Sistema Regional de Información
en línea para Revistas Científicas de América Latina,
el Caribe, España y Portugal

In this work a Luenberger observer (LO) for type 1 diabetes is established using the Hovorka’s model (HM). The HM is linearized around an operating point and the eigenvalues are calculated.  The LO is designed relocating the HM eigenvalues through the Ackermann’s methodology for linear observers where the proposed LO keeps the nonlinear structure of the model system. The LO is parameterized and tuned with the mean from six virtual patients of HM. Once the observer performance is reliable estimating the state space variables for HM, the virtual patients are changed by patients of Bergman’s model in order to test the observer behavior under unknown dynamics. These estimated variables constitute the ones corresponding to HM. The variables are estimated by the data computational processing which correspond    to the insulin (input) and glucose (output) of the virtual patients. The estimated variables by the LO are very similar for virtual patients generated by both models, where the parameter FIT is used to quantify    the performance of the observer. The computational implementation of the LO is useful tool to estimate the unmeasured variables in diabetic patients so they can be used in the artificial   pancreas.

One of the most popular current methods of improving there liability of data communication systems is error-correcting codes to correct decoding errors that occur during data transmission on communication channels by introducing some redundancy in their encoding. This paper describes the main steps of error corrections using the Redundant Residue Number System as the data encryption. The features of the implementation of algorithms in terms of reducing the time delay and hardware costs are presented.

Residue Number System (RNS) allows performing computation more efficiently. Natural parallelism of representation and processing of numbers makes this number system suitable for applying to high performance computing. We address the main features of application of RNS to high-performance parallel computing. We consider and analyze different stages of data processing in RNS. Based on this analysis, we describe the process of decomposition of algorithms using RNS

Physical, statistical models or a combination of both are used for the wind power prediction. Physical models considered meteorological and geophysical data to determine the value of the speed of the wind and with this power generation; statistical models, on the other hand, used historical data of electric generation. The latter integrated wind speed obtained from a numerical model. If the wind speed is not forecast within a range of acceptable error, power generation forecast will be affected in a critical way. This study presents the development of a hardware-software infrastructure to provide a short-term wind forecast, 4 times a day using the model Regional Atmospheric Modeling System (RAMS) and the Weather Research and Forecasting (WRF) 1 km resolution in an area of 1344 km2 located in the South of the Isthmus of Tehuantepec, Oaxaca. From the models are obtained datasets at the height of 80 m. Databases used as initial conditions and frontier models are data from the Global Forecast System (GFS) of the National Centers for Environmental Prediction (NCEP) for a period of one year. As the technique of adjustment models of prognosis numerical of weather (NWP) was implemented Kalman filter algorithm trying to eliminate systematic errors that are generated when modeling at levels close to the Earth's surface. As an option of statistical models were 3 models Autoregressive Integrated Moving Average (ARIMA) using historical data of wind speed. Forecast of the wind speed of all configurations was validated by comparing it with data measured at 80 m with a weather station located in the area, which belongs to the National Institute of electricity and clean energies (INEEL). Chai and Draxler [4] recommended using more than one metric to validate the models. The statistics were used in this study: mean absolute deviation (MAD), the mean absolute error (MAE) and the root of the mean square error (RMSE). The results show that the best model of forecasting for the period of 5 days is the WRF with an average RMSE of 2.48 and MAE average of 1.7. Forecasts 24 hours the best choice turned out to be the Kalman filter applied to the outputs of the RAMS model. This model shows the mean values of RMSE 1.74 and MAE of 1.32. Taking into account these results were operationally configured models and in a geographical information system provided the best forecasts 4 times a day every 6 hour. As future work in the short term is planned to make the forecast of wind and comparison of actual power generation with the forecasted for the wind turbine KWT300 (300 kW) located within the Regional Centre of Technology (CERTE) of the INEEL.

In Smart cities it is essential the development of information systems that collaborate in the measurement of the urban surroundings towards the cities’ sustainability. In this research, for the key performance indicators it is proposed a pattern’s visualization of efficiency metrics tool, utilizing the auto learning techniques “machine learning”. The objective is to give support to the decision making throughout the georeferenced analysis exploiting the Open Data. The research was applied to the primary public schools data study case, including four stages: the study of metrics, the search of the data model, the test of territorial dependency, and the development of the tool that applies the grouping techniques or clustering to compare the development and school resources by zone. In the tool, the kmeans algorithm is implemented with label as validation method to select the more relevant centroids to display on a map.

The Physical Properties Estimation Problem of Ionic Liquids (PPEPILs) arises from the need of designing Ionic Liquids (ILs) for specific tasks. It is important to emphasize that the synthesis of ILs is generally expensive and time-consuming. Furthermore, the number of possible ionic liquids that can be synthesized is extremely large. The purpose of PPEPILs is to avoid the experimental synthesis of Ionic Liquids (ILs) estimating their physical properties. Moreover, to estimate the melting temperature is the most difficult task. This problem has attracted the attention of interdisciplinary researchers due to their relevant applications such as their usages as catalysts and solvents. Additionally, the ILs are relevant due to their distinctive characteristics and reduced toxicity. This problem is particularly complex since the behavior of ILs is unconventional and the available information may not be accurate. This paper presents a new approach for the PPEPILs based on neuroevolutionary neural networks using molecular descriptors to predict the melting temperatures of ILs with encouraging results. Neuroevolutionary networks had been previously used in diverse areas of knowledge and present advantages over classic Neural Networks.

This paper proposes a new criterion for reducing the processing time of the assignment of data points to clusters for algorithms of the k-means family, when they are applied to instances where the number n of points is large. Our criterion allows a point to be classified in an early stage, excluding it from distance calculations to cluster centroids in subsequent iterations. The proposed criterion uses knowledge of the distance of a point to its two closest centroids and their shifts in the last two iterations. By computer experimentation using synthetic and real instances, we found that this criterion reduces execution time to approximately 2/100 of the time by k-means and generates solutions whose quality is approximately reduced by less than 3%. These findings suggest the usefulness of our criterion for problems like those found in Big Data. The NP-hardness of k-means motivates the use of this heuristics.

The Vehicle Routing Problem or VRP is an approach represented by the problems that faces a vehicle to transport goods on a route (origin-destination) under a defined time and distance. An instance is a set of data prepared specifically in order for analysis and exploration (Column 1 indicates the number of nodes, usually the number 1 is the depot, column 2 is the x coordinate, column 3 is the y-coordinate, column 4 is the demand to be covered by the node) that it was done in MATLAB R2014a software which runs the algorithm VRP with capacities (CVRP), with the structure already mentioned. For their analysis was necessary to use instances obtained from NEO. This paper presents a comparative between instances to solve CVRP and determine which one offers the best solution.