Cold weather can hide an efficiency improvement. Calculate degree days from daily means, then compare heating energy on the same weather basis.
Days 1–5, left to right. Temperature range 6–22 °C. Dashed line: 18 °C balance point.
1,000 kWh measured · 1,800 kWh heating-normalized
Move the balance temperature by editing its box. Vertical distances below the dashed line contribute heating degree days; above it contribute cooling degree days.
Heating degree days
24°C·day
Balance temperature 18 °C; daily-mean method.
Check the values to see a result.
Energy normalization assumes heating varies linearly with HDD and the reference period has the same duration, occupancy and base load.
Daily HDD = max(0, balance temperature − daily mean)sum each day
The entered 5 daily means give 24 heating and 2 cooling degree days at 18 °C.
Tip: Choose a balance temperature appropriate to the building. A standard reporting base is not automatically the best regression balance point.
Heating degree days measure how far daily mean temperatures fall below a chosen balance temperature. Cooling degree days measure how far they rise above it. Each day is clipped at zero before the period is summed, so warm days do not cancel cold days.
To compare heating energy between weather periods, first separate the energy that does not follow outdoor temperature. The simple normalization here scales only the remainder. It is useful for a transparent first comparison, not a substitute for a calibrated building-energy model.
HDD = sum(max(0, base − daily mean))
Enter one mean temperature for every day in the measured period. Averaging the entire period first hides days on opposite sides of the balance temperature.
Fahrenheit degree days = Celsius degree days × 1.8
This conversion holds for the same observations and physically equivalent balance points. Degree days are temperature differences multiplied by duration, so do not add 32 when converting their totals.
Heating kWh = measured kWh − weather-independent kWh
Hot water, electrical appliances and standing consumption may not track heating weather. Scaling the whole bill by HDD would incorrectly scale these loads as well.
Normalized kWh = base kWh + heating kWh × reference HDD ÷ actual HDD
Keep duration, occupancy, energy boundary, balance point and degree-day method consistent. Changes in operating hours or building use can otherwise look like changes in heating efficiency.
Daily HDD values are 8, 6, 0, 0 and 10, totaling 24°C·day. There are 2°C·day of cooling. Scaling 800 kWh of heating from 24 to 48 HDD and retaining 200 kWh base load gives 1,800 kWh.
Heating degree days 24 °C·day Open in the calculator
The period has 0 HDD. An energy-per-HDD relationship cannot be established, so heating-normalized energy is withheld rather than divided by zero.
Heating degree days 0 °C·day Open in the calculator
Use the specified base for a reporting comparison, or a balance point supported by building data. Insulation, internal gains and operating schedules affect the temperature at which heating becomes necessary.
No. This tool requires daily means. A monthly average can conceal heating and cooling days and underestimate both totals.
Not alone. Occupancy, operating hours, meter boundaries and base-load estimates must also be comparable. A regression with sufficient observations can help test the weather relationship and uncertainty.