Sensitivity analysis of renewable electricity price

[ejin]

Hi VEDA moderator and community,

I want to conduct a sensitivity analysis based on the stochastic analysis function in VEDA. I want to explore the effect of change in renewable electricity price on some technologies powered by renewable electricity. It is a simple two stage setup. The second stage starts in 2035 and end in 2050. In the second stage, the price has three states or three different values with the same probabilities in 2035 and 2050. I created a two-stage scenario for the sensitivity analysis but the results for three SOWs are the same. I guess I didn't choose or implement the correct parameter of renewable electricity price correctly in the model. The renewable electricity is built in as an import resource and I define the price as an external factor. The electricity price I used by attribute of IRE_PRICE of the commodity named "ELCRNWBL".

Can someone help me check if the attached file is correctly edited for the stochastic run? 

Thanks!

Enze

[AKanudia]

All attributes don't have the stage and sow indices. You can see which ones do, by filtering out the blanks from Stage column in TIMES and Veda Attributes (under Information menu).

I would create three different sources of electricity and turn them on/off in the branches using S_FLO_CUM.

[Antti-L]

As Amit pointed out, the stage and sow indexes can be used only for attributes that have these indexes. Those that have them, are called uncertain attributes, and are prefixed with S_ (or SW_ for the setup parameters). On the other hand, IRE_PRICE is a normal deterministic attribute. You can see the attribute indexes from Information → TIMES and VEDA attributes, as well as by consulting the TIMES documentation.

To model uncertain variable costs (or prices) for an exogenous trade process, you could do as Amit suggested.

For normal processes, another easy way is to use FLO_EMIS and S_FLO_FUNC for defining an uncertain dummy flow, and then to define a variable cost on that flow.  As a result, you would then have uncertain variable costs represented by the dummy flow, varying according to the uncertain S_FLO_FUNC multiplier.