Neighbor-Based Optimized Logistic Regression Machine Learning Model For Electric Vehicle Occupancy Detection
Authors:
Sayan Shaw,
Keaton Chia,
Jan Kleissl
Abstract:
This paper presents an optimized logistic regression machine learning model that predicts the occupancy of an Electric Vehicle (EV) charging station given the occupancy of neighboring stations. The model was optimized for the time of day. Trained on data from 57 EV charging stations around the University of California San Diego campus, the model achieved an 88.43% average accuracy and 92.23% maxim…
▽ More
This paper presents an optimized logistic regression machine learning model that predicts the occupancy of an Electric Vehicle (EV) charging station given the occupancy of neighboring stations. The model was optimized for the time of day. Trained on data from 57 EV charging stations around the University of California San Diego campus, the model achieved an 88.43% average accuracy and 92.23% maximum accuracy in predicting occupancy, outperforming a persistence model benchmark.
△ Less
Submitted 27 April, 2022;
originally announced April 2022.
Frequency Regulation with Heterogeneous Energy Resources: A Realization using Distributed Control
Authors:
Tor Anderson,
Manasa Muralidharan,
Priyank Srivastava,
Hamed Valizadeh Haghi,
Jorge Cortes,
Jan Kleissl,
Sonia Martinez,
Byron Washom
Abstract:
This paper presents one of the first real-life demonstrations of coordinated and distributed resource control for secondary frequency response in a power distribution grid. We conduct a series of tests with up to 69 heterogeneous active devices consisting of air handling units, unidirectional and bidirectional electric vehicle charging stations, a battery energy storage system, and 107 passive dev…
▽ More
This paper presents one of the first real-life demonstrations of coordinated and distributed resource control for secondary frequency response in a power distribution grid. We conduct a series of tests with up to 69 heterogeneous active devices consisting of air handling units, unidirectional and bidirectional electric vehicle charging stations, a battery energy storage system, and 107 passive devices consisting of building loads and photovoltaic generators. Actuation commands for the test devices are obtained by solving an economic dispatch problem at every regulation instant using distributed ratio-consensus, primal-dual, and Newton-like algorithms. The distributed control setup consists of a set of Raspberry Pi end-points exchanging messages via an ethernet switch. The problem formulation minimizes the sum of device costs while tracking the setpoints provided by the system operator. We demonstrate accurate and fast real-time distributed computation of the optimization solution and effective tracking of the regulation signal by measuring physical device outputs over 40-minute time horizons. We also perform an economic benefit analysis which confirms eligibility to participate in an ancillary services market and demonstrates up to $53K of potential annual revenue for the selected population of devices.
△ Less
Submitted 4 February, 2021; v1 submitted 15 July, 2020;
originally announced July 2020.
