romainsacchi / carculator Goto Github PK

Current

Tests with python 3.7 and 3.8 don't pass because of:

...
../../../../miniconda3/envs/premise-dev/lib/python3.8/site-packages/carculator/energy_consumption.py:135: in EnergyConsumptionModel
) -> tuple[Union[float, Any], Any, Union[float, Any]]:
E   TypeError: 'type' object is not subscriptable

/opt/conda/envs/py37/lib/python3.7/site-packages/carculator/energy_consumption.py:135: in EnergyConsumptionModel
) -> tuple[Union[float, Any], Any, Union[float, Any]]:
E   TypeError: 'type' object is not subscriptable

Expected

The readme of the repo says carculator is compatible with python 3.7 and onwards. If this is still the case, a fix for this bug is necessary.

86 parameters are defined twice in the file "default_parameters.json"

For example: ('Large', 'BEV', 'lightweighting', 2000) is defined here:

and here:

The double-defined parameters cause the following problem in carculator 1.9.0.dev1 with carculator_utils 1.2.0.dev6:

I have attached a file with all double defined parameters. Do you have a convenient way to remove the double defined parameters or is it necessary to remove them manually.
Double defined paramters.txt

Electric cars use electro-magnetic brake from the electric motor, reducing the need for friction brakes.

Hi Romain,
While I played with the custom parameters, I encountered such error as shown below:

I tried to fix it by restarting terminal and it works but again when after I restart notebook, the error comes again.

It seems that I cannot override the parameters in array (6,1) as they do not match the other elements.
Any lead or tip would be appreciated. Sorry if the question is not relevant and technical.

Best regards,
Rena

An idea while exploring the graphs of the midpoint indicators in the webtool: it might be handy to be able to (maybe hiërarchically?) group the presented graph by either year, car size or car type, for easier comparison over time for a specific group/year/car size.

Is it possible to go through / contribute to the webtool graphs? I could not find (yet) a reference in the domentation or in this repository, but if it is possible, I might be interested to look at it. I have some experience with R, Javascript, Dash/Plotly and Highcharts, if any of those (or something similar) has been used.

Thanks again for your work!

If I read this correctly, the combustion noise only kicks in above 20 km/h - but the rolling noise also only kicks in above 20 km/h. Is there no noise for slower combustion cars?

Hey,

I'm (starting to) exploring your rrreally cool LCA-modelling project (congratulations and thank you for developing this, this is what everybody needs!).

I don't know if this is the right place to ask these types of questions (I also tried on Twitter, but I have the impression that that medium is not the most frequently visited ;)), but:

On page 21 of your preprint, you mention that the tank-to-wheel E consumption of petrol fuelled ICEVs might decrease with 48 % by 2050 (180 gCO2/km -> 91 gCO2/km) due to decrease in curb mass (-8.5 %) and increase in powertrain and engine efficiency (+50%). I'm not sufficiently qualified in ICE combustion engine technology, but I was wondering on which paper/database/forecast this projection is based on (especially the powertrain/engine efficiency evolution) ? Intuitively, I would expect a rather constant engine efficiency, considering decades of ICE development. Or am I too naive and are carmakers only now starting to consider the fuel efficiency of their cars? :p

If you have time, I would be happy to get to know a bit more about this assumption, as I am intrigued by the forecast (at least for some countries) that in the long term ICEVs emission intensity might converge with those from BEV (as indicated in Figure 9 on p. 22 of the preprint):

Everytime when I run the set_all method this is what I get. Something related to pickle I can see

99.7% of the time consumed by fill_xarray_from_input_parameters is spent on:

for param in cip:
    for size in cip.metadata[param]["sizes"]:
        for powertrain in cip.metadata[param]["powertrain"]:
            array[
                size_dict[size],
                powertrain_dict[powertrain],
                parameter_dict[cip.metadata[param]["name"]],
                year_dict[cip.metadata[param]["year"]],
                :,
            ] = cip.values[param]

@romainsacchi
Good day Mr. Sacchi,
after installing the new version of carculator (1.8), unfortunately I get the value error "unsupported pickle protocol: 5" directly right after import carculator command.

The package from conda install -c romainsacchi carculator looks like has some version control problem, then I decided to install the package from PyPi but looks like the problem is still there.

I would appreciate any help regarding this issue

Hello

Impressive tool which can really be helpful!
I didn't do a dive in the data or code, but the quick results for Belgium (new cars), give only a 20% advantages for EVs compared to Diesel:
https://carculator.psi.ch/display_quick_results/BE

This would contradict other recent estimates like:
https://www.transportenvironment.org/what-we-do/electric-cars/how-clean-are-electric-cars

or:
https://www.carbonbrief.org/factcheck-how-electric-vehicles-help-to-tackle-climate-change

and also this tool:

https://climobil.connecting-project.lu/

Current

Installing carculator with pip requires "pytest"

Expected

Installing carculator doesn't bring a testing dependency

Hi there,

Currently, I am trying to import a Carculator excel file into Activity Browser (which runs on Brightway 2). However, when I try to import the excel file as a database, the whole AB environment shuts down. It might be that it's due to missing biosphere flows? @marc-vdm mentioned I could perhaps post the issue here. The reason I am trying to import the file is to get a relevant database to perform an LCA on EV technologies, as evoinvent does not include all EV related data.

What I did:

• I picked the advanced configurator on the Carculator website.
• I made two comparisons between electric and petrol vehicle for a 2022 and 2030 scenario.
• I calculated the results and exported the inventory database to an excel file (and did not modify the file).
• Then, I tried importing the file to the AB environment, however, then it shuts down and does not import the excel file.

A link to the excel file that I am trying to import to AB:
carculator_inventory_2022-04-13_for_ei_3.8_brightway2.xlsx

Do you know what I can do to fix this?

Thanks in advance!

Currently, it is assumed that the annual mileage is constant through time.
This assumption is quite important as it influences the lifetime-weighted electricity mix used to charge batteries or produce hydrogen.
But we know from data (for example here) that the annual mileage goes up after the first year of use, culminates between year 2 and 8 and slightly fades until the vehicle disposal.

It would be much better to adopt such distribution.

If the inventory is exported to an excel file, the hydrogen tank is not considered in the activity "Car, fuel cell electric, ..."

Furthermore, the wrong reference is mentioned for "Fuel tank, compressed hydrogen gas, 700bar, with carbon fiber". The reference is Benitez 2021 (https://doi.org/10.1016/j.jclepro.2020.123277) and not Boureima et al. 2009
Line 96 of https://github.com/romainsacchi/carculator_utils/blob/master/carculator_utils/data/export/references.csv:

Derived values from remind output, such as biofuel shares, moved to "intermediate" CSV files, e.g,

I could not find methods to generate these files from REMIND output, it is thus more difficult to work with scenarios different from the ones provided with PREMISE.

All activities "Car, …" (e.g., Car, diesel, Large, 2030, Euro-7) include “treatment of used glider” and “treatment of used internal combustion engine”. However, this waste treatment is already included in “glider production, passenger car” and in “internal combustion engine production, passenger car”. I would suggest to remove it from the activity "Car, …".
To be precise, you could consider "treatment of waste plastic" for the fuel tank and you could considered the reduced demand of “treatment of used glider” due to glider lightweighting directly in the activity "glider lightweighting".

following https://carculator.readthedocs.io/en/latest/validity.html#car-and-components-masses

With a new environment, installing from pip on Nov. 5, I get:

Maybe the parameter names changed?

I was wondering if it wouldn't make sense to consider equal glider base masses for all powertrains because all powertrain specific masses are considered separately.

In particular, I was wondering about the following weights:

• Large vehicle: PHEV-e weights 1650 kg all others 1550 kg
• Mini vehicle: PHEV-e weights 810 kg all others 700 kg
• Medium SUV: BEV and PHEV-e weight 1400 kg all others 1200 kg.

Based on this data: https://2020.entsos-tyndp-scenarios.eu/wp-content/uploads/2020/06/TYNDP-2020-Scenario-Datafile.xlsx.zip
Tab "MarketRun", which seems to describe the electricity generated for each scenario.

We should update the current electricity mixes used.
There are three scenarios: which one to use? All of them three?

A loop that sizes the vehicle's components and stabilizes the curb mass has been added in order to get the correct masses. It does not use so much time, but it feels inelegant. Aleksandra suggests to solve this with a matrix. To be continued...

calculate_ttw_energy() takes 99% of the time used by CarModel.set_all()