Relative Prediction Error Research Articles

Rapid and Non-Destructive Estimation of Moisture Content in Caragana Korshinskii Pellet Feed Using Hyperspectral Imaging.

Moisture content is an important parameter for estimating the quality of pellet feed, which is vital in nutrition, storage, and taste. The ranges of moisture content serve as an index for factors such as safe storage and nutrition stability. A rapid and non-destructive model for the measurement of moisture content in pellet feed was developed. To achieve this, 144 samples of Caragana korshinskii pellet feed from various regions in Inner Mongolia Autonomous Region underwent separate moisture content control, measurement using standard methods, and captured their images using a hyperspectral imaging (HSI) system in the spectral range of 935.5-2539 nm. The Monte Carlo cross validation (MCCV) was used to eliminate abnormal sample data from the spectral data for better model accuracy, and a global model of moisture content was built by using partial least squares regression (PLSR) with seven preprocessing techniques and two spectral feature extraction techniques. The results showed that the regression model developed by PLSR based on second derivative (SD) and competitive adaptive reweighted sampling (CARS) resulted in better performance for moisture content. The model showed predictive abilities for moisture content with a coefficient of determination of 0.9075 and a root mean square error (RMSE) of 0.4828 for the training set; and a coefficient of determination of 0.907 and a root mean square error (RMSE) of 0.5267 for the test set; and a relative prediction error of 3.3 and the standard error of 0.307.

Study on one-dimensional performance prediction of multi-stage axial turbine based on the blade height

Abstract As turbine operating conditions change, the blade height and tip clearance undergo continuous alterations due to the combined effects of thermal stress, aerodynamic forces and centrifugal forces, subsequently influencing the turbine performance. To take this effect into account in turbine performance prediction, this study considers the influence of fluid-heat-structure coupling on blade height and tip clearance and establishes a one-dimensional comprehensive prediction method for multi-stage axial turbine performance considering blade height. When compared with experimental results from a four-stage axial turbine, by considering the fluid-thermal-solid coupling effects, the average relative error in total pressure ratio prediction is reduced from 3.76 % to 1.99 % and the average relative error in total temperature ratio prediction is reduced from 2.03 % to 1.26 %. Compared with the traditional flow prediction method, the prediction results of turbine characteristics considering blade height and tip clearance changes in this paper are closer to the experimental results.

Experiment and Prediction of Pressure Drop in a Fiber–Powder Composite Material with Porous Structure for Energy Wheels and Air Cleaners

Energy wheels and air cleaners play crucial roles in building air conditioning systems. The former is essential for conserving energy in air conditioning systems, while the latter is necessary for ensuring the quality of indoor air. Pressure drop is a crucial parameter for both energy wheels and air cleaners, and it is essential to conduct theoretical and experimental investigations to aid in their design. In this study, we focused on the study of pressure drop in a fiber–powder composite material which can be used for both total heat exchange and air purification. Experimental tests were initially conducted to examine the impact of different parameters on the pressure drop in the material. Subsequently, based on the special fiber–powder structure of the material, two pressure drop prediction methods with different prediction strategies were proposed. The two prediction strategies were compared by analyzing the prediction accuracy of the two methods. As tested by experimental data, for both methods, the absolute prediction error was less than ±6 Pa when the pressure drop was below 50 Pa, and the relative prediction error was less than ±8% for most data sets when the pressure drop was greater than 50 Pa. Moreover, the root mean square error (RMSE) and mean absolute percentage error (MAPE) values of prediction for both methods were less than 4 Pa and 7% respectively. The test results show that although the prediction strategies are different, both prediction methods can obtain acceptable prediction results, and both methods are practical. This study is intended to serve as a valuable reference for the design of energy wheels and air cleaners.

Predicting [177Lu]Lu-HA-DOTATATE kidney and tumor accumulation based on [68Ga]Ga-HA-DOTATATE diagnostic imaging using semi-physiological population pharmacokinetic modeling

BackgroundPrediction of [177Lu]Lu-HA-DOTATATE kidney and tumor uptake based on diagnostic [68Ga]Ga-HA-DOTATATE imaging would be a crucial step for precision dosing of [177Lu]Lu-HA-DOTATATE. In this study, the population pharmacokinetic (PK) differences between [177Lu]Lu-HA-DOTATATE and [68Ga]Ga-HA-DOTATATE were assessed and subsequently [177Lu]Lu-HA-DOTATATE was predicted based on [68Ga]Ga-HA-DOTATATE imaging.MethodsA semi-physiological nonlinear mixed-effects model was developed for [68Ga]Ga-HA-DOTATATE and [177Lu]Lu-HA-DOTATATE, including six compartments (representing blood, spleen, kidney, tumor lesions, other somatostatin receptor expressing organs and a lumped rest compartment). Model parameters were fixed based on a previously developed physiologically based pharmacokinetic model for [68Ga]Ga-HA-DOTATATE. For [177Lu]Lu-HA-DOTATATE, PK parameters were based on literature values or estimated based on scan data (four time points post-injection) from nine patients. Finally, individual [177Lu]Lu-HA-DOTATATE uptake into tumors and kidneys was predicted based on individual [68Ga]Ga-HA-DOTATATE scan data using Bayesian estimates. Predictions were evaluated compared to observed data using a relative prediction error (RPE) for both area under the curve (AUC) and absorbed dose. Lastly, to assess the predictive value of diagnostic imaging to predict therapeutic exposure, individual prediction RPEs (using Bayesian estimation) were compared to those from population predictions (using the population model).ResultsPopulation uptake rate parameters for spleen, kidney and tumors differed by a 0.29-fold (15% relative standard error (RSE)), 0.49-fold (15% RSE) and 1.43-fold (14% RSE), respectively, for [177Lu]Lu-HA-DOTATATE compared to [68Ga]Ga-HA-DOTATATE. Model predictions adequately described observed data in kidney and tumors for both peptides (based on visual inspection of goodness-of-fit plots). Individual predictions of tumor uptake were better (RPE AUC –40 to 28%) compared to kidney predictions (RPE AUC –53 to 41%). Absorbed dose predictions were less predictive for both tumor and kidneys (RPE tumor and kidney –51 to 44% and –58 to 82%, respectively). For most patients, [177Lu]Lu-HA-DOTATATE tumor accumulation predictions based on individual PK parameters estimated from diagnostic imaging outperformed predictions based on population parameters.ConclusionOur semi-physiological PK model indicated clear differences in PK parameters for [68Ga]Ga-HA-DOTATATE and [177Lu]Lu-HA-DOTATATE. Diagnostic images provided additional information to individually predict [177Lu]Lu-HA-DOTATATE tumor uptake compared to using a population approach. In addition, individual predictions indicated that many aspects, apart from PK differences, play a part in predicting [177Lu]Lu-HA-DOTATATE distribution.

Electromagnetic prediction of rock thermal properties beyond boreholes: Soultz-sous-Forets (France) case study

Estimation of deep temperature, thermo-hydro-dynamic and basin simulation, heat exchange and geothermal resource assessment require knowledge of the rock thermal properties beyond boreholes. At the absence of suitable methods and instruments for such estimates we propose a new approach to prediction of thermal conductivity, specific heat capacity and heat flow density based on results of electromagnetic sounding of the study area. Artificial neural network used to this end is taught by correspondence between laboratory measurements on cores from the drilled boreholes and adjacent electrical conductivity profiles determined by inversion of the electromagnetic sounding data collected in their vicinity. This approach is used to estimate rock thermal properties along NW-SE magnetotelluric profile crossing Soultz-sous-Forêts geothermal area, Alsace, France. We demonstrate that accuracy of such prediction both in depths below the boreholes and in the interwell space is comparable with that usually achievable in the borehole scale. In particular, it is shown that relative errors of thermal conductivity prediction at depths twice as large as the borehole depth are as small as 2%, and the accuracy of cross-well thermal conductivity prediction is 6% and 10% for dry and wet rocks, respectively. On the other hand, the prediction accuracy for matrix thermal conductivity of a rock from the known profile of dry thermal conductivity is about 8%. Based on the results of our studies, we constructed dry and wet thermal conductivity sections from the electrical conductivity model determined earlier by 2-D inversion of magnetotelluric data. The temperature model of the study area built along the same profile by means of electromagnetic geothermometer and the obtained thermal conductivity sections enabled estimating the profiles of the vertical component of heat flow density at the surface from dry and wet thermal conductivity sections. They could be considered as minimum and maximum constraints, respectively, bounding the area of its possible values, which, in turn, depend on in situ thermal conductivity of fluids filling the rock pores. The range of possible vertical heat flow density variations (100–150 ± 10% mW·m− 2) agrees well with heat flow estimates in the Upper Rhine graben and, in particular, in Soultz-sous-Forêts geothermal area roughly estimated earlier from borehole measurements. Using the artificial neural network trained on the correspondence between the specific heat capacity determined in the borehole and adjacent electrical conductivity profile, we built the model of specific heat capacity. Against the background, a zone of enhanced specific heat capacity ranging between 2.05 and 2.15 MJ·m− 3·K− 1 was observed in the domain located at the base of the sedimentary cover and upper part of the granitic basement.

Adaptive Residual Useful Life Prediction for the Insulated-Gate Bipolar Transistors with Pulse-Width Modulation Based on Multiple Modes and Transfer Learning

Currently, residual useful life (RUL) prediction models for insulated-gate bipolar transistors (IGBT) do not focus on the multi-modal characteristics caused by the pulse-width modulation (PWM). To fill this gap, the Markovian stochastic process is proposed to model the mode transition process, due to the memoryless properties of the grid operation. For the estimation of the mode transition probabilities, transfer learning is utilized between different control signals. With the continuous mode switching, fractional Weibull motion (fWm) of multiple modes is established to model the stochasticity of the multi-modal IGBT degradation. The drift and diffusion coefficients are adaptively updated in the proposed RUL prediction model. In the case study, two sets of the real thermal-accelerated IGBT aging data are used. Different degradation modes are extracted from the meta degradation data, and then fused to be a complex health indicator (CHI) via a multi-sensor fusion algorithm. The RUL prediction model based on the fWm of multiple modes can reach a maximum relative prediction error of 2.96% and a mean relative prediction error of 1.78%. The proposed RUL prediction model with better accuracy can reduce the losses of the power grid caused by the unexpected IGBT failures.

Research on a Coal Seam Gas Content Prediction Method Based on an Improved Extreme Learning Machine

With the rapid advancement of artificial neural network (ANN) algorithms, many researchers have applied these methods to mine gas prediction and achieved numerous research achievements. It is of great significance to study methods that can accurately predict the gas content for the prevention of gas disasters in mining areas. In order to enhance the accuracy, stability, and generalization capability of the gas content prediction model, the GASA-KELM prediction model was established using the GASA algorithm to improve the KELM initial parameter assignment method, and the prediction model based on BPNN and SVM was established under the same conditions. The experimental results show that the GASA-BPNN model failed to achieve the desired outcome within 800 iterations. On the other hand, the GASA-SVM and GASA-KELM models accomplished the goal in significantly fewer iterations, taking only 673 and 487 iterations, respectively. Moreover, the overall average relative errors of the cross-validated gas content predictions were 15.74%, 13.85%, and 9.87% for the three models, respectively. Furthermore, the total average variance of the test set was 3.99, 2.76, and 2.05 for the GASA-BPNN, GASA-SVM, and GASA-KELM models, respectively. As a result, compared with other ANN models, the GASA-KELM model demonstrates higher accuracy, stronger prediction stability, and generalization ability in the practical application. This novel model provides a basis for accurately predicting gas content and proposing effective regional gas management measures.

Liquid-liquid microextraction method using a deep eutectic solvent for the determination of total paraben content by fluorescence spectroscopy and second-order calibration

Concerns regarding the potential harm associated with endocrine dysfunction are increasing, particularly in children, due to the immaturity of their metabolic system and skin barrier. In this work, parabens, which are esters widely used as preservatives in cosmetics, were quantified in commercial samples of diaper rash cream. For this purpose, a green and efficient liquid–liquid microextraction (LLME) method was accomplished using a deep eutectic solvent (DES). To enhance recovery rates, the extraction parameters were optimized using a central composite design (CCD) on blank matrix samples spiked. The quantification was based on excitation-emission fluorescence spectroscopy, combined with parallel factor analysis (PARAFAC) decomposition, which proved be an advantageous tool to solve the overlap of uncalibrated constituents in the complex matrix. The PARAFAC modeling strategy provided satisfactory results within a linear range response of 100–1300 ng g−1, with LOD equal to 23 ng g−1. A validation set, with various proportions of methyl, ethyl, propyl, and butyl paraben standards, was used to validate the calibration model, resulting in root mean square error (RMSE) and relative error of prediction (REP) values of 32 ng g−1 and 4.5%, respectively. Moreover, acceptable root mean square error of prediction (RMSEP) and REP values (105 ng g−1 and 15%, respectively) were obtained, when a set of blank matrix samples spiked was applied. The proposed method was successfully applied to quantify total paraben content in commercial samples of diaper rash cream, yielding recoveries ranging from 83% to 125%. The method proved to be sensitive, accurate, and green, as it significantly reduced the consumption of organic solvents. Furthermore, it could serve as an alternative for the determination of preservatives in complex cosmetic samples.

Correction method of the fluorescence intensity of petroleum hydrocarbons in different soil types based on resonance scattering spectroscopy

Soil types has an obvious impact influence on the fluorescence intensity of soil petroleum hydrocarbons. To reduce the interference caused by soil types, this paper proposes a calibration method using resonance scattering spectroscopy. To establish the correction method, 100 g/kg crude oil samples from six soil types were prepared. The mission and resonance scattering spectrum under 280 nm excitation were measured for all samples. The results showed that the fluorescence spectra and resonance scattering spectra of soil crude oil vary with different soil types. And the fluorescence peak intensity and the resonance scattering peak intensity at 360 nm of soil petroleum hydrocarbons are highly correlated in different soil types. The fluorescence peak intensity at 360 nm was divided by the resonance scattering light intensity at 360 nm to obtain the corrected fluorescence intensity, which can effectively reduce the influence of soil type on fluorescence intensity. The feasibility of correction method was further verified for different types and concentrations of soil crude oil. The results showed better linearity between the fluorescence intensity and the concentration of Petroleum Hydrocarbons after the correction (with a correlation coefficient R2 of 0.96) than before the correction (with R2 of 0.72), the mean relative prediction error of all samples decreased from 31.92 % to 4.71 % after correction. The research can provide theoretical basis and technical support for the accurate and rapid detection of Petroleum Hydrocarbons in soil by fluorescence spectroscopy.

Dynamic weighing algorithm for dairy cows based on time domain features and error compensation

Weight is an essential indicator for the growth and development of cows. Traditional static weighing method requires a lot of labor and financial resources, the existing dynamic weighing algorithm has no consideration of the influence of hidden information on weighing results during cows' movement, and fails to meet the weighing requirements in complex environments. In this paper, we propose a cow dynamic weighing algorithm based on time domain features and error compensation, quantitatively analyze the hidden information in the dynamic weighing signal, and compensate for the error of the preliminary prediction results, leading to a high-precision dynamic weighing of cows. Firstly, the dynamic weighing signal of cows is pre-processed to obtain the effective signal, and each time domain feature of the effective signal are extracted. Then, the effective signal is decomposed by using the empirical wavelet transform (EWT), and the trend component's mean value is taken as the prediction value of EWT, so that the error between the prediction weight value and the real weight can be calculated. Next, four time-domain features are used as input and the error values are used as output to construct the data set, and the sparrow search algorithm (SSA) is used to obtain the optimal parameters of support vector regression (SVR) model for the training. Finally, the error prediction values output from the model are summed with the prediction values of EWT as the final weight prediction value. To evaluate the performance of the method in this paper, we collect dynamic weighing data from 626 cows, and then randomly select 501 data as the training set while the remaining 125 data as the test set. The average relative error of weight prediction of our method is 0.68%, and the root mean square error is 4.83 kg, better than the existing dynamic weighing algorithms. The experimental results show that the method in this paper is feasible to quantitatively analyze the implicit information of the effective signal according to the time domain features, and the error compensation method can effectively reduce dynamic weighing errors, particularly under extreme conditions, making the method more robust and better meeting the practical needs of farms.

Environmentally sustainable DRS-FTIR probe assisted by chemometric tools for quality control analysis of cinnarizine and piracetam having diverged concentration ranges: Validation, greenness, and whiteness studies

A novel diffuse reflectance fourier transform infrared spectroscopic method accompanied by chemometrics was optimized to fulfill the white analytical chemistry and green analytical chemistry principles for the quantification of cinnarizine and piracetam for the first time without any prior separation in their challenging pharmaceutical preparation, which has a pretty substantial difference in the concentration of cinnarizine/piracetam (1:16). Furthermore, the suggested method was used for cinnarizine/piracetam dissolution testing as an effective alternative to traditional methods. For the cinnarizine/piracetam dissolution tests, we used a dissolution vessel with 900 mL of phosphate buffer pH 2.5 at 37 °C ± 0.5 °C, then the sampling was carried out by frequent withdrawal of 20 µl samples from the dissolution vessel at a one-minute interval, over one hour, then representative fourier transform infrared spectra were recorded. To create a partial-least-squares regression model, a fractional factorial design with 5 different levels and 2 factors was used. This led to the creation of 25 mixtures, 15 as a calibration set and 10 as a validation set, with varying concentration ranges: 1–75 and 16–1000 μg/mL for cinnarizine/piracetam, respectively. Upon optimization of the partial-least-squares regression model, in terms of latent variables and spectral region, root mean square error of cross-validation of 0.477 and 0.270, for cinnarizine/piracetam respectively, were obtained. The optimized partial-least-squares regression model was further validated, providing good results in terms of recovery% (around 98 to 102 %), root mean square error of prediction (0.436 and 3.329), relative root mean square error of prediction (1.210 and 1.245), bias-corrected mean square error of prediction (0.059 and 0.081), and limit of detection (0.125 and 2.786) for cinnarizine/piracetam respectively. Ultimately, the developed method was assessed for whiteness, greenness, and sustainability using five assessment tools. the developed method achieved a greener national environmental method index and complementary green analytical procedure index quadrants with higher eco-scale assessment scores (91), analytical greenness metric scores (0.87), and red–greenblue 12 algorithm scores (89.7) than the reported methods, showing high practical and environmental acceptance for quality control of cinnarizine/piracetam.

Towards model-informed precision dosing of piperacillin: multicenter systematic external evaluation of pharmacokinetic models in critically ill adults with a focus on Bayesian forecasting

PurposeInadequate piperacillin (PIP) exposure in intensive care unit (ICU) patients threatens therapeutic success. Model-informed precision dosing (MIPD) might be promising to individualize dosing; however, the transferability of published models to external populations is uncertain. This study aimed to externally evaluate the available PIP population pharmacokinetic (PopPK) models.MethodsA multicenter dataset of 561 ICU patients (11 centers/3654 concentrations) was used for the evaluation of 24 identified models. Model performance was investigated for a priori (A) predictions, i.e., considering dosing records and patient characteristics only, and for Bayesian forecasting, i.e., additionally including the first (B1) or first and second (B2) therapeutic drug monitoring (TDM) samples per patient. Median relative prediction error (MPE) [%] and median absolute relative prediction error (MAPE) [%] were calculated to quantify accuracy and precision.ResultsThe evaluation revealed a large inter-model variability (A: MPE − 135.6–78.3% and MAPE 35.7–135.6%). Integration of TDM data improved all model predictions (B1/B2 relative improvement vs. A: |MPE|median_all_models 45.1/67.5%; MAPEmedian_all_models 29/39%). The model by Kim et al. was identified to be most appropriate for the total dataset (A/B1/B2: MPE − 9.8/− 5.9/− 0.9%; MAPE 37/27.3/23.7%), Udy et al. performed best in patients receiving intermittent infusion, and Klastrup et al. best predicted patients receiving continuous infusion. Additional evaluations stratified by sex and renal replacement therapy revealed further promising models.ConclusionThe predictive performance of published PIP models in ICU patients varied considerably, highlighting the relevance of appropriate model selection for MIPD. Our differentiated external evaluation identified specific models suitable for clinical use, especially in combination with TDM.

Model and prediction relationship of sound velocity and porosity of seafloor sediments

The relationship between sound velocity and porosity in seafloor sediments is widely utilized for predicting or inverting sediment acoustical and physical parameters. This paper analyzes the theoretical relationship between sound velocity and porosity using the General Model of Sound Speed (GMSS). The GMSS model elucidates an empirical equation relationship in the form of a quadratic fitted polynomial based on actual measurement data from the South China Sea. The accuracy of the GMSS model and the regression empirical equation are found to be consistent, as demonstrated by the relative error, absolute error, and standard deviation of the sound velocity predictions. Our findings suggest that if the scattering of measured data is inherent to seafloor sediment, the empirical equation fails to adequately explain sediment variations and lacks broad applicability for accurate prediction and calculation. Additionally, we observe that a ± 5% relative error in porosity has minimal impact on the application of the GMSS model and empirical equations for predicting sound velocity errors. However, the empirical equation, relying solely on the porosity parameter, inadequately accounts for sound velocity scattering in seafloor sediments with the same porosity due to inherent limitations. In contrast, the GMSS model can incorporate additional physical parameters beyond porosity to better explain and predict sound velocity in seafloor sediments scattering in seafloor sediments with the same porosity, albeit with increased complexity. This paper provides an analytical method and theoretical foundation for inverting or predicting the relationship between sound velocity and porosity of seafloor sediments.

Assessment of total glycerol core aldehyde and monomer content based on NIR and PLS

Glycerol core aldehydes (GCAs) are non-volatile aldehydes residing in frying oil. In this study, near-infrared spectrum (NIR) was used in conjunction with partial least squares (PLS) to create a model for quantifying the total and monomeric content of GCAs. The total and monomer GCAs content in high oleic sunflower oil determined by the GC-MS method was used to assess the model's prediction accuracy. After comparison, the optimal spectrum ranges for both total GCAs and GCAs (9-oxo, 10-oxo-8, and 11-oxo-9) were 8800–6600 cm−1, and the optimal spectrum range for GCAs (8-oxo) was 7420–6600 cm−1. The optimal preprocessing for total GCAs and GCAs (8-oxo, 9-oxo, 10-oxo-8, and 11-oxo-9) were first derivative spectrum, first derivative, original spectrum, first derivative spectrum + NDF smoothing, and first - derivative spectrum + S-G smoothing, respectively. The results demonstrated that the NIR-PLS model had good predictive ability, and the coefficients of determination (R2) for total GCAs and four monomers in the proposed quantitative analysis models were 0.9772, 0.9383, 0.9688, 0.9674, and 0.9745, respectively, and the relative root means square errors of prediction (RMSEP) were 0.0362, 0.0071, 0.0300, 0.0034, 0.0043, and can be predicted quickly and nondestructively, non-toxic, and easy to operate and use.

Evaluation of greenness and whiteness assessment of chemometric assisted techniques for simultaneous determination of canagliflozin, sitagliptin, metformin, pioglitazone, and glimepiride in a quinary mixture

An assortment of lifestyle and genetic characteristics contribute to the eventual appearance of type 2 diabetes. Diabetes can be triggered by a variety of drugs and other health issues. Five anti-diabetic medications, Metformin, Glimepiride, Canagliflozin, Pioglitazone, and Sitagliptin, were identified for simultaneous evaluation in their synthetic mixtures applying multivariate calibration models Principal Component Regression (PCR) and Partial Least Square (PLS-2) to resolve the extensively overlapping spectrum.For extracting important information and enhancing the accuracy of the techniques, both techniques were used for deciding on the variables. The most beneficial spectral ranges and combinations were identified by considering the least values of the Correlation Coefficient (R2 ≥ 0.9998), the Root Mean Square Error of Prediction (RMSEP) values between (0.02782–0.08277), and the Relative Error of Prediction (SEC) values between (0.20693–0.90161).The two multivariate calibration approaches were successful in determining all five components in their quinary mixture simultaneously. As a result, the suggested techniques can easily be employed without a separation stage and effectively implemented in pharmaceutical formulation evaluation. In addition, statistical assessments between the suggested chemometric techniques and the reference revealed no significant differences.Furthermore, the suggested techniques succeeded in the guidelines of green analytical chemistry, and their eco-friendliness was assessed using four tools, namely, the Analytical Eco-Scale, the National Environmental Methods Index (NEMI), the Green Analytical Procedure Index (GAPI), and the Analytical Greenness Metric (AGREE), which confirmed the proposed methods' eco-friendliness. In addition, the newly generated Red-Green-Blue (RGB12 paradigm) was employed to investigate the whiteness properties. The suggested methods' acceptable observations, as well as their long-term sustainability, straightforwardness, cost-effectiveness, and inexpensive cost, stimulate adoption in quality control laboratories.

An algorithm for selecting a machine learning method for predicting nitrous oxide emissions in municipal wastewater treatment plants

This study presents an advanced algorithm for selecting machine learning (ML) models for nitrous oxide (N2O) emission prediction in wastewater treatment plants (WWTPs) employing the activated sludge process. The examined ML models comprised multivariate adaptive regression spline (MARS), support vector machines (SVM), and extreme gradient boosting (XGboost). The study explores the concept that involves new criteria to select the most appropriate ML model: (i) fitting the results of model prediction to the measurements taking into account the range of variability of the input data, and (ii) model verification applying a reference MCM to identify the input-output relationship using the global sensitivity analysis (GSA). Using the k-means method, it was shown that the relative errors (%e) of N2O prediction by ML models depend on the range of variability of the input data (nitrogen compounds concentration in the bioreactor compartments, influent flowrate, air flowrate). The smallest relative errors of N2O prediction (%e = 0.13 for MARS, %e = 0.12 for SVM and %e=0.10 for XGboost) were found for the concentrations: NH4-N = 24.14 mg N/L (anaerobic compartment), NO3-N = 5.40 mg N/L (aerobic compartment), and the largest (%e > 0.35) for the concentrations: NH4-N = 29.43 mg N/L (anaerobic compartment), NO3-N = 7.90 mg N/L (aerobic compartment). Calculations using the GSA method confirmed that the XGboost model was the only one that showed identical relationships between all the considered input variables and N2O emission rate. The ML model obtained in this way can be used as an alternative to the MCM for estimating N2O emission as a significant contributor to the carbon footprint of WWTPs.

Evaluation of Corporate Social Responsibility and Core Competitiveness Based on GA-SVM Model

Abstract The author proposes a model evaluation based on the GA-SVM model to better understand the evaluation of the company’s relationship and core competitiveness. The system index is reduced by relative gray analysis, the support vector machine model is optimized by a genetic algorithm, and the specific algorithm steps are introduced. Select models from the top 100 enterprises in China’s construction industry in 2020 published by China Construction Industry Market, using relative gray to reduce the measure, and then use the genetic algorithm support vector machine (GA-SVM) model for the training model to achieve the evaluation of the core competencies of the target construction technique business. The experimental results show that the relative error of prediction of the GA-SVM(Genetic algorithm-support vector machine) model for the evaluation of the competitive core of the business is not more than 5, which meets the should be made of accurate predictions. Therefore, choosing the Gaussian radial basis kernel function (RBF) as the kernel function of the GA-SVM intelligent evaluation model is good. It proves that the GA-SVM model can analyze the relationship between the role and the important competition.

Sustainable development of vocal music ecology in a digital ecological environment

The phenomenon of noise interference during vocal transmission can lead to the problem of poor vocal transmission quality. This paper proposes a study on the sustainability of vocal ecology in a digital ecological environment. First, the matching tracking algorithm can extract the time-frequency characteristics of the effective signal, attenuate the interference of noise, and improve the propagation quality. A sustainable development GA-BP network model is established, and the adjustment amount of each weighting coefficient is obtained according to the gradient algorithm and using the inertia adjustment strategy. The coordination is regulated through feedback control strategies to ultimately achieve ecological sustainability of vocal music. The analysis results show that the average relative error of the simulation prediction of the sustainable development GA-BP network model is 3.54%, the maximum relative error is 8.11%, and the average relative error is within 5% of 70%. It has significant superiority and high efficiency in comparison with the prediction degree of the traditional model.

DEM simulation of impact charge on polymer particles by single and repeated impact on different metal plates

Charge transfer between the particle and wall occurs after each collision in powder handling operations, called "contact electrification". In the present study, a LIGGGHTS code was developed by adding the Hertzian contact area formula and condenser model to predict the particle charge after every impact. The simulation results showed that, the particle with a higher impact velocity became saturated under fewer collisions. Considering that the charging constant depends on the impact velocity and initial particle charge, the Average Relative Error (ARE) of simulation prediction significantly decreased from 22 and 16% to 10.6 and 11% for Rubber particles at impact velocities of 4.8 and 2.8 m/s, respectively. Therefore, the current modified LIGGGHTS code has a great potential to predict the impact charge of particle-wall collision.

Thermal behaviour of Li-ion battery: An improved electrothermal model considering the effects of depth of discharge and temperature

The current paper presents a two-dimensional electrothermal model that evaluates the thermal and voltage behaviour of a LiFePO4-20 Ah Li-ion pouch cell. The electrothermal model considers the impact of depth of discharge (DoD) and temperature on heat generated inside Li-ion cells by precisely simulating the internal equivalent resistance and entropy change utilizing experimental data. The proposed electrothermal model is validated utilizing experimental data. The results for the temperature estimation utilizing the electrothermal model show a similar result for all discharge current rates. The temperature prediction for all the discharge current rates (C-rate) shows slightly more accurate and precise results toward the end of discharge of Li-ion pouch cell. Overall, the absolute maximum relative error for the temperature prediction is around 6.30 % at 4C discharge rate. Similarly, the voltage estimation shows a high level of accuracy compared to the experimental data, and the simulated voltage has slightly higher values than the experimental voltage data, but the voltage trend is similar to the experimental data. Overall, the simulated results obtained from the proposed electrothermal model exhibit similar trends and high accuracy when compared to the experimental data.