Error RMSE Research Articles

Enhancement Methods of Hydropower Unit Monitoring Data Quality Based on the Hierarchical Density-Based Spatial Clustering of Applications with a Noise-Wasserstein Slim Generative Adversarial Imputation Network with a Gradient Penalty.

In order to solve low-quality problems such as data anomalies and missing data in the condition monitoring data of hydropower units, this paper proposes a monitoring data quality enhancement method based on HDBSCAN-WSGAIN-GP, which improves the quality and usability of the condition monitoring data of hydropower units by combining the advantages of density clustering and a generative adversarial network. First, the monitoring data are grouped according to the density level by the HDBSCAN clustering method in combination with the working conditions, and the anomalies in this dataset are detected, recognized adaptively and cleaned. Further combining the superiority of the WSGAIN-GP model in data filling, the missing values in the cleaned data are automatically generated by the unsupervised learning of the features and the distribution of real monitoring data. The validation analysis is carried out by the online monitoring dataset of the actual operating units, and the comparison experiments show that the clustering contour coefficient (SCI) of the HDBSCAN-based anomaly detection model reaches 0.4935, which is higher than that of the other comparative models, indicating that the proposed model has superiority in distinguishing between the valid samples and anomalous samples. The probability density distribution of the data filling model based on WSGAIN-GP is similar to that of the measured data, and the KL dispersion, JS dispersion and Hellinger's distance of the distribution between the filled data and the original data are close to 0. Compared with the filling methods such as SGAIN, GAIN, KNN, etc., the effect of data filling with different missing rates is verified, and the RMSE error of data filling with WSGAIN-GP is lower than that of other comparative models. The WSGAIN-GP method has the lowest RMSE error under different missing rates, which proves that the proposed filling model has good accuracy and generalization, and the research results in this paper provide a high-quality data basis for the subsequent trend prediction and state warning.

Mapping Tree Mortality Caused by Siberian Silkmoth Outbreak Using Sentinel-2 Remote Sensing Data

The Siberian silkmoth is one of the most dangerous coniferous forests pests. Siberian silkmoth outbreaks cause massive defoliation and subsequent forest fires over vast areas. Remote forest disturbance assessments performed after an outbreak make it possible to assess carbon emissions and the potential for natural regeneration, estimate forest fire danger, and reveal the need to implement forest management practices. The goal of the present research was to investigate the use of modern satellite imagery of medium spatial resolution to estimate the percentage of dead trees in a given area. The subject of this study is the Siberian silkmoth outbreak that occurred in 2018–2020 and covered 42 thousand ha in the Irbey region of the Krasnoyarsk Krai. Imagery from the Sentinel-2/MSI sensor was used to calculate a number of spectral indices for images received before and after the outbreak. Field study data were used to create regression models relating the index values to the percentage of dead trees. A number of spectral indices, such as NDVI, dNDVI, NBR, dNBR, NDMI, EVI, and TCG, were used. As a result, spectral indices based on the data from NIR/SWIR bands (NBR, NDMI, dNBR) demonstrated the best correlations with field-measured tree mortality. Therefore, these indices may be used to accurately estimate the percentage of dead trees by remote sensing data. The best was the NBR index with an R2 equal to 0.87, and the lowest RMSE and MAE errors. Consequently, Sentinel-2 imagery can be successfully used for tree mortality assessment over large inaccessible areas disturbed by Siberian silkmoth outbreaks at a relatively low cost.

Efficient health indicators for the prediction of the remaining useful life of proton exchange membrane fuel cells

Proton Exchange Membrane Fuel Cells (PEMFCs) are devices that convert hydrogen into energy through a chemical process called electrolysis. They hold promise in reducing the harmful environmental effects of fossil fuels and could potentially replace current energy systems in the near future if certain challenges are overcome. One major challenge in Proton Exchange Membrane Fuel Cell (PEMFC) performance is its durability. This issue, although common in many industrial assets, holds even more significance in PEMFCs due to their high costs and the need for reliable operation. Predictive Maintenance (PM), as a part of Prognostic and Health Management (PHM), focuses on addressing the durability problem by predicting the Remaining Useful Life (RUL) of PEMFCs. In this study, machine learning was used to predict the RUL of a stack of five PEMFCs operating under both static and quasi-dynamic conditions, resulting in two distinct datasets. Multiple health indicators were tested by combining domain expertise and the characteristics of PEMFC data to estimate the RUL.The results reveal that the power-time fraction health indicator accurately predicted the RUL in the static regime with an impressive R2 value of 99% and an RMSE error of 4.59e-07, using the random forest regressor. This analysis was also extended to the quasidynamic regime for validation, achieving an R2 of 95% and an RMSE of 0.06, which demonstrates the strong potential of this health indicator. Additionally, Explainable AI was employed for the first time in this context to enhance the transparency of the results. This work serves as an excellent starting point for further exploration of new health indicators and highlights the power of machine learning in addressing the crucial issue of predicting the RUL of PEMFCs.

Supplementary cementitious materials in blended cement concrete: Advancements in predicting compressive strength through machine learning

The increasing utilization of Portland cement raises environmental concerns. Thus, leading to the exploration of supplementary cementitious materials (SCMs) as alternatives to use in the construction industry. Incorporating SCMs into blended cement concrete (BCC) has shown improved mechanical and durability properties. However, predicting the compressive strength (CS) of BCC is challenging due to its complex composite nature and nonlinear behavior. This study comprehensively evaluated eleven commonly used machine-learning techniques (MLTs) for compressive strength prediction of BCC. Gene expression programming (GEP), artificial neural network (ANN), random forest (RF), support vector (SV), adaptive neuro-fuzzy inference system (ANFIS), optimized gaussian process regression (OGPR), gradient boost (GB), extreme gradient boost (XGB), ada boost (AB), k-nearest neighbor (KNN), and bagging regressor (BR) are used for prediction of compressive strength. The performance of ML models is assessed based on regression analysis and statistical metrics. The XGB model achieved the highest accuracy with an R2 value of 0.89. The R2 values of GB, SV, and OGPR models are 0.89, 0.83, and 0.8 respectively. Moreover, all ML models achieve MAE and RMSE error values below 10 MPa, except for the random forest model. In addition, this study also employs relative importance and Shapley analysis for model explainability. Thus, providing insights into the impact of SCMs on compressive strength of BCC. Overall, this research offers an improved approach to predicting the compressive strength of blended cement concrete.

Estimating multi-scale irrigation amounts using multi-resolution soil moisture data: A data-driven approach using PrISM

Irrigated agriculture is the primary driver of freshwater use and is continuously expanding. Precise knowledge of irrigation amounts is critical for optimizing water management, especially in semi-arid regions where water is a limited resource. This study proposed to adapt the PrISM (Precipitation inferred from Soil Moisture) methodology to detect and estimate irrigation events from soil moisture remotely sensed data. PrISM was originally conceived to correct precipitation products, assimilating Soil Moisture (SM) observations into an antecedent precipitation index (API) formula, using a particle filter scheme. This novel application of PrISM uses initial precipitation and SM observations to detect instances of water excess in the soil (not caused by precipitation) and estimates the amount of irrigation, along with its uncertainty. This newly proposed approach does not require extensive calibration and is adaptable to different spatial and temporal scales. The objective of this study was to analyze the performance of PrISM for irrigation amount estimation and compare it with current state-of-the-art approaches. To develop and test this methodology, a synthetic study was conducted using SM observations with various noise levels to simulate uncertainties and different spatial and temporal resolutions. The results indicated that a high temporal resolution (less than 3 days) is crucial to avoid underestimating irrigation amounts due to missing events. However, including a constraint on the frequency of irrigation events, deduced from the system of irrigation used at the field level, could overcome the limitation of low temporal resolution and significantly reduce underestimation of irrigation amounts. Subsequently, the developed methodology was applied to actual satellite SM products at different spatial scales (1 km and 100 m) over the same area. Validation was performed using in situ data at the district level of Algerri-Balaguer in Catalunya, Spain, where in situ irrigation amounts were available for various years. The validation resulted in a total Pearson’s correlation coefficient (r) of 0.80 and a total root mean square error (rmse) of 7.19 mm∕week for the years from 2017 to 2021. Additional validation was conducted at the field level in the Segarra-Garrigues irrigation district using in situ data from a field where SM profiles and irrigation amounts were continuously monitored. This validation yielded a total bi-weekly r of 0.81 and a total rmse of −9.34 mm∕14-days for the years from 2017 to 2021. Overall, the results suggested that PrISM can effectively estimate irrigation from SM remote sensing data, and the methodology has the potential to be applied on a large scale without requiring extensive calibration or site-specific knowledge.

Inclination angles for Be stars determined using machine learning

ABSTRACT We test the viability of training machine learning algorithms with synthetic $\rm H\, \alpha$ line profiles to determine the inclination angles (the angle between the central B star’s rotation axis and the observer’s line of sight) of B emission (Be) stars from a single observed medium-resolution, moderate signal-to-noise ratio spectrum. The performances of three different machine learning algorithms were compared: neural networks tasked with regression, neural networks tasked with classification, and support vector regression. Of these three algorithms, neural networks tasked with regression consistently outperformed the other methods with a root mean squared error of 7.6° on an observational sample of 92 galactic Be stars with inclination angles known from direct $\rm H\, \alpha$ profile fitting, from the spectroscopic signature of gravitational darkening, and, in a few cases, from interferometric observations that resolved the disc. The trained neural networks enable a quick and useful determination of the inclination angles of observed Be stars, which can be used to search for correlated spin axes in young open clusters or to extract an equatorial rotation velocity from a measurement of vsin i.

PSXIV-16 Evaluation of Water Intake Equations with an Emphasis on Physiological Status and Evaluation of Current Relevance Behind Environmental Parameters

Abstract Water requirements for beef cattle are often computed according to body weight (BW) and environmental variables. Inherent variation in environmental conditions within geographical locations, in addition to a cascade of physiological changes triggered as consequence of such variability are somewhat fomented by multicollinearity amongst these variables. This conundrum generates instability of current predictive models in reaching optimization equilibrium inflating its overall predictability and robustness. The goals herein involved the identification of physiologically relevant parameters and the generation of robust, data-driven, individualistic models through high frequency data loggers. Further, we evaluate re-parametrized and current proposed equations (e.g.: Hicks et al., 1988 and Arias and Mader, 2011) for the prediction of water intake for beef cattle at different life stages and physiological conditions. The dataset utilized consisted of more than 30,000 observations, including longitudinal data on lactating animals (n = 23), grass finished steers (n = 12), grain finished steers (n = 21), and grain finished implanted steers (n = 9), steers backgrounded on a low (n = 20) or moderate (n = 20) plane of nutrition bulls under negative (n = 13), maintenance (n = 13) and positive energy balance (n = 13). Individual monitoring systems allowed for the daily collection of BW, water and feed intake, and ingestive behavior. Additionally, an on-site data-logging weather station (1-min intervals) collected relative humidity, temperature, wind speed, wind direction, solar radiation, and precipitation. All reported statistics were generated on out-of-bag randomly split portions of the dataset. For identification of relevant predictive parameters of water intake, several feature selection and penalization techniques were examined. When comparing our equations to current, the errors generated for water intake predictions using physiological classification and on-site environmental variables, explained 50% more of the variation in predicting water intake (r2 = 0.38 vs r2 = 0.22-0.27). When evaluating the errors of prediction (root mean squared error, RMSE and mean absolute error, MAE), significant decreases of up to three-fold were observed on the equations generated herein (RMSE = 9.05 vs 28.8 to 35.31, MAE = 7.04 vs 26.9 to 33.9). Accurate monitoring of individual parameters such as feed intake were better predictors of water intake than environmental parameters when evaluating re-parametrization of models. Further, inclusion of a physiological stage (categorical variable) was more significant than any of the other environmental predictors. Overall, the results discussed highlight the importance of appropriately accounting for the water requirements of beef cattle and how far-off current assumptions are. Further, it raises concern on current equations that use environmental monitoring equations and techniques that attempt to describe nutrient requirements of the animals, and potentially brings the approach to requirements rather than environment.

Using a Decision Tree Algorithm Predictive Model for Sperm Count Assessment and Risk Factors in Health Screening Population.

Approximately 20% of couples face infertility challenges and struggle to conceive naturally. Despite advances in artificial reproduction, its success hinges on sperm quality. Our previous study used five machine learning (ML) algorithms, random forest, stochastic gradient boosting, least absolute shrinkage and selection operator regression, ridge regression, and extreme gradient boosting, to model health data from 1375 Taiwanese males and identified ten risk factors affecting sperm count. We employed the CART algorithm to generate decision trees using identified risk factors to predict healthy sperm counts. Four error metrics, SMAPE, RAE, RRSE, and RMSE, were used to evaluate the decision trees. We identified the top five decision trees based on their low errors and discussed in detail the tree with the least error. The decision tree featuring the least error, comprising BMI, UA, ST, T-Cho/HDL-C ratio, and BUN, corroborated the negative impacts of metabolic syndrome, particularly high BMI, on sperm count, while emphasizing the link between good sleep and male fertility. Our study also sheds light on the potentially significant influence of high BUN on spermatogenesis. Two novel risk factors, T-Cho/HDL-C and UA, warrant further investigation. The ML algorithm established a predictive model for healthcare personnel to assess low sperm counts. Refinement of the model using additional data is crucial for improved precision. The risk factors identified offer avenues for future investigations.

Absorber Layer Thickness as a New Feature in Statistical Learning Tools of Perovskite Solar Cells

In the last decade, the development of perovskite-based solar cells has emerged as a technological alternative for the photovoltaic generation with a higher efficiency/cost ratio. Many contributions have been made in recent years, as evidenced by many academic publications with worldwide experimental results in this area. Machine learning as a tool can support the development of this technology by predicting new materials, novel solar cell configurations, and evaluating the most relevant experimental parameters, among others. For this, the automatic learning models used in predicting or classifying the information available in the literature or generated experimentally must be improved. One way to improve these models is by including new descriptors that allow improving the prediction. In this work, we evaluated the use of the absorber layer thickness as a descriptor in a linear regression model using a database of 221 literature records containing information on the bandgap, the ?HOMO (perovskite-HTL), and ?LUMO (perovskite-ETL) of different perovskite cells, together with the thickness of the absorber layer. By building two multiple linear regression models, including or not the thickness of the absorber layer, a reduction in the root mean square error RMSE of 4.4% and 2.8% was found in the prediction of the Jsc and PCE, respectively. By applying a linear regression model, an improvement in the prediction of Jsc can be seen due to the inclusion of thickness as a descriptor, which is in line with the relatively high value of the mutual information measure between thickness and Jsc.

PM2.5 Concentration Prediction Based on LightGBM Optimized by Adaptive Multi-Strategy Enhanced Sparrow Search Algorithm

The atmospheric environment is of great importance to human health. However, its influencing factors are complex and variable. An efficient technique is required to more precisely estimate PM2.5 concentration values. In this paper, an enhanced Sparrow Search Algorithm (LASSA)-optimized Light Gradient Boosting Machine (LightGBM) is proposed for PM2.5 concentration prediction. This approach can provide accurate predictions while also reducing potential losses resulting from unexpected events. LightGBM is regarded as an outstanding machine learning approach; however, it includes hyperparameters that must be optimally mixed in order to achieve the desired results. We update the Sparrow Search Algorithm (SSA) and utilize it to identify the optimal combination of the most crucial parameters, using cross-validation to increase the reliability. Using limited air quality data and meteorological data as inputs, PM2.5 concentration values were predicted. The LASSA-LGB’s output was compared to normal LGB, SSA-LGB and ISSA-LGB. The findings demonstrate that LASSA-LGB outperforms the other models in terms of prediction accuracy. The RMSE and MAPE error indices were lowered from 3% to 16%. The concordance correlation coefficient is not less than 0.91, and the R2 reached 0.96. This indicates that the proposed model has potential advantages in the field of PM2.5 concentration prediction.

Performance of equations for the longitudinal dispersion coefficient: a case study in the Orashi River

Abstract Investigation of the water quality of rivers is a key point in Water Resources Engineering. The longitudinal dispersion coefficient is one of the foremost significant parameters in river water quality monitoring. Several parameters such as hydraulic, morphology, total dissolved solids, and total suspended solids are effective parameters in the determination of the longitudinal dispersion coefficient as revealed by this study. The assessment of the river shows mean hydraulic and geometric properties such as flow, depth, velocity, longitudinal slope, and width to be 354.17 m3/s, 9.61 m, 0.69 m/s, 0.0079,101.63 m and the range of the longitudinal dispersion coefficient as (72–104.4) m2/s. Results obtained by employing the established equations revealed standard error indices and RMSE of the developed equation, and Kashefipour and Falconer equation gives correlation coefficient of about 0.819 and 4.182 and 0.421 and 12.186, respectively, as coefficient of determination and RMSE, and they are more accurate among the empirical equations. However, the newly derived equation for the longitudinal dispersion coefficient performed better when compared with others, indicating the fitness of the developed equation to estimate longitudinal dispersion coefficient.

Impurity-Based Important Features for feature selection in Recursive Feature Elimination for Stock Price Forecasting

Stock investors perform stock price forecasting based on technical indicators and historical stock prices. The large number of technical indicators and historical data often leads to overfitting and ambiguity in forecasting using machine learning. In this paper, we proposed a feature selection approach using impurity-based important features in recursive feature elimination for stock price forecasting. The data utilized includes historical data and various moving averages. Feature selection is employed to reduce the number of features and obtain important and relevant features. The recursive feature elimination with impurity-based important features is utilized as the feature selection method. The machine learning methods employed are linear regression, support vector regression, multi-layer perceptron regression, and random forest regression. The measurement results of mean squared error (mse), root mean squared error (rmse), mean absolute error (mae), and mean absolute percentage error (mape) show that the optimal feature selection and machine learning method is achieved with six features and linear regression. The average mse, rmse, mae, and mape values are 0.000279, 0.016577, 0.012843, and 1.42236%, respectively. These results validate the effectiveness of impurity-based important features for feature selection in recursive feature elimination using historical data and various moving averages in stock price forecasting.

Prediction for the Sluice Deformation Based on SOA-LSTM-Weighted Markov Model

Increasingly, deformation prediction has become an essential research topic in sluice safety control, which requires significant attention. However, there is still a lack of practical and efficient prediction modeling for sluice deformation. In order to address the limitations in mining the deep features of long-time data series of the traditional statistical model, in this paper, an improved long short-term memory (LSTM) model and weighted Markov model are introduced to predict sluice deformation. In the method, the seagull optimization algorithm (SOA) is utilized to optimize the hyper-parameters of the neural network structure in LSTM primarily to improve the model. Subsequently, the relevant error sequences of the fitting results of SOA-LSTM model are classified and the Markovity of the state sequence is examined. Then, the autocorrelation coefficients and weights of each order are calculated and the weighted and maximum probability values are applied to predict the future random state of the sluice deformation. Afterwards, the prediction model of sluice deformation on the SOA-LSTM-weighted Markov model is proposed. Ultimately, the presented model is used to predict the settlement characteristics of an actual sluice project in China. The analysis results demonstrate that the proposed model possesses the highest values of R2 and the smallest values of RMSE and absolute relative errors for the monitoring data of four monitoring points. Consequently, it concluded that the proposed method shows better prediction ability and accuracy than the SOA-LSTM model and the stepwise regression model.

Prediction of Collapsing Strength of High-Strength Collapse-Resistant Casing Based on Machine Learning

With the increasing complexity of shale gas extraction conditions, a large number of high-strength collapse-resistant casing is applied to the extraction of unconventional oil and gas resources. There are errors in the traditional API collapse strength formula. A high-precision and low-computational-cost model is needed for predicting the strength of high-collapsible casing. The key influencing factors of casing anti-collapse strength were determined as outer diameter, wall thickness, yield strength, ovality, wall thickness unevenness, and residual stress by analyzing the casing collapse mechanism. In response to the key factors mentioned above, a dataset was formed by measuring the geometric parameters of the full-size casing and collecting data on the results of the anti-collapse strength experiment, which was divided into a training set (70%) and a testing set (30%). Three machine-learning algorithms, a neural network, random forest, and support vector machine, were trained to predict the anti-extrusion strength. The correlation coefficient R2, root mean square error RMSE, and average relative MRE were used to evaluate the indexes for model preference evaluation. The results show that machine-learning algorithms have unique advantages in casing anti-collapsing strength prediction. Within which, the neural network prediction model has the best prediction effect, and its characteristics of high precision, low cost and high efficiency are more suitable for the prediction of casing extrusion strength. Its testing set R2 is 0.9733, RMSE is 0.0267 and MRE is 0.0782, and the prediction accuracy can reach 92.2% which is much higher than the API calculation result (63.3%). The network prediction model is suitable for casing anti-collapsing strength prediction and meets the actual prediction requirements.

EVALUATING A NADIR AND AN OBLIQUE CAMERA FOR 3D INFRASTRUCTURE (CITY) MODEL GENERATION

Abstract. The analysis of Earth’s surface is strongly associated with the creation of three dimensional representations. In light of this, researchers involved in any realm of research as, geological, hydrological, ecological planning, city modelling, civil infrastructure monitoring, disaster management and emergency response, require 3D information of high fidelity and accuracy. For many decades, aerial photos or satellite data and photogrammetry provided the necessary information. In recent years, high-resolution imagery acquired by Unmanned Aerial Vehicles (UAV) has become a cost-efficient and quite accurate solution. In this framework, an infrastructure-monitoring project, named called “PROION”, focuses among others on the generation of very fine and highly accurate 3D infrastructure (city) model. The specific study evaluates a high-resolution nadir camera and an oblique camera for the creation of a 3D representation of the Patras University Campus. During the project, two identical flights over a part of the campus were conducted. The flights were performed with a vertical take-off and landing (Vtol) fixed wind UAV equipped with PPK receiver on-board. Based on the conducted flights, many data sets have been evaluated regarding the accuracy and fidelity. It was proved that both nadir and oblique cameras produced very accurate 3D representations of the University campus buildings. The RMSE error of the nadir imagery is almost two times higher than the respective error of the oblique imagery reaching 30cm.

Evaluation of slake durability for carbonaceous rocks exposed to water soaking using electrical resistivity

The present study investigated the electrical conduction and slake durability behaviors of carbonaceous rocks subjected to water soaking. Several series of electrical resistivity and slake durability tests were conducted on carbonaceous rocks with different mineralogical characteristics to explore the intrinsic connection between these two property indexes. Based on the experimental data, a new empirical model was developed to estimate slake durability index of carbonaceous rocks at varied soaking times. The results showed that the clay minerals content of investigated carbonaceous rocks ranged from approximately 20 % to 50 %. The clay minerals are beneficial in enhancing the electrical conduction of carbonaceous rocks, but impose a negative impact on the slake durability, attributing to their superior performance of hydrophilicity and swelling. Electrical resistivity decreased sharply with increasing soaking time within the first 8 days and then it achieved a stable state as soaking time further increased, which was mainly due to the completion of pore saturation and salts/minerals dissolution. The slake durability index exhibited similar evolution characteristics to those of electrical resistivity, which deteriorated sharply to a low or very low level after soaking for 16 days. The clay minerals experienced water absorption and mineral expansion during the first 8 days soaking, leading to the development of micro-cracks that facilitates water intrusion and accelerates the slake durability deterioration. These two processes were essentially complete as the soaking time further increased, and the clay minerals played an insignificant role at this stage. Liner increasing trends were observed in the plot of electrical resistivity decrease versus slake durability index decrease, and the slope values presented a power function relation with the clay minerals content of carbonaceous rocks. The proposed empirical model well captured the trends in the experimental data, with an absolute difference in percentage δ below30 % and a root mean square error RMSE value of 5.613 %. The outcomes of this study provided a viable answer to assess slake durability of carbonaceous rocks exposed to adverse external environments using non-destructive testing techniques.

Effective air pollution prediction by combining time series decomposition with stacking and bagging ensembles of evolving spiking neural networks

In this article, we introduce a new approach to air pollution prediction using the CEEMDAN time series decomposition method combined with the two-layered ensemble of predictors created based on the stacking and bagging techniques. The proposed ensemble approach is outperforming other selected state-of-the-art models when the bagging ensemble consisting of evolving Spiking Neural Networks (eSNNs) is used in the second layer of the stacking ensemble. In our experiments, we used the PM10 air pollution and weather dataset for Warsaw. As the results of the experiments show, the proposed ensemble can achieve the following error and agreement values over the tested dataset: error RMSE 6.91, MAE 5.14 and MAPE 21%; agreement IA 0.94. In addition, this article provides the computational and space complexity analysis of eSNNs predictors and offers a new encoding method for spiking neural networks that can be effectively applied for values of skewed distributions.

Pharmaceutical nanonization by green supercritical processing: Investigation of Exemestane anti-estrogenic medicine solubility using machine learning

Oral administration is known as the most prevalently-employed approach of drug delivery because of possessing different benefits like simplicity of use, affordability and negligible sterility restrictions. Despite the noteworthy interests of scientists to manufacture oral therapeutic medicines, poor solubility/bioavailability of orally-administered drugs remains a big challenge in recent years. Therefore, various scientists are trying to develop novel techniques to enhance the solubility of orally-administered therapeutic agents. One of the promising ways, which has recently attracted the global attentions is the use of supercritical fluids (SCFs). This technique could overcome the unfavorable/detrimental influences of organic solvents on the environment due to its environmentally-benign characteristics, non-inflammability and low toxicity. In this research, the solubility of Exemestane (EXE) drug is modeled based on two parameters of temperature and pressure. In order to handle the modeling process, the AdaBoost method with three core models of linear regression (LR), K-Nearest Neighbor (KNN) and Gaussian Process (GPR) are considered, which are tuned by AEO method in terms of their hyper-parameters. Based on the evaluations carried out in this research, ADA GPR, ADA KNN, and ADA LR models showed 0.996, 0.991, and 0.925 values in terms of R-square score. Also, in terms of the error rate, the three models ADA GPR, ADA KNN, and ADA LR have an RMSE error equal to 1.3459, 1.8967, and 5.8875, respectively. Based on these facts and some other analyses, the Gaussian process model optimized and boosted with AdaBoost can be accepted as the most accurate model of this research.

A Simplified Pointing Model for Alt-Az Telescopes

The telescope pointing model plays a crucial role in characterizing and explaining the inaccuracies in the telescope’s pointing. It uses mathematical models to express these errors and suggests methods to calibrate and rectify them. By employing the pointing model, one can gain a deeper understanding of the discrepancies in the telescope’s pointing accuracy at various positions and orientations. This article introduces the establishment and validation of a simplified pointing model for correcting the pointing errors of an alt-az telescope. The traditional telescope pointing model, with its complex formulas, makes it difficult to quickly apply for the calibration of the telescope’s pointing. This method utilizes two-dimensional surface fitting to obtain the telescope’s pointing error function, enabling a rapid determination of the telescope’s pointing model, which is then validated on the Planetary Atmospheric Spectroscopic Telescope (PAST). Using this method, the azimuth axis and altitude axis errors (root mean square error—RMSE) of the PAST telescope are 6.8 arcs and 3.8 arcs, respectively.

Use of random forest algorithm for predictive modelling of transfer factor soil-plant for radiocaesium: A feasibility study

A German dataset with soil-plant transfer factors for radiocaesium including many co-variables was analysed and prepared for the application of the Random Forest (RF) algorithm using the R libraries 'party', and 'caret'.A RF predictive model for soil-plant transfer factor was created based on 10 co-variables. These are, for example, taxonomic plant family, plant part, soil type and the exchangeable potassium concentration in the soil. The RF model results were compared with the results of two (semi-)mechanistic models. Of the more than 3000 entries in the original dataset, only about 1200 could be used, as this was the largest complete dataset with the largest number of co-variables available.The obtained RF predictive model can reproduce the experimental observations better than the two (semi)-mechanistic models, which are based on many assumptions and fixed parameter values. Model performance was quantified using the metrics of Root Mean Square Error (rmse) and Mean Absolute Error (mae). The RF model was able to reproduce the variability of the data by up to 6 orders of magnitude. The categorical co-predictors, especially taxonomic plant family and plant part, have a greater influence than the numerical co-predictors, such as pH and exchangeable soil potassium concentration.This feasibility study shows that RF is a promising tool to obtain predictive models for transfer factors. However, to build a widely applicable predictive model, a dataset is needed that contains at least thousands of entries for transfer factors and for the most important co-variables and considers a large parameter space.