A Study on the Influence Mechanism of User Demand Response and Service Experience in Urban New Energy Charging Market Based on Multi-model Fusion: A Case Study of Urumqi

A multi-feature-based multi-model fusion method for state of health estimation of lithium-ion batteries

Prediction of soil organic carbon in an intensively managed reclamation zone of eastern China: A comparison of multiple linear regressions and the random forest model

ABSTRACT: The present descriptive quantitative research tries to find out which machine learning model is the most efficient to predict the utility of a bulk liquefied petroleum gas trading company in Metropolitan Lima. To determine daily profit, which will be a variable dependent on the output model. This dependent parameter has five independent variables: sale price, quantity sold, purchase cost, transportation cost and kilometers traveled, as well as the values with the highest correlation coefficients. There are several machine learning models, for this research the Artificial Neural Networks, Multiple Linear Regression and Random Forest models will be used, which estimated the utility through their own mathematical algorithms. To simulate the algorithms of the mentioned models, the Python program was used. These models were trained to learn and validate 70% and 30% of the database, that is, of the 235 data collected, 165 data were used to calibrate and 70 data to validate. When making the comparison between the automatic learning models for the estimation of the daily utility of the trading company, the Random Forest model was obtained as the best option, obtaining an R2 of 0.959 and also having the lowest statistical error rates with respect to the models. of Artificial Neural Networks and Multiple Linear Regression. Keywords: Machine learning, artificial neural networks, multiple linear regression, Random Forest and predictive models.

Digital soil mapping (DSM) is a field of soil science that aims to improve traditional soil maps by producing higher resolution predictive maps of soil properties using spatial environmental data. DSM has historically relied primarily on static environmental covariates—such as slope gradient, slope aspect, and other topographic variables derived from digital terrain models—for predicting static soil properties, like soil texture. Advancements in satellite imagery and statistical modeling improve the accuracy of digital soil maps by incorporating multi-temporal data that can capture landscape-scale change over relatively short periods of time. Adding these dynamic environmental covariates may be especially useful for spatial prediction of dynamic soil properties, like infiltration rate, that are strongly affected by phenomenon that satellite imagery can detect, like land use that changes rapidly due to human activity. Infiltration strongly impacts soil health and hydrologic characteristics in a watershed. Understanding infiltration for sustainable land management is vital for making best management decisions in urbanizing environments like the West Run Watershed in Morgantown, West Virginia. We hypothesized that infiltration could be predicted at a higher accuracy and a finer spatiotemporal scale using digital soil mapping techniques than is currently provided by the current official soil data and maps produced by the National Cooperative Soil Survey. Spatial predictions of infiltration rate were produced for the West Run watershed using both static and dynamic environmental covariates as inputs into multiple linear regression (MLR) and random forest (RF) models, each of which were made using 10-fold cross validation. Training and independent validation sampling locations were selected using a conditioned Latin hypercube sampling scheme and observed saturated hydraulic conductivity of the soil surface was collected using automated dual-head infiltrometers. The MLR and RF models had R 2 of 0.302 and 0.201, respectively. Validation sampling was stratified by the predicted infiltration values of the MLR model. Validation R 2 values for the MLR and RF models were 0.080 and 0.103. The results from this study will benefit the development of a dynamic soil survey and will improve hydrologic models in this and potentially other mixed-land-use watersheds.

Accurate calculation of adsorbed shale gas content is critical for gas reserve evaluation and development. However, gas adsorption and desorption experiments are expensive and time-consuming, while physics-based models and empirical correlations are unable to accurately capture the adsorption characteristics for different shales. Langmuir adsorption is one of the most commonly used model for calculating the adsorbed gas content in shale gas reservoirs. However, most existing correlations for the Langmuir pressure and Langmuir volume in the model are oversimplified based on limited experimental data points. Thus they are not representative of key geological parameters and are far from accurate for prediction in many cases. We developed a variety of machine learning models that are multivariable controlled to quantify shale gas adsorption. The data-driven method subdivides into two procedures: data compilation and machine learning regression. Over 700 data entries, composed of reservoir temperature (T, °C), total organic carbon (TOC, wt%), vitrinite reflectance (Ro,%), Langmuir pressure, and Langmuir volume are compiled from shale gas plays mainly in USA, Canada, and China. Data have been consistently curated, then machine learning approaches, including multiple linear regression (MLR), support vector machine (SVM), random forest (RF) and artificial neural network (ANN), have been built, trained and tested by partitioning the data into 75%:25%. For SVM, RF and NN models, 1000 simulations were run and averaged for performance comparison. MLR identifies non-negligible parameters and general trends for shale gas adsorption. Nonetheless, the correlation coefficients from MLR are far from satisfactory. For Langmuir pressure, RF models fit best to the data entries and the other models follow the order of SVM > ANN > MLR. Particularly, RF models show the highest performance stability with the averaged R-squared value of 0.84 and the maximum of 0.87, indicating a very strong relationship constructed for these 213 data entries. For 485 Langmuir volume data entries, RF models also perform best while the other three regression methods are comparable. It should be noted that altering machine learning model structure and parameters could significantly affect the regression results. Robust and universal machine learning models for estimating adsorbed shale gas content with high confidence level are established, which not only provide more accurate estimation and broader parameter adaptation than physics-based and empirical models, but also circumvent the high-cost and time-consuming deficiency of experimental measurements. These machine learning models can be used to estimate adsorbed gas content for shale plays with limited experimental measurements. Moreover, they can be incorporated into reservoir simulators to improve the simulation performance.

Comparative analysis of random forest, exploratory regression, and structural equation modeling for screening key environmental variables in evaluating rangeland above-ground biomass

Spatial prediction of the concentrations of soil heavy metals (HMs) in cultivated land is critical for monitoring cultivated land contamination and ensuring sustainable eco-agriculture. In this study, 32 environmental variables from terrain, climate, soil attributes, remote-sensing information, vegetation indices, and anthropogenic activities were used as auxiliary variables, and random forest (RF), regression Kriging (RK), ordinary Kriging (OK), and multiple linear regression (MLR) models were proposed to predict the concentrations of As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn in cultivated soils. In comparison to those of RK, OK, and MLR, the RF model had the best prediction performance for As, Cd, Cr, Hg, Pb, and Zn, whereas the OK and RK models had highest prediction performance for Cu and Ni, respectively, showing that R2 was the highest, and mean absolute error (MAE) and root mean square error (RMSE) were the lowest. The prediction performance of the spatial distribution of soil HMs under different prediction methods was basically consistent. The high value areas of eight HMs concentrations were all distributed in the southern plain area. However, the RF model depicted the details of spatial prediction more prominently. Moreover, the importance ranking of influencing factors derived from the RF model indicated that the spatial variation in concentrations of the eight HMs in Lanxi City were mainly affected by the combined effects of Se, TN, pH, elevation, annual average temperature, annual average rainfall, distance from rivers, and distance from factories. Given the above, random forest models could be used as an effective method for the spatial prediction of soil heavy metals, providing scientific reference for regional soil pollution investigation, assessment, and management.

Soil bulk density (ρb) is vital for assessing soil organic carbon (SOC) and nutrient stocks, as well as for modeling soil processes. Although ρb can be measured using traditional methods, these are labor-intensive and time-consuming. Consequently, there is a growing interest in developing pedotransfer functions (PTFs) to predict ρb based on more easily measured soil properties. The vast Brazilian territory, with different climates and biomes, presents a challenge for development of national-scale PTFs, particularly for ρb. In this study, a comprehensive dataset was compiled from various sources to develop ρb PTFs across Brazil. The dataset includes particle size distribution (PSD), SOC, ρb, sampling depth, soil order, land use, and geographic coordinates, allowing for the incorporation of additional numerical and categorical variables. Rigorous data preprocessing ensured quality and reliability. PTFs were developed using multiple linear regression (MLR) and Random Forest (RF) models. Model accuracy was evaluated using mean absolute error, bias, root mean square error (RMSE), and coefficient of determination. Both MLR and RF models accurately predicted ρb, with log-transformed PSD and SOC emerging as key predictors. The RF model slightly outperformed the MLR model (RMSE = 0.12 vs. 0.13 g cm−3) on the test dataset, underscoring the importance of environmental and categorical variables in predicting ρb. The developed PTFs, along with other PTFs for Brazil, were applied to estimate SOC stocks across different biomes and land uses. Best estimations were obtained with the RF model, with an R2 of 0.97, emphasizing the value of categorical variables in improving SOC stock estimations.

Prediction of soil water infiltration using multiple linear regression and random forest in a dry flood plain, eastern Iran

The comparison between multiple linear regression and random forest model in predicting environmental noise and its frequency components level in Hong Kong using a land-use regression approach.

Common bread wheat ( Triticum aestivum L.) is a key component of global diets, but the genetic improvement of wheat is not keeping pace with the growing demands of the world's population. To increase efficiency and reduce costs, breeding programs are rapidly adopting the use of unoccupied aerial vehicles to conduct high‐throughput spectral analyses. This study examined the effectiveness of multispectral indices in predicting grain yield compared to genomic prediction. Multispectral data were collected on advanced generation yield nursery trials during the 2019–2021 growing seasons in the Colorado State University Wheat Breeding Program. Genome‐wide genotyping was performed on these advanced generations and all plots were harvested to measure grain yield. Two methods were used to predict grain yield: genomic estimated breeding values (GEBVs) generated by a genomic best linear unbiased prediction (gBLUP) model and phenomic phenotypic estimates (PPEs) using only spectral indices via multiple linear regression (MLR), k‐nearest neighbors (KNNs), and random forest (RF) models. In cross‐validation, PPEs produced by MLR, KNN, and RF models had higher prediction accuracy () than GEBVs produced by gBLUP ( ). In leave‐one‐year‐out forward validation using only multispectral data for 2020 and 2021, PPEs from MLR and KNN models had higher prediction accuracy of grain yield than GEBVs of those same lines. These findings suggest that a limited number of spectra may produce PPEs that are more accurate than or equivalently accurate as GEBVs derived from gBLUP, and this method should be evaluated in earlier development material where sequencing is not feasible.

Spatial prediction of soil surface texture in a semiarid region using random forest and multiple linear regressions

Knowledge about the spatial distribution of active-layer (AL) soil thickness is indispensable for ecological modeling, precision agriculture, and land resource management. However, it is difficult to obtain the details on AL soil thickness by using conventional soil survey method. In this research, the objective is to investigate the possibility and accuracy of mapping the spatial distribution of AL soil thickness through random forest (RF) model by using terrain variables at a small watershed scale. A total of 1113 soil samples collected from the slope fields were randomly divided into calibration (770 soil samples) and validation (343 soil samples) sets. Seven terrain variables including elevation, aspect, relative slope position, valley depth, flow path length, slope height, and topographic wetness index were derived from a digital elevation map (30 m). The RF model was compared with multiple linear regression (MLR), geographically weighted regression (GWR) and support vector machines (SVM) approaches based on the validation set. Model performance was evaluated by precision criteria of mean error (ME), mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). Comparative results showed that RF outperformed MLR, GWR and SVM models. The RF gave better values of ME (0.39 cm), MAE (7.09 cm), and RMSE (10.85 cm) and higher R2 (62%). The sensitivity analysis demonstrated that the DEM had less uncertainty than the AL soil thickness. The outcome of the RF model indicated that elevation, flow path length and valley depth were the most important factors affecting the AL soil thickness variability across the watershed. These results demonstrated the RF model is a promising method for predicting spatial distribution of AL soil thickness using terrain parameters.

Structural and contextual correlates of charged behavior in product development teams

Firms increasingly use cross‐functional teams to develop new products, yet we know little about the processes that make teams excel. Although studies have focused on within‐team processes like cooperation between and integration of individuals from various functional areas, some emerging literature suggests that the processes that make teams excel are richer and more complex than cooperation and integration. In order to capture the processes that lead to excellent market performance of new products, we introduce the concept of charged team behavior, the extent to which cross‐functional product development teams are enthusiastically and jointly driven to develop superior new products. Charged team behavior captures not only the drive, commitment, and joy of team members, but also their collaborative behaviors to achieve an exceptional outcome.We propose and test a series of hypotheses concerning how charged behavior affects new product market performance and how charged behavior is, in turn, influenced by both team structural characteristics (physical proximity, team longevity, and outcome interdependence) and contextual factors (senior management encouragement to take risk, quality orientation, exposure to customer input, extent of competition, and interdepartmental connectedness). It is particularly important to examine the antecedents of charged behavior because there are concerns that some of the team‐related factors generally considered to be useful for teams may not necessarily lead to charged teams.Data from new consumer product development teams is analyzed though structural equation modeling for hypothesis testing. We find evidence that highly charged teams are more likely to develop successful new products. Results also indicate that outcome interdependence, exposure to customer input, extent of competition, and interdepartmental connectedness are positively related to charged behavior. Physical proximity, team longevity, encouragement to take risk, and quality orientation do not improve teams' charged behavior. Data suggests that charged team behavior: 1) fully mediates the effects of outcome interdependence and interdepartmental connectedness on performance, 2) partially mediates the influence of exposure to customer input and the extent of competition on performance, and 3) does not mediate the effects of quality orientation and physical proximity on performance.Our study highlights the importance of creating highly charged product development teams in order to achieve exceptional performance. Further, our results indicate that some of the factors suggested by traditional social psychology research for enhancing team effectiveness (e.g., physical proximity and team longevity) may not necessarily create charged teams. Instead, charged teams need a special arrangement, in which members are accountable to the team and where their evaluations and rewards are also linked to the performance of the team. In addition, although a strong emphasis on quality is considered to be beneficial for new products, as our results indicate, such emphasis cannot create a charged atmosphere. Moreover, our research suggests that if the organization structure does not permit frequent contact between individuals across functional boundaries, the creation of a strongly charged team and development of a successful new product will be hindered.