DJU (Unified Degree Days)

Heating degree days (HDD) measure, over a given period, the cumulative difference engineering consultancyween the outside temperature and a chosen base temperature. They are used to quantify the severity of a winter (heating HDD) or a summer (cooling HDD), to normalize energy consumption, and to compare years/sites. In practice, the higher the heating HDD, the colder the season and the greater the heating requirements; the higher the cooling HDD, the hotter the summer and the greater the air conditioning requirements.

DJU: operational definition and calculation

For a heating base temperature (often 18°C ​​in France), the following is calculated daily:

- DJU₍heating₎(day) = max(0, Tbase − Tmoy.ext), where Tmoy.ext is the average (Tmin+Tmax)/2.

We then accumulate this over the month or year. Quick example: Base temperature 18°C, Average temperature 8°C → 10 heating degree days (HDD) for the day.

- For cold, it's the opposite : DJU₍cold₎(day) = max(0, Tmoy.ext − Tbase_cold), with a base often 22–24 °C depending on the use.

Choosing the baseline depends on the setpoint and the building's thermal inertia (18°C ≠ 20°C). Heating degree days (HDD) are available for each weather station and are broken down monthly/annually. They allow for the creation of correlation curves engineering consultancyween kWh and HDD (regressions) to isolate the weather-related contribution to energy consumption, estimate a reference consumption under "normal weather conditions," or compare two sites/climates.

DJU in planning and operation: advantages, limitations and points to consider

Interests

• Normalizing consumption: comparing a harsh winter to a mild winter.
• Monitor performance: if kWh/DJU decreases, operation improves (adjustments, balancing, heating curve).
• Forecast: estimate heating/cooling kWh according to weather scenarios (budget).
• Multi-site benchmarking: bringing consumption down to a common indicator.

Boundaries

• Assumption of linearity of needs ↔ DJU (reality impacted by solar inputs, wind, occupancy).
• Base temperature sometimes unsuitable for the building (over/underestimation).
• It does not deal with fine summer comfort: the RE2020 rather uses DH (degree-hours) for overheating.

Points to consider

• Consistent basis: choose 18 °C (or 20 °C) for heating according to setpoint/inertia; 22–24 °C for cooling according to use.
• Single source: using the same station and the same method from one month to the next.
• Scope: clean the kWh of non-climate uses (process, office automation) before correlation.
• Analysis: follow kWh/DJU and slope/intercept of the regression (the interception reveals the non-weather basis).
• Clarification: if kWh/DJU drifts, check heating curve, balancing, insulation, BMS schedules.
• Summer: Cold heating degree days are useful for air conditioning, but supplemented by heating factor, humidity and solar gains.

Anecdote — “DJU and invoice handed over in Gardanne”

In Gardanne, an office building saw its energy bills rise despite a predicted "average" winter. A comparison of kWh versus 18-day heating degree days (HDD) revealed a steeper slope than the previous year and an increase in intercepted consumption (consumption excluding weather-related factors). An audit revealed an excessively steep heating curve, inadequate insulation in the ductwork, and "temporarily" extended operating hours. By reducing the slope, re-insulating the risers, and tightening the operating hours, the kWh/HDD ratio returned to the previous year's level, confirming real energy savings independent of the weather.