Computer Control and Optimization

I see that one of my advertisers is AspenTech. This is an interesting development, since AspenTech is one of the companies I worked with at BP. In fact, if you dig deep enough into the white papers on their website you'll find reference to their working together with BP on developing and applying software for olefins processes. I was involved to some small extent with this work when I was at BP's Chocolate Bayou plant.

This is apparently in response to a casual mention in a previous post that computer control was one of the main tools for energy conservation, and the AspenTech work is an example of the cutting edge of this trend, and serves as a good example of the potential in this area.

Olefins, as with many industrial processes, has a large number of variables which effect the overall efficiency and economics of the process. In particular, several different feedstocks, varying heat input flux and temperatures and pressures at numerous points of the process, as well as heat and power sources work together in an almost infinite number of possibilities with dramatically different outcomes. The inability to handle optimization of all these variables manually can result in the plant running in a generic condition which may well be inefficient for the given conditions, and on feedstocks that are not optimal for today's price environment. By generating a computer model that simulates the process, it is possible to determine the optimal conditions and either manually or automatically control these conditions as required for best economics. By doing so, it is possible to input the optimum feedstocks and fuels based on real time prices and optimize the process for maximum yield for real time conditions. Doing so can result in millions of dollars in feedstock and fuel savings, as well as top quality products for a typical olefins plant.

The above is just one example of the potential of computer control and optimization. It is becoming more widely used in industrial processes, and has also been widely used in automobiles. A large part of the efficiency and performance gains in automobiles in the past 20 years has been a result of this computer control and optimization. With the low fuel costs prevailing for most of the past two decades, the optimization has been primarily on the performance side, but with the higher recent fuel prices I expect more focus on efficiency.

Obviously, viability of this type optimization is largely related to scale, but if fuel prices continue to move up the potential moves down the scale to smaller applications. Did you know your furnace would operate more efficiently with a variable, rather than an on/off control, or that your refrigerator would be more efficient with a power factor correction? That a significant part of the power consumption of your electronic equipment is consumed when you are not using it? That your air conditioner could operate considerably more efficiently if could detect and optimize both humidity and temperature, or that it would be more efficient with variable speeds rather that on/off controls? Come to think of it, why are you air conditioning rooms you are not occupying today? All these efficiencies could be captured with better controls. And, indeed they will when economics and the market demand it.