2.90E-06, 1.98E-06, and 1.10E-06, respectively. One more metric to consider is MAPE, in which the error rates for the three regions were 3.15E-01, 2.27E-01, and 1.16E-01, respectively. The DT method was chosen as the best method, and can be used to find optimal amounts, which is summarized as a vector: (×1 = 383, X2 = 333.15, Y = 6.004e-05).

In order to reduce the doctor's subjective misjudgments, it is necessary to explore an effective predictive model to predict heparin dose during continuous renal replacement therapy surgery. In this paper, we use a combination of random forest and genetic algorithm to extract features and use EasyEnsemble algorithm to deal with unbalanced data. When training the model, this paper takes the ln transformation of the targeted variables, and then uses the Gradient Boosting Regression model and the Decision Tree Regression model to train. By comparing their mean absolute error, mean square error and square of R, finally, this paper chooses the Gradient Boosting Regression model as the final predictive model. The purpose of this study is to assist doctors in making more accurate judgments by this predictive model.

Enhancing biomass Pyrolysis: Predictive insights from process simulation integrated with interpretable Machine learning models

Deep learning hyperparameter optimization: Application to electricity and heat demand prediction for buildings

<abstract> <b><sc>Abstract.</sc></b> Anaerobically digested swine wastewater has high concentrations of total ammoniacal nitrogen (TAN) (approximately 1,000 mg N L^-1) and a low chemical oxygen demand (COD) to total nitrogen (TN) (COD/TN) ratio of <3.0, making it difficult to achieve a high TN removal rate using traditional biological nitrogen removal processes. The purpose of this study was to investigate the effectiveness of combining ammonia stripping (AS) and membrane bioreactor (MBR) processes to remove nitrogen from anaerobically digested swine wastewater. The coupling parameters that would lead to a high TN removal rate and cost effectiveness were also explored. Ammonia stripping was conducted at two pH levels of 10.0-10.5 (High-pH) and 9.0-9.5 (Low-pH). The MBR had the configuration of anoxic biofilm and aerobic MBR, and it was operated at two influent COD/TN ratios of 6.0-8.0 and 5.0-6.0 coupled with two TAN sludge loading rates in the oxic zone of 0.04-0.05 and 0.12-0.16 kg N kg^-1 mixed liquid suspended solid (MLSS) d^-1. Total nitrogen removal rate for the High-pH and Low-pH AS was 83% and 65%, respectively. In the MBR at a dissolved oxygen (DO) level of 6.12 ±0.69 mg L^-1, increasing the TAN sludge loading rate from 0.04 to 0.16 kg N kg^-1 MLSS d^-1 led to an increase in free ammonia from 0 to 4-9 mg L^-1 and an increase in nitrite accumulation rate (NAR) from 0% to 71%. At a TAN sludge loading rate in the oxic zone of 0.12 kg N kg^-1 MLSS d^-1 and NAR of 53%, the TN removal rate reached 82% in the MBR through shortcut nitrification and denitrification (SND). The influent biochemical oxygen demand (BOD₅) to TN (BOD₅/TN) ratio was 1.41, COD/TN was 5.41, and the oxidation-reduction potential (ORP) of the denitrification zone was as low as -300 to -400 mV. Among the experimental conditions examined, the combined Low-pH AS+SND process provided the highest TN removal rate of 94% with the least cost and HRT.

Dynamic properties, such as shear modulus and damping ratio, are critical for civil engineering applications and essential for accurate dynamic response analysis. This study introduces a novel Deep Neural Network (DNN) approach to predict the normalized shear modulus (G/Gmax ) and damping ratio (D) of granular soils over a wide strain range. Utilising a comprehensive dataset from cyclic triaxial (CT) and resonant column (RC) tests, we developed a Deep Feed-Forward Neural Network (DFFNN) model. The model incorporates grading characteristics, shear strain, void ratio, mean effective confining pressure, consolidation stress ratio, and specimen preparation method as inputs. The DFFNN demonstrated high accuracy with testing results of 0.9830 for G/Gmax and 0.9396 for D, outperforming traditional empirical models and other intelligent techniques such as Shallow Neural Network (SNN), Support Vector Regression (SVR), and Gradient Boosting Regression (GBR). This data-driven approach offers a robust and adaptable method for predicting the dynamic properties of granular soils across diverse conditions.

Exploring the relationship between 2D/3D landscape pattern and land surface temperature based on explainable eXtreme Gradient Boosting tree: A case study of Shanghai, China

The increasing penetration of renewable energy sources introduces variability and uncertainty into power systems, creating a growing need for demand-side flexibility (DSF). Forecasting the availability of demand-side flexibility, while maintaining consumer comfort, remains a central challenge in supporting aggregators with efficient energy management. This study presents an assessment of DSF using one year of household load consumption data, analysed through statistical methods and machine learning. Hourly appliancelevel consumption patterns were examined to identify shiftable loads and quantify the flexibility potential. An XGBoost (Extreme Gradient Boosting) regression model was developed to predict flexible demand, including Root Mean Square Error (RMSE), Determination Coefficient ($\mathbf{R}^{\mathbf{2}}$ Score) & total shiftable load. The demonstrated proposed model delivers high forecasting accuracy and outperforms several conventional prediction methods. A comparative analysis of pre-shifting and post-shifting load profiles demonstrated the potential to flatten evening peaks and redistribute demand to off-peak hours, improving system efficiency without reducing overall energy consumption. This work highlights the role of data-driven approaches in quantifying household flexibility and provides a foundation for demand-response strategies and smart energy management systems.

PurposeEuropean Parliament adopted a new directive on gender balance in corporate boards when by 2026, companies must employ 40% of the underrepresented sex into non-executive directors or 33% among all directors. Therefore, this study aims to analyze the impact of gender diversity (GD) on board of directors and the shareholders’ structure and their impact on the likelihood of company bankruptcy during the COVID-19 pandemic.Design/methodology/approachThe data sample consists of 1,351 companies for 2019 and 2020, of which 173 were large, 351 medium-sized companies and 827 small companies. Three bankruptcy indicators were tested for each company size, and extreme gradient boosting (XGBoost) and logistic regression models were developed. These models were then cross-validated by a 10-fold approach.FindingsXGBoost models achieved area under curve (AUC) over 98%, which is 25% higher than AUC achieved by logistic regression. Prediction models with GD features performed slightly better than those without them. Furthermore, this study indicates the existence of critical mass between 30% and 50%, which decreases the probability of bankruptcy for small and medium companies. Furthermore, the representation of women in ownership structures above 50% decreases bankruptcy likelihood.Originality/valueThis is a pioneering study to explore GD topics by application of ensembled machine learning methods. Moreover, the study does analyze not only the GD of boards but also shareholders. A highly innovative approach is GD analysis based on company size performed in one study considering the COVID-19 pandemic perspective.

Nickel-Titanium (NiTi) shape memory alloys (SMAs) are smart materials able to recover their original shape under thermal stimulus. Near-net-shape NiTi SMA foils of 2 meters in length and width of 30 mm have been successfully produced by a planar flow casting facility at CSIRO, opening possibilities of wider applications of SMA foils. The study also focuses on establishing a fully automated experimental system for the characterisation of their reversible actuation, significantly improving SMA foils adaptation into real applications. Artificial Intelligence involving Computer Vision and Machine Learning based methods were successfully employed in the development of the automation SMA characterization process. The study finds that an Extreme Gradient Boosting (XGBoost) Regression model based predictive system experimented with over 175,000 video samples could achieve 99% overall prediction accuracy. Generalisation capability of the proposed system makes a significant contribution towards the efficient optimisation of the material design to produce high quality 30 mm SMA foils.