There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

The purpose of this paper is to provide a comprehensive presentation and interpretation of the Ensemble Kalman Filter (EnKF) and its numerical implementation. The EnKF has a large user group, and numerous publications have discussed applications and theoretical aspects of it. This paper reviews the important results from these studies and also presents new ideas and alternative interpretations which further explain the success of the EnKF. In addition to providing the theoretical framework needed for using the EnKF, there is also a focus on the algorithmic formulation and optimal numerical implementation. A program listing is given for some of the key subroutines. The paper also touches upon specific issues such as the use of nonlinear measurements, in situ profiles of temperature and salinity, and data which are available with high frequency in time. An ensemble based optimal interpolation (EnOI) scheme is presented as a cost-effective approach which may serve as an alternative to the EnKF in some applications. A fairly extensive discussion is devoted to the use of time correlated model errors and the estimation of model bias.

The Ensemble Kalman Filter: Theoretical formulation and practical implementation

Economic Control of Quality of Manufactured Product.

https://users.encs.concordia.ca/~ymzhang/publications/ARC32-2-98Zhang_pp.229-252.pdf

Bibliographical review on reconfigurable fault-tolerant control systems

Lexicographic Optimization Based Sensor Network Design for Robust Fault Diagnosis

Lithium ion batteries are one of the most commercially used batteries. Though they are widely used in mobile phones, laptops and other consumer electronics, concerns related to their safety and efficient operation still persists. One of the major issues in Li-ion batteries is the capacity degradation with aging due to various mechanisms such as solid electrolyte interphase (SEI) formation and dissolution, thermal runaway, and Li-plating. In this work, we describe a capacity fade minimizing model predictive control approach for identification and realization of optimal charge-discharge cycles for Li-ion batteries. Optimum charging profiles are obtained such that the reduction in charge carrying capacity with cycling is minimized, while still obtaining required charging. We expect the proposed strategy to improve battery capacity and prolong lifespan. Examples that demonstrate the significance of the proposed framework by comparing battery performance with and without the presence of controller are discussed. Extensions to this work in terms of addressing various battery failure mechanisms, on-line identification of failure mechanisms, and designs for on-line implementation in real battery systems are also outlined.

Capacity Fade Minimizing Model Predictive Control Approach for the Identification and Realization of Charge-Discharge Cycles in Lithium Ion Batteries

Droplet-based microfluidics is an emerging area that has important applications in chemical engineering and several other disciplines. For example, these devices have applications in the areas of bubble computers, cancer diagnostics and development of microfactories on a chip. Some of the challenges that need to be overcome to realize the potential of this area are related to experimental uncertainties and development of devices that offer multiple functionalities, thereby reducing the footprint and aiding in miniaturization. Achieving precise functionalities in microfluidics is challenging because droplets exhibit complex dynamic behavior in these devices due to hydrodynamic interactions and discontinuities that are a result of discrete decision-making at junctions. Remarkably, transitions from periodic to aperiodic/chaotic behavior based on input conditions can be witnessed even in a simple loop device. Hence rational design frameworks that handle this complexity are required to make significant progress. Further, passive designs that achieve such functionalities might entail large footprint. As a result, active control might become a necessity for developing efficient device designs. The main purpose of this paper is the identification of the challenges and opportunities for the PSE community in this area of emerging importance. Further, our work in the area model predictive control (MPC) for active control in a microfluidic loop device will also be briefly described.

Modeling and Control Challenges in the development of Discrete Microfluidic Devices

Quick fault detection and diagnosis is very important for effective process operations management. Qualitative trend analysis (QTA), a data-driven semiquantitative technique, has been widely used for fault diagnosis (FD). Though QTA provides quick and accurate diagnosis - the increase in computational complexity of QTA with the increase in the number of sensors used for diagnosis - may prohibit its real-time application for very large-scale plants. In most of the chemical plants, the measurements are highly redundant and this redundancy can be exploited by performing principal component analysis (PCA) on the measured data. In this paper, we present a PCA-QTA technique for fault diagnosis in large-scale plants to reduce the computation time. Essentially, QTA is applied on the principal components rather than on the sensor data. The results are quite promising for the Tennessee Eastman (TE) process

Raghunathan Rengaswamy

Papers

Lexicographic Optimization Based Sensor Network Design for Robust Fault Diagnosis

Capacity Fade Minimizing Model Predictive Control Approach for the Identification and Realization of Charge-Discharge Cycles in Lithium Ion Batteries

Modeling and Control Challenges in the development of Discrete Microfluidic Devices

Qualitative trend analysis of the principal components: application to fault diagnosis

Reinforcement-Learning designs droplet microfluidic networks