Skip to main content

v8 to v8.1

In this migration guide you will find:

  1. YAML changes

Yaml migration​

Migration overview​

This doc guides you through migrating an existing eCalc™ model from version v8 to v8.1.

We try to make this as easy as possible, and provide a step-by-step migration guide.

1. Changes to TIME_SERIES​

  • RATE_INTERPOLATION_TYPE is renamed to INTERPOLATION_TYPE
  • New time series type: DEFAULT with default RIGHT interpolation
  • RESERVOIR type is removed

Previously, it looked like this:

TIME_SERIES:
  - NAME: <time series reference name>
    TYPE: <MISCELLANEOUS/RESERVOIR>
    FILE: <path to file>
    INFLUENCE_TIME_VECTOR: <True/False>
    EXTRAPOLATION: <True/False>
    RATE_INTERPOLATION_TYPE: <LEFT/RIGHT/LINEAR>

But the new valid definition of time series in the yaml is now:

TIME_SERIES:
  - NAME: <time series reference name>
    TYPE: <MISCELLANEOUS/DEFAULT>
    FILE: <path to file>
    INFLUENCE_TIME_VECTOR: <True/False>
    EXTRAPOLATION: <True/False>
    INTERPOLATION_TYPE: <LEFT/RIGHT/LINEAR>
important

We have understood that the previous definitions lead to misunderstandings and errors. It is therefore of high importance that you now go through the reservoir data timeseries one-by-one and make sure that the correct type and interpolation type is set. See below for the 2 possibilities of migrating from the old RESERVOIR type.

If you previously used the RESERVOIR type, you know have 2 options. See below.

TIME_SERIES:
  - NAME: <time series reference name>
    TYPE: RESERVOIR
    ...

If you know that the reservoir data comes from Centuries, and/or that the data is right-interpolated, then you should change to the new DEFAULT type, like this:

TIME_SERIES:
  - NAME: <time series reference name>
    TYPE: DEFAULT
    ...
important

If you do NOT know the origin of the timeseries or whether it uses LEFT or RIGHT interpolation, you MUST contact somebody that can help you out to make this right. The consequences of doing it wrong is that the calculations is easily offset by 1 year.

2. Not possible to have different interpolation types for vectors within one file​

Previously eCalc™ tried to recognize vectors as rates- or non-rates for time series. Hence, rate-vectors were set to use LEFT interpolation type, or if the type was MISCELLANEOUS it was set to the method given by the user. If the vector was not recognized as a rate, the interpolation method was automatically set to LINEAR, regardless of user input.

This behaviour is now changed:

important

eCalc™ will no longer auto-detect rates and set interpolation type based on recognized rate vectors. If different vectors (e.g. pressures and rates) should be interpolated differently, they now need to be in separate files.

3. Empty data in time series columns no longer allowed​

Each column in a time series resource should have data for all time steps, eCalc™ will now show an error if this is not the case. The reason behind this is that it can be ambiguous to know whether missing data should be interpolated or considered as 0. Now users will have to be explicit, and this will lead to fewer ambiguities and errors.

4. New LTP Category: STEAM-TURBINE-GENERATOR​

A new LTP requirement to report steam turbine generator consumption/generation. This affects the total generator production negatively (reduced load), as some energy is provided through this steam turbine generator. It is therefore modelled as a consumer with a negative load in order to subtract from the total energy provided by the generator set.

The load on the steam turbine generator is reported separately in a new column in LTP Export.

Added new CATEGORY with name STEAM-TURBINE-GENERATOR to report power generated by a steam turbine. Should be negative load to deduct from genset. See excerpt example below:

    - NAME: steamgen
      CATEGORY: STEAM-TURBINE-GENERATOR
      ENERGY_USAGE_MODEL:
        TYPE: DIRECT
        LOAD: -1.1 # MW
important

Set negative load for STEAM-TURBINE-GENERATOR (similar to the way OFFSHORE-WIND is being used)