updated tclp test

[purboday]

ABM_CE_PV_Model.py

• Added new model parameters tclp_params for both “fresh” and “aged” modules (mean, std)
• Introduced age-dependent parameterization for hazard testing, allowing module hazard probability to vary smoothly with age. Implemented logistic interpolation between fresh and aged parameters.
• Updated the hazard classification rule to use a Weibull distribution, with a configurable threshold.

test_model.py

• Added new unit tests to validate the updated TCLP hazard logic. Tests cover age-dependent hazard classification and ensure correct behavior for both fresh and aged module scenarios.

[jwalzberg]

@purboday how is start_year determined? I imagine module_age_year should be dependent on the installation year rather than a fixed start_year? Some module have been installed prior to 2020 and some module will be installed after 2020? Is the code accounting for that by letting start_year be an input to this method?

[purboday]

Yes that is the idea that when this function is called the installation year would be passed as an argument. But I was not able to find a way to trace the origin of the waste produced and its installation date so for now this input is not really useful.

[jwalzberg]

What we could do is having a new agent variable self.installation_year that is initialized with the USPVDB p_year column. Then we could have a function that is more or less like that:

def update_install_year():
       # When reaching installation end-of-life, agent install new PV panels 
       if self.installation_year + self.model.product_lifetime == self.model.current_date.year:
                self.installation_year = self.current_date.year

In the consummer agent step function we can add this new function.

[jwalzberg]

@purboday looks all good to me, thanks for your hard work! I am just wondering about start_year but I am sure it is not fixed at 2020. Will look at the code more extensively to see what is fed to the tclp method for the start_year.

[jwalzberg]

@purboday looks all good to me, thanks for your hard work! I am just wondering about start_year but I am sure it is not fixed at 2020. Will look at the code more extensively to see what is fed to the tclp method for the start_year.

Actually... It seems that nothing is passed to the tclp_test method:

else: # If the TCLP test is applicable, check if the waste is hazardous # based on the TCLP test results. self.hazardous = self.model.tclp_test()

We need to find a way to pass installation year of waste. Let's discuss...

[purboday]

@purboday looks all good to me, thanks for your hard work! I am just wondering about start_year but I am sure it is not fixed at 2020. Will look at the code more extensively to see what is fed to the tclp method for the start_year.

Actually... It seems that nothing is passed to the tclp_test method:

else: # If the TCLP test is applicable, check if the waste is hazardous # based on the TCLP test results. self.hazardous = self.model.tclp_test()

We need to find a way to pass installation year of waste. Let's discuss...

Yes you are right @jwalzberg . I thought about it and there could be two options we can take:

1. Define some sort of probability distribution to estimate the installation year.
2. Just go with a static mean and sd considering all modules to belong to the "field aged" category.

Let's revisit this after the paper.