This README should explain how all of these files work

TRNSYS reads a TRD file, along with some supporting files, and writes hourly data to several for_##.dat files. The TRD has equations and connections between Components, which are written in Fortran and compiled seperately. Everything else here either produces inputs to TRNSYS or processes the outputs.

The buildings directory has files which describe the building dimensions and enclosure materials to TRNSYS. TRNSYS Type56 reads these files and models thermal and moisture loads. There is one set of files for each combination of HERS level (ie, energy efficiency), climate zone, and building size. The .bui files can be edited using TRNBuild.exe or as text files. The .bld, .inf, and .trn files are created by TRNBuild.exe when saving a .bui file. Single_Zone_Buildings.txt is a lookup file used to associate a given TRD with a given building file.

TRN_Resdh4.py takes an input TRD file, changes some variables within it, and calls TRNSYS on the output file. In its --batch mode, it reads a CSV file specifying source TRD, variable names, and values, and calls TRNSYS once per line in the CSV file. --batch mode will also accept multiple CSV files. TRN_Resdh4.py will place the outputs of each CSV file in a directory named after that CSV file. For example,

python TRN_Resdh4.py --batch sim-specs/s1.csv

will create a directory s1 with a subdirectory for each row in s1.csv.

python TRN_Resdh4.py --batch sim-specs/s1.csv sim-specs/s2.csv

will create two directories, s1 and s2.

For this project, all scenarios use 1449.TRD as the source file. All differences between scenarios are expressed in the CSV files. Examples include:

• TMY weather file (building location)
• building file (enclosure specification)
• AC capacity (tons)
• dehumidification system (affects many variables)
• RH setpoint

More information about the CSV file format which TRN_Resdh4 uses is given in the sim-specs section.

The CSV files are kept in the sim-specs/ directory (short for simulation specifications). TRN_Resdh4.py expects one row per scenario, and one column per variable. The variable names are taken from the first row of the CSV. The script gensim.py (also in sim-specs/) generates the CSV files.

Each scenario is described by 5 values, and these are combined into a compact name of the form z#h#s#rh#v#-# where each # is replaced by a number (except the last). The fields are:

• z: Climate Zone (0-5) (as a special case, 0 refers to Orlando, FL)
• h: HERS level, (50-130) a proxy for performance of enclosure and mechanical systems
• s: system, refering to the dehumidification systems enumerated in the Task 4 Report (under docs/)
• rh: Relative Humidity setpoint (50, 60)
• v: ventilation system, as follows:
  • 0: none
  • 1: Exhaust Only
  • 2: CFIS
  • 3: HRV
  • 4: ERV
• sz: building size, one of sm, md, lg (last # in name)

The main function, sim_line, takes six parameters corresponding to the bullet points above. Multiple CSV files are produced, one per system.

This directory contains scripts to summarize and display the outputs from TRNSYS. TRNSYS, as configured by the TRD files, writes several .dat files for each scenario. Each file has one row per hour, and one column per variable. Columns are separated by tabs. The variables are spread over several files; each file has a row for every hour of the year.

process.py is used to produce tables of annual values from the hourly data. It uses the same specification files as TRN_Resdh4.py, to correctly name each row of the summary. The fields used in the name are also used to sort the output data. process.py calls several functions in graphs.py, which uses the python library matplotlib to graph the hourly data.

process.py reads all of the hourly output for a given simulation run and creates a python object, which is passed to each output function. The member variables of this object are named after the column heads in the TRNSYS .dat files, while the values are numpy arrays of length 8760 (one floating point value per hour). The object is dynamically defined, so any fields appearing in the .dat files will be available. Modifying the hourly_data function, which reads the .dat files, is not necessary. New summary columns and summary files can be created by adding output_row lines to the summarize_csv function, which take input in the format described above.

physics.py contains psychrometric functions to convert between various units of temperature and humidity.

parametrics.py contains functions to parse the TRNSYS output format, and utility functions which are less general than those in physics.py, but likely to be useful in other TRNSYS projects. (The functions in process.py are mostly less general still, with notable exceptions.)

weather.py reads a TMY2 file and converts it into a numpy recarray including many of the most useful fields. It is not complete, however.

Several subdirectories within postprocessing contain the (incomplete) analysis of field data from an AAON heat pump with hot gas reheat, installed in New Orleans. 2011-graphs contains graphs produced early in the debugging of the RP-1449 project.

Several parts of the model take additional parameters. Because these are reused in many scenarios, across multiple projects, they are defined in separate lookup files. Examples, with some relevant parameters, include:

• Air Conditioner (CFM/ton, W/CFM, EER)
• Dehumidifier (CFM, moisture removal rate, W/CFM)
• Infiltration (shelter class, wind coefficient)

These are exposed as pulldown menus in the graphical interface to TRN_Resdh4, and are specified by row number in the CSV simulation specifications. TRN_Resdh4.py reads these lookups and writes the values into the TRD.

For each zone, HERS level, and building size, a Manual J calculation was performed using Elite RHVAC. These are the input files for RHVAC. The results are summarized in the Excel file excel/RP-1449 Calculations.xls. gensim.py writes the calculated tonnage for each scenario into the CSV simulation specifications.

This directory contains the Fortran code for TRNSYS components (types) which Hugh Henderson provided. During the RP-1449 project, only Hugh compiled these components, and the code in this directory is probably out of date --- not the code used for the final version of the report.