PATH

Gated Ensemble Learning Method for Demand-Side Electricity Load Forecasting

Burger, Eric M.
Scott Moura
2015

The 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 building-level electricity demand 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 demand forecasting across geographic locations, seasons, and use-...

Piecewise Linear Thermal Model and Recursive Parameter Estimation of a Residential Heating System

Burger, Eric M.
Perez, Hector E.
Scott Moura
2016

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...

Battery Charge Control with an Electro-Thermal-Aging Coupling

Hu, Xiaosong
Perez, Hector E.
Scott Moura
2016

Efficient and safe battery charge control is an important prerequisite for large-scale deployment of clean energy systems. This paper proposes an innovative approach to devising optimally health-conscious fast-safe charge protocols. A multi-objective optimal control problem is mathematically formulated via a coupled electro-thermal-aging battery model, where electrical and aging sub-models depend upon the core temperature captured by a two-state thermal sub-model. The Legendre-Gauss-Radau (LGR) pseudo-spectral method with adaptive multi-mesh-interval collocation is employed to solve the...

Building Electricity Load Forecasting via Stacking Ensemble Learning Method with Moving Horizon Optimization

Burger, Eric M.
Scott Moura
2015

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...

Optimal Routing and Charging of Electric Ride-Pooling Vehicles in Urban Networks

Nicolas, Léa
Scott Moura
2016

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...

Nonlinear Predictive Energy Management of Residential Buildings with Photovoltaics & Batteries

Sun, Chao
Sun, Fengchun
Scott Moura
2016

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...

Optimal Charging of Batteries via a Single Particle Model with Electrolyte and Thermal Dynamics

Perez, H. E.
Hu, X.
Scott Moura
2016

This paper seeks to derive insight on battery charging control using electrochemistry models. Directly using full order complex multi-partial differential equation (PDE) electrochemical battery models is difficult and sometimes impossible to implement. This paper develops an approach for obtaining optimal charge control schemes, while ensuring safety through constraint satisfaction. An optimal charge control problem is mathematically formulated via a coupled reduced order electrochemical-thermal model which conserves key electrochemical and thermal state information. The Legendre-Gauss-...

Coordination of V2G and Distributed Wind Power Using the Storage-like Aggregate PEV Model

Zhang, Hongcai
Hu, Zechun
Song, Yonghua
Scott Moura
2016

A plug-in electric vehicle (PEV) fleet utilizing vehicle-to-grid (V2G) technology, i.e., a V2G fleet, can behave as a storage system, e.g., promoting integration of distributed wind power resources. However, because the PEVs' behaviors are stochastic and a V2G fleet's population is large, three technical difficulties hinder the utilization of V2G: charging demand forecasting; ahead-of-time charge and discharge scheduling; real-time charge and discharge power dispatching. This paper utilizes a storage-like aggregate model (SLAM) of a V2G fleet that employs aggregated parameters to represent...

Minimizing Cost Uncertainty with a New Methodology for Use in Policy Making: China's Electricity Pathways

Avrin, Anne-Perrine
Scott Moura
Kammen, Daniel M.
2016

Planning the long-term expansion of a power sector requires anticipating future technologies, fuel costs, and new carbon policies. Many state-of-the-art models rely on exogenous data for cost and performance projections where the inherent uncertainty is either ignored or addressed only with sensitivity analysis and scenarios. For the few models accounting for uncertainty, the transition from the research field to policy making has not occurred because of important practical barriers in the latter field: higher reliance on time-tested models, impossibility to constantly adopt new models,...

Stochastic Control of Smart Home Energy Management with Plug-in Electric Vehicle Battery Energy Storage and Photovoltaic Array

Wu, Xiaohua
Hu, Xiaosong
Scott Moura
Yin, Xiaofeng
Pickert, Volker
2016

Energy management strategies are instrumental in the performance and economy of smart homes integrating renewable energy and energy storage. This article focuses on stochastic energy management of a smart home with PEV (plug-in electric vehicle) energy storage and photovoltaic (PV) array. It is motivated by the challenges associated with sustainable energy supplies and the local energy storage opportunity provided by vehicle electrification. This paper seeks to minimize a consumer's energy charges under a time-of-use tariff, while satisfying home power demand and PEV charging requirements...