MODELLING TEMPERATURE SENSATION IN A POULTRY BARN

Chickens and their thermal environment

Meat chickens, also called broilers, in intensive rearing live in closed, controlled environment for about 36-42 days (depending on the target market) before they are brought to the abattoir. During this time, the thermal environment is very important, as chickens have no sweat glands and are prone to suffer from both cold and hot temperatures. Chickens manage their core temperature through heat exchange via their feet and their respiration, but are able to do so only after about 10 days of life. The so-called thermo-neutral zone of a chicken, i.e. the thermal range that allows a chicken to maintain a constant core temperature grows over time – the exact ranges are not currently known.

Temperature isn’t everything

Chickens like humans are affected not only by temperature: humidity also plays a great role. The higher the humidity, the more extreme temperature is perceived, because the higher water contents in the air conducts the heat better. Air flow or air speed also pays an important role: the higher the air speed, the colder it feels. This is called the windchill effect. For chickens, our best measure currently is called the effective temperature and is a combination of temperature, relative humidity and airspeed.

In order to cater for the different needs of birds during their lifetime, farmers have targets for temperature, relative humidity and some also for airspeed.

How is the climate controlled

Modern barns operate heating and cooling elements automatically via sensor input. The so-called “controller”, essentially a computer, uses inputs from sensors and translates them into actions: starting/stopping the heater, ventilators, air flaps and cooling elements. In essence, the controller compares the sensor input to the targets and decides on the action to be taken based on that information. Given that farms are quite large spaces (2,500m$^2$ x 3m height = 7,500m$^3$ is not atypical), controllers try to anticipate changes in the ambient conditions to avoid reaching suboptimal conditions for the animals.

The cost of climate control

In cold weather, farms try to ventilate less – because then they have to heat the entering cold air which increases the cost of production. In hot weather, farmers use laminar ventilation (called “tunnel ventilation”) to employ the windchill effect in cooling down the birds. Farms in hot climate also operate heat exchangers or coolers to mitigate outside conditions.

Are controllers perfect?

If controllers were perfect and barns were perfect, the ambient conditions inside would be indentical with the targets. However, that is not true. The sun, for example, has significant impact by heating up the barn; this triggers higher ventilation rates and lower temperatures (or effective temperatures) in the afternoons. At the same time, while heating does change the relative humidity (it dries the air), heating is expensive. If the outside relative humidity is too high, but the temperature is acceptable, controllers will allow for a higher relative humidity. As controller typically do not operate on effective temperature, this means that the thermal sensation will be stronger (colder or hotter).

The goal: How much impact do external conditions have on the thermal sensation of the animals?

The goal of this notebook is to investigate the impact of outside conditions on the effective temperature inside the barn. For this purpose, a data set is being made available from a particular in southern Europe for one production cycle of 40 days. The data was created from an IoT device and preprocessed so that there is a data point roughly every 8s. The task is to use correlation analysis and linear regression to identify whether outside weather conditions have an impact on the thermal sensation inside the barn and attempt to quantify this influence.

Results

We have attempted to understand the impact of outside weather conditions on the internal thermal sensation through a correlation analysis and a linear regression.

The regression results are reasonable (R2 = 0.87). From the analysis of the residuals we can observe that there is still information contained in them and therefore the model can be improved – very likely by adding confounding features to which we might not have direct access.

The inspection of the simulated results compared to the real results show that the model is able to reproduce to some degree the modulation of the controller of the outside conditions. In some cases, the model underestimates the capacity of the controller to reduce the strong variation of the outside thermal environment (see 2022-08-16 above). The model is also not able to reproduce a strong variation from the target temperature on 4th and 5th of September 2022 where the controller seemingly was not able to properly react to a strong increase of outside temperature. The model is by no means perfect but at the same time it is a good first order approximation of the interplay between outside and inside conditions.

From the correlation analysis and the correlation coefficients we can conclude that clearly the controller is not able to keep the effective temperature independent of the outside conditions at all times. While the effect of outside absolute humidity is smaller, outside temperature has about half the impact of the internal temperature target. The barn has a reaction time of roughly 3-5 hours. This means that with the outside temperature 4 hours ago, the outside absolute humidity and the current inside conditions, it should be possible to predict reasonably well the inside temperature.

The residuals have a standard deviation of about 1ºC, meaning that in 68% of cases the predicted temperature with this simplistic model is +/- 1ºC off the real value (and in 95% of cases +/- 2ºC). This is in the same range of the temporal and spatial differences in effective temperature in the barn.

Next steps

– Given that the data are time series, the toolbox for time series analysis (in particular autocorrelation analysis, trend decomposition etc) should be undertaken to improve the quality of the fit and of the prediction