Optimization Strategies for Electric Vehicle Charging Schedules Master Thesis Defense Kassel, the 8th of June, 2015 Erkki Juha Matti Rautiainen Under the Supervision of M.Sc. Chenjie Ma Fraunhofer IWES, Kassel 1 Committee: Prof. Dr. Mohamed El Sobki (Jr.) Prof. Dr. Adel Khalil Prof. Dr. Hafez El-Salamawy Prof. Dr. -sc.techn. Dirk Dahlhaus 1 This thesis is conducted as part of project Systemintegration Elektromobilität funded by the Federal Ministry for Economic Affairs and Energy in Germany
Agenda Motivation Objectives & Justification Charging Strategies Test Environment Case Studies & Results Conclusions & Future Research Erkki Juha Matti Rautiainen 2(20)
Motivation EVs are important to mitigate CO 2 emissions Uncontrolled EV charging poses a risk Realistic implementation for controlled charging is difficult due to multiple stakeholders EV users EV fleet aggregators DSOs Traffic light concept as a common rule set for these stakeholders Network status defined by three different colors: green, amber and red DSOs are supervising networks and interract with market players if needed Erkki Juha Matti Rautiainen 3(20)
Objectives of the Thesis 1. A new charging strategy for realistic implementation. Separated DSO and the EV fleet aggregator Unidirectional communication between the entities Communication carried out by technical signal for power capacities 2. Compare against existing charging strategies 3. Investigate the effects of two EV fleet aggregators Erkki Juha Matti Rautiainen 4(20)
Justification of the Topic The implementations in reviewed publications are not realistic Combined entities or iterative solution finding for EV charging schedules Only global power capacity accounted as a constraint No comprehensive comparison No multiple EV fleet aggregators Erkki Juha Matti Rautiainen 5(20)
Communication Based Strategy CBS Start CS Network data Load profiles DSO Calculate the power capacities Capacity profiles P f,τ and P tot,τ End CS Price profile EV user data EV fleet aggregator EV schedules Figure: Inputs and outputs for the CBS Erkki Juha Matti Rautiainen 6(20)
Charging Strategies Strategy Communication Based Strategy (CBS) Uncontrolled Charging Scenario(UCS) Aggregator Based Strategy (ABS) DSO Based Strategy (DBS) System Based Strategy (SBS) Objective & Constraints Cost minimization, EV & power capacities Immediate EV charging Cost minimization, EV related Load flattening, EV & network related Cost minimization, EV & network related Erkki Juha Matti Rautiainen 7(20)
Test Environment Household loads EV schedules Voltages EV model EV Pac, EV Qac SoC EV,n Electrical network model ephasorsim Voltages Currents Figure: Simulink model for the simulation system Erkki Juha Matti Rautiainen 8(20)
Test Environment The electric network has a 630 kva distribution transformer, accommodates 146 households and has 10 main feeders The used data covers 48 hour time periods for Wednesday-Thursday Saturday-Sunday Two different EV penetration rates are applied 75% EV penetration rate 100% EV penetration rate Simulations are performed in On-line mode to attain the dynamic behavior of the models Erkki Juha Matti Rautiainen 9(20)
Comparison of Single Charging Strategies Table: Simulation results on Wednesday with 75% EV penetration rate Scenario: UCS ABS DBS SBS CBS Total charging cost [e] 85.7 23.2 59.6 28.2 24.9 EV user Change of charging cost[%] 0-73 -30.4-67.1-71.0 comfort Net number of interruptions 0-13 79 16-7 Time flexibility [h] 12.1 2.6 1.5 3.4 3.2 Maximum current [A] 348 441 257 250 257 Electric Violation magnitude [%] 22.4 55.2 0 0 0 network Violation duration [min] 8 32 0 0 0 impacts Change of network losses [%] 0.0-19.6-50.0-35.3-27.6 Cost of network losses [e] 8.0 1.6 3.1 1.6 1.6 Erkki Juha Matti Rautiainen 10(20)
Comparison of Single Charging Strategies Currents in the most heavily loaded feeder [A] 400 300 200 100 0 12 18 0 6 12 Time of the day [h] I max CBS SBS DBS ABS UCS BS Electricity price [e/kwh] 0.4 0.2 0 12 18 0 6 12 Time of the day [h] Figure: Wednesday, electric network impacts Erkki Juha Matti Rautiainen 11(20)
Comparison of Single Charging Strategies Strategy: Pros Cons Uncontrolled Charging Scenario Flexibility Highest charging costs & Large network impacts Aggregator Based Strategy Lowest charging costs Large network impact DSO Based Strategy System Based Strategy Communication Based Strategy Smallest network impact Low charging costs Small network impact Low charging costs Small network impact High charging costs Combined entities DSO calculation accuracy Erkki Juha Matti Rautiainen 12(20)
Multiple EV Fleet Aggregators With Different Strategies Start CS CBS - multi aggregator Capacity profiles P f,τ and P tot,τ Network data Load profiles DSO Capacity profiles P f,τ and P tot,τ Price profile EV1 user data Agg1 Surpluss capacities or power requests EV1 schedules and EV2 schedules Agg2 Price profile EV2 user data Figure: Multi aggregator flowchart End CS Erkki Juha Matti Rautiainen 13(20)
Total charging cost of both EV fleets [e] Change of charging cost[%] 100 50 0 0 20 40 60 80 Multiple EV Fleet Aggregators With Different Strategies Scenario 1 30% share Scenario 1 30% share Scenario 2 50% share Scenario 2 50% share Scenario 3 70% share Scenario 3 70% share Scenario 4 100% share Scenario 4 100% share UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS Figure: Wednesday, charging costs and change percentages Erkki Juha Matti Rautiainen 14(20)
Net number of interruptions 100 50 0 Multiple EV Fleet Aggregators With Different Strategies UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS Time flexibility [h] 15 10 5 Scenario 1 30% share Scenario 2 50% share Scenario 3 70% share Scenario 4 100% share UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS 0 Scenario 1 30% share Scenario 2 50% share Scenario 3 70% share Scenario 4 100% share Figure: Wednesday, net number of interruption and time flexibility Erkki Juha Matti Rautiainen 15(20)
Current limit violation magnitude [%] Current limit violation duration [min] 100 50 0 50 40 30 20 10 0 Multiple EV Fleet Aggregators With Different Strategies Scenario 1 30% share Scenario 1 30% share Scenario 2 50% share Scenario 2 50% share Scenario 3 70% share Scenario 3 70% share Scenario 4 100% share Scenario 4 100% share UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS Figure: Wednesday, current violation magnitude and duration Erkki Juha Matti Rautiainen 16(20)
Change of network losses [%] Cost of network losses [e] 0 10 20 30 40 50 10 5 0 Multiple EV Fleet Aggregators With Different Strategies Scenario 1 30% share Scenario 1 30% share Scenario 2 50% share Scenario 2 50% share Scenario 3 70% share Scenario 3 70% share Scenario 4 100% share Scenario 4 100% share UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS UCS-CBS ABS-CBS DBS-CBS SBS-CBS CBS-CBS Figure: Wednesday, costs and changes of network losses Erkki Juha Matti Rautiainen 17(20)
Conclusions Robust and reliable Flexible for both entities Ideas successfully adopted Shortcomings avoided Potential for further development Erkki Juha Matti Rautiainen 18(20)
Future Research Enhance the DSO calculation More comprehensive tests Implement V2G Power capacities to market based signals Erkki Juha Matti Rautiainen 19(20)
Questions and Comments? Thank you for your time. Contact: Juha Rautiainen juha.rautiainen(at)hotmail.fi Erkki Juha Matti Rautiainen 20(20)