Energy, Controls, and Applications Lab

We present an architecture for peer-to-peer energy markets which can guarantee that operational constraints are respected and payments are fairly rendered, without relying on a centralized utility or microgrid aggregator. We demonstrate how to address trust, security, and transparency issues by using blockchains and smart contracts, two emerging technologies which can facilitate decentralized coordination between non-trusting agents.  While blockchains are receiving considerable interest as a platform for distributed computation and data management, this is the first work to examine their use to facilitate distributed optimization and control. Using the Alternating Direction Method of Multipliers (ADMM), we pose a decentralized optimal power flow (OPF) model for scheduling a mix of batteries, shapable loads, and deferrable loads on an electricity distribution network. The DERs perform local optimization steps, and a smart contract on the blockchain serves as the ADMM coordinator, allowing the validity and optimality of the solution to be verified. The optimal schedule is securely stored on the blockchain, and payments can be automatically, securely, and trustlessly rendered without requiring a microgrid operator.

Lithium-ion (Li-ion) battery faults or failure mechanisms are potentially hazardous to battery health, safety and performance. Thermal fault mechanisms represent a critical subset of such failures. To ensure safety and reliability, battery management systems must have the capability of diagnosing these thermal failures. In line with this requirement, we present a Partial Differential Equation (PDE) model-based scheme for diagnosing thermal faults in Li-ion batteries. For this study, we adopt a distributed parameter one-dimensional thermal model for cylindrical battery cells. The diagnostic scheme objective is to detect and estimate the size of the thermal fault. The scheme consists of two PDE observers arranged in cascade with measured surface temperature feedback. The first observer, denoted as Robust Observer, estimates the distributed temperature inside the cell under nominal (healthy) and faulty conditions. The second observer, denoted as Diagnostic Observer, receives this estimated temperature distribution, and in turn outputs a residual signal that provides the fault information. Furthermore, the residual signal is evaluated against non-zero thresholds to achieve robustness against modeling and measurement uncertainties. Lyapunov stability theory has been utilized to verify the analytical convergence of the observers under heathy and faulty conditions. Simulation studies are presented to illustrate the effectiveness of the proposed scheme.

In this project, we study an Electric Vehicle Routing Problem with Pick-ups and Deliveries, Time Windows, and Recharging Stations on New York City Taxicab data. In order to solve this problem, we divide the problems into three phases: (i) grouping similar customer requests by identifying geographic zones and time slots; (ii) determine groups of passengers to be transported together; (iii) complete the vehicle itinerary between these groups of passengers. The first phase uses the clustering method k-means  on the locations of pick-ups and deliveries of New York City taxicabs in january 2013. The second phase uses exact optimization methods, while the third uses metaheuristics methods.

Model predictive control (MPC) strategies show great potential for improving the performance and energy efficiency of building heating, ventilation, and air-conditioning (HVAC) systems. A challenge in the deployment of such predictive thermostatic control systems is the need to learn accurate models for the thermal characteristics of individual buildings. This necessitates the development of online and data-driven methods for system identification. In this paper, we propose a piecewise linear thermal model of a building. To learn the model, we present a Kalman filter based approach for estimating the parameters. Finally, we fit the piecewise linear model to data collected from a residential building with a forced-air heating and ventilation system and validate the accuracy of the trained model.

Energy systems (e.g. ventilation fans, refrigerators, and electrical vehicle chargers) often have binary or discrete states due to hardware limitations and efficiency characteristics. Typically, such systems have additional programmatic constraints, such as minimum dwell times to prevent short cycling. As a result, non-convex techniques, like dynamic programming, are generally required for optimization. Recognizing developments in the field of distributed convex optimization and the potential for energy systems to participate in ancillary power system services, it is advantageous to develop convex techniques for the approximate optimization of energy systems. In this manuscript, we develop the alternative control trajectory representation -- a novel approach for representing the control of a non-convex discrete system as a convex program. The resulting convex program provides a solution that can be interpreted stochastically for implementation.

The short-term forecasting of building electricity demand is certain to play a vital role in the future power grid. Given the deployment of intermittent renewable energy sources and the ever increasing consumption of electricity, the generation of accurate demand-side electricity forecasts will be valuable to both grid operators and building energy management systems. The literature is rich with forecasting models for individual buildings. However, an ongoing challenge is the development of a broadly applicable method for electricity forecasting across geographic locations, seasons, and use-types. This paper addresses the need for a generalizable approach to electricity demand forecasting through the formulation of a stacking ensemble learning method. Rather than using a single model to predict electricity demand, our method uses a weighted linear combination of forecasts from multiple sub-models. By learning the model weights in real-time using electricity demand data streams and a moving horizon training technique, the method is more robust than a single model approach. By applying our method to electricity demand data sets for 8 different buildings, we show that this data-driven approach is capable of producing accurate multivariate forecasts for building level applications.

For thermostatically controlled loads (TCLs) to perform demand response services in real-time markets, recursive methods for parameter estimation are needed.  As the physical characteristics of a TCL change (e.g. the contents of a refrigerator or the occupancy of a conditioned room), it is necessary to update the parameters of the TCL model. Otherwise, the TCL will be incapable of accurately predicting its potential energy demand, thereby decreasing the reliability of a TCL aggregation to perform demand response. In this paper, we investigate the potential of an unscented Kalman filter (UKF) algorithm to identify a TCL model that is non-linear in the parameters. Experimental results demonstrate the parameter estimation of two residential refrigerators.

A fundamental requirement of the electric power system is to maintain a continuous and instantaneous balance between generation and load. The intermittency and uncertainty introduced by renewable energy generation requires the expansion of ancillary power system services to maintain such a balance. In this paper, we examine the potential of thermostatically controlled loads (TCLs), such as refrigerators and electric water heaters, to provide generation following services in real-time energy markets (1 to 5 minutes). To simulate the non-linear dynamics of hysteretic dead-band systems in a manner suitable for convex optimization, we introduce an alternative control trajectory representation of the TCLs and their discrete input signals. To perform distributed optimization across large populations of TCLs, we propose a variation of the alternating direction method of multipliers (ADMM) algorithm. Based on our simulation results, we have demonstrated the potential for controlling a population of TCLs within an error tolerance of 10 kW.

This paper studies a nonlinear predictive energy management strategy for a residential building with a rooftop photovoltaic (PV) system and second-life lithium-ion battery energy storage. A key novelty of this manuscript is closing the gap between building energy management formulations, advanced load forecasting techniques, and nonlinear battery/PV models. Additionally, we focus on the fundamental trade-off between lithium-ion battery aging and economic performance in energy management. The energy management problem is formulated as a model predictive controller (MPC). Simulation results demonstrate that the proposed control scheme achieves 96%-98% of the optimal performance given perfect forecasts over a long-term horizon. Moreover, the rate of battery capacity loss can be reduced by 25% with negligible losses in economic performance, through an appropriate cost function formulation.

This paper studies a state estimation scheme for a reduced electrochemical battery model, using voltage and current measurements. Real-time electrochemical state information enables high-fidelity monitoring and high-performance operation in advanced battery management systems, for applications such as consumer electronics, electrified vehicles, and grid energy storage. This paper derives a single particle model with electrolyte (SPMe) that achieves higher predictive accuracy than the single particle model (SPM). Next, we propose an estimation scheme and prove estimation error system stability, assuming the total amount of lithium in the cell is known. The state estimation scheme exploits dynamical properties such as marginal stability, local invertibility, and conservation of lithium. Simulations demonstrate the algorithm's performance and limitations.