Reusing the code for Flexibility calculation: Questions about the OPF formulation

[Ingrid Munné Collado]

Dear Valentin,

Thank you for this repository, it is very helpful. I'm a PhD student currently working on flexibility services for distribution networks. At the moment, I'm trying to use your code to calculate the flexibility request the DSO needs (considering flexibility activation costs) in order to avoid overvoltages and overcurrents. I think it can be a good exercise and I would be happy to share the code when it works, if you think it can be useful! I am mainly modifying the objective function and some of the constraints.

I would like to ask you some questions about the OPF formulation (OPF_model_creator_v01.py):

1. I have been mainly working with the polar power-voltage formulation for AC-OPF applied to distribution networks. Is the one you propose equivalent to it?
2. When I've been working with pyomo before, I always specified the sense of the objective function (minimize, maximize), however I get an error if I try to specify the objective function sense here. Do we assume is always "minimize"?
3. When you define the variables, you say the voltage is in V, however, when defining the constraints, the variable P_control is [0,1], but why are you including a 1e3 factor in the objective function? is it to consider that in kW? What is the meaning of alpha in the objective function? It is not specified on the paper.

Thank you in advance!

Best regards,

Íngrid

[valentin...@gmail.com]

Hi Ingrid,

I'm happy that you find the model useful. Maybe this research can be of interest "Coordinating Demand Response Aggregation With LV Network Operational Constraints" if you want to simplify the network model.

https://www.researchgate.net/publication/343318295_Coordinating_Demand_Response_Aggregation_With_LV_Network_Operational_Constraints

In response to the questions:

1- I do not use polar components. Everything in the model is phase-to-ground. You can transform from one representation to the other.

2- You need to define the sense in the objective function definition. See this https://pyomo.readthedocs.io/en/stable/pyomo_modeling_components/Objectives.html

3- The objective function is multi-objective. It maximizes P_control and minimizes tanphi (reactive power). alpha is there as a weight coefficient so you give more importance to the P_control term. In other words, you want the solver to avoid doing curtailment as much as possible over reactive power support.

Best regards,

V