README.md

In tinnaing347/bRema: Building Energy Modelling and Analysis Tool in R

bRema: Building Energy Modeling and Analysis in R

Overview

bRema is Building Energy Modeling and Analysis package developed in R by CUNY Building Performance Lab. bRema allows users to call multiple functions and build desired outside air temperature-dependent linear change-point regression models using monthly energy consumption, and use their own statistical metric thresholds for choosing the best model. Users can also take advantage of built-in plotting functions to visualize models and perform additional statistical analysis using built-in statistical functions. bRema can be used to create models for portfolio analysis or for measurement and verification activities.

Installation

install.packages('devtools')
#bRema uses plotly to plot parameter model graphs and timeseries
install.packages('plotly')
devtools::install_github('cunybpl/bRema')

Usage

On the very basic level, given a simple dataset of outside air temperature and energy consumption (either electricity or thermal energy) of a single building, bRema will perform segmented linear regression, generate five different linear change-point regression models that fit the dataset to varying degrees, and select the best model based upon statistical parameters. On a more advanced level, if input data is structured in a similar format as the included sample dataset (See bRema::unretrofit_utility), bRema can batch-run hundreds of buildings in a matter of a few minutes and generate a LEAN analysis (Kissock, 2004).

Following is a simple example that shows how bRema can model a data frame of outside air temperature and energy consumption:

#monthly average outside air temperature
x <- c(78.61290, 78.77419, 73.73333, 58.06452, 53.46667, 50.35484,
        35.64516, 38.24138, 48.35484, 52.46667, 62.87097, 73.13333,
        80.25806, 80.80645, 73.56667, 60.70968, 52.13333, 40.45161,
        39.54839, 42.21429, 40.64516, 57.21212, 62.70968, 73.40000)
#monthly average electricity usage normalized by square footage
y <- c(0.05996058, 0.05959213, 0.05592924, 0.03800914, 0.03652459,
        0.03190152, 0.03549965, 0.03550670, 0.03681476, 0.03813243,
        0.04048421, 0.05412409, 0.06184637, 0.06345264, 0.05440611,
        0.03750763, 0.03192324, 0.03497471, 0.03475597, 0.03480360,
        0.03242668, 0.03065114, 0.03763059, 0.04835995)
#constructing utility data frame
util <- data.frame(OAT = x, usage = y, energy_type = 'Elec')
#segmented linear regression, modeling
result = bRema::run_model(util, plot_flag = TRUE)
#finding the best fitted model
best_model_list = bRema::main_best_model_func(result, energy_n = 'Elec')
best_model_list$figure

To model a dataset of multiple buildings similar to sample data provided in the package, use the bRema::batch_run function to batch run multiple buildings for analysis.

library(bRema)
#batch running and modelling
batch_result = batch_run(unretrofit_utility)
#a data frame that includes information about models and statistics
best_model_df = batch_result$best_result_df
#performing post model calculations such as heating load, and lean analysis ranking 
post_df = main_post_model(unretrofit_utility, best_model_df, rank_flag = TRUE)

License

This project is licensed under the MIT License.

tinnaing347/bRema documentation built on Aug. 23, 2019, 9:15 p.m.