- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Prediction model optimization using full model selection with regression trees demonstrated with FTIR data from bovine milk
摘要： Predictive modeling is the development of a model that is best able to predict an outcome based on given input variables. Model algorithms are different processes that are used to define functions that transform the data within models. Common algorithms include logistic regression (LR), linear discriminant analysis (LDA), classification and regression trees (CART), na?ve Bayes (NB), and k-nearest neighbor (KNN). Data preprocessing option, such as feature extraction and reduction, and model algorithms are commonly selected empirically in epidemiological studies even though these decisions can significantly affect model performance. Accordingly, full model selection (FMS) methods were developed to provide a systematic approach to select predictive modeling methods; however, current limitations of FMS, such as its dependency on user-selected hyperparameters, have prevented their routine incorporation into analyses for model performance optimization. Here we present the use of regression trees as an innovative method to apply FMS. Regression tree FMS (rtFMS) requires the development of a model for every combination of predictive modeling method options under consideration. The iterated, cross-validation performances of these models are then passed through a regression tree for selection of a final model. We demonstrate the benefits of rtFMS using a milk Fourier transform infrared spectroscopy dataset, wherein we build prediction models for two blood metabolic health parameters in dairy cows, nonesterified fatty acids (NEFA) and β-hydroxybutyrate acid (BHBA). The goal for building NEFA and BHBA prediction models is to provide a milk-based screening tool for metabolic health in dairy cattle that can be incorporated automatically in milk analysis routines. These models could be used in conjunction with physical exams, cow side tests, and other indications to initiate medical intervention. In contrast to previously reported FMS methods, rtFMS is not a black box, is simple to implement and interpret, it does not have hyperparameters, and it illustrates the relative importance of modeling options. Additionally, rtFMS allows for indirect comparisons among models developed using different datasets. Finally, rtFMS eliminates user bias due to personal preference for certain methods and rtFMS removes the dependency on published comparisons of methods. Thus, rtFMS provides clear benefits over the empirical selection of data preprocessing options and model algorithms.
关键词： Prediction model,Fourier-transform infrared spectra,Regression tree,Preprocessing,Full model selection
更新于2025-09-23 15:23:52
-
Performance evaluation of a MPPT controller with model predictive control for a photovoltaic system
摘要： Efficiency has been a major factor in the growth of photovoltaic (PV) systems. Different control techniques have been explored to extract maximum power from PV systems under varying environmental conditions. This paper evaluates the performance of a new improved control technique known as model predictive control (MPC) in power extraction from PV systems. Exploiting the ability of MPC to predict future state of controlled variables, MPC has been implemented for tacking of maximum power point (MPP) of a PV system. Application of MPC for maximum power point tracking (MPPT) has been found to result into faster tracking of MPP under continuously varying atmospheric conditions providing an efficient system. It helps in reducing unwanted oscillations with an increase in tracking speed. A detailed step by step process of designing a model predictive controller has been discussed. Here, MPC has been applied in conjunction with conventional perturb and observe (P&O) method for controlling the dc-dc boost converter switching, harvesting maximum power from a PV array. The results of MPC controller has been compared with two widely used conventional methods of MPPT, viz. incremental conductance method and P&O method. The MPC controller scheme has been designed, implemented and tested in MATLAB/Simulink environment and has also been experimentally validated using a laboratory prototype of a PV system.
关键词： maximum power point tracking (MPPT),prediction model,Model predictive control (MPC),cost function,photovoltaic (PV),renewable energy
更新于2025-09-23 15:21:01
-
Reliable energy prediction method for grid connected photovoltaic power plants situated in hot and dry climatic condition
摘要： This paper presents a mathematical model to predict the energy generation of photovoltaic power plant in hot and humid climatic condition. This model is based on meteorological data and laboratory tested solar module parameters with twenty-four inputs and one output. In addition the twenty-four inputs drive an equation to calculate final energy generation from photovoltaic power plant. Validation of the proposed model was done by comparing the results of predicted energy generation using proposed model and PVWATT software model for two existing PV power plants of India. Monthly and annual energy production and errors will be the main criteria for the selection of batter model. The result shows that in comparison with PVWATT software proposed model was found to be more efficient and accurate to predict energy generation and proposed model also reduces mean absolute percentage error and root mean square error significantly compared to PVWATT software for hot and humid climatic condition.
关键词： PV power plant,Climatic condition,India,Mathematical method,Prediction model,Energy generation
更新于2025-09-19 17:13:59
-
Accurate prediction model of bead geometry in crimping butt of the laser brazing using generalized regression neural network
摘要： There are few researches that concentrate on the prediction of the bead geometry for laser brazing with crimping butt. This paper addressed the accurate prediction of the bead profile by developing a generalized regression neural network (GRNN) algorithm. Firstly GRNN model was developed and trained to decrease the prediction error that may be influenced by the sample size. Then the prediction accuracy was demonstrated by comparing with other articles and back propagation artificial neural network (BPNN) algorithm. Eventually the reliability and stability of GRNN model were discussed from the points of average relative error (ARE), mean square error (MSE) and root mean square error (RMSE), while the maximum ARE and MSE were 6.94% and 0.0303 that were clearly less than those (14.28% and 0.0832) predicted by BPNN. Obviously, it was proved that the prediction accuracy was improved at least 2 times, and the stability was also increased much more.
关键词： bead geometry,generalized regression neural network,prediction model,laser brazing
更新于2025-09-11 14:15:04
-
Detection of parameters in solid state fermentation of Monascus by near infrared spectroscopy
摘要： The prediction model was constructed using the near-infrared spectroscopy combined with the interval least squares support vector machine method (siLS-SVM) of moisture content and pH value change during the solid fermentation of Monascus. The predictive model was established with partial least squares regression (PLS), and the comprehensive performance of the model was evaluated by cross-validating the mean square error, absolute error value and relative error value. The findings suggest that the LS-SVM model established by siLS-SVM algorithm owns superior predictability and stability for the changes of water content and pH value in the solid fermentation of Monascus (the average relative error is 1.52% and 1.55%, respectively), which can be used for the accurate quantitative prediction. The results showed that near infrared spectroscopy could be used for rapid and non-destructive determination of water content and PH value in solid-state fermentation of Monascus, which provided a new way for optimization of solid-state fermentation process of Monascus under bran substrate.
关键词： Near-infrared spectroscopy,Prediction model,Monascus,Solid state fermentation
更新于2025-09-10 09:29:36
-
AIP Conference Proceedings [Author(s) GREEN DESIGN AND MANUFACTURE: ADVANCED AND EMERGING APPLICATIONS: Proceedings of the 4th International Conference on Green Design and Manufacture 2018 - Ho Chi Minh, Vietnam (29–30 April 2018)] - Quantification of acidity and total soluble solids in guavas by near infrared hyperspectral imaging
摘要： In order to provide premium quality for marketing of guavas the titratable acidity (TA) and total soluble solids (TSS) levels should be determined. A reflectance near infrared hyperspectral imaging (NIR-HSI) unit in the wavelength range of 936-1696 nm, which is a nondestructive technique, was tested for use in predicting TA and TSS. Samples of 100 guavas were scanned by NIR-HIS as a group for calibration (N=67) and as a group for prediction (N=33). The average spectra from the region of interest ( ROI) of samples were used to establish the calibration models for TA and TSS by using partial least squares regression ( PLSR) to establish calibration models. The calibration model for TA gave a coefficient of determination (R2) of 0.972 and the root mean square error of prediction (RMSEP) of 0.010% and for TSS the R2 was 0.801 and the RMSEP was 0.437oBx. The accuracies of these results indicate that NIR-HSI has potential for use in measuring TA and TSS of guavas.
关键词： calibration,prediction,model,nondestructive,near infrared
更新于2025-09-09 09:28:46
-
[Institution of Engineering and Technology 12th European Conference on Antennas and Propagation (EuCAP 2018) - London, UK (9-13 April 2018)] 12th European Conference on Antennas and Propagation (EuCAP 2018) - Numerical Weather Prediction Models for the Estimate of Clear-Sky Attenuation Level in Alphasat Beacon Measurement
摘要： With the move of satellite systems towards Ka and QIV bands, the Alphasat TDPS Aldo Paraboni scientific experiment aims to characterize atmospheric attenuation in those bands. However, during the retrieval of the attenuation from the measured beacon signal, the clear-sky contribution to attenuation is lost. Microwave radiometers give the clear sky absolute reference level, but these are costly and not always available. This paper proposes Numerical Weather Prediction models as an alternative source of clear-sky attenuation. Three months of beacon and radiometric data from Spino d' Adda the Alphasat receiving station are used as benchmark for validation of the method. A preliminary conclusion is that gaseous attenuation is well predicted, but cloud attenuation is underestimated.
关键词： microwave radiometer,Numerical Weather Prediction model,radiowave propagation,clear-sky attenuation,Alphasat
更新于2025-09-04 15:30:14
-
Reduction of Forced Outages in Islanded Microgrids by Compensating Model Uncertainties in PV Rating and Battery Capacity
摘要： Energy management systems for islanded microgrids often rely on predictions of energy availability and usage. Such predictions can be used to plan actions, such as shedding non-essential loads, so that critical loads continue to be served. However, uncertainties in the prediction models may lead to incorrect decisions, and subsequently jeopardize reliable operation of the microgrid. For a photovoltaic (PV) and battery based microgrid, uncertainties in the PV rating and the battery capacity model parameters can lead to otherwise avoidable outages. In this paper, techniques have been developed to identify and compensate for such model uncertainties. The approach uses differences between the actual and predicted data sequences to determine compensation factors to improve prediction accuracy. The developed techniques account for operating condition changes automatically, and no additional sensors are needed for their implementation. The method has been evaluated using data from rooftop irradiance and temperature sensors and the corresponding forecasts. It has been shown that the proposed techniques can improve the accuracy of the predictions and hence lead to more effective energy management decisions. Together with a pre-emptive load shedding strategy, the total outage time of the microgrid can be shortened by as much as 11% for the chosen scenario.
关键词： Energy management,photovoltaic systems,prediction model uncertainties,microgrids
更新于2025-09-04 15:30:14