SPORES¶
A normal optimisation gives you one answer: the cheapest system. Usually there are many other systems that cost only a little more but look very different, for example with wind in other regions or less transmission. SPORES (Spatially explicit Practically Optimal REsultS) finds a set of these alternatives.
The method comes from Lombardi, Pickering, Colombo and Pfenninger (2020), Policy Decision Support for Renewables Deployment through Spatially Explicit Practically Optimal Alternatives, Joule 4, 2185–2207.
How it works¶
- TEMPO solves the model normally to find the cheapest cost.
- It then solves the model again N more times. Each time, total cost may be at most slack percent above the cheapest cost, and the solver is pushed away from the locations and technologies used in earlier solutions.
The result is N + 1 solutions with similar cost and different layouts. Every solution is a full solver run, so SPORES takes roughly N + 1 times as long as a normal run.
Which engines¶
| Engine | SPORES |
|---|---|
| Calliope 0.6.8 | Yes, on your computer |
| Calliope 0.7 | Only on a remote MEME server |
| PyPSA, OSeMOSYS, AdOpT-NET0 | No |
Running SPORES¶
- Open Run and choose the Calliope engine.
- Set Mode to the SPORES option.
- Set the two options in the SPORES panel:
| Option | Range | Default | Meaning |
|---|---|---|---|
| Cost Slack (%) | 1 to 30 | 10 | How much more the alternatives may cost than the cheapest plan. Higher values give more variety but stray further from the optimum |
| Number of SPORES | 5 to 100 | 20 | How many alternatives to generate. 5 to 10 is enough for testing; 20 to 50 gives a better picture. The original study used 178 |
- Click Run.
To run SPORES at several cost slacks, or across sensitivity cases, use a SPORES alternatives card in Scenario Studio. Each slack becomes a separate run.
Reading the results¶
A SPORES run adds a SPORES tab to the Results screen:
| Section | What it shows |
|---|---|
| Cost Summary | Total cost of every solution and how far above the cheapest it is |
| Fig 1: Technology Mix & System Cost per Solution | Installed capacity by technology for each solution, next to its cost |
| Fig 2: Technology Deployment Variability | How much each technology's capacity changes across solutions |
| Fig 3: Pairwise Technology Trade-off | Capacity of one technology against another, one dot per solution. A falling trend means they substitute for each other, a rising trend means they are built together |
| Fig 4: Near-Optimal Space | Parallel coordinates across all solutions |
| Fig 5: Geographic Capacity Distribution per Solution | Map of where capacity is built in a chosen solution |
| Technology Deployment Correlation | Which technologies tend to appear together (dark) or replace each other (light) |
| Technology Classification | Each technology labelled as a must-have (in every solution), preferred (in at least 70 % of solutions) or a real choice (missing from at least one) |
The classification is often the most useful output for policy questions. A must-have is needed whatever else happens; a real choice can be avoided at an acceptable extra cost.