Theoretically, chiller plant optimization can seem simple. Practically speaking, however, experience shows that it can be really challenging. Chillers are an excellent case study for this. On the chilled water (evaporator) side of a chiller, optimization is “simple”. Although it is easier said than done, the most efficient operating point of the system should provide the highest supply and return temperatures possible to meet the cooling loads. However, on the warm (condenser) side of the chiller, optimization is usually difficult both on paper and in the field.
In a condenser water system, there are three main components: the chiller, the pumps, and the cooling tower (see simplified diagram). All else equal, chillers are more efficient with colder water entering their condensers. To provide that, the cooling tower fan has to speed up (using more energy) to increase evaporative cooling. Greater evaporative cooling then allows pumps to decrease in speed (using less energy) to carry the same thermal load.
The relationship between these moving parts varies with the cooling tower approach, which is the difference between the cooling tower water supply (entering condenser temperature) and the outdoor-air wet-bulb temperature (the theoretical minimum temperature the cooling tower can provide). The chart shows one possible combination of pump, cooling tower fan, and chiller operations. Under this system, as the approach goes up under any given set of operating conditions, cooling tower energy and pumping energy use increases, but cooling tower fan energy decreases, and vice versa. This creates a minimum in the total operating cost curve, which is where systems should operate for maximum savings.
Generally speaking, the chiller will consume quite a bit more energy than the cooler tower pump and fans, so when in doubt, it pays to defer to what reduces the chiller load. But in most systems, running the cooling tower fans faster has a negative return at some point.
Modeling and the Machines
Identifying the “right” answer here is not simple. There are a few different options to make a condenser water system more efficient. First, approximated condenser water formulas are likely the most cost effective approach to optimizing a chilled water plant. But those are not specific to the building, so there will be room to improve there.
A calibrated energy model can help engineers select a theoretically optimum strategy, although even this may produce less than optimal practical results. The ultimate way to optimize these systems is to monitor their power consumption over a range of conditions, and use that data to generate an operating profile for the system. Engineers can do that today, but maybe controls systems will be able to do this in the not-so-distant future.