The power of OEE and a few discrete energy meters
25 Nov 2010
Alan France, operations director of Idhammar Systems, has worked out exactly how much to charge the kids for cooking pizzas in the oven. Here, he links this concept to the use of OEE and smart metering to reduce energy bills in the process industries:
The energy companies are giving away small meters to accurately measure domestic electricity consumption; I guess you will have seen the advertisements.
It’s a simple process - you place a small clamp around a cable near the electricity meter which is then cabled to a transmitter. That transmitter links via wireless to a small screen which is probably placed in the lounge.
The screen then displays kilowatt hour (kWh), cost or CO2 usage, and you can set targets to help persuade the kids to put on an extra jumper, rather than firing up the heating system.
The meters seem to work very well and give you an insight into what appliances zap the most energy. As I have replaced most of my traditional bulbs with LED lamps, my base house load is quite low at a few hundred watts, but start up the big screen TV and surround sound system and you can see just where the money goes.
When my kids bring round with a couple of pizzas for Saturday lunch it costs a surprising 9 pence to heat them in the electric oven. (I now collect the money at the door.)
Taking this concept to the factory floor presents some interesting opportunities. It is pretty easy to measure overall factory kWh and providing there are enough meters it is possible to measure each zone or area. But until recently, unless you spent big money, it wasn’t so easy to measure the energy consumed by individual bits of equipment.
But is it a good idea?
Let’s think through the normal energy reduction process. Most manufacturers have already implemented a wide-range of energy related programmes with varying degrees of success, including switch-off campaigns or installing sensors to turn lights off automatically.
In fact, I remember training at IBM in the early 80s, and after 5pm I had to use a phone in order to turn on the lights for my desk while the rest of the office remained in total darkness.
Other energy-reduction measures have included intelligent production scheduling at a lower time-tariff, the installation of new energy-efficient equipment, and improving maintenance procedures to reduce leaks.
All of these programmes are necessary to reduce energy consumption but they are not sufficient on their own to provide step-changes in energy efficiency. We need to address a bigger issue – equipment effectiveness in relation to products made.
A meter in each area will show how much electricity the entire area uses, and give a good indication of energy costs per product, but what is the true cost and level of usage if for instance we suffer a breakdown and the kit is left running at the end of production?
Can we use OEE & loss analysis to reduce slow running, and prevent breakdowns?
There is significant logic in linking simple current transformer (CT) type meters to an OEE application. OEE software already includes the asset structure, running hours and a product count.
If we then include the output from a number of meters, at least at line level, we can ‘read’ the electricity consumption at defined intervals and compare the value to products produced. We can then consolidate the values at higher levels for effective site reporting.
If we go further and link meters at asset level we provide even more opportunities for analysis. We can then see which machines consume the most energy and which should be switched off during a breakdown. Also think about fixing plates to each machine to show operators the energy consumed by each piece of equipment.
And let’s not forget the impact of slow running; a filling machine running at 150 bottles per minutes instead of the 300bpm it’s designed to run at uses virtually the same amount of electricity.
But for that same energy cost you only have 9,000 bottles to sell at the end of an hour instead of 18,000. Quite some loss if the product is Vodka.
These types of meters are becoming very economical and they can be easily accessed via an Ethernet IP Address. Surprising what you could learn, and improve, from a few discrete meters and a product count.
.