In the general context of the technology assessment of power system technologies expected to impact grid architecture studies at 2050, a techno-economic database displaying data (i.e. technical performances, costs, environmental impact, etc.) that characterizes the different technologies for the next four decades, i.e. from today to 2050 has been produced.
Uncertainty due to the remoteness of time horizon and to the diversity of configurations was managed by providing ranges of values for some critical variables needed for the system simulations.
A simple and direct approach was used to project likely values from these uncertainty intervals according to each of the five e-Highway2050 scenarios. This process is called contextualization of data per scenario.
The main challenge relates to the management of uncertainty and data contextualization in the context of technology assessment for the grid architecture of the pan-European transmission system at 2050:
- How to contextualize technology cost and performance data consistent with a given energy scenario at 2050?
Background and assumptions
This knowledge article is to be read in the general context of the e-Highway2050 project. In this project, an energy-scenario approach was adopted and five energy scenario projections of likely futures are expected to grasp all likely evolutions of the power system at 2050. The methodologies to define these scenarios and their quantification are detailed in parallel knowledge articles. The five retained eHW scenarios are described in the knowledge article “e-Highway2050: Challenging energy scenarios for the pan European transmission system by 2050”.
More particularly, it is written in the particular context of the power system technology characterization. The scope of power system technologies covers the whole electricity value chain from generation and storage, transmission (passive and active transmission technologies) to demand. See related knowledge article on database characterization and its technological scope definition.
It is reminded that the most impacting variables describing technologies correspond to the two data types technology performance characteristics and costs1. They are the ones that have been detailed in-depth since of the highest interest for the power system simulations to be performed by the project.
Description of the result
The result includes two interrelated methodologies dealing with the management of data uncertainties in the techno-economic database of technologies and their contextualization to fit to each energy scenario at 2050.
Defining uncertainties and data contextualization
Uncertainties and contextualization are two closely related concepts that are defined as follows.
Uncertainties refer to the intervals of confidence of the values for given variables
For example, the value of a given variable at 2050 cannot be determined with certitude, i.e. 2100 MW (if one considers for instance the maximum power for a VSC station at 2050); it should rather be 2100 MW (+/- 10%) or [2080-2265] MW (it may vary within a min max interval).
The increasing uncertainty over time has been a major difficulty when assessing numerical values for several data types, such as costs or technical performances.
Contextualization refers to the different values that might be taken by a variable depending on the e-Highway2050 scenario
For example, in the scenario 100% RES with a high penetration of large scale renewables at 2050, one can expect that the installation costs of a VSC substation might be different from the ones in a scenario where renewables reach a lower penetration level and the thermal electricity generation is roughly at the same level as today (Large fossil & nuclear). In the latter, one could expect that the installation costs of a VSC station would be higher than in the former.
Origin of uncertainties and uncertainty management
When considering nature of uncertainties in the technology database, it is observed that uncertainties might result from three main different sources, cf. Table 1. The first two components are easy to address by focusing on a particular technological variant under given installation conditions. The remaining uncertainty will then be dealt with by considering ranges for the studied variables.
Table 1. Main sources of uncertainties in the technology database
The three natures of uncertainty are progressively managed by each stage of the general data construction process. Figure 1below shows how each successive step contributes to address the data uncertainty challenges.
- The first three steps (A, B, C) of the data construction process deal with the gathering by the experts of performances and costs data of a given technological variant (step A stands for an Archetypal representative of the technology family) and under a reference installation context (step B for Base case identification). Under these assumptions, typical trajectories of cost and technical performances (step C for Cost and performance trajectories) are built including uncertainty margins. The result is the production of min-max intervals for a selection of variables.
- The fourth step (step D for Data adjustment) deals with the fine-tuning of these min-max ranges according to the five scenarios defined by the project.
- The final one (step E for Explanation) consists in the quality and transparency issues of the whole process (a “Technology Assessment Report” is produced for each technology family gathering the construction assumptions and introducing the corresponding datasheet).
Figure 1: The data construction process for each technology area
Influence of the energy scenario: a simple “data contextualization” approach
The impact of scenarios is obtained through the so-called “Data Contextualization”. The key underlying assumption is that the main driver for contextualization is the penetration rate of the considered technology (characterized by an indicator such as the cumulated number of units at a given time). It is indeed assumed that the cost and performance trends of a technology by 2050 are directly correlated to its level of deployment.
A generic methodology has been developed for all technologies; the successive steps are displayed hereunder in the particular case of electric vehicles (EVs) for the sake of clarity:
- an overall qualitative assessment is made, which reflects for the given scenario, the deployment level of EVs, on a three degree scale (Low, Medium, High),
- in parallel, a subset of key technology variables describing EVs is selected, for example the penetration level (number of units by 2050), performances (driving range) and costs (battery and vehicle),
- from the value ranges attached to the selected key technology variables, the minimum, average, and maximum values are extracted,
- by combining the scenario assessments made at step 1 and the EV value tables built at step 3, specific values are allocated to the subset of EV variables (key technology variables) according to each given scenario (mapping of the minimum, average, maximum values with the Low, Medium, High scale depending on the type of variable).
The table below displays the results of data contextualization for Battery EVs (BEVs). Each scenario corresponds to a given penetration rate (High, Medium, Low) according to the analysis described above. The values of specific variables for each scenario are presented.
Table 2. Data contextualization for Battery Electric Vehicles
As an example, this means that for a scenario where the penetration of BEVs is Low, it is expected that the battery costs (250 €/kWh) will be higher than in a scenario where the penetration of BEVs is High (140 €/kWh), because of economies of scale, higher investment in R&D, etc., effects which result from a growth in the vehicle production.
Assessment of the methodology use and limitations
The methodology relies on the central assumption that the deployment of a technology is directly correlated to the cost and performance trends of a technology over a period of several decades. This might be tuned to the maturity of the technology. A high maturity technology should reach a kind of asymptotic performance while lower maturity technologies should be subject to higher rates of adoption and thus of cost reduction.
A second limitation is that the data contextualization does not take into account breakthrough type innovations which might brutally interrupt a technology trajectory.
Despite such limitations, the methodology allows some fine tuning of generic cost and technical data to the specificity of very different futures whose complexity is modeled by the five identified energy scenarios.
This article is connected to the following related complementary knowledge articles:
Deliverable D3.1 eHighway2050
The possible open access release of the database is currently under discussion.
1 the construction of future cost trajectories for transmission equipment based on their today’s estimated cost which is detailed in the knowledge article “e-Highway: a methodology for building future costs of transmission technologies in the framework of the technology assessment of power system technologies expected to impact grid architecture studies at 2050”.
2 For DEMAND technology area some wording adjustments are necessary (e.g. installation site replaced by consumer site).