Lowest Root Mean Square Error Research Articles

Multiple faults detection in doubly-fed induction generator wind turbine using artificial neural network

The development of fault detection methods in wind turbine (WT), especially for single fault detection, is continuously increasing. However, the rapid growth of fault detection in WT leads to another challenge where multiple faults can occur. The single fault detection method in WT is no longer reliable, especially when multiple faults occur simultaneously. Therefore, multiple faults detection in doubly-fed induction generators (DFIG) WT was proposed using an artificial neural networks (ANN) model. These multiple faults include internal and external stator faults happening simultaneously. Internal stator faults cover inter-turn short circuit faults and open circuit faults, while external stator faults cover loss of excitation and external short circuit faults. The performance of the developed multiple faults detection model was measured using accuracy and the root mean square error (RMSE) value. The results show that the developed model performs well with high accuracy and a low RMSE value. Thus, the developed model can accurately detect the coexistence of multiple faults in DFIG WT.

Measured and predicted resting metabolic rate of Dutch and Norwegian Paralympic athletes.

While the resting metabolic rate (RMR) is crucial for understanding athletes' energy requirements, limited information is available on the RMR of Paralympic athletes. To determine RMR and its predictors in a diverse cohort of Paralympic athletes and evaluate the agreement between measured and predicted RMR from both newly developed and preexisting equations. This cross-sectional study, conducted between September 2020 and September 2022 in the Netherlands and Norway, assessed RMR in Paralympic athletes by ventilated hood indirect calorimetry and body composition by dual energy X-ray absorptiometry. Sixty-seven Paralympic athletes (male: n=37; female: n=30) competing in various sports, with a spinal cord disorder (SCD; n=22), neurological condition (n=8), limb deficiency (n=18), visual or hearing impairment (n=7) or other disability (n=12) participated. RMR, fat-free mass (FFM), body mass, and triiodothyronine (T3) concentrations were assessed. Multiple regression analyses were conducted with height, FFM, body mass, sex, T3 concentration, and disabilities as potential predictors of RMR. Differences between measured and predicted RMRs were analyzed for individual accuracy, root mean square error (RMSE), and intraclass correlation (ICC). Mean RMR was 1386±258 kcal/day for females and 1686±302 kcal/day for males. Regression analysis identified FFM, T3 concentrations and the presence of a spinal cord disorder (SCD), as the main predictors of RMR (adjusted R2=0.71; F=50.3; P<0.001). The novel prediction equations based on these data, as well as pre-existing equations of Chun et al. and Nightingale and Gorgey performed well on accuracy (>60% of participants within 10% of measured RMR), had a good reliability (ICC >0.78), and low RMSE (≤141 kcal). FFM, total T3 concentrations, and the presence of SCD are the main predictors of RMR in Paralympic athletes. Both the current study's prediction equations and those by Chun et al. and Nightingale and Gorgey align well with measured RMR, offering accurate prediction equations for the RMR of Paralympic athletes.

Characterizing the Exponential Profile of W' Recovery Following Partial Depletion.

The aim of this study was to characterize W' recovery kinetics in response to a partial W' depletion. We hypothesized that W' recovery following partial depletion would be better described by a biexponential than by a monoexponential model. Nine healthy men performed a ramp incremental exercise test, three to five constant load trials to determine critical power and W', and ten experimental trials to quantify W' depletion. Each experimental trial consisted of two constant load work bouts (WB1 + WB2) interspersed by a recovery interval. WB1 was designed to evoke a 25% or 75% W' depletion (DEP25% + DEP75%). Subsequently, participants recovered for 30, 60, 120, 300 or 600 s, and then performed WB2 to exhaustion in order to calculate the observed W' recovery (W'OBS). W'OBS data were fitted using monoexponential and biexponential models, both with a variable and a fixed model amplitude. Root mean square error (RMSE) and Akaike information criterion (AICc) were calculated to evaluate the models' goodness-of-fit. The biexponential model fits were associated with overall lower RMSE values (0.4-5.0%) compared to the monoexponential models (2.9-8.0%). However, ΔAICc resulted in negative values (-15.5 and -23.3) for the model fits where the amplitude was free, thereby favoring the use of a monoexponential model for both depletion conditions. For the model fits where the amplitude was fixed at 100%, ΔAICc was negative for DEP25% (-15.0), but positive for DEP75% (11.2). W'OBS values were strongly correlated between both depletion conditions (r = 0.92), and positively associated with V̇O2peak, CP and GET (r = 0.67-0.77). The present study results did not provide evidence in favor of a biexponential modeling technique to characterize W' recovery following partial depletion. Moreover, we demonstrated that fixed t values were insufficient to model W' recovery across different depletion levels, and that W' recovery was positively associated with aerobic fitness. These findings underline the importance of employing variable and individualized t values in future predictive W' models.

Shear strength model of FRP-reinforced concrete beams without shear reinforcement

The current design codes for evaluating the shear strength of concrete beams reinforced with fiber-reinforced polymer (FRP) bars predominantly rely on a semi-empirical approach through test data calibration without a robust theoretical background. This approach may lead to either overestimation or underestimation, resulting in significant scattering in strength predictions. This study proposed a modified approach to enhance the existing unified shear-strength model developed for steel-reinforced concrete (RC) beams based on the compression zone failure mechanism, for predicting the shear strength of FRP-RC beams without shear reinforcement. The proposed model integrates strain and stress approaches, accounting for the distinctive characteristics associated with FRP reinforcement. Additionally, the overall shear resistance capacity of the FRP-RC beam is considered as the sum of contributions from the intact concrete of the compression zone and the aggregate interlock mechanism. The reliability of the proposed model is verified by comparing its predictions with a comprehensive database comprising 264 FRP-RC beam specimens without shear reinforcement reported by 41 research groups. Furthermore, the accuracy of the proposed model was assessed through a comparative analysis with existing evaluation methods, encompassing machine learning-based models, empirical models, theoretical models, and design codes. The results indicate that the proposed model satisfactorily predicts the shear strength of FRP-RC beams within a wide range of design parameters. In comparison with machine learning-based models, the proposed theoretical model exhibits a similar level of accuracy in terms of the mean shear strength ratio and coefficient of determination (R2) values but with less scattering, indicated by lower coefficient of variation (COV) and root mean square error (RMSE) values. In addition, the proposed model outperforms existing empirical and theoretical models, as well as current design codes, by significantly enhancing prediction accuracy, demonstrated by lower COV, RMSE, and higher R2 values. Finally, a parametric study was conducted to investigate the effects of the key parameters on the shear strength of FRP-RC beams using both the proposed model and representative existing evaluation methods.

A comparative survey between cascade correlation neural network (CCNN) and feedforward neural network (FFNN) machine learning models for forecasting suspended sediment concentration

Suspended sediment concentration prediction is critical for the design of reservoirs, dams, rivers ecosystems, various operations of aquatic resource structure, environmental safety, and water management. In this study, two different machine models, namely the cascade correlation neural network (CCNN) and feedforward neural network (FFNN) were applied to predict daily-suspended sediment concentration (SSC) at Simga and Jondhara stations in Sheonath basin, India. Daily-suspended sediment concentration and discharge data from 2010 to 2015 were collected and used to develop the model to predict suspended sediment concentration. The developed models were evaluated using statistical indices like Nash and Sutcliffe efficiency coefficient (NES), root mean square error (RMSE), Willmott’s index of agreement (WI), and Legates–McCabe’s index (LM), supplemented by a scatter plot, density plots, histograms and Taylor diagram for graphical representation. The developed model was evaluated and compared with CCNN and FFNN. Nine input combinations were explored using different lag-times for discharge (Qt-n) and suspended sediment concentration (St-n) as input variables, with the current suspended sediment concentration as the desired output, to develop CCNN and FFNN models. The CCNN4 model with 4 lagged inputs (St-1, St-2, St-3, St-4) outperformed the other developed models with the lowest RMSE = 95.02 mg/l and the highest NES = 0.0.662, WI = 0.890 and LM = 0.668 for the Jondhara Station while the same CCNN4 model secure as the best with the lowest RMSE = 53.71 mg/l and the highest NES = 0.785, WI = 0.936 and LM = 0.788 for the Simga Station. The result shows the CCNN model was better than the FFNN model for predicting daily-suspended sediment at both stations in the Sheonath basin, India. Overall, CCNN showed better forecasting potential for suspended sediment concentration compared to FFNN at both stations, demonstrating their applicability for hydrological forecasting with complex relationships.

Application of machine learning on hydrate formation prediction of pure components with water and inhibitors solution

AbstractThe present work investigates the use of machine learning approaches for the prediction of hydrate formation pressure (HFP) in gas hydrate systems. Advanced machine learning models, including the decision tree regressor (DTR), random forest regressor (RFR), extreme gradient boosting (XGB), gradient boosting regressor (GBR), histogram gradient boosting regressor (HGBR), and CatBoost regressor (CB), are trained and evaluated on a large dataset consists of 3137 experimental data points. The models are evaluated using R‐squared (R2), mean squared error (MSE), mean absolute error (MAE), and root mean squared error (RMSE). The study indicates that for the intent of HFP prediction, CatBoost outperformed all other machine learning models. It demonstrated high accuracy on the testing set with an R2 value of 0.9922, and with the lowest RMSE (1.61 × 10−3), MAE (7.90 × 10−4), and MSE (2.58 × 10−6), CatBoost strengthened its prediction ability.

The impact of base design and restoration type on the resin consumption, trueness, and dimensional stability of dental casts additively manufactured from liquid crystal display 3D printers.

To evaluate the effects of two base types and three restoration designs on the resin consumption and trueness of the 3D-printed dental casts. Additionally, the study explored the dimensional stability of these 3D-printed dental casts after 1 year of storage. Various types of reference dental casts were specifically designed to represent three types of dental restoration fabrications, including full-arch (FA), long-span (LS), and single-unit (SU) prostheses. The reference casts were digitized with a dental laboratory scanner and used to create flat and hollow base designs (N = 18) for the 3D-printed study casts. The 3D-printed study casts were digitized and evaluated against their corresponding references immediately after 3D printing and again after 1 year of storage, with the trueness quantified using the root mean square error (RMSE) at both time points. Volumes of resin used were recorded to measure resin consumption, and the weights of the 3D-printed study casts were also measured. The data were analyzed using two-way ANOVA and a Tukey post hoc test, α = 0.05. Volumetric analysis showed the flat-base design had significantly higher resin consumption with weights for the FA group at 42.51 ± 0.16g, the LS group at 31.64 ± 0.07g, and the SU group at 27.67 ± 0.31g, as opposed to 26.22 ± 1.01g, 22.86± 0.93g, and 20.10 ± 0.19g for the hollow designs respectively (p < 0.001). Trueness, assessed through two-way ANOVA, revealed that the flat-base design had lower RMSE values indicating better trueness in the LS (54 ± 6 µm) and SU (59 ± 7 µm) groups compared to the hollow-base design (LS: 73 ± 5, SU: 99 ± 11 µm, both p < 0.001), with no significant difference in the FA group (flat-base: 50 ± 3, hollow: 47 ± 5 µm, p= 0.398). After 1 year, the flat-base design demonstrated superior dimensional stability in the LS (flat base: 56 ± 6 µm, hollow base: 149 ±45 µm, p < 0.001) and SU groups (flat base: 95 ± 8 µm, hollow base: 183 ±27 µm, p < 0.001), with the FA group showing no significant difference in the base design (flat base: 47 ± 9, hollow base: 62 ± 12 µm, p = 0.428). The hollow-base design group showed lower resin consumption than the flat-base design group. However, the flat-base designs exhibited superior trueness and less distortion after 1 year of storage. These findings indicate that despite the higher material usage, flat-base designs provide better initial accuracy and maintain their dimensional stability over time for most groups.

Comparing the performance of pediatric weight estimation methods

BackgroundWeight estimation is essential in the care of ill children when a weight cannot be obtained. This is particularly important for children with medical complexity, who are at higher risk for adverse drug events. Our objective was to compare the accuracy of different methods of weight estimation in children and stratify by the presence of medical complexity. MethodsWe performed a retrospective cross-sectional study of children (<18 years) seen in the emergency department (ED) or ambulatory clinic from January 1, 2013 to December 31, 2022 at a tertiary academic pediatric center. We compared the performance of nine age-based formulae and two length-based methods using metrics of mean average error (MAE), root mean square error (RMSE), and agreement within 10% and 20% of measured weight. We additionally evaluated the approaches stratified by body mass index (BMI) and the presence of medical complexity. ResultsOf 361,755 children (median age 8.2 years, IQR 2.5–14.2 years; 51.5% male), 59,283 (16.4%) were seen in the ED. Length was measured or available in 21,330 (36.0%) patients in the ED and 293,410 (97%) patients in clinics. The Broselow tape outperformed all methods, with 50.7% estimates within 10% of measured weight, 80.0% estimates within 20% of measured weight, the lowest MAE (2.5 kg), and lowest RMSE (4.5 kg). The Antevy formula was the most accurate age-based formula, with 49.2% estimates within 10% of measured weight, 80.1% estimates within 20% of measured weight, MAE of 2.8 kg, and RMSE of 4.7 kg. Estimates became less accurate as BMI and estimated weight increased for all methods. Among children with medical complexity (14.1%), the Broselow tape consistently outperformed age-based formulae, with 47.7% estimates within 10% of measured weight, 77.1% estimates within 20% of measured weight, MAE of 2.6 kg, and RMSE of 5.4 kg. The Antevy formula remained the most accurate age-based method among children with medical complexity. ConclusionThe Broselow tape predicted weight most accurately in this large sample of children, including among those with medical complexity. The Antevy formula is the most accurate age-based method for pediatric weight estimation.

Texture prediction of natural soils in the Brazilian Amazon through proximal sensors

Proximal sensors provide fast, low-cost, environmentally friendly, and reliable analyses for the characterization of soils and other materials. Numerous studies have been conducted on soils in temperate regions, but there are knowledge gaps regarding the use of these devices in tropical soils, especially in the Amazon region. In this regard, this study utilized portable proximal sensors of X-ray fluorescence spectroscopy (pXRF) and diffuse reflectance spectroscopy in the visible to near-infrared region (Vis-NIR) for predicting the texture of natural soils in 61 municipalities in the state of Pará, Amazon region, Brazil. The objectives were: i) to investigate the accuracy of soil texture prediction based on data from sensors separately and sensor fusion (pXRF and Vis-NIR data) using two supervised algorithms (Random Forest, RF, and Support Vector Machine, SVM) and ii) to assess the effect of soil horizon (superficial and subsuperficial horizons, and their combination) in predicting the texture of tropical natural soils. In total, 233 soil samples were collected in the 0–20 cm and 80–100 cm depths, equivalent to superficial and subsuperficial horizons in areas with primary or secondary forest cover with at least 20 years of natural regeneration and approximately 20 ha of coverage area. The hydrometer method was used for soil texture analysis. In parallel, a portion of the soil samples was analyzed by pXRF and Vis-NIR, in triplicate, under laboratory conditions. The predictive models with RF were more robust compared to the models obtained with SVM, according to ratio performance interquartile distance (RPIQ), coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The R2 values obtained by pXRF, Vis-NIR, and sensor data combination were, respectively, 0.89, 0.87, and 0.93 for sand; 0.92, 0.90, and 0.93 for clay; and 0.91, 0.67, and 0.93 for silt. Overall, clay prediction models achieved higher R2 values compared to sand and silt models. Soil texture prediction using sensor fusion showed lower RMSE values and higher R2 and RPIQ values, respectively (sand: 7.79, 0.93, 4.69; clay: 5.58, 0.93, 3.86; and silt: 5.72, 0.92, 2.92) compared to the best-performing sensor individually (Vis-NIR). With regard to the optimal model utilizing individual sensor data, Vis-NIR models exhibited reduced error for clay and sand prediction. The effect of combining horizons to a single and bigger dataset was minimally important for the models. The results demonstrate confidence in the use of proximal sensors for soil texture assessment in natural Amazon soils, aiming to reduce costs and the time required for analyses.

Artificial intelligence applications for accurate geothermal temperature prediction in the lower Friulian Plain (north-eastern Italy)

Geothermal energy as a sustainable and clean energy source depends on the accurate estimation of reservoir temperatures. Understanding aquifer temperatures is crucial for optimizing low-enthalpy geothermal system exploitation. Advances in predictive algorithms can improve geothermal efficiency, while conventional methods of indirect temperature measurement and assumptions in geochemical analysis lead to uncertainties. As a solution, this study presents a comprehensive evaluation of six machine learning algorithms including eXtreme gradient boosting (XGBoost), decision tree, generalized regression neural network, extreme randomized trees, radial basis function, and elastic net. We employed essential performance metrics including coefficient of determination (R2) score, root mean square error (RMSE), mean absolute error, mean absolute percentage error (MAPE), and variance accounted for (VAF) to elucidate their predictive accuracy and generalization potential in the lower Friulian Plain (north-eastern Italy) where a geothermal reservoir is present. Among the algorithms scrutinized, XGBoost emerges as a predictive exemplar, achieving a remarkable R2 score of 0.9930 on the test dataset, with consistently low RMSE of 0.788, MAE of 0.587, MAPE of 1.909, and high VAF of 99.30, reaffirming its exceptional precision and robustness. It is worth noting that the other four models show slightly weaker performance than XGBoost, while Elastic Net shows moderate predictive power, which illustrates the complexity of the database. The Wilcoxon signed-rank test confirmed the superior performance of XGBoost in estimating geothermal temperatures compared to other algorithms, with statistical evidence supporting its precision and reliability. A Monte Carlo simulation for uncertainty analysis underlined the importance of model selection, accuracy and uncertainty management in the planning of geothermal projects in the lower Friulian Plain. A sensitivity analysis was performed to identify the main factors influencing the temperature prediction. Among the parameters considered, hydrogen carbonate the highest significance at 0.51, which is essential for accurate temperature prediction because of its buffering capacity which directly influences water's thermal properties. Magnesium and electrical conductivity each contribute with 0.11, also play significant roles due to their impact on the water's heat retention and distribution capabilities. Water depth, with a value of 0.08, also has a significant influence on the temperature profiles in prediction models. In summary, the accurate prediction of XGBoost for the temperature of aquifer in carbonate reservoirs in the lower Friulian Plain, underline its value for optimizing geothermal resources and highlight most important influences on temperature.

Simulation of soil temperature under maize: An inter-comparison among 33 maize models

Accurate simulation of soil temperature can help improve the accuracy of crop growth models by improving the predictions of soil processes like seed germination, decomposition, nitrification, evaporation, and carbon sequestration. To assess how well such models can simulate soil temperature, herein we present results of an inter-comparison study of 33 maize (Zea mays L.) growth models. Among the 33 models, four of the modeling groups contributed results using differing algorithms or “flavors” to simulate evapotranspiration within the same overall model family. The study used comprehensive datasets from two sites - Mead, Nebraska, USA and Bushland, Texas, USA wherein soil temperature was measured continually at several depths. The range of simulated soil temperatures was large (about 10–15 °C) from the coolest to warmest models across whole growing seasons from bare soil to full canopy and at both shallow and deeper depths. Within model families, there were no significant differences among their simulations of soil temperature due to their differing evapotranspiration method “flavors”, so root-mean-square-errors (RMSE) were averaged within families, which reduced the number of soil temperature model families to 13. The model family RMSEs averaged over all 20 treatment-years and 2 depths ranged from about 1.5 to 5.1 °C. The six models with the lowest RMSEs were APSIM, ecosys, JULES, Expert-N, SLFT, and MaizSim. Five of these best models used a numerical iterative approach to simulate soil temperature, which entailed using an energy balance on each soil layer. whereby the change in heat storage during a time step equals the difference between the heat flow into and that out of the layer. Further improvements in the best models for simulating soil temperature might be possible with the incorporation of more recently improved routines for simulating soil thermal conductivity than the older routines now in use by the models.

Estimation of Small-Stream Water Surface Elevation Using UAV Photogrammetry and Deep Learning

Unmanned aerial vehicle (UAV) photogrammetry allows the generation of orthophoto and digital surface model (DSM) rasters of terrain. However, DSMs of water bodies mapped using this technique often reveal distortions in the water surface, thereby impeding the accurate sampling of water surface elevation (WSE) from DSMs. This study investigates the capability of deep neural networks to accommodate the aforementioned perturbations and effectively estimate WSE from photogrammetric rasters. Convolutional neural networks (CNNs) were employed for this purpose. Two regression approaches utilizing CNNs were explored: direct regression employing an encoder and a solution based on prediction of the weight mask by an autoencoder architecture, subsequently used to sample values from the photogrammetric DSM. The dataset employed in this study comprises data collected from five case studies of small lowland streams in Poland and Denmark, consisting of 322 DSM and orthophoto raster samples. A grid search was employed to identify the optimal combination of encoder, mask generation architecture, and batch size among multiple candidates. Solutions were evaluated using two cross-validation methods: stratified k-fold cross-validation, where validation subsets maintained the same proportion of samples from all case studies, and leave-one-case-out cross-validation, where the validation dataset originates entirely from a single case study, and the training set consists of samples from other case studies. Depending on the case study and the level of validation strictness, the proposed solution achieved a root mean square error (RMSE) ranging between 2 cm and 16 cm. The proposed method outperforms methods based on the straightforward sampling of photogrammetric DSM, achieving, on average, an 84% lower RMSE for stratified cross-validation and a 62% lower RMSE for all-in-case-out cross-validation. By utilizing data from other research, the proposed solution was compared on the same case study with other UAV-based methods. For that benchmark case study, the proposed solution achieved an RMSE score of 5.9 cm for all-in-case-out cross-validation and 3.5 cm for stratified cross-validation, which is close to the result achieved by the radar-based method (RMSE of 3 cm), which is considered the most accurate method available. The proposed solution is characterized by a high degree of explainability and generalization.

Evaluation of equations for predicting ileal nutrient digestibility and digestible nutrient content of broiler diets based on their gross chemical composition

The coefficient of apparent ileal digestibility (CAID) and ileal digestible contents (IDC) of nutrients of 56 diets using 10 feed ingredients were measured in broilers (21–24 d post-hatch). Diets contained varying inclusion levels of traditional and non-traditional ingredients and differed widely in chemical composition. The chemical composition and in vivo digestibility values were used to establish prediction equations for CAID and IDC of nutrients using stepwise multiple regression. The strength and accuracy of the developed equations were evaluated by root mean square error (RMSE), coefficient of determination (R2), adjusted R2 (adj. R2), and Akaikie’s Information Criteria (AIC). The bootstrap method was used to validate the choice of variables by stepwise selection method in the original equation based on their frequencies of selection. Selection of variables was validated if the variables that appear in the original stepwise model were selected in more than 30% of the 1000 bootstrap samples. A close agreement between the original equations and bootstrap resampling was observed for CAID of nitrogen (N) and energy and IDC of energy, starch, and calcium (Ca). Additionally, the original data was subjected to another run of stepwise regression analysis using the selected variables by bootstrapping. The initial regression showed that the CAID of N and energy was highly dependent on crude fibre (CF) and energy contents of the diets. The CAID of energy can be predicted (R2 = 0.89 and RMSE = 0.035) by CF, gross energy (GE), CF2, and starch-to-CF ratio (starch:CF). Calcium content had a positive influence, while phosphorus (P) content had a negative influence on the prediction of CAID of fat. The main variable to predict CAID and IDC of most nutrients was the dietary CF content. Based on the lowest RMSE and AIC, the best predictors for IDC of N were ash, N, fat, CF, CF2, and starch:CF, while the best predictors for IDC of energy were CF, GE, CF2, and starch:CF. The results of the original stepwise regression models and the stepwise regression with the selected variables from the bootstrap results for CAID of N, energy, fat, and DM, as well as IDC of energy, starch, and Ca, were the same with no differences in R2, Adj. R2, RMSE, and AIC. This method can be useful for developing stable and reproducible models using stepwise regression. However, an external validation is needed to confirm the use of these equations in commercial settings.

Prediction of jumbo drill penetration rate in underground mines using various machine learning approaches and traditional models.

Estimating penetration rates of Jumbo drills is crucial for optimizing underground mining drilling processes, aiming to reduce costs and time. This study investigates various regression and machine learning methods, including Multilayer Perceptron (MLP), Support Vector Regression (SVR), and Random Forests (RF), to predict the penetration rates (ROP) using multivariate inputs such as operation parameters and rock mass characteristics. The Rock Mass Drillability Index (RDi), incorporating both intact rock properties and structural parameters, was utilized to characterize the rock mass. The dataset was split into 80% for training and 20% for testing. Performance metrics including correlation coefficient (R2), variance accounted for (VAF), mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean square error (RMSE) were calculated for each method to evaluate the accuracy of the predictions. SVR exhibited the best prediction performance for ROP, achieving the highest R2, lowest RMSE, MAE, and MAPE, as well as the largest VAF values of 0.94, 0.15, 0.11, 4.84, and 94.13 during training, and 0.91, 0.19, 0.13, 6.02, and 91.11 during testing, respectively. With this high accuracy, we conclude that the proposed machine learning algorithms are valuable and efficient predictors for estimating jumbo drill penetration rates in underground mining operations.

Jowar and Wheat Yield Prediction using a Wavelet based Fusion of Landsat and Sentinel Data with Meteorological Parameters

Objectives: The objective of this study is to improve the accuracy of crop yield prediction models, specifically focusing on wheat and jowar crops in Maharashtra during the Rabi season, by integrating Landsat and Sentinel satellite data with meteorological parameters. Methods: The study utilizes Landsat 8 and Sentinel satellite datasets covering Maharashtra State. Atmospheric correction is applied to extract surface properties, followed by wavelet-based fusion to combine the images. Normalized Difference Vegetation Index (NDVI) is calculated and combined with meteorological parameters using ensemble learning techniques, including Random Forest and Ada-Boost algorithms. Comparative analysis is conducted against existing models, considering parameters such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). Findings: Significant findings reveal that the proposed methodology outperforms existing models, achieving lower MAE, MSE, and RMSE values for wheat and jowar yield predictions. Additionally, our research highlights the superiority of wheat production over jowar in the Rabi season, based on comprehensive analysis of crop yield predictions. Novelty: This study introduces a novel approach that integrates multiple data sources and employs ensemble learning techniques to enhance crop yield prediction accuracy. By combining Landsat and Sentinel satellite data with meteorological parameters, our methodology provides a more comprehensive understanding of crop growth dynamics, leading to more reliable predictions compared to existing methods. Keywords: Satellite imagery, Machine learning, Normalized Difference Vegetation Index, Fusion, Ensemble learning

Ethical Considerations in AI-driven Dynamic Pricing in the USA: Balancing Profit Maximization with Consumer Fairness and Transparency

Organizations in the USA are progressively employing AI-driven dynamic pricing as a strategic intervention to flexibly modify their prices based on competition, market demand, and various other factors. This research paper focused on the ethical dimensions of AI-driven dynamic pricing and the crucial interplay between profitability and the establishment of unwavering consumer transparency and fairness. The recommended models for dynamic pricing solutions entailed ensemble learning methods, notably, XG-Boost, Light-GBM, Cat-Boost, and X-NGBoost models. Particularly, the proposed model consolidated the XG-Boost algorithm and the NG-Boost model, resulting in a novel methodology termed the X-NGBoost. To compare and contrast the performance of the proposed models, these algorithms were trained and subjected to the same dataset. The comparison between the models was mainly grounded on the root-mean-square error (RMSE) metric, which was quantified in meters. The results indicated that X-NGBoost had the lowest RMSE on both the testing and training sets, at 4.23 and 5.34 respectively. This indicated that X-NGBoost performed very well on both seen and unseen data. Therefore, from the outcomes it was deduced that, for the provided data set, the X-NGBoost model provided the accurate pricing solution.

Machine learning methods to improve spatial predictions of coastal wind speed profiles and low-level jets using single-level ERA5 data

Abstract. Observations of the wind speed at heights relevant for wind power are sparse, especially offshore, but with emerging aid from advanced statistical methods, it may be possible to derive information regarding wind profiles using surface observations. In this study, two machine learning (ML) methods are developed for predictions of (1) coastal wind speed profiles and (2) low-level jets (LLJs) at three locations of high relevance to offshore wind energy deployment: the US Northeastern Atlantic Coastal Zone, the North Sea, and the Baltic Sea. The ML models are trained on multiple years of lidar profiles and utilize single-level ERA5 variables as input. The models output spatial predictions of coastal wind speed profiles and LLJ occurrence. A suite of nine ERA5 variables are considered for use in the study due to their physics-based relevance in coastal wind speed profile genesis and the possibility to observe these variables in real-time via measurements. The wind speed at 10 ma.s.l. and the surface sensible heat flux are shown to have the highest importance for both wind speed profile and LLJ predictions. Wind speed profile predictions output by the ML models exhibit similar root mean squared error (RMSE) with respect to observations as is found for ERA5 output. At typical hub heights, the ML models show lower RMSE than ERA5 indicating approximately 5 % RMSE reduction. LLJ identification scores are evaluated using the symmetric extremal dependence index (SEDI). LLJ predictions from the ML models outperform predictions from ERA5, demonstrating markedly higher SEDIs. However, optimization utilizing the SEDI results in a higher number of false alarms when compared to ERA5.

Long Term Forecasting of Peak Demand and Annual Electricity Consumption of the West African Power Pool Interconnected Network by 2032

The uneven distribution of primary sources of electric power generation in Economic Community of West African States (ECOWAS) compelled the heads of states to create the West African Power Pool (WAPP). The vision of this system is to set up a common electrical energy market to satisfy the balance between supply and demand at an affordable price using the interconnected network. Forecasting maximum power demand and energy consumption is essential for planning and the coordination of new power plant and transmission lines building. This work consists of predicting maximum power demand and total energy that must transit through the WAPP interconnected network by the year 2032. We compare the performances of three time series models namely the Long Short-Term Memory (LSTM), Auto-Regressive Integrated Moving Average (ARIMA) and Fb Facebook Prophet. Electric power and energy data used for training the systems comes from the WAPP authorties. The results show that, for monthly peaks, the Facebook (Fb) Prophet model is the best, with a MAPE (mean absolute error percentage) of 3.1% and a low RMSE (root mean square error) of 1.225 GW. For energy prediction, ARIMA performances are the best compared to others with (RMSE 1.20 TWh, MAPE 1.00%). Thus, the forecast for total annual energy consumption and annual peak demand will be, respectively, 96.85TWh and 13.6 GW in 2032.

Application of the improved dung beetle optimizer, muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area, China

Due to the lack of accurate data and complex parameterization, the prediction of groundwater depth is a challenge for numerical models. Machine learning can effectively solve this issue and has been proven useful in the prediction of groundwater depth in many areas. In this study, two new models are applied to the prediction of groundwater depth in the Ningxia area, China. The two models combine the improved dung beetle optimizer (DBO) algorithm with two deep learning models: The Multi-head Attention–Convolution Neural Network–Long Short Term Memory networks (MH-CNN-LSTM) and the Multi-head Attention–Convolution Neural Network–Gated Recurrent Unit (MH-CNN-GRU). The models with DBO show better prediction performance, with larger R (correlation coefficient), RPD (residual prediction deviation), and lower RMSE (root-mean-square error). Compared with the models with the original DBO, the R and RPD of models with the improved DBO increase by over 1.5%, and the RMSE decreases by over 1.8%, indicating better prediction results. In addition, compared with the multiple linear regression model, a traditional statistical model, deep learning models have better prediction performance.摘要本研究将两个新模型应用于位于中国西北干旱半干旱区的宁夏地区地下水深度预测. 这两个模型将改进的蜣螂优化 (DBO) 算法与两个深度学习模型相结合, 即多头注意力-卷积神经网络-长短期记忆网络和多头注意力-回旋神经网络-门控递归单元. 带有DBO的模型预测结果表现出更大的相关系数 (R) , 残差预测偏差 (RPD) 和较低的均方根误差 (RMSE) , 预测结果更好. 此外, 与DBO模型相比, 改进后的DBO模型的R和RPD增加了1.5%以上, RMSE降低了1.8%以上, 表明预测结果更好. 与传统的统计模型多元线性回归模型相比, 深度学习模型具有更好的预测性能.

Improving ventilation classification in under-actuated zones: a k-nearest neighbor and data preprocessing approach

This study investigates the use of k-nearest neighbors (k-NN) for classifying occupant positions in under-actuated zones, aiming to enhance ventilation control. The focus is on evaluating different data preprocessing techniques, particularly cumulative moving average (CMA), Kalman filtering (KF), and their combination, to boost the k-NN model's reliability and accuracy. The research uses received signal strength indicator (RSSI) data in a controlled setting. The methodology involves dividing the dataset into training and testing subsets and using root mean squared error (RMSE) to determine the best k value for model validation. The study performs a comparative analysis of the k-NN model's performance with both original and preprocessed RSSI data, focusing on metrics such as accuracy, precision, recall, F1-score, and RMSE. The findings emphasize the significant impact of the combined CMA-KF preprocessing technique in improving the model's accuracy and reliability. Specifically, this approach achieved an accuracy of 98.58%. The RMSE values are particularly noteworthy, exhibiting a perfect fit (RMSE of 0) for training data and a remarkably low RMSE of 0.119 for testing data, confirming the model's high accuracy and predictive capability.