faq 108654942

Problem integrate charging to drt

by Marco Bohmbach on 2017-03-29 12:51:34

---

Hi, i'm still trying to combine drt with ev like the example (taxi and ev) from michalm playground but i'm stuck implementing "bestchargerfinder" to like "bestchargerinsertion". 

I don't fully understand how the drt thing works and where to put the charging information into schedule. In words it is just like:

If Battery is low, finish trip and go to charger. If charged, go back to work.

thanks

Marco

---

Comments: 1

---

Re: Problem integrate charging to drt

by Michal Maciejewski on 2017-03-30 07:50:48

Hi,

Let's assume that for a non-EV the schedule would look like:

Serve A → go to B → serve B → go to C → serve C

If we use EV we need to: monitor the SOC and predict its level when planning new tasks. In the simplest approach the charging strategy is purely reactive: once SOC gets below (for instance) 30%, send the vehicle to the charger. (Not really optimal, but should work as a good starting point for writing something more complex).

Let's assume, SOC gets down while serving B. So we need first to recharge the vehicle and then to serve C.

Serve A → go to B → serve B → go to Charger → charge → go to C → serve C

The decision about inserting a trip to a charger and recharging can be done at one place in the optimizer, e.g.

public void notifyMobsimBeforeSimStep(MobsimBeforeSimStepEvent e) {
 	sendUnderchargedVehiclesToChargers();


 	//...standard non-EV code
 }

The method notifyMobsimBeforeSimStep is called every second. But it is usually enough to handle undercharged vehicles every minute or so.

Please have a look here https://github.com/matsim-org/matsim/blob/0056453c98536d4c1f726b2d9aa52b0c88e8c1b7/playgrounds/michalm/src/main/java/playground/michalm/taxi/optimizer/rules/RuleBasedETaxiOptimizer.java (a rather simple rule-based strategy).

If you need more specific answers then I would need to have a look at your code.

Hope that helps,

Michal