How to perform sensitivity analysis using eplusr? #553
-
|
Transferred from one user's email:
|
Beta Was this translation helpful? Give feedback.
Replies: 5 comments 82 replies
-
|
Here I use the
Below is a code snippet showcasing the whole workflow: # use an example file from EnergyPlus v22.1 for demonstration here
path_idf <- eplusr::path_eplus_example(22.1, "5Zone_Transformer.idf")
path_epw <- eplusr::path_eplus_weather(22.1, "USA_CA_San.Francisco.Intl.AP.724940_TMY3.epw")
# create a parametric job
param <- eplusr::param_job(path_idf, path_epw)
# define a measure about how to modify the IDF
measure <- function(idf, efficiency, thickness, conductivity) {
idf$set(
`Supply Fan 1` = list(Fan_Total_Efficiency = efficiency),
Material := list(Thickness = thickness, Conductivity = conductivity)
)
idf
}
# define parameter ranges and levels
vars <- tibble::tribble(
~name, ~min, ~max, ~levels,
"efficiency", 0.1, 1.0, 5,
"thickness", 0.01, 0.08, 6,
"conductivity", 0.1, 0.6, 6
)
# use {sensitivity} package to create parameters
mo <- sensitivity::morris(
model = NULL, factors = vars$name, r = 12,
design = list(type = "oat", levels = vars$levels, grid.jump = 4),
binf = vars$min, bsup = vars$max, scale = TRUE
)
#> Registered S3 method overwritten by 'sensitivity':
#> method from
#> print.src dplyr
# create parametric models
param$apply_measure(
measure,
efficiency = mo$X[, "efficiency"],
thickness = mo$X[, "thickness"],
conductivity = mo$X[, "conductivity"]
)
#> [7/48] | 15% ■■■■■ [Elapsed: 1.1s]
#> [25/48] | 52% ■■■■■■■■■■■■■■■■■ [Elapsed: 3.7s]
#> [44/48] | 92% ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ [Elapsed: 6.6s]
#> [48/48] | 100% ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ [Elapsed: 7.2s]
#> Measure 'measure' has been applied with 48 new models created:
#> [01]: measure_1
#> [02]: measure_2
#> [03]: measure_3
#> [04]: measure_4
#> [05]: measure_5
#> [06]: measure_6
#> [07]: measure_7
#> [08]: measure_8
#> [09]: measure_9
#> [10]: measure_10
#> [11]: measure_11
#> [12]: measure_12
#> [13]: measure_13
#> [14]: measure_14
#> [15]: measure_15
#> [16]: measure_16
#> [17]: measure_17
#> [18]: measure_18
#> [19]: measure_19
#> [20]: measure_20
#> [21]: measure_21
#> [22]: measure_22
#> [23]: measure_23
#> [24]: measure_24
#> [25]: measure_25
#> [26]: measure_26
#> [27]: measure_27
#> [28]: measure_28
#> [29]: measure_29
#> [30]: measure_30
#> [31]: measure_31
#> [32]: measure_32
#> [33]: measure_33
#> [34]: measure_34
#> [35]: measure_35
#> [36]: measure_36
#> [37]: measure_37
#> [38]: measure_38
#> [39]: measure_39
#> [40]: measure_40
#> [41]: measure_41
#> [42]: measure_42
#> [43]: measure_43
#> [44]: measure_44
#> [45]: measure_45
#> [46]: measure_46
#> [47]: measure_47
#> [48]: measure_48
# run parametric simulations (in a temporary folder for demonstration)
param$run(dir = tempdir())
#> ── EnergPlus Parametric Simulation Job ─────────────────────────────────────────
#> • Path: 'C:\EnergyPlusV22-1-0\ExampleFiles\5Zone_Transformer.idf'
#> • Version: 'C:\EnergyPlusV22-1-0\WeatherData\USA_CA_San.Francisco.Intl.AP.7249…
#> • EnergyPlus Version: '22.1.0'
#> • EnergyPlus Path: 'C:\EnergyPlusV22-1-0'
#> Applied Measure: 'measure'
#> Parametric Models [48]:
#> [01]: 'measure_1.idf' <-- SUCCEEDED
#> [02]: 'measure_2.idf' <-- SUCCEEDED
#> [03]: 'measure_3.idf' <-- SUCCEEDED
#> [04]: 'measure_4.idf' <-- SUCCEEDED
#> [05]: 'measure_5.idf' <-- SUCCEEDED
#> [06]: 'measure_6.idf' <-- SUCCEEDED
#> [07]: 'measure_7.idf' <-- SUCCEEDED
#> [08]: 'measure_8.idf' <-- SUCCEEDED
#> [09]: 'measure_9.idf' <-- SUCCEEDED
#> [10]: 'measure_10.idf' <-- SUCCEEDED
#> [11]: 'measure_11.idf' <-- SUCCEEDED
#> [12]: 'measure_12.idf' <-- SUCCEEDED
#> [13]: 'measure_13.idf' <-- SUCCEEDED
#> [14]: 'measure_14.idf' <-- SUCCEEDED
#> [15]: 'measure_15.idf' <-- SUCCEEDED
#> [16]: 'measure_16.idf' <-- SUCCEEDED
#> [17]: 'measure_17.idf' <-- SUCCEEDED
#> [18]: 'measure_18.idf' <-- SUCCEEDED
#> [19]: 'measure_19.idf' <-- SUCCEEDED
#> [20]: 'measure_20.idf' <-- SUCCEEDED
#> [21]: 'measure_21.idf' <-- SUCCEEDED
#> [22]: 'measure_22.idf' <-- SUCCEEDED
#> [23]: 'measure_23.idf' <-- SUCCEEDED
#> [24]: 'measure_24.idf' <-- SUCCEEDED
#> [25]: 'measure_25.idf' <-- SUCCEEDED
#> [26]: 'measure_26.idf' <-- SUCCEEDED
#> [27]: 'measure_27.idf' <-- SUCCEEDED
#> [28]: 'measure_28.idf' <-- SUCCEEDED
#> [29]: 'measure_29.idf' <-- SUCCEEDED
#> [30]: 'measure_30.idf' <-- SUCCEEDED
#> [31]: 'measure_31.idf' <-- SUCCEEDED
#> [32]: 'measure_32.idf' <-- SUCCEEDED
#> [33]: 'measure_33.idf' <-- SUCCEEDED
#> [34]: 'measure_34.idf' <-- SUCCEEDED
#> [35]: 'measure_35.idf' <-- SUCCEEDED
#> [36]: 'measure_36.idf' <-- SUCCEEDED
#> [37]: 'measure_37.idf' <-- SUCCEEDED
#> [38]: 'measure_38.idf' <-- SUCCEEDED
#> [39]: 'measure_39.idf' <-- SUCCEEDED
#> [40]: 'measure_40.idf' <-- SUCCEEDED
#> [41]: 'measure_41.idf' <-- SUCCEEDED
#> [42]: 'measure_42.idf' <-- SUCCEEDED
#> [43]: 'measure_43.idf' <-- SUCCEEDED
#> [44]: 'measure_44.idf' <-- SUCCEEDED
#> [45]: 'measure_45.idf' <-- SUCCEEDED
#> [46]: 'measure_46.idf' <-- SUCCEEDED
#> [47]: 'measure_47.idf' <-- SUCCEEDED
#> [48]: 'measure_48.idf' <-- SUCCEEDED
#> Simulation started at '2023-05-15 19:06:02.362011' and completed successfully after 7.68 secs.
# extract output
# here use total site energy for demonstration
eng <- param$tabular_data(
table_name = "site and source energy",
column_name = "energy per total building area",
row_name = "total site energy")
eng <- as.numeric(eng$value)
# calculate sensitivity metrics
mo <- sensitivity::tell(mo, eng)
sen <- data.frame(
mu = apply(mo$ee, 2, mean),
mu.star = apply(mo$ee, 2, function(x) mean(abs(x))),
sigma = apply(mo$ee, 2, sd)
)
# plot the results
plot(mo)Created on 2023-05-15 with reprex v2.0.2 |
Beta Was this translation helpful? Give feedback.
-
|
Thank you again for the scripts. Following the script you demonstrated on modifying the IDF. I want to ask if it's possible to modify the IDF, thesame way you did, but with measures/material properties, like the thermal transmittance, thermal capacitance which are not included in EnergyPlus IDD and IDF editor (contrary to conductivity, and thickness you used which are present in Energyplus IDF editor). |
Beta Was this translation helpful? Give feedback.
-
Yes, it is possible. But since everything at the end is converted to an IDF, you must define the relationship between your custom parameters and IDD fields. E.g., calculate the specific heat of a material given its thermal capacitance and mass by yourself.
As the function name suggests, Currently EnergyPlus v23.1.0 support is not included in the CRAN version of {eplusr}. You can install the development version via: |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the reply. That means the c(paste()) function in R can be to define the files' path when the custom EPW and IDF files are being used? For the variables and IDD file, I would like more shed of light. I already have minimum and maximum values of all my input variables and also their corresponding units, but those not in the IDD file and how to go about them still remain a draw back for me. I really appreciate your time and help. |
Beta Was this translation helpful? Give feedback.
-
|
Good morning Hongyuanjia, For example, in my custom input data, I have thermal capacity of a material (KJ/m2K) which was not given in Energyplus , and a field, specific heat (J/KgK) was given. I am sure you meant I should convert the values of the heat capacity (my custom) to the specific heat right? Similarly, in the "Advanced Construction, Surface, Zone Concepts" object in Energyplus, under the |
Beta Was this translation helpful? Give feedback.
-
|
Hi honggyuanjia Do the quote and without quote make any difference? Because in my case, I have Materials and also WindowMaterial:SingleGlazingSystem as new objects to modify. |
Beta Was this translation helpful? Give feedback.
-
|
It should be quoted if it is not a valid R object name. |
Beta Was this translation helpful? Give feedback.
-
That's right!
For this one, I am not sure. In EnergyPlus, it does not use thermal transmittance to calculate thermal heat gains and cooling/heating loads. Each component of heat transfer for surfaces is calculated separately, e.g. convection, conduction and radiation. The actual used value of the convection coefficient value of the inside and outside surface is based on the algorithm defined in However, EnergyPlus does report construction U values in the HTML tabular report under "Envelope Summary" section. Those values are calculated based on the conductivity of each material of construction and a fixed value of convection heat transfer coefficient value of 0.5 W/m2.K, if I recall correctly. Those values are just for demonstration and are not used in the actual calculation. If you want to specify the thermal transmittance of a construction, what you will normally do is modify the corresponding material of the wall/roof/floor, plus consider a fixed convection heat transfer coefficient. You can use the same value as in EnergyPlus tabular report. |
Beta Was this translation helpful? Give feedback.
-
|
This is just a sample before I input the real variables in the work I am doing, Yes I have quite some parameters to modify. Ideally in the work I am doing, I have the following parameters to modify: thermal transmittance of the wall, thermal transmittance of the roof, thermal transmittance of the windows, solar heat gain coefficient of the window, solar absorptance of the walls, solar absorptance of the roof, thermal capacity of wall, thermal capacity of the roof, thermal capacity of the floor, thermal transmittance of the floor, fraction of ventilation area by the opening and so other variables. Please what do you think about modifying these variables? |
Beta Was this translation helpful? Give feedback.
-
|
So in total, I am passing at least 15 variables for modification. |
Beta Was this translation helpful? Give feedback.
-
No. As the names indicate, they are just calculated design cooling/heating loads, which are different from cooling/heating energy consumption. If you want to calculate CEC or HEC, you have to have values about:
You can easily get total pump and fan energy in tabular HTML output. What is little bit tricky here is how to find an output variable that can indicate the system working condition, i.e. heating or cooling. You can output
|
Beta Was this translation helpful? Give feedback.
-
Then you could write a measure function in some way like this: measure <- function(
wall_trans, wall_thick, wall_solar_abs, wall_therm_capacity,
roof_trans, roof_thick, roof_solar_abs, roof_therm_capacity,
win_trans, win_shgc,
...
) {
# calculate corresponding material conductivity here using wall_thick and
# wall_trans
wall_conduct <- ...
# calculate corresponding material specific heat using wall_therm_capacity
# wall_thick, wall area and the material density
wall_specific_heat <- ...
idf$set(
YourWallMaterialName = list(
thickness = wall_thick,
conductivity = wall_conduct,
specific_heat = wall_specific_heat,
solar_absorptance = wall_solar_abs
),
YourRoofMaterialName = list(
# similar process
)
)
}
|
Beta Was this translation helpful? Give feedback.
-
|
Okay what I am asking specifically here is, in yesterday's codes that you suggested, we have something like this: Here, I assume wall_conduct would hold a value, right? I believed you mean multiplying the Or do you mean right from my I believe you also mean I should calculate the (# calculate corresponding material specific heat using wall_therm_capacity I understood how you tried to relate the u-value and the heat capacity to the conductivity and specific heat in EnergyPlus, for modifying the different layers of materials. But the values to use among either min or max custom variables that I am not sure. |
Beta Was this translation helpful? Give feedback.
-
|
Dear Hongyuanjia,
I want to apologize for my persistent emails to you lately. I have been
finding your efforts on github truly helpful, and very much fit my learning
process in EnergyPlus and R. I understand your very tight schedule which
has most probably caused you not responding to all my questions there.
I eagerly await your valuable responses soon.
Thanks once again.
Best regards,
Ibrahim
…On Fri, May 19, 2023, 8:01 AM Hongyuan Jia ***@***.***> wrote:
It should be quoted if it is not a valid R object name. Supply Fan 1 is
not a valid R object name and should be quoted using `, since it contains
spaces. WindowMaterial:SingleGlazingSystem is not a valid R object name,
since it contains a :. If you are not familiar with the naming rules in
R, you can check Names and values
<https://adv-r.hadley.nz/names-values.html#non-syntactic> chapter in
Hadley's *Advanced R* book.
—
Reply to this email directly, view it on GitHub
<#553 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A3UH7YXUZU2PVAW7SOOO4YLXG4LETANCNFSM6AAAAAAYBS4ZF4>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Thank you very much.
I will try the codes and get back to you on the outcome.
Much appreciated.
…On Fri, May 19, 2023, 4:09 PM Hongyuan Jia ***@***.***> wrote:
So in total, I am passing at least 15 variables for modification.
Then you could write a measure function in some way like this:
measure <- function(
wall_trans, wall_thick, wall_solar_abs, wall_therm_capacity,
roof_trans, roof_thick, roof_solar_abs, roof_therm_capacity,
win_trans, win_shgc,
...
) {
# calculate corresponding material conductivity here using wall_thick and
# wall_trans
wall_conduct <- ...
# calculate corresponding material specific heat using wall_therm_capacity
# wall_thick, wall area and the material density
wall_specific_heat <- ...
wall_condct <- wall_trans /
idf$set(
YourWallMaterialName = list(
thickness = wall_thick,
conductivity = wall_conduct,
specific_heat = wall_specific_heat,
solar_absorptance = wall_solar_abs
),
YourRoofMaterialName = list(
# similar process
)
)
}
—
Reply to this email directly, view it on GitHub
<#553 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/A3UH7YXTRGAFDQVF4XVYY5LXG6ELXANCNFSM6AAAAAAYBS4ZF4>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Sorry, but without a minimal reproducible example, I could not help anymore on my side. |
Beta Was this translation helpful? Give feedback.
-
|
Okay! Sorry about that. What I am saying is, can you kindly use the "code snippet" below for annual simulation for reproducibility? With library(eplusr)
library(sensitivity)
idf=eplusr::read_idf("C:/Users/DELL USER/Documents/pasta/IDF_file.idf")
epw = eplusr::read_epw("C:/Users/DELL USER/Documents/pasta/NGA_KN_Kano.Intl.AP.650460_TMYx.2007-2021.epw")
# add output variables
vars <- c(
# cooling/heating indicator for each coil
coil_clg = "Cooling Coil Total Cooling Energy",
# all kinds of energy
fan = "Fan Electricity Energy",
pump = "Pump Electricity Energy",
chiller = "Chiller Electricity Energy",
boiler = "Boiler NaturalGas Energy"
)
idf$add(`Output:Variable` := list("*", vars, "Timestep"))
# make sure annual simulation is triggered by removing all existing run periods
# and adding a new one expanding the whole year
idf$RunPeriod <- NULL
idf$add(RunPeriod = list("AnnualSim", 1, 1, NULL, 12, 31))
# save the model
idf$save(tempfile(fileext = ".idf"))
# create a parametric job
param <- param_job(idf,epw)
# Define a measure to modify IDF
measure <- function(idf,
Wall_therm_trans,Wall_therm_cap, Wall_sol_abs,Roof_therm_trans,
Roof_therm_cap,Roof_sol_abs, Floor_therm_trans,Floor_therm_cap,
Win_therm_trans,Cooling_Setpoint, SHGC_win){
idf$set(
`8IN Concrete HW` = list( conductivity = Wall_therm_trans,
specific_heat = Wall_therm_cap,
solar_absorptance = Wall_sol_abs
),
`Roof Membrane` = list(
conductivity = Roof_therm_trans,
specific_heat = Roof_therm_cap,
solar_absorptance = Roof_sol_abs
),
`MAT-CC05 4 HW CONCRETE` = list(
conductivity = Floor_therm_trans,
specific_heat = Floor_therm_cap
),
`Uglass` = list(
U_Factor = Win_therm_trans,
Solar_Heat_Gain_Coefficient = SHGC_win),
`Thermostat` = list(
Constant_Cooling_Setpoint = Cooling_Setpoint)
)
idf
}
pasta=c("pasta")
factors.path=c(paste(pasta,"/Input_file - Copy.csv",sep=""))
factors.classes=c("character","character","numeric","numeric")
variables=read.table(file=factors.path,
header=TRUE,sep=",",colClasses=factors.classes)
print(variables)
variables.names=variables[[1]]
variables.lower=variables[[3]]
variables.upper=variables[[4]]
mo <- sensitivity::morris(model=NULL,factors=variables.names,r=12,
design=list(type="oat",levels=8,grid.jump=4),
binf=variables.lower,bsup=variables.upper,scale=TRUE)
param$apply_measure(
measure,
Wall_therm_trans = mo$X[, "Wall_therm_trans"],
Wall_therm_cap = mo$X[, "Wall_therm_cap"],
Wall_sol_abs = mo$X[, "Wall_sol_abs"],
Roof_therm_trans = mo$X[, "Roof_therm_trans"],
Roof_therm_cap = mo$X[, "Roof_therm_cap"],
Roof_sol_abs = mo$X[, "Roof_sol_abs"],
Floor_therm_trans = mo$X[, "Floor_therm_trans"],
Floor_therm_cap = mo$X[, "Floor_therm_cap"],
Win_therm_trans = mo$X[, "Win_therm_trans"],
SHGC_win = mo$X[, "SHGC_win"],
Cooling_Setpoint = mo$X[, "Cooling_Setpoint"]
)
#Run parametric simulations
param$run(dir = "D:/File", echo = FALSE)
calc_energy <- function(param) {
# only extract data for annual simulation
rp <- "AnnualSim"
coil_clg <- param$report_data(name = vars["coil_clg"], environment_name = rp)
elec_fan <- param$report_data(name = vars["fan"], environment_name = rp)
elec_pump <- param$report_data(name = vars["pump"], environment_name = rp)
elec_chiller <- param$report_data(name = vars["chiller"], environment_name = rp)
gas_boiler <- param$report_data(name = vars["boiler"], environment_name = rp)
# check if the system is running in cooling mode
is_clg <- coil_clg[, .(is_clg = any(value > 0)), .(index, datetime)]
# split fan energy for each model based on the operation mode
elec_fan[is_clg, on = .(index, datetime), is_clg := i.is_clg]
elec_fan_sum <- elec_fan[, .(electricity = sum(value)), .(index, is_clg)]
# split pump energy
elec_pump_sum <- elec_pump[
J(c("CW CIRC PUMP", "HW CIRC PUMP")), on = .(key_value),
.(electricity = sum(value)), .(index, is_clg = key_value == "CW CIRC PUMP")
]
# calculate chiller and boiler energy per model
elec_chiller_sum <- elec_chiller[, .(is_clg = TRUE, electricity = sum(value)), .(index)]
gas_boiler_sum <- gas_boiler[, .(is_clg = FALSE, natural_gas = sum(value)), .(index)]
# calculate total electricityand natural gas energy
energy <- data.table::copy(elec_fan_sum)[
elec_pump_sum, on = .(index, is_clg), electricity := electricity + i.electricity][
elec_chiller_sum, on = .(index, is_clg), electricity := electricity + i.electricity][
gas_boiler_sum, on = .(index, is_clg), natural_gas := i.natural_gas
]
# unit conversion: J -> kWh
data.table::set(energy, NULL, "electricity",
units::set_units(units::set_units(energy$electricity, "J / yr"), "kWh / yr")
)
# unit conversion: J -> GJ
data.table::set(energy, NULL, "natural_gas",
units::set_units(units::set_units(energy$natural_gas, "J / yr"), "GJ / yr")
)
# add a new column `mode`
energy[is_clg == TRUE, mode := "cooling"]
energy[is_clg == FALSE, mode := "heating"]
# transform the data table into "wide" format
data.table::dcast(energy, index ~ mode, value.var = c("electricity", "natural_gas"))
}
energy <- calc_energy(param)
print(energy)
# calculate Morris Sensitivity for Comfort Metrics
Source_Comfort = param$tabular_data(table_name = "Comfort and Setpoint Not Met Summary",
column_name = "Facility",
row_name = "Time Setpoint Not Met During Occupied Cooling")
Source_Comfort<- as.numeric(Source_Comfort$value)
mo<-sensitivity::tell(mo, Source_Comfort)
sen_indices <- data.frame(
mu = apply(mo$ee, 2, mean),
mu.star = apply(mo$ee, 2, function(x) mean(abs(x))),
sigma = apply(mo$ee, 2, sd)
)``` |
Beta Was this translation helpful? Give feedback.
-
|
In addition to that, I as well did a quick check with "your code snippet" below, to see if the results on the summary tables would return NA or not for the I hope these information are sufficient. library(eplusr)
path <- path_eplus_example(max(avail_eplus()), "5Zone_Transformer.idf")
idf <- read_idf(path)
#> IDD v23.1.0 has not been parsed before.
#> Try to locate 'Energy+.idd' in EnergyPlus v23.1.0 installation folder 'C:\EnergyPlusV23-1-0'.
#> IDD file found: 'C:\Users\hongy\AppData\Local\EnergyPlusV23-1-0\Energy+.idd'.
#> Start parsing...
#> Parsing completed.
# add output variables
vars <- c(
# cooling/heating indicator for each coil
coil_clg = "Cooling Coil Total Cooling Energy",
# all kinds of energy
fan = "Fan Electricity Energy",
pump = "Pump Electricity Energy",
chiller = "Chiller Electricity Energy",
boiler = "Boiler NaturalGas Energy"
)
idf$add(`Output:Variable` := list("*", vars, "Timestep"))
#> $<NA>
#> <IdfObject: 'Output:Variable'> [ID:323]
#> Class: <Output:Variable>
#> ├─ 1 : "*", !- Key Value
#> │─ 2*: "Cooling Coil Total Cooling Energy", !- Variable Name
#> └─ 3 : "Timestep"; !- Reporting Frequency
#>
#> $<NA>
#> <IdfObject: 'Output:Variable'> [ID:324]
#> Class: <Output:Variable>
#> ├─ 1 : "*", !- Key Value
#> │─ 2*: "Fan Electricity Energy", !- Variable Name
#> └─ 3 : "Timestep"; !- Reporting Frequency
#>
#> $<NA>
#> <IdfObject: 'Output:Variable'> [ID:325]
#> Class: <Output:Variable>
#> ├─ 1 : "*", !- Key Value
#> │─ 2*: "Pump Electricity Energy", !- Variable Name
#> └─ 3 : "Timestep"; !- Reporting Frequency
#>
#> $<NA>
#> <IdfObject: 'Output:Variable'> [ID:326]
#> Class: <Output:Variable>
#> ├─ 1 : "*", !- Key Value
#> │─ 2*: "Chiller Electricity Energy", !- Variable Name
#> └─ 3 : "Timestep"; !- Reporting Frequency
#>
#> $<NA>
#> <IdfObject: 'Output:Variable'> [ID:327]
#> Class: <Output:Variable>
#> ├─ 1 : "*", !- Key Value
#> │─ 2*: "Boiler NaturalGas Energy", !- Variable Name
#> └─ 3 : "Timestep"; !- Reporting Frequency
# make sure annual simulation is triggered by removing all existing run periods
# and adding a new one expanding the whole year
idf$RunPeriod <- NULL
idf$add(RunPeriod = list("AnnualSim", 1, 1, NULL, 12, 31))
#> $AnnualSim
#> <IdfObject: 'RunPeriod'> [ID:328] `AnnualSim`
#> Class: <RunPeriod>
#> ├─ 1*: "AnnualSim", !- Name
#> │─ 2*: 1, !- Begin Month
#> │─ 3*: 1, !- Begin Day of Month
#> │─ 4 : <Blank>, !- Begin Year
#> │─ 5*: 12, !- End Month
#> │─ 6*: 31, !- End Day of Month
#> └─ 7 : <Blank>; !- End Year
# save the model
idf$save(tempfile(fileext = ".idf"))
# DO YOUR PARAMETRIC SIMULATIONS AS USUAL
# here I use a measure changing building orientation as an example
param <- param_job(
idf,
path_eplus_weather(max(avail_eplus()), "USA_CA_San.Francisco.Intl.AP.724940_TMY3.epw")
)
#> Adding an object in class 'Output:SQLite' and setting its 'Option Type' to 'SimpleAndTabular' in order to create SQLite output file.
param$apply_measure(
function(idf, ori) {
idf$set(Building := .(north_axis = ori))
idf
},
seq(30, 180, 30)
)
#> Measure function has been applied with 6 new models created:
#> [1]: case_1
#> [2]: case_2
#> [3]: case_3
#> [4]: case_4
#> [5]: case_5
#> [6]: case_6
param$run(echo = FALSE)
#> ── EnergPlus Parametric Simulation Job ─────────────────────────────────────────
#> • Path: 'C:\Users\hongy\AppData\Local\Temp\RtmpuMI4yG\file3d503eaa4a4b.idf'
#> • Version: 'C:\Users\hongy\AppData\Local\EnergyPlusV23-1-0\WeatherData\USA_CA_…
#> • EnergyPlus Version: '23.1.0'
#> • EnergyPlus Path: 'C:\Users\hongy\AppData\Local\EnergyPlusV23-1-0'
#> Parametric Models [6]:
#> [1]: 'case_1.idf' <-- SUCCEEDED
#> [2]: 'case_2.idf' <-- SUCCEEDED
#> [3]: 'case_3.idf' <-- SUCCEEDED
#> [4]: 'case_4.idf' <-- SUCCEEDED
#> [5]: 'case_5.idf' <-- SUCCEEDED
#> [6]: 'case_6.idf' <-- SUCCEEDED
#> Simulation started at '2023-06-08 14:34:13.93588' and completed successfully after 8.81 secs.
calc_energy <- function(param) {
# only extract data for annual simulation
rp <- "AnnualSim"
coil_clg <- param$report_data(name = vars["coil_clg"], environment_name = rp)
elec_fan <- param$report_data(name = vars["fan"], environment_name = rp)
elec_pump <- param$report_data(name = vars["pump"], environment_name = rp)
elec_chiller <- param$report_data(name = vars["chiller"], environment_name = rp)
gas_boiler <- param$report_data(name = vars["boiler"], environment_name = rp)
# check if the system is running in cooling mode
is_clg <- coil_clg[, .(is_clg = any(value > 0)), .(index, datetime)]
# split fan energy for each model based on the operation mode
elec_fan[is_clg, on = .(index, datetime), is_clg := i.is_clg]
elec_fan_sum <- elec_fan[, .(electricity = sum(value)), .(index, is_clg)]
# split pump energy
elec_pump_sum <- elec_pump[
J(c("CW CIRC PUMP", "HW CIRC PUMP")), on = .(key_value),
.(electricity = sum(value)), .(index, is_clg = key_value == "CW CIRC PUMP")
]
# calculate chiller and boiler energy per model
elec_chiller_sum <- elec_chiller[, .(is_clg = TRUE, electricity = sum(value)), .(index)]
gas_boiler_sum <- gas_boiler[, .(is_clg = FALSE, natural_gas = sum(value)), .(index)]
# calculate total electricityand natural gas energy
energy <- data.table::copy(elec_fan_sum)[
elec_pump_sum, on = .(index, is_clg), electricity := electricity + i.electricity][
elec_chiller_sum, on = .(index, is_clg), electricity := electricity + i.electricity][
gas_boiler_sum, on = .(index, is_clg), natural_gas := i.natural_gas
]
# unit conversion: J -> kWh
data.table::set(energy, NULL, "electricity",
units::set_units(units::set_units(energy$electricity, "J / yr"), "kWh / yr")
)
# unit conversion: J -> GJ
data.table::set(energy, NULL, "natural_gas",
units::set_units(units::set_units(energy$natural_gas, "J / yr"), "GJ / yr")
)
# add a new column `mode`
energy[is_clg == TRUE, mode := "cooling"]
energy[is_clg == FALSE, mode := "heating"]
# transform the data table into "wide" format
data.table::dcast(energy, index ~ mode, value.var = c("electricity", "natural_gas"))
}
energy <- calc_energy(param)
param$cases()[energy, on = "index"]
#> index case ori electricity_cooling electricity_heating natural_gas_cooling
#> 1: 1 case_1 30 5742.600 [kWh/yr] 1143.754 [kWh/yr] NA [GJ/yr]
#> 2: 2 case_2 60 6148.594 [kWh/yr] 1238.821 [kWh/yr] NA [GJ/yr]
#> 3: 3 case_3 90 6507.838 [kWh/yr] 1288.575 [kWh/yr] NA [GJ/yr]
#> 4: 4 case_4 120 6659.757 [kWh/yr] 1285.288 [kWh/yr] NA [GJ/yr]
#> 5: 5 case_5 150 6494.502 [kWh/yr] 1216.239 [kWh/yr] NA [GJ/yr]
#> 6: 6 case_6 180 6440.916 [kWh/yr] 1196.517 [kWh/yr] NA [GJ/yr]
#> natural_gas_heating
#> 1: 28.84122 [GJ/yr]
#> 2: 28.59514 [GJ/yr]
#> 3: 28.93842 [GJ/yr]
#> 4: 28.83044 [GJ/yr]
#> 5: 28.76399 [GJ/yr]
#> 6: 29.30865 [GJ/yr]
Source_Comfort = param$tabular_data(table_name = "Comfort and Setpoint Not Met Summary",
column_name = "Facility",
row_name = "Time Setpoint Not Met During Occupied Cooling")
Source_Comfort<- as.numeric(Source_Comfort$value)
mo<-sensitivity::tell(mo, Source_Comfort)
sen_indices <- data.frame(
mu = apply(mo$ee, 2, mean),
mu.star = apply(mo$ee, 2, function(x) mean(abs(x))),
sigma = apply(mo$ee, 2, sd)
print (sen_indices)
)``` |
Beta Was this translation helpful? Give feedback.
-
|
Hy Hongyuan, There was a remarkable progress! I modified the > library(eplusr) > library(sensitivity)
> idf=eplusr::read_idf("C:/Users/DELL USER/Documents/pasta/IDF_file.idf") IDD v9.4.0 has not been parsed before. Try to locate 'Energy+.idd' in EnergyPlus v9.4.0 installation folder 'C:\EnergyPlusV9-4-0'. IDD file found: 'C:\EnergyPlusV9-4-0\Energy+.idd'.
Start parsing...
Parsing completed.
> epw = eplusr::read_epw("C:/Users/DELL USER/Documents/pasta/NGA_KN_Kano.Intl.AP.650460_TMYx.2007-2021.
#Annual simulation
idf$RunPeriod <- NULL
`idf$add(RunPeriod = list("AnnualSim", 1, 1, NULL, 12, 31))`
#save the model
`idf$save(tempfile(fileext = ".idf"))`
#create a parametric job
`param <- param_job(idf,epw)`
#Define a measure to modify IDF
`measure <- function(idf,
Wall_therm_trans,Wall_therm_cap, Wall_sol_abs,Roof_therm_trans,
Roof_therm_cap,Roof_sol_abs, Floor_therm_trans,Floor_therm_cap,
Win_therm_trans,Cooling_Setpoint, SHGC_win){
idf$set(`
`8IN Concrete HW` = list( conductivity = Wall_therm_trans,
specific_heat = Wall_therm_cap,
solar_absorptance = Wall_sol_abs
),`
`Roof Membrane` = list(
conductivity = Roof_therm_trans,
specific_heat = Roof_therm_cap,
solar_absorptance = Roof_sol_abs
),`
`MAT-CC05 4 HW CONCRETE` = list(
conductivity = Floor_therm_trans,
specific_heat = Floor_therm_cap
),`
` `Uglass` = list(
U_Factor = Win_therm_trans,
Solar_Heat_Gain_Coefficient = SHGC_win),
`Thermostat` = list(
Constant_Cooling_Setpoint = Cooling_Setpoint)
)
idf
}`
`factors.path=c(paste(pasta,"/Input_file - Copy.csv",sep=""))`
`factors.classes=c("character","character","numeric","numeric")`
`variables=read.table(file=factors.path,
header=TRUE,sep=",",colClasses=factors.classes)`
`print(variables)`
`variables.names=variables[[1]]`
`variables.lower=variables[[3]]`
`variables.upper=variables[[4]]`
`mo <- sensitivity::morris(model=NULL,factors=variables.names,r=12,
design=list(type="oat",levels=8,grid.jump=4),
binf=variables.lower,bsup=variables.upper,scale=TRUE)`
`param$apply_measure(
measure,
Wall_therm_trans = mo$X[, "Wall_therm_trans"],
Wall_therm_cap = mo$X[, "Wall_therm_cap"],
Wall_sol_abs = mo$X[, "Wall_sol_abs"],
Roof_therm_trans = mo$X[, "Roof_therm_trans"],
Roof_therm_cap = mo$X[, "Roof_therm_cap"],
Roof_sol_abs = mo$X[, "Roof_sol_abs"],
Floor_therm_trans = mo$X[, "Floor_therm_trans"],
Floor_therm_cap = mo$X[, "Floor_therm_cap"],
Win_therm_trans = mo$X[, "Win_therm_trans"],
SHGC_win = mo$X[, "SHGC_win"],
Cooling_Setpoint = mo$X[, "Cooling_Setpoint"]
)`
#Run parametric simulations
param$run(echo = FALSE)
Measure 'measure' has been applied with 144 new models created:
[001]: measure_1
[002]: measure_2
[003]: measure_3
[004]: measure_4
[005]: measure_5
[006]: measure_6
[007]: measure_7
[008]: measure_8
[009]: measure_9
[010]: measure_10
[011]: measure_11
[012]: measure_12
[013]: measure_13
[014]: measure_14
[015]: measure_15
[016]: measure_16
[017]: measure_17
[018]: measure_18
[019]: measure_19
[020]: measure_20
[021]: measure_21
[022]: measure_22
[023]: measure_23
[024]: measure_24
[025]: measure_25
[026]: measure_26
[027]: measure_27
[028]: measure_28
[029]: measure_29
[030]: measure_30
[031]: measure_31
[032]: measure_32
[033]: measure_33
[034]: measure_34
[035]: measure_35
[036]: measure_36
[037]: measure_37
[038]: measure_38
[039]: measure_39
[040]: measure_40
[041]: measure_41
[042]: measure_42
[043]: measure_43
[044]: measure_44
[045]: measure_45
[046]: measure_46
[047]: measure_47
[048]: measure_48
[049]: measure_49
[050]: measure_50
[051]: measure_51
[052]: measure_52
[053]: measure_53
[054]: measure_54
[055]: measure_55
[056]: measure_56
[057]: measure_57
[058]: measure_58
[059]: measure_59
[060]: measure_60
[061]: measure_61
[062]: measure_62
[063]: measure_63
[064]: measure_64
[065]: measure_65
[066]: measure_66
[067]: measure_67
[068]: measure_68
[069]: measure_69
[070]: measure_70
[071]: measure_71
[072]: measure_72
[073]: measure_73
[074]: measure_74
[075]: measure_75
[076]: measure_76
[077]: measure_77
[078]: measure_78
[079]: measure_79
[080]: measure_80
[081]: measure_81
[082]: measure_82
[083]: measure_83
[084]: measure_84
[085]: measure_85
[086]: measure_86
[087]: measure_87
[088]: measure_88
[089]: measure_89
[090]: measure_90
[091]: measure_91
[092]: measure_92
[093]: measure_93
[094]: measure_94
[095]: measure_95
[096]: measure_96
[097]: measure_97
[098]: measure_98
[099]: measure_99
[100]: measure_100
[101]: measure_101
[102]: measure_102
[103]: measure_103
[104]: measure_104
[105]: measure_105
[106]: measure_106
[107]: measure_107
[108]: measure_108
[109]: measure_109
[110]: measure_110
[111]: measure_111
[112]: measure_112
[113]: measure_113
[114]: measure_114
[115]: measure_115
[116]: measure_116
[117]: measure_117
[118]: measure_118
[119]: measure_119
[120]: measure_120
[121]: measure_121
[122]: measure_122
[123]: measure_123
[124]: measure_124
[125]: measure_125
[126]: measure_126
[127]: measure_127
[128]: measure_128
[129]: measure_129
[130]: measure_130
[131]: measure_131
[132]: measure_132
[133]: measure_133
[134]: measure_134
[135]: measure_135
[136]: measure_136
[137]: measure_137
[138]: measure_138
[139]: measure_139
[140]: measure_140
[141]: measure_141
[142]: measure_142
[143]: measure_143
[144]: measure_144
Measure 'measure' has been applied with 144 new models created:
> param$run(dir = "D:/File", echo = FALSE)
[144/144 | 100%]
======================================================================================================================= [Elapsed: 01:45:17] ── EnergPlus Parametric Simulation Job
───────────────────────────────────────────────────────────────────────────────────────────────────────
* Seed: 'C:\Users\DELL USER\AppData\Local\Temp\RtmpKcZYqZ\file29bc2ca172e7.idf'
* Weather: 'C:\Users\DELL USER\Documents\pasta\NGA_KN_Kano.Intl.AP.650460_TMYx.2007-2021.epw'
* EnergyPlus Version: '9.4.0'
* EnergyPlus Path: 'C:\EnergyPlusV9-4-0'
Applied Measure: 'measure'
Parametric Models [144]:
[001]: 'measure_1.idf' <-- SUCCEEDED
[002]: 'measure_2.idf' <-- SUCCEEDED
[003]: 'measure_3.idf' <-- SUCCEEDED
[004]: 'measure_4.idf' <-- SUCCEEDED
[005]: 'measure_5.idf' <-- SUCCEEDED
[006]: 'measure_6.idf' <-- SUCCEEDED
[007]: 'measure_7.idf' <-- SUCCEEDED
[008]: 'measure_8.idf' <-- SUCCEEDED
[009]: 'measure_9.idf' <-- SUCCEEDED
[010]: 'measure_10.idf' <-- SUCCEEDED
[011]: 'measure_11.idf' <-- SUCCEEDED
[012]: 'measure_12.idf' <-- SUCCEEDED
[013]: 'measure_13.idf' <-- SUCCEEDED
[014]: 'measure_14.idf' <-- SUCCEEDED
[015]: 'measure_15.idf' <-- SUCCEEDED
[016]: 'measure_16.idf' <-- SUCCEEDED
[017]: 'measure_17.idf' <-- SUCCEEDED
[018]: 'measure_18.idf' <-- SUCCEEDED
[019]: 'measure_19.idf' <-- SUCCEEDED
[020]: 'measure_20.idf' <-- SUCCEEDED
[021]: 'measure_21.idf' <-- SUCCEEDED
[022]: 'measure_22.idf' <-- SUCCEEDED
[023]: 'measure_23.idf' <-- SUCCEEDED
[024]: 'measure_24.idf' <-- SUCCEEDED
[025]: 'measure_25.idf' <-- SUCCEEDED
[026]: 'measure_26.idf' <-- SUCCEEDED
[027]: 'measure_27.idf' <-- SUCCEEDED
[028]: 'measure_28.idf' <-- SUCCEEDED
[029]: 'measure_29.idf' <-- SUCCEEDED
[030]: 'measure_30.idf' <-- SUCCEEDED
[031]: 'measure_31.idf' <-- SUCCEEDED
[032]: 'measure_32.idf' <-- SUCCEEDED
[033]: 'measure_33.idf' <-- SUCCEEDED
[034]: 'measure_34.idf' <-- SUCCEEDED
[035]: 'measure_35.idf' <-- SUCCEEDED
[036]: 'measure_36.idf' <-- SUCCEEDED
[037]: 'measure_37.idf' <-- SUCCEEDED
[038]: 'measure_38.idf' <-- SUCCEEDED
[039]: 'measure_39.idf' <-- SUCCEEDED
[040]: 'measure_40.idf' <-- SUCCEEDED
[041]: 'measure_41.idf' <-- SUCCEEDED
[042]: 'measure_42.idf' <-- SUCCEEDED
[043]: 'measure_43.idf' <-- SUCCEEDED
[044]: 'measure_44.idf' <-- SUCCEEDED
[045]: 'measure_45.idf' <-- SUCCEEDED
[046]: 'measure_46.idf' <-- SUCCEEDED
[047]: 'measure_47.idf' <-- SUCCEEDED
[048]: 'measure_48.idf' <-- SUCCEEDED
[049]: 'measure_49.idf' <-- SUCCEEDED
[050]: 'measure_50.idf' <-- SUCCEEDED
[051]: 'measure_51.idf' <-- SUCCEEDED
[052]: 'measure_52.idf' <-- SUCCEEDED
[053]: 'measure_53.idf' <-- SUCCEEDED
[054]: 'measure_54.idf' <-- SUCCEEDED
[055]: 'measure_55.idf' <-- SUCCEEDED
[056]: 'measure_56.idf' <-- SUCCEEDED
[057]: 'measure_57.idf' <-- SUCCEEDED
[058]: 'measure_58.idf' <-- SUCCEEDED
[059]: 'measure_59.idf' <-- SUCCEEDED
[060]: 'measure_60.idf' <-- SUCCEEDED
[061]: 'measure_61.idf' <-- SUCCEEDED
[062]: 'measure_62.idf' <-- SUCCEEDED
[063]: 'measure_63.idf' <-- SUCCEEDED
[064]: 'measure_64.idf' <-- SUCCEEDED
[065]: 'measure_65.idf' <-- SUCCEEDED
[066]: 'measure_66.idf' <-- SUCCEEDED
[067]: 'measure_67.idf' <-- SUCCEEDED
[068]: 'measure_68.idf' <-- SUCCEEDED
[069]: 'measure_69.idf' <-- SUCCEEDED
[070]: 'measure_70.idf' <-- SUCCEEDED
[071]: 'measure_71.idf' <-- SUCCEEDED
[072]: 'measure_72.idf' <-- SUCCEEDED
[073]: 'measure_73.idf' <-- SUCCEEDED
[074]: 'measure_74.idf' <-- SUCCEEDED
[075]: 'measure_75.idf' <-- SUCCEEDED
[076]: 'measure_76.idf' <-- SUCCEEDED
[077]: 'measure_77.idf' <-- SUCCEEDED
[078]: 'measure_78.idf' <-- SUCCEEDED
[079]: 'measure_79.idf' <-- SUCCEEDED
[080]: 'measure_80.idf' <-- SUCCEEDED
[081]: 'measure_81.idf' <-- SUCCEEDED
[082]: 'measure_82.idf' <-- SUCCEEDED
[083]: 'measure_83.idf' <-- SUCCEEDED
[084]: 'measure_84.idf' <-- SUCCEEDED
[085]: 'measure_85.idf' <-- SUCCEEDED
[086]: 'measure_86.idf' <-- SUCCEEDED
[087]: 'measure_87.idf' <-- SUCCEEDED
[088]: 'measure_88.idf' <-- SUCCEEDED
[089]: 'measure_89.idf' <-- SUCCEEDED
[090]: 'measure_90.idf' <-- SUCCEEDED
[091]: 'measure_91.idf' <-- SUCCEEDED
[092]: 'measure_92.idf' <-- SUCCEEDED
[093]: 'measure_93.idf' <-- SUCCEEDED
[094]: 'measure_94.idf' <-- SUCCEEDED
[095]: 'measure_95.idf' <-- SUCCEEDED
[096]: 'measure_96.idf' <-- SUCCEEDED
[097]: 'measure_97.idf' <-- SUCCEEDED
[098]: 'measure_98.idf' <-- SUCCEEDED
[099]: 'measure_99.idf' <-- SUCCEEDED
[100]: 'measure_100.idf' <-- SUCCEEDED
[101]: 'measure_101.idf' <-- SUCCEEDED
[102]: 'measure_102.idf' <-- SUCCEEDED
[103]: 'measure_103.idf' <-- SUCCEEDED
[104]: 'measure_104.idf' <-- SUCCEEDED
[105]: 'measure_105.idf' <-- SUCCEEDED
[106]: 'measure_106.idf' <-- SUCCEEDED
[107]: 'measure_107.idf' <-- SUCCEEDED
[108]: 'measure_108.idf' <-- SUCCEEDED
[109]: 'measure_109.idf' <-- SUCCEEDED
[110]: 'measure_110.idf' <-- SUCCEEDED
[111]: 'measure_111.idf' <-- SUCCEEDED
[112]: 'measure_112.idf' <-- SUCCEEDED
[113]: 'measure_113.idf' <-- SUCCEEDED
[114]: 'measure_114.idf' <-- SUCCEEDED
[115]: 'measure_115.idf' <-- SUCCEEDED
[116]: 'measure_116.idf' <-- SUCCEEDED
[117]: 'measure_117.idf' <-- SUCCEEDED
[118]: 'measure_118.idf' <-- SUCCEEDED
[119]: 'measure_119.idf' <-- SUCCEEDED
[120]: 'measure_120.idf' <-- SUCCEEDED
[121]: 'measure_121.idf' <-- SUCCEEDED
[122]: 'measure_122.idf' <-- SUCCEEDED
[123]: 'measure_123.idf' <-- SUCCEEDED
[124]: 'measure_124.idf' <-- SUCCEEDED
[125]: 'measure_125.idf' <-- SUCCEEDED
[126]: 'measure_126.idf' <-- SUCCEEDED
[127]: 'measure_127.idf' <-- SUCCEEDED
[128]: 'measure_128.idf' <-- SUCCEEDED
[129]: 'measure_129.idf' <-- SUCCEEDED
[130]: 'measure_130.idf' <-- SUCCEEDED
[131]: 'measure_131.idf' <-- SUCCEEDED
[132]: 'measure_132.idf' <-- SUCCEEDED
[133]: 'measure_133.idf' <-- SUCCEEDED
[134]: 'measure_134.idf' <-- SUCCEEDED
[135]: 'measure_135.idf' <-- SUCCEEDED
[136]: 'measure_136.idf' <-- SUCCEEDED
[137]: 'measure_137.idf' <-- SUCCEEDED
[138]: 'measure_138.idf' <-- SUCCEEDED
[139]: 'measure_139.idf' <-- SUCCEEDED
[140]: 'measure_140.idf' <-- SUCCEEDED
[141]: 'measure_141.idf' <-- SUCCEEDED
[142]: 'measure_142.idf' <-- SUCCEEDED
[143]: 'measure_143.idf' <-- SUCCEEDED
[144]: 'measure_144.idf' <-- SUCCEEDED
Simulation started at '2023-06-21 17:48:57.035339' and completed successfully after 1.76 hours.
> Source_Comfort = param$tabular_data(table_name = "Comfort and Setpoint Not Met Summary", column_name = "Facility", row_name = "Time Setpoint Not Met During Occupied Cooling")
> Source_Comfort<- as.numeric(Source_Comfort$value)
> sen_ind <- data.frame( + mu = apply(morris$ee, 2, mean),
mu.star = apply(morris$ee, 2, function(x) mean(abs(x))),
sigma = apply(morris$ee, 2, sd)
)
> morris<-sensitivity::tell(mo,Source_Comfort)
> sen_ind <- data.frame(
mu = apply(morris$ee, 2, mean),
mu.star = apply(morris$ee, 2, function(x) mean(abs(x))),
sigma = apply(morris$ee, 2, sd)
)
> print(sen_ind)
mu mu.star sigma
Wall_therm_trans -4.958333 7.510417 2.711934
Wall_therm_cap -31.682292 31.682292 31.890587
Wall_sol_abs 0.000000 0.000000 0.000000
Roof_therm_trans -2.114583 3.645833 5.830515
Roof_therm_cap -3.682292 4.994792 6.180032
Roof_sol_abs -9.515625 9.515625 7.076947
Floor_therm_trans 45.390625 45.390625 37.909789
Floor_therm_cap -7.984375 7.984375 9.453765
Win_therm_trans 0.000000 0.000000 0.000000
SHGC_win 0.000000 0.000000 0.000000
Cooling_Setpoint 63.984375 63.984375 43.914796 |
Beta Was this translation helpful? Give feedback.
-
|
Dear Hongyuan. Happy to share the this with you. More simulations for more library(ggplot2)
library(gridExtra)
library(cowplot)
# Define the colors for each variable
variable_colors <- c("Wall_therm_trans" = "red",
"Wall_therm_cap" = "blue",
"Wall_sol_abs" = "green",
"Roof_therm_trans" = "purple",
"Roof_therm_cap" = "orange",
"Roof_sol_abs" = "pink",
"Floor_therm_trans" = "brown",
"Floor_therm_cap" = "cyan",
"Win_therm_trans" = "gray",
"SHGC_win" = "yellow",
"Cooling_Setpoint" = "black")
# Create separate plots for each combination
plot1 <- ggplot(Kano_current) +
geom_point(aes(x = setpoint_MU., y = setpoint_SIGMA, color = Variables), size = 3) +
geom_abline(aes(intercept = 0, slope = 1, linetype = "Line 1"), color = "black", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.5, linetype = "Line 2"), color = "red", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.1, linetype = "Line 3"), color = "green", linewidth = 0.85) +
labs(x = expression(paste(mu^"*", " [Hrs]")),
y = expression(paste(sigma, " [Hrs]")),
title = "Cooling Time Setpoint Not Met (Kano TMY)") +
scale_color_manual(values = variable_colors, guide = "none") +
scale_linetype_manual(values = c("solid", "dashed", "dotted"),
breaks = NULL,
labels = NULL,
guide = "none") +
theme_bw() +
theme(plot.title = element_text(size = 16),
axis.text = element_text(size = 12),
axis.title = element_text(size = 12))
plot2 <- ggplot(Kano_current) +
geom_point(aes(x = comfort_MU., y = comfort_SIGMA, color = Variables), size = 3) +
geom_abline(aes(intercept = 0, slope = 1, linetype = "Line 1"), color = "black", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.5, linetype = "Line 2"), color = "red", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.1, linetype = "Line 3"), color = "green", linewidth = 0.85) +
labs(x = expression(paste(mu^"*", " [Hrs]")),
y = expression(paste(sigma, " (Hrs)")),
title = "Uncomfort Time (Kano TMY)") +
scale_color_manual(values = variable_colors, guide = "none") +
scale_linetype_manual(values = c("solid", "dashed", "dotted"),
breaks = NULL,
labels = NULL,
guide = "none") +
theme_bw() +
theme(plot.title = element_text(size = 16),
axis.text = element_text(size = 12),
axis.title = element_text(size = 12))
plot3 <- ggplot(Kano_current) +
geom_point(aes(x = cooling_MU., y = cooling_SIGMA, color = Variables), size = 3) +
geom_abline(aes(intercept = 0, slope = 1, linetype = "Line 1"), color = "black", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.5, linetype = "Line 2"), color = "red", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.1, linetype = "Line 3"), color = "green", linewidth = 0.85) +
labs(x = expression(paste(mu^"*", " [KWh/yr]")),
y = expression(paste(sigma, " [KWh/yr]")),
title = "Cooling Electricity (Kano TMY)") +
scale_color_manual(values = variable_colors, guide = "none") +
scale_linetype_manual(values = c("solid", "dashed", "dotted"),
breaks = NULL,
labels = NULL,
guide = "none") +
theme_bw()+
theme(plot.title = element_text(size = 16),
axis.text = element_text(size = 12),
axis.title = element_text(size = 12))
plot4 <- ggplot(Kano_current) +
geom_point(aes(x = heating_MU., y = heating_SIGMA, color = Variables), size = 3) +
geom_abline(aes(intercept = 0, slope = 1, linetype = "Line 1"), color = "black", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.5, linetype = "Line 2"), color = "red", linewidth = 0.85) +
geom_abline(aes(intercept = 0, slope = 0.1, linetype = "Line 3"), color = "green", linewidth = 0.85) +
labs(x = expression(paste(mu^"*", " [KWh/yr]")),
y = expression(paste(sigma, " [KWh/yr]")),
title = "Heating Electricity (Kano)") +
scale_color_manual(values = variable_colors) +
scale_linetype_manual(values = c("solid", "dashed", "dotted"),
breaks = c("Line 1", "Line 2", "Line 3"),
labels = c("non-linear/non-monotonic", "monotonic", "linear"))+
theme_bw() +
theme(plot.title = element_text(size = 16),
axis.text = element_text(size = 12),
axis.title = element_text(size = 12),
legend.text = element_text(10))
# Combine the plots using cowplot
final_plot <- cowplot::plot_grid(plot1, plot2, plot3, plot4, ncol = 2) +
theme(legend.title = element_blank(),
legend.position = "none")
# Print the final plot
final_plot
|
Beta Was this translation helpful? Give feedback.
Here I use the
morris()function from {sensitivity} package for demonstration. There are 5 steps to perform sensitivity analyses using {eplusr}ParametricJobobject usingeplusr::param_job()by specifying a seed model and a weather fileBelow is a code snippet showcasing the whole workflow: