Root Mean Square Error Of Prediction Research Articles

Non-destructive prediction of ready-to-eat kiwifruit firmness based on Fourier transform near-infrared spectroscopy

There is a growing demand for ready-to-eat kiwifruit in the world. However, ready-to-eat kiwifruit has a rather narrow range of firmness (e.g. 10–30 N), and it remains challenging to predict this firmness in a non-destructive manner. Here, we report a strategy for non-destructive prediction of kiwifruit firmness based on Fourier transform near-infrared (FT-NIR) spectroscopy. The radial basis function (RBF) model displayed superior performance, with a coefficient of determination (Rc2) of 0.83, a cross-validation coefficient of determination (Rp2) of 0.73, a root mean square error of calibration (RMSEC) of 0.58, a root mean square error of prediction (RMSEP) of 0.72, and a ratio of performance to deviation (RPD) of 1.92. To enhance the accuracy of kiwifruit firmness prediction, we optimized the FT-NIR algorithm through data preprocessing, feature selection, and dimensionality reduction. The results showed that the FD-CARS-SVR (RBF) algorithm exhibited the best performance in predicting kiwifruit firmness during the shelf life with impressive values of Rc2 (0.99), Rp2 (0.92), RMSEC (0.15), RMSEP (0.40), and RPD (3.48). To further evaluate the applicability of the FT-NIR model, we compared the data predicted by the model and acquired from the Kiwifirm™ and penetrometer GY-4. The results revealed pronounced superiority of the FT-NIR model for the firmness ranging from 10 to 40 N to replace Kiwifirm™, providing a new non-destructive model for the prediction of the firmness of ready-to-eat kiwifruit.

Detection of Hardening in Mangosteens Using near-Infrared Hyperspectral Imaging

Mangosteens can develop a postharvest physiological disorder, called “hardening”, which affects their marketability and is not detectable using visual inspection. The hardening disorder of mangosteens was determined by firmness value using the texture analyzer. Near-infrared hyperspectral imaging (NIR-HSI) in the region of 935–1720 nm was tested as a possible rapid and non-destructive method to detect this disorder. The spectra from a region of interest of mangosteens were acquired and used for analysis. Calibration models for firmness of a similarly sized group and a mixed-size group were established using partial least squares regression (PLSR) and support vector machine regression (SVMR). Chemometric algorithms were investigated in order to determine the optimal conditions for establishing the models for firmness. The optimum model was obtained when the fruit were graded into similarly sized groups. Using partial least squares regression (PLSR), the correlation coefficient of prediction (Rp) was 0.87 and the root mean square error of prediction (RMSEP) was 6.25 N. The predictive images for firmness of the fruit were created by interpreting predicted firmness visualized as colors in every pixel. From the data, it was concluded that NIR-HSI can potentially be used to visualize hardening of individual mangosteens based on their predictive images.

Fourier Transform Infrared Spectroscopy Tracking of Fermentation of Oat and Pea Bases for Yoghurt-Type Products

The fermentation process of plant-based yoghurt (PBY)-like products must be followed for consistency by monitoring, e.g., the pH, temperature, and lactic acid concentration. Spectroscopy provides an efficient multivariate in situ quality monitoring method for tracking the process. Therefore, quality monitoring methods for pea- and oat-based yoghurt-like products using Fourier transform infrared (FT-IR) spectroscopy and high-performance liquid chromatography (HPLC) were developed and modeled. Plant-based yoghurt (PBY) was formulated by fermenting pea and oat plant drinks with a commercial starter culture based on Lactobacillus and Streptococcus strains. The main variance during fermentation was explained by spectral carbohydrate and protein bands with a notable shift in protein band peaks for the amide II band at 1548 cm−1 to 1576 cm−1. In addition to the identification of changed spectral bands during fermentation, FT-IR efficiently tracked the variation in oat and pea fermentation using pH as the main indicator. Prediction models with an R2 for the predicted value of pH as a fermentation indicator (R2 = 0.941) with a corresponding root-mean-squared error of prediction (RMSEP) of 0.247 was obtained when compared to the traditional pH method.

Determining Moisture Content of Basil Using Handheld Near-Infrared Spectroscopy

Accurate and rapid determination of moisture content is essential in crop production and decision-making for irrigation. Near-infrared (NIR) spectroscopy has been shown to be a promising method for determining moisture content in various agricultural products, including herbs and vegetables. This study tested the hypothesis that NIR spectroscopy is effective in accurately measuring the moisture content of Genovese basil (Ocimum basilicum L.), with the objective of developing a respective calibration model. Spectral data were obtained from a total of 120 basil leaf samples over a period of six days. These included freshly harvested and detached leaves, as well as those left in ambient air for 1–6 days. Five spectra were taken from each leaf using a handheld NIR spectrophotometer, which covers the first and second overtones of the NIR spectral region: 950–1650 nm. After the spectral acquisition, the leaves were weighed for fresh mass and then put in an oven for 72 h at 80 °C to determine the dry weight and calculate the reference moisture content. The calibration model was developed using multivariate analysis in MATLAB, including preprocessing and regression modeling. The data obtained from 75% of the samples were used for model training and 25% for validation. The final model demonstrates strong performance metrics. The root mean square error of calibration (RMSEC) is 2.9908, the root mean square error of cross-validation (RMSECV) is 3.2368, and the root mean square error of prediction (RMSEP) reaches 2.4675. The coefficients of determination for calibration (R2C) and cross-validation (R2CV) are consistent, with values of 0.829 and 0.80, respectively. The model’s predictive ability is indicated by a coefficient of determination for prediction (R2P) of 0.86. The range error ratio (RER) stands at 11.045—highlighting its predictive performance. Our investigation, using handheld NIR spectrophotometry, confirms NIR’s usefulness in basil moisture determination. The rapid determination offers valuable insights for irrigation and crop management.

Greenness-sustainability metrics for assessment smart-chemometric spectrophotometric strategy for evaluation of the combination of six gastric proton-pump inhibitors with two selected impurities

Green analytical approaches are employed for the determination of active pharmaceutical ingredients, in conjunction with their impurities. Smart chemometric spectrophotometric techniques, including orthogonal partial least square (OPLS), variable selection such as genetic algorithm (GA-OPLS), and interval selection (i-OPLS), were utilized. These chemometric models were implemented for assessing six proton-pump inhibitors Omeprazole, Esomeprazole, Lansoprazole, Pantoprazole, Rabeprazole, and Dexlansoprazole along with two selected official impurities, namely 4-Desmethoxy omeprazole impurity and Rabeprazole-impurity B. Experimental design was implemented to separate impurities, in the process of multivariate calibration, a five-level eight-factor calibration design consisting of 25 samples was selected. This design was deliberately selected to guarantee that the components were mutually orthogonal to assess the model's performance and reliability, a separate validation set of 15 samples was constructed. The best-performing of the proposed techniques were identified by considering the least favorable values of the Correlation Coefficient (R ≥ 0.9995), the Root Mean Square Error of Prediction (RMSEP) values between (0.0102–0.5622), and the Relative Error of Prediction (REP) values between (0.2961–1.1917). The proposed and reported methods' greenness-sustainability was quantitatively evaluated, and a comparative study of the greenness profile was established through a spider chart, the National Environmental Method Index tool, advanced and modified NEMI along with the Hexagon tool, and the whiteness qualities of the presented approaches were assessed by implementing the recently adopted Red-Green-Blue paradigm and White Analytical Chemistry tool. These approaches are well-suited for use in quality control laboratories due to their observed acceptance, long-term sustainability, simplicity, and affordability.

Accurate identification of methanol and ethanol gasoline types and rapid detection of the alcohol content using effective chemical information

Methanol and ethanol gasoline are two emerging clean energy sources with different characteristics. To achieve the qualitative identification and quantitative analysis of the alcohols present in methanol and ethanol gasoline, effective chemical information (ECI) models based on the characteristic spectral bands of the near-infrared (NIR) spectra of the methanol and ethanol molecules were developed using the partial least squares discriminant analysis (PLS-DA) and partial least squares (PLS) algorithms. The ECI model was further compared with models built from the full wavenumber (Full) spectra, variable importance in projection (VIP) spectra, and Monte Carlo uninformative variable elimination (MC-UVE) spectra to determine the predictive performance of ECI model. Among the various qualitative identification models, it was found that the ECI-PLS-DA model, which is built using the differences in molecular chemical information between methanol and ethanol, exhibited sensitivity, specificity and accuracy values of 100%. The ECI-PLS-DA model accurately identified methanol gasoline and ethanol gasoline with different contents. In the quantitative analysis model for methanol gasoline, the methanol gasoline and ethanol gasoline ECI-PLS models exhibited the smallest root mean squared error of predictions (RMSEP) of 0.18 and 0.21% (v/v), respectively, compared to the other models. Meanwhile, the F-test and T-test results revealed that the NIR method employing the ECI-PLS model showed no significant difference compared to the standard method. Compared with other spectral models examined herein, the ECI model demonstrated the highest recognition success and determination accuracy. This study therefore established a highly accurate and rapid determination model for the qualitative identification and quantitative analysis based on chemical structures. It is expected that this model could be extended to the NIR analysis of other physicochemical properties of fuel.

Enhancing quality control in emulsion-type sausage production: Predicting chemical composition of intact samples with near infrared spectroscopy

This research actively explores the potential of near infrared spectroscopy (NIR) for analyzing the chemical composition of emulsion-type sausages, focusing on critical factors like residual nitrite, moisture, protein, and fat content. To establish robust and generalizable models, we utilized a dataset of 100 experimentally prepared sausages encompassing a wide range of pork back fat replacement levels (5%, 15%, 30%, 45%, and 60%) and added sodium nitrite amounts (0, 80, 125, 250, and 375 ppm). An external validation set of 20 commercially sourced sausages further assessed the model’s real-world applicability. Partial least squares (PLS) regression calibration models with multiplicative scatter correction (MSC) pre-treatment demonstrated impressive accuracy for moisture (RMSECV = 0.57%, RPD = 9.8), fat (RMSECV = 1.17%, RPD = 9.5), and protein (RMSECV = 0.30%, RPD = 7.6). While residual nitrite prediction presented challenges due to its inherent complexity, the external validation yielded a competitive root mean square error of prediction (RMSEP) of 12.02 ppm, surpassing the average performance reported in similar studies (RMSEP ∼15 ppm) by 3 ppm. Importantly, sample homogenization did not significantly affect parameter prediction, highlighting the robustness of the NIR-based approach. These findings suggest that NIR spectroscopy, with its non-destructive, rapid, and cost-effective nature, could provide valuable tools for quality control and monitoring in the emulsion-type sausage industry. More importantly, improved nitrite prediction could pave the way for enhanced precision and control in sausage production, ultimately contributing to improved food safety and sustainability.

VARIABLE SELECTION METHODS APPLIED IN HTGC DATA TO DETERMINE PHYSICOCHEMICAL PROPERTIES OF CRUDE OILS

High-temperature gas chromatography (HTGC) is an analytical technique employed in the petroleum industry for component separation. By incorporating chemometrics, HTGC data can be effectively utilized to predict various properties of crude oil. However, HTGC chromatograms generate a substantial number of variables, some of which may lack pertinent chemical information. Consequently, employing variable selection methods becomes crucial to reduce the number of variables and enhance the predictive capability of calibration models. In this study, the interval partial least squares (iPLS), synergy interval partial least squares (siPLS), and ordered predictors selection (OPS) methods were applied for variable selection to construct linear regression models. The main objective was to investigate the potential of these methods in predicting eight properties of crude oil: American Petroleum Institute (API) gravity, standardized kinematic viscosity at 50 °C (VISp), flash point (FP), Reid vapor pressure (RVP), micro carbon residue (MCR), saturates (SAT), aromatics (ARO), and polar (POL) content. While all variable selection methods yielded satisfactory results, the OPS-PLS regression models consistently exhibited the best performance in estimating these properties, achieving root mean squared error of prediction (RMSEP) values of 1.244 for API, 0.029 for VISp, 15.356 °C for FP, 0.324 kPa for RVP, 0.629 wt% for MCR, 3.691 wt% for SAT, 2.939 wt% for ARO, and 3.374 wt% for POL. Variable selection demonstrated remarkable effectiveness, significantly improving the accuracy of the models, and allowing for the creation of concise models with a focused set of variables.

“Reagent-free determination of γ-butyrolactone in beverages”

A green partial least square (PLS) attenuated total reflectance infrared spectroscopy (ATR-FTIR) method has been developed for the prediction of the γ-butyrolactone (GBL) in all type of soft and alcoholic beverages. The method involves the use of a drop (less than 5 µL) of sample and a single calibration model based on aqueous and alcoholic solutions of GBL and beverages free from GBL. A selected wavenumber range from 1225 to 1171 cm−1 and from 999 to 972 cm−1 of the second derivative (SD) spectra of samples and standards were employed. The validation set provided root mean square error of prediction (RMSEP) of the order of 300 µg g−1 GBL with a ratio of standard error of performance to standard deviation value (RPD) of 13.4. The prediction of unknown samples not considered in the calibration nor in the validation sets offered accuracy relative errors between −1.0 and 2.9 % GBL. The proposed PLS-FTIR based method allows a fast and direct determination of GBL in all type of beverages, in less than 2 min and avoiding the use of neither reagents nor solvents, without any sample treatment required, being in agreement with the principles of Green Analytical Chemistry.

Assessment of Persian export pomegranate quality: A reliable non-destructive method based on spectroscopy and chemometrics

The present study investigated the feasibility of visible near-infrared (Vis-NIR) spectroscopy method as a low-cost, quick and reliable method for the prediction of qualitative characteristics (fruit juice percentage, soluble solids, acidity, taste index and vitamin C) of an export variety of Persian pomegranate. For this purpose, pomegranate samples were first analyzed by Vis-NIR spectroscopy, followed by destructive reference methods for quality determination. Subsequently, the data from Vis-NIR were evaluated with different pre-processing methods, among which the combination of the Savitzky–Golay (SG) smoothing model and standard normal variation (SNV) normalization model, generates the best outcomes in predicting six desired qualitative characteristics by partial least square regression (PLSR) model. Fruit juice percentage and pH presented the best prediction models with a correlation coefficient (rP) of 0.98 for both, and root mean square errors of prediction (RMSEP) equal to 0.036 and 0.041, respectively. The values of rp for all desired parameters in this study were higher than 0.95, which pointed to the excellent accuracy of the PLSR models in the prediction of pomegranate qualitative features. It can be concluded from the present study that, Vis-NIR spectroscopy using the PLSR model as a non-destructive functional approach is capable of verifying the quality of Persian export variety pomegranate as an online technique with the aim of grading products before the export phase in order to augment the product marketability.

Determination of Trace Levels of Fenbendazole in Milk, Yogurt and Curd Using Surface Enhanced Raman Spectroscopy (SERS)

A simple, rapid and sensitive surface enhanced Raman spectroscopy (SERS) method is described for the determination of fenbendazole (FBZ) residues in milk and fermented dairy products using yogurt and curd as examples. An extraction method was used to purify the samples from interfering components of the matrix. 5-Nitrothiabendazole, which has similar physicochemical properties, was used as an internal standard. The detection limits for milk were 1.8 nM, 8.8 nM for yogurt, and 13.5 nM for curd. Using partial least squares (PLS), a calibration model was built for the determination of fenbendazole from 100 to 1000 nM. The root-mean square error of calibration (RMSEC) and the root-mean square error of prediction (RMSEP) were 39.4 and 46.5 nM, respectively. The recovery rate was from 97% to 120% and the relative standard deviation was less than 5.0%. The difference between this technique from the widely used liquid chromatography–mass spectrometry (LC–MS) method is in the reduction in time for preliminary sample preparation and significantly lower costs for reagents and consumables. The article examines the influence of order when mixing reagents in a SERS experiment, which is an important tool for optimizing the sensitivity and reproducibility of the method. It was shown for the first time that when various alkaline earth metal chlorides were used for agglomeration of silver nanoparticles (AgNPs), a change in the ratio of peak intensities of the two analytes in the solution was observed. This phenomenon may be used to increase the selectivity of SERS analysis of multicomponent samples.

Intelligent color image analysis of sintered ores for simple and rapid determination of Fe3O4 concentration

A simple determination of the Fe3O4 concentrations of sintered ores using color images of the samples has been explored. Sintered ore is mainly composed of Fe2O3 (red), Fe3O4 (black), and other white inorganic oxides, so the color of sintered ore could be representative of the relative abundance of the constituents. Two important challenges were addressed to achieve reliable quantitative color image analysis. First, minute dents and bumps (embosses) exist on the sample surface due to inconsistent particle sizes and particle agglomeration, thereby generating dark shadows. Second, small white spots corresponding to inorganic oxide particles were spread throughout acquired images. The white spots yield very high RGB values, which would hamper the translation of the real sample color originating from the iron oxides. Therefore, the segmentations of particle agglomeration-induced shadows and white spots in the sample images were separately executed using Otsu's method and modified fuzzy C-means (MFCM), respectively. Then, color moments and derived variables from the segmented images were employed to determine Fe3O4 concentrations (6.5–10.5 wt%) using extreme gradient boosting (XGBoost). The predicted concentrations from the color analysis correlated well with reference concentrations determined using conventional titration, with a root mean square error of prediction (RMSEP) of 0.39 wt%.

Near-infrared spectroscopy based on colorimetric sensor array coupled with convolutional neural network detecting zearalenone in wheat

Wheat is a vital global cereal crop, but its susceptibility to contamination by mycotoxins can render it unusable. This study explored the integration of two novel non-destructive detection methodologies with convolutional neural network (CNN) for the identification of zearalenone (ZEN) contamination in wheat. Firstly, the colorimetric sensor array composed of six selected porphyrin-based materials was used to capture the olfactory signatures of wheat samples. Subsequently, the colorimetric sensor array, after undergoing a reaction, was characterized by its near-infrared spectral features. Then, the CNN quantitative analysis model was proposed based on the data, alongside the establishment of traditional machine learning models, partial least squares regression (PLSR) and support vector machine regression (SVR), for comparative purposes. The outcomes demonstrated that the CNN model had superior predictive performance, with a root mean square error of prediction (RMSEP) of 40.92 μ g ∙ kg−1 and a coefficient of determination on the prediction (RP2) of 0.91. These results affirmed the potential of integrating colorimetric sensor array with near-infrared spectroscopy in evaluating the safety of wheat and potentially other grains. Moreover, CNN can have the capacity to autonomously learn and distill features from spectral data, enabling further spectral analysis and making it a forward-looking spectroscopic tool.

Optimizing hydropower scheduling through accurate power load prediction: A practical case study

Hydropower stations that are part of the grid system frequently encounter challenges related to the uneven distribution of power generation and associated benefits, primarily stemming from delays in obtaining timely load data. This research addresses this issue by developing a scheduling model that combines power load prediction and dual-objective optimization. The practical application of this model is demonstrated in a real-case scenario, focusing on the Shatuo Hydropower Station in China. In contrast to current models, the suggested model can achieve optimal dispatch for grid-connected hydropower stations even when power load data is unavailable. Initially, the model assesses various prediction models for estimating power load and subsequently incorporates the predictions into the GA-NSGA-II algorithm, specifically an enhanced elite non-dominated sorting genetic algorithm. This integration is performed while considering the proposed objective functions to optimize the discharge flow of the hydropower station. The outcomes reveal that the CNN-GRU model, denoting Convolutional Neural Network-Gated Recursive Unit, exhibits the highest prediction accuracy, achieving R-squared and RMSE (i.e., Root Mean Square Error) values of 0.991 and 0.026, respectively. The variance between scheduling based on predicted load values and actual load values is minimal, staying within 5 (m3/s), showcasing practical effectiveness. The optimized scheduling outcomes in the real case study yield dual advantages, meeting both the demands of ship navigation and hydropower generation, thus achieving a harmonious balance between the two requirements. This approach addresses the real-world challenges associated with delayed load data collection and insufficient scheduling, offering an efficient solution for managing hydropower station scheduling to meet both power generation and navigation needs.

Advanced Machine Learning application for Permeability Prediction for (M) Formation in an Iraqi Oil Field

Permeability estimation is a vital step in reservoir engineering due to its effect on reservoir's characterization, planning for perforations, and economic efficiency of the reservoirs. The core and well-logging data are the main sources of permeability measuring and calculating respectively. There are multiple methods to predict permeability such as classic, empirical, and geostatistical methods. In this research, two statistical approaches have been applied and compared for permeability prediction: Multiple Linear Regression and Random Forest, given the (M) reservoir interval in the (BH) Oil Field in the northern part of Iraq. The dataset was separated into two subsets: Training and Testing in order to cross-validate the accuracy and the performance of the algorithms. The random forest algorithm was the most accurate method leading to lowest Root Mean Square Prediction Error (RMSPE) and highest Adjusted R-Square than multiple linear regression algorithm for both training and testing subset respectively. Thus, random Forest algorithm is more trustable in permeability prediction in non-cored intervals and its distribution in the geological model.

Cross-column density functional theory-based quantitative structure-retention relationship model development powered by machine learning.

Quantitative structure-retention relationship (QSRR) modeling has emerged as an efficient alternative to predict analyte retention times using molecular descriptors. However, most reported QSRR models are column-specific, requiring separate models for each high-performance liquid chromatography (HPLC) system. This study evaluates the potential of machine learning (ML) algorithms and quantum mechanical (QM) descriptors to develop QSRR models that can predict retention times across three different reversed-phase HPLC columns under varying conditions. Four machine learning methods-partial least squares (PLS) regression, ridge regression (RR), random forest (RF), and gradient boosting (GB)-were compared on a dataset of 360 retention times for 15 aromatic analytes. Molecular descriptors were calculated using density functional theory (DFT). Column characteristics like particle size and pore size and experimental conditions like temperature and gradient time were additionally used as descriptors. Results showed that the GB-QSRR model demonstrated the best predictive performance, with Q2 of 0.989 and root mean square error of prediction (RMSEP) of 0.749min on the test set. Feature analysis revealed that solvation energy (SE), HOMO-LUMO energy gap (∆E HOMO-LUMO), total dipole moment (Mtot), and global hardness (η) are among the most influential predictors for retention time prediction, indicating the significance of electrostatic interactions and hydrophobicity. Our findings underscore the efficiency of ensemble methods, GB and RF models employing non-linear learners, in capturing local variations in retention times across diverse experimental setups. This study emphasizes the potential of cross-column QSRR modeling and highlights the utility of ML models in optimizing chromatographic analysis.

Rapid Quality Assessment of Polygoni Multiflori Radix Based on Near-Infrared Spectroscopy

The precise and prompt determination of quality control indicators such as moisture, stilbene glycosides, and anthraquinone glycosides is crucial in assessing the quality of Polygoni Multiflori Radix. Near-infrared spectroscopy is a nondestructive analytical technique that offers a more desirable approach than traditional methods for assessing content levels. In this study, various spectral preprocessing techniques were used to preprocess the raw spectral data. The spectral data were correlated with the determination of three-component contents using the partial least squares regression (PLSR) method. Then different algorithms, such as competitive adaptive weighted sampling (CARS), Monte Carlo uninformative variable elimination (MCUVE), and random frog hopping (RF), were used for model simplification and feature selection. The data suggest that the first-order deconvolution derivative (1st Dev.) processing of the spectral data is superior to other methods in all three model evaluation metrics. The PLSR model for moisture, stilbene glycosides, and anthraquinone glycosides produced the calibration coefficient of determination (R2C) of 0.82, 0.52, and 0.58, the root mean square error of cross validation (RMSECV) of 0.91%, 0.77%, and 0.69%, the prediction coefficient of determination (R2P) of 0.72, 0.28, and 0.54, the root mean square error of prediction (RMSEP) of 0.65%, 0.81%, and 0.75%, and relative percentage differences (RPDs) of 1.7, 1.0, and 0.8. After optimizing the model using CARS, R2C increased by 0.15%, 0.41%, and 0.34%, RMSECV decreased by 0.53%, 0.32%, and 0.24%, R2P increased by 0.21%, 0.63%, and 0.35%, RMSEP decreased by 0.36%, 0.41%, and 0.31%, and RPD increased by 1.1, 0.9, and 0.6, significantly improving the predictive capacity of the model. This research provides a feasible method for rapid compliance testing of Polygoni Multiflori Radix. To further improve the model’s performance and applicability, it is necessary to continuously expand the sample set with different varieties and locations for wide variation.

Two chironomid-inferred mean July air temperature reconstructions in the South Carpathian Mountains over the last 2000 years

We present chironomid-based reconstructions of mean July air temperature changes over the last 2000 years from Lake Latoriței (1530 m a.s.l.) in the Southern Carpathians. A multi-proxy analysis was performed along a 58 cm long sediment core and two training sets were used for quantitative July air temperature reconstructions: the Eastern-European (EE, 212 lakes) and the Finnish-Polish-Carpathian (FPC, 273 lakes). The transfer functions had a coefficient of determination ( r2) 0.88 and 0.91 with a root mean squared error of prediction (RMSEP) 0.88°C and 1.02°C. Despite possible biases resulting from methodological problems and the ecological complexity of the chironomid response to both climatic and environmental changes, the agreement of the temperature reconstruction of Lake Latoriței with other alpine records suggests that the transfer function successfully reconstructed past summer temperatures between 750 and 1830 CE. Biases in the temperature reconstruction in the period before 750 and after 1830 CE were likely caused by increased abundance of rheophilic and semi-terrestrial chironomid species related to increased inflow activity before 750 CE and local land use changes after 1830 CE, which was also indicated by increasing deforestation and increasing lake productivity in the pollen and diatom records. Our results suggest that the region experienced a warm period between 750 and 1360 CE, and a cold period between 1360 and 1600 CE followed by fluctuating summer temperatures until 1830 CE. These events were associated with the so-called ‘Mediaeval Warm Period’ (MWP) and the ‘Little Ice Age’ (LIA), respectively. The inference models reconstructed a decrease in July air temperatures by 0.7°C–1.1°C during the LIA relative to the warmer MWP. We also demonstrated that the FPC training set gives better results, supporting that local/continental training sets are efficient to detect weak amplitude summer temperature changes in the Late-Holocene.

Study on rapid prediction of low concentration o-nitrotoluene in mononitrotoluene mixture by near infrared spectroscopy combined with novel calibration strategies

The determination of the o-nitrotoluene (o-MNT) content in separation process of mononitrotoluene (MNT) is of interest, since it affects the purity of m-nitrotoluene (m-MNT) and p-nitrotoluene (p-MNT). However, the analytical techniques traditionally used for its content determination are tedious and time consuming. Therefore, we explored the analysis of spectral data based on near-infrared spectroscopy (NIRS) and chemometrics, and extracted the spectral features of the o-MNT based on the interval selection algorithm. The calibration models for the o-MNT content based on samples with different concentration ranges were developed by PLS. Among them, the calibration model based on samples with 0.01–0.5 % concentration range has the best prediction performance. The calibration model was established with the determination coefficient of prediction (R2) of 0.959, root mean squared error of prediction (RMSEP) of 0.011 and ratio of standard deviation of the calibration set to standard error of prediction (RPD) of 4.899 for o-MNT. It is sufficient to meet the fast detection needs of the o-MNT content for process control in chemical industry. In addition, in order to reduce the demand of the model on samples and corresponding reference values, we explored the virtual sample generation method based on background difference compensation. The calibration model based on virtual samples was established with R2 of 0.74, RMSEP of 0.028 and RPD of 1.951. This study shows that the method based on NIRS and chemometrics has strong prediction performance for o-MNT in separation process of MNT, which is a guideline for controlling product purity of m-MNT and p-MNT. And the virtual sample generation method proposed in this study can significantly reduce the sample demand of calibration model.

Application of Multivariate Regression and Artificial Neural Network Modelling for Prediction of Physicochemical Properties of Grape-Skin Compost.

The reusability of by-products in the food industry is consistent with sustainable and greener production; therefore, the aim of this paper was to evaluate the applicability of multiple linear regression (MLR), piecewise linear regression (PLR) and artificial neural network models (ANN) to the prediction of grape-skin compost's physicochemical properties (moisture, dry matter, organic matter, ash content, carbon content, nitrogen content, C/N ratio, total colour change of compost samples, pH, conductivity, total dissolved solids and total colour change of compost extract samples) during in-vessel composting based on the initial composting conditions (air-flow rate, moisture content and day of sampling). Based on the coefficient of determination for prediction, the adjusted coefficient of determination for calibration, the root-mean-square error of prediction (RMSEP), the standard error of prediction (SEP), the ratio of prediction to deviation (RPD) and the ratio of the error range (RER), it can be concluded that all developed MLR and PLR models are acceptable for process screening. Furthermore, the ANN model developed for predicting moisture and dry-matter content can be used for quality control (RER >11). The obtained results show the great potential of multivariate modelling for analysis of the physicochemical properties of compost during composting, confirming the high applicability of modelling in greener production processes.