As discussed in the previous parts, Power Bi provides plenty of features to clean, monitor, and report your utility data. This article would be the last from this series and I would like to draw your attention to the baseload analysis and what we learn by looking at the consumption graphs.
On 17 Aug 2021, New Zealand prime minister announced the country enforcing lockdown followed by finding 1 new community case of Covid-19 with delta variation. The alert levels 4 and then 3 applied, forcing all the non-essential businesses and services to remain closed until further notices. Today that I am writing this, it has been almost 70 days that these restrictions have been put in place. Many businesses and commercial buildings had only a few hours after receiving the announcement to shut the energy users equipment hence leaving many of them running unintentionally during the whole lockdown period…causing not only extra cost for the business but also hundreds and thousands of kWh of energy and therefore emission. Below, are photos I took from Sylvia Park, the country’s largest shopping mall a night during alert level 4. The lights, sound system, and perhaps ventilations were running days and nights for no one! This is just an example of many across the country. And yet we are loudly talking these days about climate change and sustainability…
So, who will be going to pay for this? the shopping mall owners? the shopkeepers? the government? you and me? or the whole world? Yes! During my evening walks not only during the lockdown, but I also notice many buildings and malls and shops are shining unreasonably at night!
The amount of energy we use while no one uses a building or facility is called ‘baseload’ and managing this is a simple task so-called a low-hanging-fruit in energy management system implying a handy action plan to be taken without a huge effort. Let’s go back to the school (our example) and see how the fancy Power Bi graphs can help us understand this baseload and its reduction. The following displays the main site buildings usage in Aug. 2021.
What we are observing here? a sharp drop since 17 Aug. as expected due to the lockdown from 2400kWh to 900kWh a day. This means if no one uses the buildings this is the baseload would be still around 37% of normal daily usage. This has been almost close to the weekend load (look at those drops on days 7-8 and 14-15 which are Sat. & Sun.). Knowing that on a normal weekend, there are still people who use the buildings and some facilities, it appears that the usage in lockdown time is higher than expected as nobody has been allowed to get into the buildings according to the restrictions. The school has an energy management system put in place and they have been successful in implementing some good practices for example switching off unnecessary equipment and energy users after working hours. So the flat rate of usage could be optimistically due to the servers, emergency lightings, and cameras that would have been running all 247. The following is the sum of 24hour usage for Aug. 2021. This very flat rate (before and after the working hours) which has been also appeared in the previous months, proves this claim.
This is where having a good check metering system can be helpful to drill down to the users. The above data were collected from the retailer meter (revenue meter). The following illustrates the data collected from the check meters across the same site. Note in Power Bi we can add an interactive site map highlighting the selected buildings showing more details like where they are located and how big they are. I selected 4 buildings out of 13 which are being monitored by the check meters. They appeared to be large users.
Further analysis on this finding, can be really useful, for example, we know that the data server is located at one of these 4 buildings, so it makes sense that the user remains nearly the same as a holiday or weekend. I would not expect to see a higher usage in a building like ‘Technology and art”. Take a closer look! the lockdown daily usage is even bigger than a weekend! In the next step, I only choose and focus on two buildings that contributed to remarkable higher usages than an ordinary weekend.
Both selected buildings have been using more energy during the lockdown compared to a weekend day. They are not even hosting any data server so the question remains as to why their usage has gone up comparing the weekend before lockdown (days 14-15)? There might be some devices left on! Pay more attention! the red building has relatively a flat load while the yellow was fluctuating! knowing that none of these has been in use in all those days, more investigation is required. Next time when we go back to school, we can try listing the energy equipment in both buildings and have clear instructions to turn them off when another lockdown happens!!
In the other buildings, such as the boarding house where the students live, the lockdown drop is even sharper. Depicted below, from 18 Aug. which is the first day of lockdown, a 100kWh daily usage is a baseload. Apparently, for this facility, being on weekend does not necessarily mean a lower usage, instead, the occupancy rate does! Listing the energy users and grouping them to base users and variable users can be a good practice to further assess why some equipment is running all the time. Ask is it possible to switch them off or control them remotely when no one uses the facility?
The next figure shows the total baseload for a different month in the summer. A load of 200kW has been nearly in use from midnight to 6 AM. This 1200kWh per month is very similar to the same baseload in a winter month. It means the lighting or another use which are not affected by outside temperature, are the main contributing ones.
In many cases even including the school buildings, measuring the occupancy rate which is potentially an influential driver of the energy usage is not easy. We used to take the weekend load as the baseload of our energy baseline but we knew that people can still use the facilities during the weekend. Before the lockdown time, we never had a chance to monitor our fully vacant building/facility! Let’s say the bright side of Covid-pandemic is the opportunity of investigating our energy usage during the lockdown periods which has been long enough to learn from!
Summary:
There are many ways to interpret and analyze the energy/utility data if you manage to visualize them in meaningful graphics. What is important is doing these reviews systematically, discussing them with the team regularly, documenting the reasoning and findings, and consulting with energy advisers. If you have an energy management system, these practices will be central and inevitable parts of your routine tasks. The energy management system is still the most achievable effort towards reducing our energy cost and making tangible while strong steps towards abatement of emission and climate change.
In the end, I appreciate once again the Epsom Girls Grammar School’s manager, Bronwyn McGill, who granted me permission to use their data for these publications. I am also grateful for Simon Ross that I had the opportunity to work with him during my career at Economech Analytics. I hope you have found this series of articles comprehensible and informative.
Managers would usually love to take a quick look at the overall performance indicators rather than spending time on worksheets, graphs, and other forms of showing historical usages. Performance indicators are essential when you want to simply and smoothly monitor how your business is going. In an energy management system, it is required to identify and adopt EnPi (energy performance indicators) to represent and reflect how energy is being used in your organization. Setting the appropriate and relevant EnPi does not depend on which software you use. However, as I am going to explain, using Power Bi is a huge relief when someone has to regularly calculate and report these metrics.
Use the ‘Gauge’ visual in the Power Bi desktop to set a calculated ‘measures’ that are to be updated automatically at the end of each period and indicate performance indicators. The metrics could be defined based upon the context of the organization as well as what the stakeholders (e.g. managers) would like to track (and share). The above indicators calculate the per capita energy/water usage, energy-related emission, and renewable energy generation of a university campus on a monthly basis. Creating and updating this with ‘Excel’ would involve much more effort and time.
In the figure above, another way of showing the indicators is using the ‘Tableua’ concept. This way, as you go forward, you see how different types of energy across different buildings are performing. The metric here is the ‘deviation % of Year to Date usage’ to be compared with the same period last time. For instance, if the total electricity usage of campus 1 from Jan-Aug 2021 exceeds over 10% of that of the same period last year (Jan-Aug 2020), the color would turn to red for 10%, yellow to below 10% increase, and green for less usage. This is a quick awareness for a manager to further investigate (or drill down on the other graphs) for the red areas and keep looking at the yellow ones in the future hence save a lot of time to analyze. Needless to say, green is not showing necessarily a saving because of your good practice and can be an effect of a long shutdown. A regression model with appropriate variables is a rather reliable reference to compare the actual usage with and conclude the savings. The concept of showing the colorful tableau can be simply applied to your other indicators.
The above chart is depicting the EII indicator for the energy performance of a site. It is worked out by dividing the ‘actual daily usage’ by ‘expected daily usage’ calculated from the regression equation (energy baseline model Ref. part 4 of this article). This is simple while a robust EnPi. It is supposed to be smoothly moving around 1 and if you have any action plan, you would expect the values to fall less than 1. The unusual changes or outliers can be further investigated and are easily detected with this graph.
Alert notification:
It is apparent that we cannot expect the managers to constantly monitor the graphs and indicators every day and hour. So, it makes more sense we notify them when something goes off the track! Here, we could manage to set up an alert to send to a school manager once the hourly (or total daily) water usage in the site reading from a smart meter exceeds a certain threshold. With this simple feature of Power Bi, the school could save a significant amount of water that would have been wasted from unnoticed leakages many of which occurred during night or weekend.
In the context of energy management, having a baseline model as a reference for energy consumption has been highly recommended. An energy baseline is a mathematical (simple or complex) model representing the relationship between energy usage vs. energy drivers or the variables that influence this usage. Using the ‘Regression’ visual in Microsoft Power Bi provides you a lot of insights and ideas about how energy is being used and why! Using this technique with excel might end up with the same outcomes but the level of interaction and attraction of the graphs in Power Bi is outstanding. You may easily toggle the energy users, variables, timeline, and other inputs., include or exclude them in the model with a flick. Of course, this visual will be viewable when only one variable comes to play. A multi-variable regression, unfortunately, cannot be handled with this visual however the opportunity of trying the different variables and see which becomes the most single relevant one is something not to be missed.
The above example is borrowed from an urban water service project. On the x-axis, you may choose a variable like the amount of water delivered (the energy driver). And, the y-axis could be the pumping energy user. A very fine and neat correlation has come out. The line slope (0.43) is the amount of additional energy required to pump one more cubic meter of water. Isn’t it amazing? The dots represent each day of these 366 data (one year). The equation on the top left of the graph is the energy baseline given that the correlation ratio is significantly high in this case.
The outcome of regression modeling is not always promising. The following depicts how energy is used against the outside temperature (HDD) for a school building during the weekends. 106 dots have been picked in this sample. Even a poor correlation like this has something to learn from. It shows the energy varies indifferently to the heating required. This is obvious as, during a weekend, no heating equipment like a heat pump is used so the variation is likely due to other factors. The level of dispersion of the dots around the line (which is automatically added by this feature) may also give us some ideas of why on some weekends the energy behavior is different. The fitting line even in a poor regression is an indicator of the average daily usage (900kWh/weekend in this case). Interestingly if we switch to ‘weekday’ we observe a better correlation with HDD (heating degree days) which makes sense. Hence, we might end up with different baselines for different day types.
Needless to say that clustering the data is a useful technique. You may look at the regression outcome for your hourly/daily/monthly energy usage and group the variables (e.g. outside temperature, occupancy rate, et.) accordingly. Also grouping the data based on the daytype.
In the graph below, we managed to slice the timeline to include only the daily data for 2 years. The total energy usage versus the outside temperature. Clustering the data on this graph is helpful in such a way that we can look at the weekday and weekends separately. Listed below, the lessons I would learn from this graph:
I expect this site uses relatively higher electricity on a weekday.
Energy usage behaves differently from ‘weekday’ to a ‘weekend day’ with or without respect to the outside temperature. There is also a non-linear curve-shaped data form observed in the black chart so maybe a linear regression is not a useful method for showing this relationship.
Those 4 or 5 black dots that are closer to the green line (the fitting line of the weekend) are worthwhile for further investigation so are the green ones in the proximity of the black dots! the former set delineates weekdays with good performance while the latter indicates more than expected usage.
Outside temperature (HDD in this case) is definitely not an influential variable on the energy consumption of this specific site. So one needs to further seek the other variables to incorporate into the model.
Talking about regression and its application in energy management may take days and nights! and we are here just to quickly show you there is a powerful feature in Power Bi (named Craydec Regression Chart) you can plug and play to your report as an analytical tool for energy baseline modeling and performance tracking. In the presence of an acceptable correlation, it can be also used as a tool for the evaluation of action plans and calculation of savings. Please write to me if have any inquiry.
