Encouraging engineers to push beyond minimum compliance toward genuinely excellent systems.

For several years now I have been working to improve the energy metrics associated with refrigerated warehouses. In fact, this has been a long term fixation; the third column that I wrote in this series (Improving Efficiency, May 2012) defined efficiency as “the ratio of ‘what you want to do’ to ‘what you need to do.’” I pointed out that if you are a pie maker then the suitable metric for you will be “pies per kWh.”

Despite this, I have never seen a process refrigeration plant where information on production throughput is gathered by the refrigeration control system and used to assess plant performance. My recent work has concentrated on warehouses because a suitable metric can be created based on store volume. This is the specific energy consumption (SEC) measured in kilowatt hours per cubic meter per year, or kpma for short. This has been relatively easy to implement in the plant control system because the store volume is a constant that can be noted at the time of commissioning and should not change over the life of the store. It’s a bit of an oversimplification because it ignores the level of activity in the store and assumes that the temperature level is satisfactory, but it has proved to be an extremely valuable tool in driving energy efficiency improvements.

However, when the facility includes an element of process refrigeration (for example, blast or spiral freezers or chillers), the oversimplification is too severe and a different approach is required. It is essential to know how much product was processed otherwise the SEC kpma will be meaningless. When the kWh number is high in a central plant system, the operator has the tough task of figuring out whether that is caused by poor operation of the warehouse or inefficient control of the process. Often it’s impossible to tell and so improvement opportunities are missed.

The first step in getting more analytical about this is to decide whether the system is a freezer plant with a bit of storage or storage with a bit of freezing capability. The latter can be addressed by using the regular SEC metric with an adjustment for the process throughput. This makes allowance for the process load by deducting a portion of the energy use equivalent to the ideal required to achieve the process requirement. However, in the former case, where the process load far outweighs the storage requirement, we need a different approach.

Process energy consumption (PEC) is the kWh per tonne of product processed and is a more appropriate measure if the store volume is small compared to the product throughput. Switching between the two measures is tricky because SEC is “per annum” whereas PEC is “per tonne.” However, in my experience most systems are clearly one or the other—there are very few in the middle ground where the energy needs of the process are about the same as the energy needs of the store. The important thing is to pick a method, start gathering data as soon as possible and be consistent. Surprisingly, consistency is much more important than accuracy in this type of exercise.

The information on process energy consumption benchmarking is readily available and yet this methodology is still not incorporated into refrigeration plant control systems. Is this because we think it is too difficult or because we just don’t care enough?