3.4 Quantify the effects of relevant variables on energy consumption
What is this?
The energy consumption of all organisations is affected by different activities to different extents. Two typical significant driversare discussed in this guide but specific circumstances may require the investigation of others. The two driversconsidered here are production activity and weather, which experience has shown to be the most common drivers. Drivers are those activities and/or factors which cause a change in energy consumption, e.g. in colder weather we normally consume more heat energy than in warmer weather. Drivers are also known as energy factors or energy variables. In ISO 50001 they are called relevant variables and that is the term we will use in this guide.
How to implement
As a simple example we can know the energy performance of an automobile by knowing its fuel consumption in litres per 100 km. We can predict the fuel consumption by multiplying distance by documented fuel consumption. This will normally give an accurate result. However there are parameters that will cause differences including driving style, driving conditions, vehicle age and condition, etc.
Production Activity
In simple cases with only one product then the amount of the product produced will affect the energy consumption. There are often more than one product and then the effects of each need to be considered.
Weather
If a significant amount of the organization’s energy is used for space heating or cooling then this should drive a significant part of its energy use. This will involve the application of the concept of degree days. It is beyond the scope of this Guide to fully explain this concept other than to say that degree days are a measure of how hot or cold it has been and thus should be related to how much space cooling and heating are required. At a very simple level, if outside temperatures are equal to or higher than the required internal temperatures then it should be not needed to add more heat. It is not uncommon to find space heating systems operating in warm or summer conditions and vice versa with space cooling systems.
Weather also has a significant effect on the energy performance of refrigeration systems and if these are a SEU then it will need to be considered. This is due to the effect on condensing temperature, where ambient dry bulb temperature affects air cooled systems and ambient wet bulb temperature affects evaporative condensers and cooling towers.
It is beyond the scope of this Guide to give all the information on the use of spreadsheets and the interpretation of the statistics involved. It is recommended that more research and training is completed in this critical area. It is necessary to understand how to carry out regression analysis including multivariate regression and to interpret the results. Multivariate analysis is usually needed as there is often more than one driver of energy variability.
For each energy source and each SEU you need to try to identify and quantify the relevant variables affecting their consumption.
There are various methods of measuring energy performance in organisations. The 3 most common are:
Trends in absolute consumption. As stated earlier this is very simple but does not take account of the effects of relevant variables on the amount of energy consumed.
Specific energy consumption: this is also simple being a ratio of energy consumption per unit of production or other output. It is not an accurate or helpful method as it does not take account of the relevant variables and typically gives a wrong and overly optimistic estimation of energy performance improvement. It is affected by baseload. but takes no account of this. It will not be used in this guide for these reasons. It is in common use in many large organisations and it is necessary to learn its weaknesses in order to move towards better methods. In a large number of organisations, there is a high energy base load or a complex mixture of products. In the case of the large base load the level of production activity has a large effect on this indicator, i.e. if production volumes increase the ratio decreases and appears to show an improvement in energy performance when none has occurred. These indices are very popular during times of growth as they tend to show improving performance but do the opposite in times of falling output.
Methods based on regression analysis: these take account of the relevant variables and will be used throughout this guide as the preferred method. There is extensive experience in the UNIDO energy management program of companies learning to use this method and using it to identify savings opportunities.
The best method of establishing a baseline is to use the relevant variables which have been established earlier to predict the amount of energy that should have been used and to compare it with what has actually been achieved. In this method the baseline is the best fit straight line on the scatter chart of variable against energy consumption. As performance improves this line will move downwards.
The most common and simplest energy performance indicator is conformance to financial budgets. In many organisations this might be interpreted as successful energy management. It is not! The overall purpose of the energy management system is to improve energy performance and to continually improve this performance.
Ideally you will have at least one high level EnPI for each energy source (electricity, fuel, etc) at the top level to indicate that overall you are in control. This is often very difficult depending on product and energy variable mix. You should also try to have an EnPI for each of your significant energy users.
It is important that you develop these indicators while you are planning and that you monitor them routinely once they are developed. The indicators may require modification once you start using them in order to improve their effectiveness in showing you how your organization is performing.
An example of a simple ratio which is commonly used and normally of little value is the specific energy consumption (SEC) of various utilities. As an example the SEC of compressed air in terms of kWh/Nm3 of air produced is used. This can be very misleading as, for example, if we repair leaks or reduce our air consumption, we will almost always increase the SEC. Thus increasing SEC can be an indicator of improving or worsening performance depending on the reason for the increase. The use of these ratios can divert attention away from truly indicative indicators of energy performance. Even simple annualised trends of energy use are often of more real value.
Note that the use of SEC is often perceived to be of value in plants where SEC of individual compressors can be established and their performance compared. However the cost of instrumentation (especially the air flow meters) involved will usually make this level of information uneconomical to establish. Note also that the SEC will vary depending on the output of the compressors making comparison even more difficult. The same applies to thermal efficiency of boilers and coefficient of performance of refrigeration plants.
You want to improve performance and demonstrate this performance improvement. In order to demonstrate it the slope of the regression line and/or its intercept with the Y axis will need to be reduced. It is beyond the scope of this guide to go into much detail on this topic. You will also use the principles of cumulative sum (CUSUM) to track performance on an ongoing basis.
Being able to show performance improvement is essential for getting or gaining management commitment. Management wants to see clear results and return on investment for the measures taken. When the energy performance results are polluted with other determinants that can’t be controlled, the shown performance result gives a false representation. In practise this is one of the causes for management not to invest in energy saving measures any longer since they do not seem to pay out. When it can be made clear that they do, but that other determinants affect the energy performance of the organization negatively, this will help to keep management commitment.
What is this?
The energy consumption of all organisations is affected by different activities to different extents. Two typical significant driversare discussed in this guide but specific circumstances may require the investigation of others. The two driversconsidered here are production activity and weather, which experience has shown to be the most common drivers. Drivers are those activities and/or factors which cause a change in energy consumption, e.g. in colder weather we normally consume more heat energy than in warmer weather. Drivers are also known as energy factors or energy variables. In ISO 50001 they are called relevant variables and that is the term we will use in this guide.
How to implement
As a simple example we can know the energy performance of an automobile by knowing its fuel consumption in litres per 100 km. We can predict the fuel consumption by multiplying distance by documented fuel consumption. This will normally give an accurate result. However there are parameters that will cause differences including driving style, driving conditions, vehicle age and condition, etc.
Production Activity
In simple cases with only one product then the amount of the product produced will affect the energy consumption. There are often more than one product and then the effects of each need to be considered.
Weather
If a significant amount of the organization’s energy is used for space heating or cooling then this should drive a significant part of its energy use. This will involve the application of the concept of degree days. It is beyond the scope of this Guide to fully explain this concept other than to say that degree days are a measure of how hot or cold it has been and thus should be related to how much space cooling and heating are required. At a very simple level, if outside temperatures are equal to or higher than the required internal temperatures then it should be not needed to add more heat. It is not uncommon to find space heating systems operating in warm or summer conditions and vice versa with space cooling systems.
Weather also has a significant effect on the energy performance of refrigeration systems and if these are a SEU then it will need to be considered. This is due to the effect on condensing temperature, where ambient dry bulb temperature affects air cooled systems and ambient wet bulb temperature affects evaporative condensers and cooling towers.
It is beyond the scope of this Guide to give all the information on the use of spreadsheets and the interpretation of the statistics involved. It is recommended that more research and training is completed in this critical area. It is necessary to understand how to carry out regression analysis including multivariate regression and to interpret the results. Multivariate analysis is usually needed as there is often more than one driver of energy variability.
For each energy source and each SEU you need to try to identify and quantify the relevant variables affecting their consumption.
There are various methods of measuring energy performance in organisations. The 3 most common are:
The best method of establishing a baseline is to use the relevant variables which have been established earlier to predict the amount of energy that should have been used and to compare it with what has actually been achieved. In this method the baseline is the best fit straight line on the scatter chart of variable against energy consumption. As performance improves this line will move downwards.
The most common and simplest energy performance indicator is conformance to financial budgets. In many organisations this might be interpreted as successful energy management. It is not! The overall purpose of the energy management system is to improve energy performance and to continually improve this performance.
Ideally you will have at least one high level EnPI for each energy source (electricity, fuel, etc) at the top level to indicate that overall you are in control. This is often very difficult depending on product and energy variable mix. You should also try to have an EnPI for each of your significant energy users.
It is important that you develop these indicators while you are planning and that you monitor them routinely once they are developed. The indicators may require modification once you start using them in order to improve their effectiveness in showing you how your organization is performing.
An example of a simple ratio which is commonly used and normally of little value is the specific energy consumption (SEC) of various utilities. As an example the SEC of compressed air in terms of kWh/Nm3 of air produced is used. This can be very misleading as, for example, if we repair leaks or reduce our air consumption, we will almost always increase the SEC. Thus increasing SEC can be an indicator of improving or worsening performance depending on the reason for the increase. The use of these ratios can divert attention away from truly indicative indicators of energy performance. Even simple annualised trends of energy use are often of more real value.
Note that the use of SEC is often perceived to be of value in plants where SEC of individual compressors can be established and their performance compared. However the cost of instrumentation (especially the air flow meters) involved will usually make this level of information uneconomical to establish. Note also that the SEC will vary depending on the output of the compressors making comparison even more difficult. The same applies to thermal efficiency of boilers and coefficient of performance of refrigeration plants.
You want to improve performance and demonstrate this performance improvement. In order to demonstrate it the slope of the regression line and/or its intercept with the Y axis will need to be reduced. It is beyond the scope of this guide to go into much detail on this topic. You will also use the principles of cumulative sum (CUSUM) to track performance on an ongoing basis.
Being able to show performance improvement is essential for getting or gaining management commitment. Management wants to see clear results and return on investment for the measures taken. When the energy performance results are polluted with other determinants that can’t be controlled, the shown performance result gives a false representation. In practise this is one of the causes for management not to invest in energy saving measures any longer since they do not seem to pay out. When it can be made clear that they do, but that other determinants affect the energy performance of the organization negatively, this will help to keep management commitment.