• Energy Research
• arXiv.org e-Print Archive close

In this paper, we study the interactions among interconnected autonomous microgrids, and propose a joint energy trading and scheduling strategy. Each interconnected microgrid not only schedules its local power supply and demand, but also trades energy with other microgrids in a distribution network. Specifically, microgrids with excessive renewable generations can trade with other microgrids in deficit of power supplies for mutual benefits. Since interconnected microgrids operate autonomously, they aim to optimize their own performance and expect to gain benefits through energy trading. We design an incentive mechanism using Nash bargaining theory to encourage proactive energy trading and fair benefit sharing. We solve the bargaining problem by decomposing it into two sequential problems on social cost minimization and trading benefit sharing, respectively. For practical implementation, we propose a decentralized solution method with minimum information exchange overhead. Numerical studies based on realistic data demonstrate that the total cost of the interconnected-microgrids operation can be reduced by up to 13.2% through energy trading, and an individual participating microgrid can achieve up to 29.4% reduction in its cost through energy trading. To appear in IEEE Transactions on Smart Grid

In this paper, we address the long-term system's requirement reserve sizing due to the high-level of variable renewable energy (VRE) sources penetration, inside the expansion planning model. The increase in the insertion of this kind of energy source will also bring an increase in the reserve requirements. A higher requirement will be translated into additional costs to the system since the system operator will need to allocate generators for reserve purposes. The VRE sources implicitly cause these costs, so besides the investment cost, the expansion planning models should consider those costs on the expansion decision process. The methodology proposed here aims to provide a probabilistic and dynamic evaluation of the forecast errors of VRE sources generation, translating these errors into the system's requirement reserve. This evaluation is done inside the expansion planning optimization model, treating the reserve requirement as an endogenous variable. Finally, a real case study of the Mexican system is presented, so that we can analyze the results of the methodology and the impacts of considering the reserve requirement along with the expansion planning decision.

The assessment of voltage stability margins is a promising direction for wide-area monitoring systems. Accurate monitoring architectures for long-term voltage instability are typically centralized and lack scalability, while completely decentralized approaches relying on local measurements tend towards inaccuracy. Here we present distributed linear algorithms for the online computation of voltage collapse sensitivity indices. The computations are collectively performed by processors embedded at each bus in the smart grid, using synchronized phasor measurements and communication of voltage phasors between neighboring buses. Our algorithms provably converge to the proper index values, as would be calculated using centralized information, but but do not require any central decision maker for coordination. Modifications of the algorithms to account for generator reactive power limits are discussed. We illustrate the effectiveness of our designs with a case study of the New England 39 bus system. 10 pages, submitted for publication

We investigate the effect of slowly-varying parameter on the energy transfer in a system of weakly coupled nonlinear oscillators, with special attention to a mathematical analogy between the classical energy transfer and quantum transitions. For definiteness, we consider a system of two weakly coupled oscillators with cubic nonlinearity, in which the oscillator with constant parameters is excited by an initial impulse, while a coupled oscillator with slowly-varying parameters is initially at rest. It is proved that the equations of the slow passage through resonance in this system are identical to equations of the nonlinear Landau-Zener (LZ) tunneling. Three types of dynamical behavior are distinguished, namely, quasi-linear, moderately nonlinear and strongly nonlinear. The quasi-linear systems exhibit a gradual energy transfer from the excited to the attached oscillator, while the moderately nonlinear systems are characterized by an abrupt transition from the energy localization on the excited oscillator to the localization on the attached oscillator. In the strongly nonlinear systems, the transition from the energy localization to strong energy exchange between the oscillators is revealed. A special case of the rapid irreversible energy transfer in the strongly nonlinear system with slowly-varying parameters is also investigated. The conditions providing different types of the dynamical behavior are derived. Explicit approximate solutions describing the transient processes in moderately and strongly nonlinear systems are suggested. Correctness of the constructed approximations is confirmed by numerical results.

We present Vehicle Energy Dataset (VED), a novel large-scale dataset of fuel and energy data collected from 383 personal cars in Ann Arbor, Michigan, USA. This open dataset captures GPS trajectories of vehicles along with their time-series data of fuel, energy, speed, and auxiliary power usage. A diverse fleet consisting of 264 gasoline vehicles, 92 HEVs, and 27 PHEV/EVs drove in real-world from Nov, 2017 to Nov, 2018, where the data were collected through onboard OBD-II loggers. Driving scenarios range from highways to traffic-dense downtown area in various driving conditions and seasons. In total, VED accumulates approximately 374,000 miles. We discuss participant privacy protection and develop a method to de-identify personally identifiable information while preserving the quality of the data. After the de-identification, we conducted case studies on the dataset to investigate the impacts of factors known to affect fuel economy and identify energy-saving opportunities that hybrid-electric vehicles and eco-driving techniques can provide. The case studies are supplemented with a number of examples to demonstrate how VED can be utilized for vehicle energy and behavior studies. Potential research opportunities include data-driven vehicle energy consumption modeling, driver behavior modeling, machine and deep learning, calibration of traffic simulators, optimal route choice modeling, prediction of human driver behaviors, and decision making of self-driving cars. We believe that VED can be an instrumental asset to the development of future automotive technologies. The dataset can be accessed at https://github.com/gsoh/VED. 11 pages, 15 figures

Building energy flexibility has been increasingly demonstrated as a cost-effective solution to respond to the needs of energy networks, including electric grids and district cooling and heating systems, improving the integration of intermittent renewable energy sources. Adjusting zonal temperature set-points is one of the most promising measures to unlock the energy flexibility potential of central air conditioning systems in complex commercial buildings. However, most existing studies focused on quantifying the energy flexibility on the building level since only building-level energy consumption is normally metered in commercial buildings. This study aims to investigate the impacts of temperature set-point adjustment strategies on zone-level thermal and energy performance by developing a non-intrusive data-driven load disaggregation method (i.e., a virtual zonal power meter). Three university buildings in Northern California were selected to test the proposed load disaggregation method. We found that heterogeneities of energy use and energy flexibility existed across not only buildings but also air handling units (AHUs) and zones. Moreover, a small number of zones accounted for a large amount of energy use and energy flexibility; and the most energy-intensive zones are not necessarily the most energy-flexible zones. For the three tested buildings, the top 30% most energy-intensive zones accounted for around 60% of the total energy use; and the top 30% most energy-flexible zones provided around 80% of the total energy flexibility. The proposed method enables the electric grid or district energy system operators to regard the controlled energy-flexible entities as a fleet of AHUs or zones instead of a fleet of buildings and helps unlock the possibility for targeted demand flexibility strategies that balance zone-by-zone energy reduction with zone-by-zone costs to occupants. 33 pages, 18 figures

Reliable uncertainty estimation for time series prediction is critical in many fields, including physics, biology, and manufacturing. At Uber, probabilistic time series forecasting is used for robust prediction of number of trips during special events, driver incentive allocation, as well as real-time anomaly detection across millions of metrics. Classical time series models are often used in conjunction with a probabilistic formulation for uncertainty estimation. However, such models are hard to tune, scale, and add exogenous variables to. Motivated by the recent resurgence of Long Short Term Memory networks, we propose a novel end-to-end Bayesian deep model that provides time series prediction along with uncertainty estimation. We provide detailed experiments of the proposed solution on completed trips data, and successfully apply it to large-scale time series anomaly detection at Uber. To appear in DSBDA-2017 @ ICDM'17

Hosting capacity (HC) and dynamic operating envelopes (DOEs), defined as dynamic, time-varying HC, are calculated using three-phase optimal power flow (OPF) formulations. Due to the computational complexity of such optimisation problems, HC and DOE are often calculated by introducing certain assumptions and approximations, including the linearised OPF formulation, which we implement in the Python-based tool ppOPF. Furthermore, we investigate how assumptions of the distributed energy resource (DER) connection phase impact the objective function value and computational time in calculating HC and DOE in distribution networks of different sizes. The results are not unambiguous and show that it is not possible to determine the optimal connection phase without introducing binary variables since, no matter the case study, the highest objective function values are calculated with mixed integer OPF formulations. The difference is especially visible in a real-world low-voltage network in which the difference between different scenarios is up to 14 MW in a single day. However, binary variables make the problem computationally complex and increase computational time to several hours in the DOE calculation, even when the optimality gap different from zero is set.