MORP: Data-Driven Multi-Objective Route Planning and Optimization for Electric Vehicles Ankur Sarker, Haiying Shen, and John A. Stankovic Department of Computer Science, University of Virginia Charlottesville, Virginia, USA
Outline • Introduction • System Design • Performance Evaluation • Conclusion 2
Introduction Wireless power transfer system • The Online Electric Vehicle (OLEV) is an electric vehicle that charges wirelessly while moving using electromagnetic induction. 3
Introduction Wireless power transfer system • The Online Electric Vehicle (OLEV) is an electric vehicle that charges wirelessly while moving using electromagnetic induction. • The Korean Advanced Institute of Science and Technology (KAIST) developed first recharging road on March 9, 2010. 4
Introduction Wireless power transfer system Long Queue 5
Introduction Wireless power transfer system Long Queue Time-Consuming 6
Introduction Wireless power transfer system Long Queue Time-Consuming Range Anxiety 7
Introduction Wireless power transfer system Long Queue Time-Consuming Range Anxiety Maintain SoC 8
Introduction Wireless power transfer system B EV k RSU EV i RSU: Road Side Unit A EV: Electrical Vehicle 9
Introduction Wireless power transfer system B EV k RSU Charging sections EV i RSU: Road Side Unit A EV: Electrical Vehicle 10
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k RSU Charging sections EV i RSU: Road Side Unit A EV: Electrical Vehicle 11
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k i) Arrive at its destination with sufficient power supply on the way RSU Charging sections EV i RSU: Road Side Unit A EV: Electrical Vehicle 12
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k i) Arrive at its destination with sufficient power supply on the way RSU Charging ii) Consider current traffic flow and sections minimize: the driver’s range anxiety EV i RSU: Road Side Unit A EV: Electrical Vehicle 13
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k i) Arrive at its destination with sufficient power supply on the way RSU Charging ii) Consider current traffic flow and sections minimize: the driver’s range anxiety, charging monetary cost EV i RSU: Road Side Unit A EV: Electrical Vehicle 14
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k i) Arrive at its destination with sufficient power supply on the way RSU Charging ii) Consider current traffic flow and sections minimize: the driver’s range anxiety, charging monetary cost, travel time EV i RSU: Road Side Unit A EV: Electrical Vehicle 15
Introduction Wireless power transfer system For a given EV driving from a source to B a destination, how to choose a route so that: EV k i) Arrive at its destination with sufficient power supply on the way RSU Charging ii) Consider current traffic flow and sections minimize: the driver’s range anxiety, charging monetary cost, travel time, EV i and energy consumption RSU: Road Side Unit A EV: Electrical Vehicle 16
Introduction State-of-the-Art Plug-in charging station IEEE TSG’12 IEEE TPS’14 IEVC’14 IEEE TSG’14 IEEE TPD’13 IEEE TPS’12 IEEE TPS’14 17
Introduction State-of-the-Art Wireless power transfer Plug-in charging station IEEE TSG’12 IEEE TPS’14 Annals of Physics’08 IEVC’14 IEEE TSG’14 IEEE Systems Journal’16 IEEE TPD’13 IEEE TPS’12 ICPP’16 IEEE ICDCS ’17 IEEE TPS’14 18
Introduction State-of-the-Art Wireless power transfer Annals of Physics’08 1 IEEE Systems Journal’16 ICPP’16 IEEE ICDCS ’17 Not applicable for dynamic wireless charging 19
Introduction State-of-the-Art 2 1 Cannot maintain the SoC of Not applicable for dynamic vehicles in a metropolitan road wireless charging network 20
Outline • Introduction • System Design • Performance Evaluation • Conclusion 21
System Design overview Gridded Roadmap Data Cleaning Traffic Data 1 22
System Design overview Gridded Roadmap Data Cleaning Temporal Analysis Traffic Data 1 23
System Design overview Gridded Roadmap Spatial Analysis Data Cleaning Temporal Analysis Traffic Data 1 24
System Design overview Spatio- Temporal Gridded Correlation Roadmap Spatial Analysis Data Cleaning Temporal Analysis Traffic Data 1 25
System Design overview Spatio- Traffic Temporal Gridded Prediction Correlation Roadmap Spatial Analysis Data Cleaning Temporal Analysis Traffic Data 2 1 26
System Design overview Spatio- Traffic Temporal Gridded Prediction Correlation Roadmap Velocity Spatial Prediction Analysis Data Cleaning Temporal Analysis Traffic Data 2 1 27
System Design overview Spatio- Traffic Temporal Gridded Prediction Correlation Roadmap Velocity Spatial Prediction Analysis Data Cleaning Temporal Power Analysis Consumption Traffic Data 2 1 28
System Design overview Spatio- Traffic Objective Temporal Gridded Prediction Functions Correlation Roadmap Velocity Spatial Prediction Analysis Data Cleaning Temporal Power Analysis Consumption Traffic Data 2 3 1 29
System Design overview Spatio- Traffic Objective Temporal Gridded Prediction Functions Correlation Roadmap Velocity Spatial Prediction Multi- Analysis Objective Data Cleaning Route Temporal Power Planning Analysis Consumption Traffic Data 2 3 1 30
System Design Traffic data analysis We collected 212 consecutive day- long historical hourly traffic flow data: 31
System Design Traffic data analysis We collected 212 consecutive day- long historical hourly traffic flow data: 1. From December 1, 2016 to June 30, 2017 32
System Design Traffic data analysis We collected 212 consecutive day- long historical hourly traffic flow data: 1. From December 1, 2016 to June 30, 2017 2. 20 locations in 3 interstate routes, 9 US routes, and 6 state routes 33
System Design Traffic counts prediction • Traffic locations with historical traffic data • Traffic locations without historical traffic data 34
System Design Traffic counts prediction Traffic locations with historical traffic data 35
System Design Traffic counts prediction Traffic locations with historical traffic data Hourly traffic counts 36
System Design Traffic counts prediction Traffic locations with historical traffic data Correlation of two locations 37
System Design Traffic counts prediction Traffic locations with historical traffic data Weekly traffic counts 38
System Design Traffic counts prediction Traffic locations with historical traffic data Change rate of traffic counts 39
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical traffic data 40
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical traffic data 41
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical Autoregressive traffic data term w.r.t. day 42
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical Autoregressive Moving average traffic data term w.r.t. day term w.r.t. day 43
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical Autoregressive Moving average traffic data term w.r.t. day term w.r.t. day Autoregressive term w.r.t. location 44
System Design Traffic counts prediction Horizontal spatio-temporal autoregressive integrated moving average model Traffic locations with historical Autoregressive Moving average traffic data term w.r.t. day term w.r.t. day Autoregressive Moving average term w.r.t. location term w.r.t. location 45
System Design Traffic counts prediction • Traffic locations with historical traffic data • Traffic locations without historical traffic data 46
System Design Traffic counts prediction Spatio-temporal ordinary-kriging Traffic locations without historical traffic data 47
System Design Traffic counts prediction Spatio-temporal ordinary-kriging Traffic locations without historical traffic data 48
System Design Traffic counts prediction Spatio-temporal ordinary-kriging Traffic locations without historical traffic data Weights chosen to minimize the prediction error variance 49
Recommend
More recommend
