Miguel Velez-Reyes

Nonlinear control of a heating, ventilating, and air conditioning system with thermal load estimation

This paper presents a nonlinear controller for a heating, ventilating, and air conditioning (HVAC) system capable of maintaining comfort conditions under time varying thermal loads. The controller consist of a regulator and a disturbance rejection component designed using Lyapunov stability theory. The mitigation of the effect of thermal loads other than design loads on the system is due to an online thermal load and state estimator. The availability of the thermal load estimates allows the controller to keep comfort regardless of the thermal loads affecting the thermal space being heated or cooled. Simulation results are used to demonstrate the potential for keeping comfort and saving energy of this methodology on a variable-air-volume HVAC system operating on cooling mode.

Unsupervised Linear Feature-Extraction Methods and Their Effects in the Classification of High-Dimensional Data

This paper presents an analysis and a comparison of different linear unsupervised feature-extraction methods applied to hyperdimensional data and their impact on classification. The dimensionality reduction methods studied are under the category of unsupervised linear transformations: principal component analysis, projection pursuit (PP), and band subset selection. Special attention is paid to an optimized version of the PP introduced in this paper: optimized information divergence PP, which is the maximization of the information divergence between the probability density function of the projected data and the Gaussian distribution. This paper is particularly relevant with current and the next generation of hyperspectral sensors that acquire more information in a higher number of spectral channels or bands when compared to multispectral data. The process to uncover these high-dimensional data patterns is not a simple one. Challenges such as the Hughes phenomenon and the curse of dimensionality have an impact in high-dimensional data analysis. Unsupervised feature extraction, implemented as a linear projection from a higher dimensional space to a lower dimensional subspace, is a relevant process necessary for hyperspectral data analysis due to its capacity to overcome some difficulties of high-dimensional data. An objective of unsupervised feature extraction in hyperspectral data analysis is to reduce the dimensionality of the data maintaining its capability to discriminate data patterns of interest from unknown cluttered background that may be present in the data set. This paper presents a study of the impact these mechanisms have in the classification process. The impact is studied for supervised classification even on the conditions of a small number of training samples and unsupervised classification where unknown structures are to be uncovered and detected

Subset selection in identification, and application to speed and parameter estimation for induction machines

A method to determine which parameters of a model are numerically identifiable is presented. With this method, parameters are separated into ill-conditioned and well-conditioned parameters. Prior information about ill-conditioned parameters can be incorporated into the estimation process resulting in sensitivity reduction and improved numerical performance of estimation algorithms. The method is an extension to nonlinear models of subset selection methods developed in linear regression. The results are illustrated by application to the case of speed and parameter estimation for induction machine. The insights provided by our parameter subset selection approach are of decisive value in this application.

Comparative Study of Semi-Implicit Schemes for Nonlinear Diffusion in Hyperspectral Imagery

Nonlinear diffusion has been successfully employed over the past two decades to enhance images by reducing undesirable intensity variability within the objects in the image, while enhancing the contrast of the boundaries (edges) in scalar and, more recently, in vector-valued images, such as color, multispectral, and hyperspectral imagery. In this paper, we show that nonlinear diffusion can improve the classification accuracy of hyperspectral imagery by reducing the spatial and spectral variability of the image, while preserving the boundaries of the objects. We also show that semi-implicit schemes can speedup significantly the evolution of the nonlinear diffusion equation with respect to traditional explicit schemes

Interest Points for Hyperspectral Image Data

Interest points are widely used as point-features for image matching. This paper describes robust and efficient algorithms to extract multiscale interest points in hyperspectral images in which structural information is distributed across several spectral bands. The formulation is based on a Gaussian scale-space representation of the hyperspectral data cube, and the use of a principal components decomposition to combine information efficiently across spectral bands. A spectral distance measure is used to characterize spatial relations between neighboring hyperspectral pixels. In addition, we describe methods for preprocessing a pair of hyperspectral images, clustering the spectral signatures of interest points, and using the resulting data for matching points under simple geometric transformations. The stability of the resulting interest points in time-lapse satellite images was determined to be in the range of 52% to 75% in the testing data set that were acquired from variety of landforms like coastal islands of La Parguera, Chesapeake Bay, the Cuprite Mining District of Nevada, and agricultural field images of Kansas and Oklahoma, and thus, they can be used as a foundation for image matching and related image analysis tasks.

Multiscale Representation and Segmentation of Hyperspectral Imagery Using Geometric Partial Differential Equations and Algebraic Multigrid Methods

A fast algorithm for multiscale representation and segmentation of hyperspectral imagery is introduced in this paper. The multiscale/scale-space representation is obtained by solving a nonlinear diffusion partial differential equation (PDE) for vector-valued images. We use algebraic multigrid techniques to obtain a fast and scalable solution of the PDE and to segment the hyperspectral image following the intrinsic multigrid structure. We test our algorithm on four standard hyperspectral images that represent different environments commonly found in remote sensing applications: agricultural, urban, mining, and marine. The experimental results show that the segmented images lead to better classification than using the original data directly, in spite of the use of simple similarity metrics and piecewise constant approximations obtained from the segmentation maps.

Unmixing Analysis of a Time Series of Hyperion Images Over the Guánica Dry Forest in Puerto Rico

This paper presents the analysis of a time series of hyperspectral images collected with the Hyperion sensor on board EO-1 to demonstrate how hyperspectral imaging can be used for studying seasonal variations of vegetation cover over the Guánica Dry Forest in Puerto Rico. The approach is based on a local unmixing procedure that splits the hyperspectral scene into tiles and performs endmember extraction on each tile. The main assumption is that within a tile, a single spectral signature is an adequate representation of an endmember. Local endmember signatures from each tile are then clustered to extract endmember classes that better account for endmember spectral variability across the scene and provide a better global description of the full forest scene. Within a scene, abundances are computed using all extracted spectral endmembers and the abundance of an endmember class is computed as the sum of the abundances for the spectral endmembers belonging to that class. Variations in abundance maps are used to understand seasonal changes in forest cover. The procedure was performed using eleven near-cloud-free Hyperion images collected in different months in 2008. Results from the analysis agreed with published knowledge of the phenological changes for this forest. Correlation analyses with NDVI and rainfall time series are used to understand variations in coverage of certain endmember classes with weather. Mangrove was shown to be uncorrelated with rainfall, whereas the upland forest endmember was highly correlated with rain. This study shows the potential for unmixing methods to exploit hyperspectral data for temporal analysis.

Decoupled control of temperature and relative humidity using a variable-air-volume HVAC system and non-interacting control

In this paper, we develop a nonlinear noninteracting control system for temperature and relative humidity in a thermal-space conditioned by a variable-air-volume (VAV) heating, ventilating, and air conditioning (HVAC) system. In some industrial processes it is desirable to be able to control temperature and relative humidity independently and accurately. When the controller does not take into account the coupling dynamics between these variables, it is impossible to set one without affecting the other, therefore the importance of decoupling techniques to achieve accurate control. We demonstrate how decoupled control of temperature and relative humidity is possible using a multivariable cascade control with two loops. The inner-loop is the noninteracting control law used for decoupling, and the outer-loop is a PD controller used for stabilization and control. Simulations are presented at the end of the paper in order to validate the theoretical results.

Subset selection analysis for the reduction of hyperspectral imagery

Presents the formulation of the dimension reduction problem using subset selection as a matrix approximation problem. A heuristic algorithm to solve this problem is presented. Numerical results using LANDSAT and AVIRIS images show that the selected bands are contained in a space that is almost aligned with the first few principal components.

Hierarchical control of Hybrid Power Systems

This paper presents a hierarchical controller for a DC bus configuration hybrid power system consisting of a wind turbine, photovoltaic panels, a battery bank, capability for connection to grid, and load. The proposed hierarchical control system consisted of lower level controllers for each generation and storage component designed using sliding mode control and a supervisory controller that had the objective of maximizing use of the renewable sources while minimizing connection to the grid. The supervisory controller changed sliding surfaces and set points according to the operation conditions and performed load shedding based on load priorities when available generation is insufficient. The functionality and performance of the system is studied by means of simulations. The paper will present HPS model development, hierarchical controller design, and simulation results under the different operating scenarios. Simulation results show that the proposed controller has the capability to meet the design objectives with acceptable performance.