I'm working with LEAP 2018 version and my task is to analyse electricity generation mix for different countries from the point of different scenarios. One of the scenarios is Optimum scenario, which I'm trying to generate using LEAP optimisation feature (other two scenarios I've created myself). The inheritance of the Optimum scenario is current accounts as said in the guide book. After I've generated the results of three scenarios (transformation-capacity) I've noticed that Optimum scenario has the same installed capacity as it was in the current accounts.

"Optimized new capacity" section has "0 ?Calculated with GLPK on 18/02/18 00:11" in the expression lines for two scenarios (not Optimum) (please see attached, p.1).

In the diagnostics section (please see attached, p.2) it says "Dispatch exceeded plants rated availablity" and indicats the amounts in GJ. I hope this is the problem. For the dispatch rule I've used "Running cost". As I understand the numbers for dispatch come from "Historical production", but I cannot find the numbers indicated by LEAP as "rated availablity".

Could you please let me know why the Optimum scenario is similar to the current accounts situation and how I can fix this error.

Thnak you!

M.

The year which triggers the diagnostic messages (2015) falls inside your Current Accounts period. So, you are likely correct - unless you have modified the First Simulation Year variable, the amount of electricity which your processes produce in this year will be controlled by the Historical Production variable. The "rated availability" is not a single variable, but the product of the Maximum Availability and the installed capacity (Exogenous and/or Endogenous). The diagnostic message you see means that according to your capacity and availability assumptions, there was insufficient available capacity for the Historical Production that you specified. I don't think this is related to your optimization question.

You mentioned that your optimized scenario does not construct any additional capacity, but I see from your screen shot that it does - at least in 2016, for "New Wind". Can you clarify what you are expecting to see in your model?

Taylor

As to the Optimization, my idea is the following. I have to build an Optimum scenario for generation mix in different countries. In my model I've created two scenarios - Reference (REF) and European (EUR), and like in the Optimization exercise I've added Optimization scenario (OPT) which inherits from current accounts and CO2 Limit which inherits from OPT. In the branch "Electricity Generation-Processes" I've entrered different data on Exogenous capacities for REF and EUR, but not for OPT (I thought LEAP would calculate it itself). In the results I've got optimized capacity section for REF and EUR, but not for OPT.

If to compare my results with the Optimization exercise, things are vice versa: in Optimization exercise there is a calculated optimized capacity for OPT only. So my question is why there are such differences and whether I should create assumptions for OPT capacity changes as well (its not done by LEAP)?

Another issue I've got is connected to "Effects" brunch. When I entered data on emissions for current account and then constraints on emissions in CO2 Limit scenario the results generate an error message "no primal feasible solution in optimization calculation".

Please see the file attached.

Would highly appreciate any help on this!

Maya

From your screenshots, it looks like your REF scenario is the one in which you have enabled the optimization variables, but not in your OPT scenario. Are you sure you've set the Transformation\Electricity Generation:Optimize variable to "Yes" under the correct scenario? I also see that your model contains more than one region. Make sure that when you do change the Transformation\Electricity Generation:Optimize variable, you do so in the region from which all other regions inherit. LEAP will not permit you to use the optimization features in some regions but not in others.

Finally, I would recommend including one common "parent" scenario (other than Current Accounts) that both REF and OPT inherit from. This is so that you can be sure both REF and OPT share the same assumptions or projections in the future. Right now, based on your screenshots, I can't rule out that your energy demands are different in the REF and OPT scenarios (which would cause them to add capacity differently), since they do not inherit from a common baseline scenario.

Let's wait on including any emission constraints under the Effects branches until you have the rest of your model working properly. But note that the Annual Emissions Constraint applies to all sources of emissions (demand and supply) which appear above your optimized transformation module in the tree. So if your demand-only emissions exceed the allowed emissions target, there will be no more "emission budget" for your electricity generation module to operate - hence, you will find that no solution can be found. This message is almost always the result of two or more constraints which logically conflict with one another.

Taylor

Thank you very much for your recommendations. You were right - I didnt have YES to optimize under the Optimum scenario. I've also created a parent scenario for both the REF and OPT - it helps a lot.

Now I have another issue - I'm a bit lost in what the LEAP results actually show. I've indicated both Exogenous and Endogenous capacities for REF and EUR scenarios. In case I put YES to "optimize" for Optimum scenario only (like in Optimization exercise), the results do not show endogenous capacities for the scenarios other than the Optimum scenario. In case I put YES to "optimize" for all the scenarios the endogenous capacity is shown and can be extremely high for some regions (for example, for the European scenario for Germany given that the demand projection is much lower than that for the Basis scenario (Current accounts)). The Optimization scenario shows the same results in both cases and is very different (smaller amount of total generating capacity with only one type of capacity optimized) compared to other scenarios. I wonder whether these results are logical and whether I have to optimize all the scenarios in order to see what is the optimum generation mix suggested by the LEAP.

Thanks a lot for your time,

Maya

For non-optimized scenarios, you need to specify the minimum amount of process capacity to be added in the "Endogenous Capacity" tab if you want LEAP to control when new capacity is added.

Refer to the following help pages for more information on the endogenous capacity variable and how endogenous capacity is added in non-optimized scenarios:

https://www.energycommunity.org/Help/Transformation/Endogenous_Capacity.htm.

https://www.energycommunity.org/Help/Transformation/Endogenous_Capacityu_Expansion_Calculations.htm

I also highly recommend reading the following help page to understand the different approaches of specifying capacity data in LEAP: https://www.energycommunity.org/Help/Transformation/Specifying_Capacty_Data_in_LEAP.htm

Note that endogenous capacity is added to meet the planning reserve margin, so check this value to make sure it is specified correctly. LEAP also considers the capacity credit when adding new endogenous capacity, so a process with a low capacity credit (like wind power) may be added at a greater rate under the optimization scenario unless you constrain the capacity through the "Maximum Capacity" or "Maximum Capacity Addition" variables.

Thanks,

Emily

Thank you a lot for your recommendations and links (some of them were extremely useful for understanding the calculations). It seems that the model is now working, except for:

1) As I understand the Optimization scenario shows the least cost scenario among the scenarios which I've created in my model. As I understand this is evaluated by "social costs". In my case in some countires the social costs in Optimization scenario are higher than in other scenarios. Can it be appropriate or something is wrong in my calculations?

2) Is there any documentation or link describing the constranits for OSEMOSYS modelling used for LEAP?

Thank you a ot in advance,

Maya

To answer your questions:

Emily

I'm sorry for the late reply. I've attached a file containing the input data and the results for my capacity planinng modelling. In non-optimized scenarios (reference and european), I've indicated endogenous capacity, but its not shown in the results unless I set the planning reserve margin close to 100% (currently it is 30% (in reality-10-15%)). At the same time, the capacity results for the Optimum scenario are the highest and the social costs in the Optimum scenario are the highest as well. However, there are no limits on max and min capacity, no RES/carbon limit tragets. The reserve margin in the results shows high volatility between the scenarios, even above 100%. I'm not sure whether it has any logic.

Would be grateful for any suggestion.

Kind regards,

Maya

Thanks for providing the additional information. It appears that you have sufficient existing capacity to meet demands and reserve requirements for the Basis/European/Reference scenarios which is why there is no endogenous capacity being added. Did you intend to model exogenous capacity this way? Are you incorporating plant retirements when specifying the exogenous capacity? I would check the exogenous capacity variable first.

Also, because of the way the exogenous capacity is specified, the Basis Scenario and Optimisation Scenario have higher costs. Most of the costs are related to the operations and maintenance of existing plants.

It is unclear why there is some endogenous capacity being added under the Optimisation Scenario - does it have higher demands/energy requirements than the other scenarios? Is there endogenous capacity specified for the Basis Scenario? I recommend taking a look through some of the other results, including the Energy Balance, to better understand what is happening in each scenario.

Thanks,
Emily