Tutorial 1: EVs and Scenarios

See tutorial/tutorial_1_EVs_and_Scenarios.mlx. This tutorial includes advanced topics on EV and Scenario. You can refer to EV and Scenario while following this tutorial to better understand the methods and properties discussed here.

An `EV` object

1. Create an `EV` object

An EVobject contains information such as the arrival/departure time of an EV, battery capacity, initially stored energy, reference energy. The following syntax creates EV objects.

ev1 = EV();                         % Create an EV object with default parameters.
ev2 = EV(Arr_time=2, Dep_time=6);   % Create an EV object with name-value arguments.

Once created, the properties of an EV object can be modified using set() method. (Directly changing the properties is forbidden.)

ev1 = set(ev1,E_ini=40);
ev2 = ev2.set(E_ref=80);

2. Validate an `EV` object

Each EV object has a property named valid, which indicates whether the object has passed a validation process.

ev1.valid
ev2.valid

The following syntax executes a validation process and chage valid true if all the requirements (e.g., Arr_time < Dep_time, E_ref <= E_cap) for an EV object are satisfied.

ev1 = validate(ev1);
ev2 = ev2.validate();
ev1.valid
ev2.valid

However, modifying a property afterward requires an additional validation.

ev1 = set(ev1,U_max=15);
ev1.valid

A `Scenario` object

A Scenario object is an object that contains all information describing a charging scheduling problem, such as the base load profile, electricity price, limits on charging powers, EV behaviors, and topology of communication infrastructure.

1. Import a scenario from json

Navigate to the tutorial/Sample_Lf folder. Inspect the following files.

“Sample_LF_cent.json” and “Sample_LF_dist.json” that contain all key information

“Baseload.xlsx” that contains a base load profile
“EV_Lf.xlsx” that contains all EV information: arrival/departure times, battery capacities, etc

To create a scenario object by importing the information provided in these files, execute the following.

scenario1 = import_scenario('Sample_Lf_cent.json');

2. Create a scenario manually

Indeed, it is possible to create a scenario object manually.

Problem_type = "Lf";                % We focus on the load flattening problem
T            = 3;                   % Consider three time slots
Delta        = 10;                  % of length 10 minutes.
D            = [100; 150; 100];     % The base load at each time slot is 100kW, 150kW, 100kW.
N_c          = 2;                   % We consider two chargers
N_ev         = 2;                   % and two EVs.
EVs          = [ev1, ev2];          % EVs contains information about the two EVs.

scenario2 = Scenario(Problem_type=Problem_type, ...
                      T=T, ...
                      Delta=Delta, ...
                      D=D, ...
                      N_c=N_c, ...
                      N_ev=N_ev,...
                      EVs=EVs);     % Now we can create a scenario object.

3. Generate a scenario from ScenarioGenerator object

Another method is to instanciate a ScenarioGenerator object and generate a scenario from it.

gen = PoissonScenarioGenerator(T=24);               % Instantiate a PoissonScenarioGenerator
                                                    % object that generates scenarios of length 24.
                                                    % Here, we set other parameters to default values.

scenario3 = generate(gen);                          % Then, generate a scenario from it.

4. Validate a scenario

A Scenario object also has a property named valid which indicates whether the object has passed the validation process.

scenario1.valid
scenario2.valid
scenario3.valid

One can see that both scenario_1 and scenario_3 are valid, whereas scenario_2 is not. This is because the functions import_scenario() and generate() of PoissonScenarioGenerator automatically perform a validation process. Here, we demonstrate the validation process for scenario_2.

scenario2 = validate(scenario2);
scenario2.valid

All scheduling functions provided in SH-V2G-Simulatior undergo this scenario validation process before executing the scheduling.

Note

validate() includes the validation of all contained EV objects.

5.Modifying properties of a scenario

The properties of a scenario object can be modified by set() method.

scenario2 = set(scenario2,Delta=30);
scenario2 = scenario2.set(N_c=3);

A single EV object can be replaced using set_ev() method.

ev2_new = EV(Charger_ind=3);
scenario2 = scenario2.set_ev(2,ev2_new);

However, again, modifying a property afterward requires an additional validation.

scenario2.valid

6. Obtaining a subscenario from a scenario

In some cases, it is required to generate a subscenario from a given scenario, for example, when performing scheduling in a receding horizon manner or when analyzing the results.

The method get_subscenario() supports this functionality with various options.

The name-value arguments ini and fin specify the initial time slot and the final time slot for obtaining a subscenario.

subscenario_1a = scenario1.get_subscenario(ini=2,fin=5);

subscenario_1a corresponds to the time slot 2, 3, 4 of the scenario1.
subscenario_1a.EVs consists of the EVs that is connected to a charger at some time slot among 2, 3, 4.

Setting future_evs=false makes the subscenario exclude the EVs that arrives in the future (i.e., at time slots after ini).

subscenario_1b = scenario1.get_subscenario(ini=2,fin=5,future_evs=false);

subscenario_1b.EVs consists of the EVs that is connected to a charger at time slot 2.

By providing E_ini argument we can set the E_ini of the EVs in the subscenario.

subscenario_1c = scenario1.get_subscenario(ini=2,fin=5,E_ini=zeros(1,50));
subscenario_1a.EVs(1).E_ini
subscenario_1c.EVs(1).E_ini

For each i, subscenario_1c.EVs(i).E_ini is set to 0.