https://hdl.handle.net/10259/6164 Mon, 20 Apr 2026 08:10:15 GMT 2026-04-20T08:10:15Z https://hdl.handle.net/10259/8281 Principal component regression that minimizes the sum of the squares of the relative errors: Application in multivariate calibration models Valencia García, Olga; Ortiz Fernández, Mª Cruz; Sarabia Peinador, Luis Antonio Relative errors are typically used in chemometrics to evaluate the performance of a multivariate predictive model. However, these models are not obtained through the criterion of minimizing relative errors, as would be expected in a model whose response is the concentration of an analyte. There are no studies in chemometrics on the use of a principal component regression that minimizes the sum of the squares of the relative errors. This work proposes a model, which serves this purpose. The suggested model, wPCR, has been applied to 7 datasets with 12 predicted responses, 10 of which are multivariate calibrations of analytes in complex mixtures based on instrumental signals coming from various analytical techniques. As PCR and wPCR are methods seeking to optimize different criteria, each one achieves a better performance with respect to its own criterion. Therefore, the new model wPCR leads to better results insofar as the relative errors are considered, especially for the smallest responses. In this sense, the wPCR model also outperforms PCR with logarithmic transformation of the response (logPCR). In addition, as for the performance of the method using Joint Confidence Regions for the intercept and the slope of the accuracy line, it is shown that the application of wPCR does not introduce bias, neither constant nor proportional for the models built, nor a systematic alteration of the achievable accuracy. Fri, 01 Jan 2021 00:00:00 GMT https://hdl.handle.net/10259/8281 2021-01-01T00:00:00Z https://hdl.handle.net/10259/7127 Integrated Design of a Supermarket Refrigeration System by Means of Experimental Design Adapted to Computational Problems Sarabia Ortiz, Daniel; Ortiz Fernández, Mª Cruz; Sarabia Peinador, Luis Antonio In this paper, an integrated design of a supermarket refrigeration system has been used to obtain a process with better operability. It is formulated as a multi-objective optimization problem where control performance is evaluated by six indices and the design variables are the number and discrete power of each compressor to be installed. The functional dependence between design and performance is unknown, and therefore the optimal configuration must be obtained through a computational experimentation. This work has a double objective: to adapt the surface response methodology (SRM) to optimize problems without experimental variability as are the computational ones and show the advantage of considering the integrated design. In the SRM framework, the problem is stated as a mixture design with constraints and a synergistic cubic model where a D-optimal design is applied to perform the experiments. Finally, the multi-objective problem is reduced to a single objective one by means of a desirability function. The optimal configuration of the power distribution of the three compressors, in percentage, is (50,20,20). This solution has an excellent behaviour with respect to the six indices proposed, with a significant reduction in time oscillations of controlled variables and power consumption compared with other possible power distributions. Tue, 01 Nov 2022 00:00:00 GMT https://hdl.handle.net/10259/7127 2022-11-01T00:00:00Z https://hdl.handle.net/10259/6322 Pixel-Based Image Processing for CIE Standard Sky Classification through ANN Granados López, Diego; García Rodríguez, Ana; García Rodríguez, Sol; Suárez García, Andrés; Diez Mediavilla, Montserrat; Alonso Tristán, Cristina Digital sky images are studied for the definition of sky conditions in accordance with the CIE Standard General Sky Guide. Likewise, adequate image-processing methods are analyzed that highlight key image information, prior to the application of Artificial Neural Network classification algorithms. Twenty-two image-processing methods are reviewed and applied to a broad and unbiased dataset of 1500 sky images recorded in Burgos, Spain, over an extensive experimental campaign. The dataset comprises one hundred images of each CIE standard sky type, previously classified from simultaneous sky scanner data. Color spaces, spectral features, and texture filters image-processing methods are applied. While the use of the traditional RGB color space for image-processing yielded good results (ANN accuracy equal to 86.6%), other color spaces, such as Hue Saturation Value (HSV), which may be more appropriate, increased the accuracy of their global classifications. The use of either the green or the blue monochromatic channels improved sky classification, both for the fifteen CIE standard sky types and for simpler classification into clear, partial, and overcast conditions. The main conclusion was that specific image-processing methods could improve ANN-algorithm accuracy, depending on the image information required for the classification problem. Wed, 01 Dec 2021 00:00:00 GMT https://hdl.handle.net/10259/6322 2021-12-01T00:00:00Z https://hdl.handle.net/10259/6218 Modelling Photosynthetic Active Radiation (PAR) through meteorological indices under all sky conditions García Rodríguez, Ana; Granados López, Diego; García Rodríguez, Sol; Diez Mediavilla, Montserrat; Alonso Tristán, Cristina In this study, ten-minute meteorological data-sets recorded at Burgos, Spain, are used to develop models of Photosynthetic Active Radiation () following two different procedures: multilinear regression and Artificial Neural Networks. Ten Meteorological Indices (MIs) are chosen as inputs to the models: clearness index (), diffuse fraction (), direct fraction (), Perez's clear sky index (ɛ), brightness index (), cloud cover (), air temperature (), pressure (), solar azimuth cosine (), and horizontal global irradiation (). The experimental data are clustered according to the sky conditions, following the CIE standard sky classification. A previous feature selection procedure established the most adequate MIs for modelling in clear, partial and overcast sky conditions. was the common MI used by all models and for all sky conditions. Additional variables were also included: the geometrical parameter, , and three variables related to the sky conditions, , and Both modelling methods, multilinear regression and ANN, yielded very high determination coefficients () with very close results in the models for each of the different sky conditions. Slight improvements can be observed in the ANN models. The results underline the equivalence of multilinear regression models and ANN models of PAR following previous feature selection procedures. Mon, 01 Nov 2021 00:00:00 GMT https://hdl.handle.net/10259/6218 2021-11-01T00:00:00Z