Save energy, save money — energy management
Energy management is relatively new, but is being strongly driven by increased energy costs and the rising insecurity of supplies. It is not a ‘green’ or ‘carbon management’ issue, but a business concern and, in many cases, a matter of survival. For most materials processes, energy is the third largest variable cost (after materials and direct labour) and in some cases, it is the second highest. Getting the measurements wrong can be fatal, but most materials processors have a poor understanding of energy costs and how to manage them.
However, it is possible to use easily available information to gain an understanding of how any materials processing site is performing and use this to improve operations and performance.
For most materials processing sites, plotting a scatter chart of monthly energy use and production volume will give a graph as shown below (real data for an injection moulding site). Fitting this with a linear best-fit line gives the performance characteristic line (PCL). This can be used to find the ‘base’ and ‘process’ energy loads.
The intersection of the PCL with the vertical axis is the base load – the average energy use when no effective production is taking place. The example site has a base load of 152,440kWh/month, showing the energy ‘overhead’ of the site.
The slope of the PCL is the process load – the average energy used to process a kilogramme of material. It shows the process efficiency of the site. The example site has a process load of 1.5751kWh/kg.
The PCL can be employed in energy performance assessment. Using this example, if the production volume is 200,000kg, then the target energy use will be –
kWh = 1.5751 x 200,000 + 152,440 = 467,460kWh.
Production accountability for energy use is now possible by comparing the target and actual energy use for the monthly production volume. This provides a tool to set targets and assess performance.
Many companies calculate a monthly ‘kWh/kg’ from energy use and production volume information. Decreasing values are good and increasing values are bad. This monthly value is the slope of a line from the origin to the individual monthly point on the graph (below) and is naturally affected by both the base and process loads. Simply increasing production volume will reduce the ‘kWh/kg’ value because the fixed base load will be amortised over a greater variable process load – this will lead to the impression that energy efficiency is improving.
With rising production volumes, management sees a decreasing ‘kWh/kg’ value and is congratulated for doing nothing at all. With falling production volumes, management sees an increasing ‘kWh/kg’ value and is criticised for failing to reduce costs – a situation they are often less than happy with!
The PCL can also be used to forecast future energy use. For example, the sales forecast can be converted into production volume and the PCL, which helps predict the energy use.
The PCL can also show the extent of management control over energy use – high data scatter generally indicates poor process management and process changes. Alterations to processes will be revealed by changes in the PCL.
The PCL method applies to any materials conversion process and provides an insight into operational efficiency, as well as being a tool for assessment and prediction. Not a bad result from information already collected but not processed.
External site benchmarking
Most materials processing sites have a fixed base power load and a variable process power load that is directly proportional to the production rate
Total load = Base Load + Process Load where Process Load = Process specific energy consumption x Production rate.
The can be transformed to give the site specific energy consumption (SEC) as a function of the production rate and this is plotted above,left, with the ‘average’ process SEC and a power law (y=Axb) fitted to the model data.
This shows that at most materials processing sites, the global SEC will decrease with increasing production rate, so the often quoted ‘average’ SEC value for any process is misleading. As an example, production data from 98 injection moulding sites can be used to produce the graph (below) to show that real sites behave broadly as predicted by a simple model.
The average of the individual SEC values is also shown, but tells nothing about a sites relative energy efficiency. An operating curve is needed for external benchmarking against similar sites with the same production rate.
Materials conversion processes are not equally energy intensive and this is reflected in the coefficients of the power requirement equation and the shape of the operating curve. For example, site data for extrusion gives an operating curve of – Site SEC = 2.8794 x (Production rate)-0.4239.
External site benchmarking provides valuable information but only when corrected for the production rate.
External machine benchmarking
Many companies would like to benchmark specific machines and as with sites, the power use of most materials conversion machinery is rate dependent and a similar equation operates –
Total load = Base Load + Process Load.
As an example, machine monitoring data from 141 hydraulic injection moulding machines were used to show that real machines also behave broadly as predicted by a simple model. A single number SEC for machine efficiency is misleading. An operating curve is needed to benchmark the external machine against similar machines with the same production rate.
The energy intensity of other conversion machinery also behaves similarly but produces different operating curves. External machine benchmarking provides valuable information, but only when corrected for the effect of production rate.
Efficiency versus production
Unlike cars, where you drive slowly to improve energy efficiency, in materials processing the harder you push the site and the machine, the better the relative energy efficiency.
Understanding what we are doing in energy management can start with the use of existing internal measurements for assessment and prediction. It can then include external site benchmarking (corrected for production rate) and external machine benchmarking (corrected for production rate). This logical and easily understood framework can be used to generate real improvements rather than paper and statistics.
To view detailed images, graphs and diagrams referred to in this feature, please download the PDF of the original article using the link below.