ASoSEPOC

This code base is using the Julia Language and DrWatson to make a reproducible scientific project named

ASoSEPOC

It is authored by Sebastian Gonzato.

To (locally) reproduce this project, do the following:

0. Download this code base. Notice that raw data are typically not included in the git-history and may need to be downloaded independently.
1. Open a Julia console and do:

julia> using Pkg
julia> Pkg.add("DrWatson") # install globally, for using `quickactivate`
julia> include("path/to/this/project/instantiate.jl")

This will install all necessary packages for you to be able to run the scripts and everything should work out of the box, including correctly finding local paths.

Code description

• src - source files (i.e. individual functions)
• scripts - main scripts used for generating results
• scripts_debug - scripts used for debugging or testing purposes (may not be functional)
• papers - latex report of work done for this adequacy and operational security interaction.
• pbs - useful files for running on the Flemish super computer.

For the scripts:

• process_input - converts input data from Belderbos paper to format required for GEPPR.jl.
• check_and_fix_scenarios - fixes infeasible wind, solar and load forecast errors (e.g. wind output less than that which is forecasted).
• main_model_runs - DUC-PR model runs to investigate whether tradeoff between adequacy and security is possible for day 285.
• copperplate_adequacy_security_tradeoff - same as above but without network constraints, to illustrate that the trade-off is possible in the absence of these.
• nodal_imbalance_investigation - illustration of how to tighten reserve activation network constraints.

General description

Model

The Deterministic Unit Commitment with Probabilistic (Operating) Reserves (DUC-PR) model is described in papers/main.pdf. In brief:

• Network, unit commitment and storage are included in the model.
• Operating reserves are modeled using "reserve levels", which have an associated activation probability.
  • These reserve levels can essentially be thought of as different reserve types, e.g. FCR, RR, etc in that they have different probabilities of activation.
  • No flexibility requirements (e.g. ramping) are associated with reserve level activation though they are taken into account in the day ahead scheduling.
  • An attempt at including the network in the reserve level activation constraints is made, though this is a non-trivial task.

Data

Download

Input data can be found here.

Description

Grid, load, renewables and capacity mix data come from this paper. Brief description:

• It is a Belgium-like system with a very high share of renewables (~80%).
• The nuclear reactors in Doel and Tihange have been replaced by gas fired power plants.

More information can be found in papers/main.pdf.

Model runs

Questions to be answered in preliminary investigation

• Is there load shedding with an entirely linear model with no network constraints?
• Above, but with unit commitment constraints?
• Above, but with the network?
• Above, but with simultaneous charging/discharging not allowed?
• Same 3 questions as above, but with operating reserves and for different reserve level shedding limits, i.e. attempting to trace a adequacy / security pareto curve? + additional run with reserve activation network constraints?
• Starting from the most "inflexible" system, does adding absolute limits on the nodal imbalance change anything?
• Same, but with convex hull constraints?
• Same, but with load multiplier

Table below outlines this better. Acronyms are:

• UC = Unit commitment constraints
• DANet = Day ahead network constraints
• PSCD = Prevent simultaneous charge and discharge
• OR = Include operating reserves and reserve shedding limits
• RANet = Reserve activation network constraints
• AbsImb = absolute limits on nodal imbalance
• ConvImb = convex hull limits on nodal imbalance
• L1.5 = Load times 1.5

All this for just one of the days selected for investigation.

Trade-off between load shedding and reserve shedding

There should be a tradeoff between shedding load in the day ahead stage and shedding reserves which would jeopardise real-time operational security (since there are fewer reserves available to deal with unforeseen situations). The DUC-PR model can make this tradeoff coarsely, hence the real-time operational security is then analysed using a Quasi Steady State Simulator (QSSS) developed by ULiege. This verifies how well the DUC-PR model can perform this trade-off (if at all).