在R中将小时级电力负荷数据插值为半小时级数据

Here's a clean, reproducible approach using dplyr and tidyr to transform your daily hourly load data into the required half-hourly format:

Step 1: Load Required Libraries #

First, make sure you have these packages installed (if not, run install.packages(c("dplyr", "tidyr"))):

library(dplyr)
library(tidyr)

Step 2: Prepare the Data with Previous Day's HR24 #

We need the previous day's HR24 value to calculate HR0030 (the 00:30 AM reading). We'll use dplyr::lag() to fetch this:

load_demand <- load_demand %>%
  mutate(prev_HR24 = lag(HR24))

Step 3: Generate All Half-Hourly Columns #

Now we'll create each required half-hourly column by either mapping directly to existing hourly values or calculating the linear interpolation (average, since intervals are equal) between adjacent hours:

halfhourly_load <- load_demand %>%
  # Calculate cross-day 00:30 reading
  mutate(HR0030 = (prev_HR24 + HR1)/2) %>%
  # Map existing hourly columns to their 00-ending half-hour names
  mutate(HR0100 = HR1,
         HR0200 = HR2,
         HR0300 = HR3,
         HR0400 = HR4,
         HR0500 = HR5,
         HR0600 = HR6,
         HR0700 = HR7,
         HR0800 = HR8,
         HR0900 = HR9,
         HR1000 = HR10,
         HR1100 = HR11,
         HR1200 = HR12,
         HR1300 = HR13,
         HR1400 = HR14,
         HR1500 = HR15,
         HR1600 = HR16,
         HR1700 = HR17,
         HR1800 = HR18,
         HR1900 = HR19,
         HR2000 = HR20,
         HR2100 = HR21,
         HR2200 = HR22,
         HR2300 = HR23,
         HR2400 = HR24) %>%
  # Calculate intra-day 30-minute interpolations
  mutate(HR0130 = (HR1 + HR2)/2,
         HR0230 = (HR2 + HR3)/2,
         HR0330 = (HR3 + HR4)/2,
         HR0430 = (HR4 + HR5)/2,
         HR0530 = (HR5 + HR6)/2,
         HR0630 = (HR6 + HR7)/2,
         HR0730 = (HR7 + HR8)/2,
         HR0830 = (HR8 + HR9)/2,
         HR0930 = (HR9 + HR10)/2,
         HR1030 = (HR10 + HR11)/2,
         HR1130 = (HR11 + HR12)/2,
         HR1230 = (HR12 + HR13)/2,
         HR1330 = (HR13 + HR14)/2,
         HR1430 = (HR14 + HR15)/2,
         HR1530 = (HR15 + HR16)/2,
         HR1630 = (HR16 + HR17)/2,
         HR1730 = (HR17 + HR18)/2,
         HR1830 = (HR18 + HR19)/2,
         HR1930 = (HR19 + HR20)/2,
         HR2030 = (HR20 + HR21)/2,
         HR2130 = (HR21 + HR22)/2,
         HR2230 = (HR22 + HR23)/2,
         HR2330 = (HR23 + HR24)/2)

Step 4: Reorder Columns to Match the Required Sequence #

Finally, we'll keep only the Date column and the 48 half-hourly columns in the exact order you specified:

# Define the required column order
halfhour_cols <- c("HR0030", "HR0100", "HR0130", "HR0200", "HR0230", "HR0300", 
                   "HR0330", "HR0400", "HR0430", "HR0500", "HR0530", "HR0600", 
                   "HR0630", "HR0700", "HR0730", "HR0800", "HR0830", "HR0900", 
                   "HR0930", "HR1000", "HR1030", "HR1100", "HR1130", "HR1200", 
                   "HR1230", "HR1300", "HR1330", "HR1400", "HR1430", "HR1500", 
                   "HR1530", "HR1600", "HR1630", "HR1700", "HR1730", "HR1800", 
                   "HR1830", "HR1900", "HR1930", "HR2000", "HR2030", "HR2100", 
                   "HR2130", "HR2200", "HR2230", "HR2300", "HR2330", "HR2400")

# Select and reorder columns
halfhourly_load <- halfhourly_load %>%
  select(Date, all_of(halfhour_cols))

Notes on Edge Cases #

• The first row's HR0030 will be NA because there's no prior day's HR24 data. You can handle this by either:
  • Setting it to HR1 (if you assume the previous day's 24:00 is same as current day's 01:00)
  • Removing the first row if you don't need incomplete data
  • Manually inputting the missing value if you have it

Let me know if you need adjustments for specific edge cases!

内容的提问来源于stack exchange，提问作者Nick HK