Agriculture Reference
In-Depth Information
development of relationships to describe the behavior of the plant over the varying
range of conditions observed in animal housing and production environments, and as
such animal responses are typically treated as perturbations to the system.
Models for prediction of animal response vary in complexity from simple predic-
tion of performance decline to profit optimization, with a recent review provided by
Bridges and Gates (2009). Thermal comfort indices, such as temperature-humid-
ity index (THI) and black-globe THI (BGTHI) have been used to assess thermal
stress (Hahn et al., 2009; Bridges and Gates, 2009) and thereby predict performance
decline in swine (Nienaber et al., 1987), dairy (Johnson, 1962, 1963; Buffington
1981), and poultry (Zulovich and Deshazer, 1990). They have been used to assess
the suitability of different regions and housing systems (Gates et al., 1995) and as
an objective function in numerical and analytical assessments of poultry ventilation
systems (Gates et al., 1991b, 1991c).
Thermal comfort measures (Chao et al., 1995, 2000) as a control variable or a pro-
cess variable has found little application, particularly in heating applications. Lower
critical temperatures are not well defined for most modern genotypes. Increases in
heat and moisture production indicating increases in growth rate have been mea-
sured in swine (Brown-Brandl and Nienaber, 2008; Brown-Brandl et al., 2011) and
also in poultry (Chepete and Xin, 2001, 2004; Chepete et al., 2004). This increased
metabolic rate allows for lower temperatures to be used as animals mature and
growth rate slows to maintain productivity. Dozier (2007) reported increased gains
when broilers were kept at lower temperatures (12.8°C) as opposed to that consid-
ered to be within the thermoneutral zone (21°C), indicating that the thermoneutral
zone of the broiler has shifted, and reduced the upper critical temperature. Hamrita
and Hoffacker (2008) noted that deep body temperature in broilers shows promise
for use as an input parameter for ventilation system control, but traditional closed
loop control algorithms such as PID were unable to reach stability and suggest that
improved models relating physiological response to environmental stimuli are neces-
sary to fully implement feedback control of body temperature.
Indirect measures of thermal comfort as control variables have been evaluated by
a number of researchers, and some commercial applications have been attempted.
Widespread adoption, however, has not occurred. Notably, image analysis can pro-
vide substantial information on animal behavior and thermal status, and has been
applied to automate individual weight capture by a number of groups (Shao and Xin,
2008; Chedad et al., 2003; Brandl and Jorgensen, 1996; Mingawa and Ichikawa,
1994; Minagawa and Murakami, 2001; Schofield, 1990). Additionally, Shao and Xin
(2008) developed a metric for the relative distance separating individuals to deter-
mine current thermal status and used that information to adjust building tempera-
ture. Another example of a commercial approach of indirect measurement, Time
Integrated Variable (TIV) control, was applied to dairy and broiler housing for con-
trol of heat stress (Timmons and Gates, 1996; Timmons et al., 1995a, 1995b). In
these heat stress applications, the substantial heat stored by the animal population
was extracted during cooler nighttime conditions by the use of a longer-term run-
ning average temperature and separate TIV setpoint. Effectively similar to an inte-
grator controller, the resultant ventilation system and/or cooling system operation
was difficult for operators to understand because it behaved quite differently from
