WGEN Weather Simulator

• 

  Jason

  • January 04, 2019 19:32

  Model version 1.015 released on 4 Jan, 2019

  Version Report

  1. Added clarification to some element descriptions
  2. Changed the random number generator for determining wet/dry states in the Markov model from a Random Choice element to a stochastic with Uniform distribution.
  3. Added ability to override the random numbers with a user-defined value for the purposes of validation, if it is needed (so you can predict the outcome).

  1

  Comment actions Permalink
• 

  Jason

  • January 11, 2019 18:46
  • Edited

  Model version 1.020 released on 11 Jan, 2019

  Version Report

  1. Corrected the label for observed rain, which was mislabeled as "observed radiation" and placed in the radiation portion of the inputs. This change doesn't affect the model output.
  2. Hard-code in deterministic values so that user only needs to turn it on or off to debug in deterministic mode
  3. Minor clarifications in documentation of temperature calculations
  4. Some textbooks define the scale factor for the gamma distribution differently. Some use the notation of lambda and this is 1/beta. Clarified in the documentation.

  1

  Comment actions Permalink
• 

  Jason

  • January 26, 2019 00:14
  • Edited

  Model version 1.024 released on 25 Jan, 2019

  Version Report

  The harmonic value "x" in the temperature container should have an initial condition of vector(0) unless you are running it deterministically for verification purposes. This has been fixed using an IF statement: 

  if(Temp_Determ, vector(-0.3513, -1.754, 0.6244), vector(0))

  Where "Temp_Determ" is set to False by default. Only set this to True if you need to run a deterministic validation run.

  I was using a Random Choice element for determining the wet/dry state but it's not possible to use this if you want to switch to a deterministic run for validation. I rolled this back to an earlier version of the markov function that uses a stochastic with uniform distribution instead. This is how the FORTRAN code is implemented.

  1

  Comment actions Permalink
• 

  Jason

  • February 08, 2019 00:28

  Model version 1.025 released on 7 Feb, 2019

  Version Report

  1. Fixed random number generator in markov rain model. The stochastic element used to generate random numbers in the rainfall container was not set to trigger daily. The trigger was added.

  1

  Comment actions Permalink
• 

  Jason

  • March 06, 2019 18:03

  Model version 1.026 released on 6 Mar, 2019

  Version Report

  1. Precipitation chart was missing from the dashboard. This Result was added back in and placed on the dashboard.

  1

  Comment actions Permalink
• 

  Jason

  • May 13, 2019 21:00

  Model version 1.029 released on 13 May, 2019

  Version Report:

  1. Added support for units so that data can be entered in terms of metric or English units.

  2. Added results to support the differences in units

  1

  Comment actions Permalink
• 

  Jason

  • September 11, 2019 22:26
  • Edited

  Model version 1.031 released on 11 September, 2019

  Version Report:

  1. The x axis for the top 2 charts on the dashboard is redundant because the bottom chart axis can apply to all 3. Removed x axis on top to charts.
  2. Changed the title shown on the dashboard.

  0

  Comment actions Permalink
• 

  Jason

  • December 17, 2019 22:58

  Model version 1.035 released on 17 December, 2019

  Version Report:

  1. Applied units to main outputs (precip, min/max temp, and radiation)
  2. Result plots were updated
  3. Verification tests and charts were updated with units change

  These changes did not affect the underlying calculations in any way but make it easier to port the model to another model, which will need to carry those unit dimensions forward.

  0

  Comment actions Permalink

Please sign in to leave a comment.