Soiling#

Soiling Module

This module is for analyzing soiling trends in performance index (PI) data and daily energy data.

class solardatatools.algorithms.soiling.SoilingAnalysis(data_handler)#

Bases: object

plot_analysis(figsize=None)#
run(**config)#
solardatatools.algorithms.soiling.soiling_seperation(observed, index_set=None, degradation_term=True, tau=0.85, w1=2, w2=0.03, w3=0.2, w4=500.0, iterations=5, solver=None, period=365, verbose=False)#

Apply signal decomposition framework to Performance Index soiling estimation problem. The PI signal is a daily performance index, typically daily energy normalized by modeled or expected energy. PI signal assumed to contain components corresponding to

  1. a soiling loss trend (sparse 1st-order differences)

  2. a seasonal term (smooth, yearly periodic)

  3. linear degradation

  4. residual

Parameters:

  • observed

  • index_set

  • degradation_term

  • tau

  • w1 – PWL weight - soiling term

  • w2 – sparseness weight - soiling term

  • w3 – asymmetric slopes - soiling term

  • w4 – smoothness weight - seasonal term

  • iterations

solardatatools.algorithms.soiling.soiling_seperation_old(observed, index_set=None, degradation_term=False, period=365, tau=0.85, w1=2, w2=0.01, w3=100, iterations=5, soiling_max=1.0, solver='MOSEK')#

Apply signal decomposition framework to Performance Index soiling estimation problem. The PI signal is a daily performance index, typically daily energy normalized by modeled or expected energy. PI signal assumed to contain components corresponding to

  1. a soiling loss trend (sparse 1st-order differences)

  2. a seasonal term (smooth, yearly periodic)

  3. linear degradation

  4. residual

Parameters:

  • observed

  • index_set

  • degradation_term

  • tau

  • w1 – PWL weight - soiling term

  • w2 – sparseness weight - soiling term

  • w3 – smoothness weight - seasonal term

  • iterations