Simulation results

[JoachimBenner]

I think the EnergyADE should be able to represent also the output data of simulation systems. At the moment, it is possible to include simulated energy demands into an EnergyADE file, but not, e.g., the simulated actual temperature of a Thermal Zone.

Because the ADE supports different types of simulations and mutiple simulation systems, it is quite difficult to model arbitrary simulation results. Therefore, I propose to start with a very simple and generic data type SimulationData: A text attribute describing the simulated quantity (e.g. actual temperature) and a Time Series with time dependent, scalar data. This data structure can be referenced by every city object (ADE attribute of _CityObject). In future versions of the standard, it might be extended for other types of simulation data, or for supporting references to data bases.

[RomainNouvel]

I wouldn't create an Object specific to simulation data. In TimeSeries, we have this attribute "AcquisitionMethod" which can be either measures, simulation, calibrated simulation. The rest of the attributes is similar if it is measurement or simulation data. When the DynamizerADE will be introduced, you will be able to integrate all your simulation results in the CityGML file (if you think it makes sense). In between, I wouldn't charge the Energy ADE with a further temporary object...

[JoachimBenner]

I did not attend the meeting in Ferrara, and minutes of the meeting are not (yet?) available. When will this "Dynamizer ADE" be available, and what will be the relation of EnergyADE and Dynamizer ADE? Is it planned to eliminate the TimeSeries classes of the EnergyADE, because time series are also represented in Dynamizer? This would have (among others) the consequence that in future the EnergyADE could not be used stand-alone, but only in combination with the DynamizerADE.

[RomainNouvel]

Neither Alexandru nor Kanishk were there in Ferrara unfortunately, then we haven't discuss much about Dynamizer there. Both are working on the compaibility of Dynamizer development with the Energy ADE. In any case, we should have a strategical discussion with them and maybe other participants early next year about a possible integration ...

[JoachimBenner]

I agree, let's postpone the issue until next year.

[JoachimBenner]

I want to point out once more that the next version of the Energy ADE needs a feature type to represent the results of a simulation in the XML file. Please notice that gbXML also has such a concept with the class Results, aggregating regular time series.

From my point of view, this issue has very little to do with the Dynamizer ADE. It is unnecessary and much to complicated to use the central Dynamizer concept, i.e. the transformation of static properties into time dependent ones, for the representation of simulation results. If Dynamizer provides additional concepts to represent time series, we can check whether thay have benefits compared to the existing Energy ADE classes, but the Energy ADE in any case must be usable without the Dynamizer ADE.

The first proposal for the feature type SimulationResults comprises (1) meta data of the simulation, (2) aggregated simulation resultes and (3) time series of simulation results. Comments are welcome!

[RomainNouvel]

Hi Joachim,
What about organising a conf call with the contributors of WG time data ( @kanishk-chaturvedi and Alexandru Nichersu) as well as building Simulation users ( @mlauster , @jaykae26 ...) to discuss about a final solution to this long issue... 1st week of April?

[JoachimBenner]

No problem for me, when we avoid Monday afternoon.

[JoachimBenner]

This is a second and extended version of the proposed Feature Type. In Grenoble, I can present and justfy my proposal, which of course needs to be discussed and further enhanced.
EnergySimulationResults.pdf

[JoachimBenner]

Taking into account the discussions during the last workshop im Grenoble, I generated a new and much more general proposal. It now represents Observation Data, which comprises simulation results as well as sensor or measurement data.

In Grenoble, we discussed that concepts of the Dynamizer ADE, esp. the OGC standard TimeSeriesML, could be used alternatively. For me, it is very difficult to assess the consequences without a more detailled knowlege about the Dynamizer data model. @kanishk-chaturvedi, can you provide the UML model of Dynamizer for the Energy ADE workking group? At least for me, this would be very helpful.

I will setup a doodle to find the date for a Web Session, where we can further discuss the representation of Observation Data in the Energy ADE.

ObservationData.pdf

[JoachimBenner]

The Energy ADE plenum decided at December 6th, 2017 to postpone the issue for later versions of the Energy ADE