I was skimming Make magazine in the bookstore today, and found out that Google has a .org project to provide power meter data to the web for free. They call it PowerMeter, and it works with a pilot of a handful of power companies at the moment, but the potential for rollout is huge.
I especially like this quote on the top of the PowerMeter site:
“If you cannot measure it, you cannot improve it.” — Lord Kelvin
A genius and great engineer.
This meshes with something else in Make: Using the DMAIC analysis process to find energy “leaks” before taking large action action. From memory:
Define, Measure, Analyze, Improve, Control
Tell me that wasn’t drilled into my head at my former employer.
The DMAIC Process
The DMAIC process is an engineering tool, and I learned via Six Sigma courses. It closely parallels the scientific method, and it can be a great razor of the unnecessary in any problem.
Define the problem:
We start the place we start with any problem. We define it:
My energy use seems high in the summer.
My power use is above the national average per year (10,656 kWh)
That wasn’t so hard. With the problem defined as broadly as possible, we are now free to go on to the next step.
Measure as much as possible:
Use tools like stuff like the Powerkuff or TED, and Google’s real-time charts to find typical power usage. Use a Kill a Watt on individual appliances. Use an IR thermometer to look for hot spots around your home. Use punks to check for drafts. Look around for inefficiencies in the specifications of things you can’t measure directly.
Collect all of this data somewhere. You’ll need most of it later.
Analyze and filter:
What uses 80% of the energy in your home? Find the answer to that question and you’ll know what you need to correct. You can also find the real cause of your problem though Root Cause Analysis, Ishikawa (fishbone) diagrams, and process tools like SIPOC. None of these tools are any harder to use than DMAIC, and they’re easy to put into practice.
I find that visualizing your data is the quickest way to make intuitive leaps. Using Google spreadsheet, you can plot the consumption, and in the case of real-time data, you can tag certain spikes in your use to devices around the house. A spike of 45 watts at 15 minute intervals may be your refrigerator, for example.
It’s critical to filter your data for meaningful results. Figure out if your central air system is healthy before replacing all your windows. Check to see if your windows are leaking before increasing your central air system’s capacity. Using the old adages:
Spend less than you earn
Eat less than you exert
Let’s call it “Less loss; Use drops” Start with that razor and cut back the noise.
Improve the system:
The fun part, which everyone skips to without research, is the Improve step. Now that you’ve done your analysis, and you know what your need to improve, you have to find ways of setting about improving the situation.
If you’re talking about a minor appliance or light drawing too much energy, simple replacement or careful use will do. I have some decision tools for changing out an appliance, and a personal example involving my bathroom lights, with calculations.
Once you have an improvement planned, you set your plan to action. You make your purchases, your replacements, and you habit changes. This is not the end of the process.
Control and refine:
You have to continue to Control the situation. Keep taking those measurements, analyzing your powermeter account, checking your TED or PowerKuff screen. Take more physical measurements whether directly or indirectly. Do this regularly, and continue the moving average of your consumption to find trends.
Staying Under Control
The important thing is you take your measurements in the same way as you did before, whether you think they are needed or not. You’d be surprised that sometimes replacing a refrigerator affects your home heating bill. The waste heat has to go somewhere, doesn’t it?
Indirect methods include the built-in analysis tools provided by your power company. For example, I include a side-by-side comparison of last year to this year’s June power bill. I’ve made a number of improvements in the last year, the biggest of which was 8 new windows:
The financial result of controlling one's spending
The net result was a 25% drop in my bill, and 260 less total kWh drawn for the billing period. As I mentioned, this method is indirect. It took half a year to start gathering results, and the drop can be attributed to many improvements made in the interim. That’s only part of the story, however. I created a plot of the past 24 months as well, including a 12 month rolling average:
My electricity use from July 07 to Jun 09
This is the process I try to employ in all improvements at home. I can’t wait to get my hands on Google’s PowerMeter, or Microsoft’s Hohm, but I may end up purchasing a TED since neither of those options will be available for some time.
Using the DMAIC Process to Improve Power Consumption
I was skimming Make magazine in the bookstore today, and found out that Google has a .org project to provide power meter data to the web for free. They call it PowerMeter, and it works with a pilot of a handful of power companies at the moment, but the potential for rollout is huge.
I especially like this quote on the top of the PowerMeter site:
A genius and great engineer.
This meshes with something else in Make: Using the DMAIC analysis process to find energy “leaks” before taking large action action. From memory:
Define, Measure, Analyze, Improve, Control
Tell me that wasn’t drilled into my head at my former employer.
The DMAIC Process
The DMAIC process is an engineering tool, and I learned via Six Sigma courses. It closely parallels the scientific method, and it can be a great razor of the unnecessary in any problem.
Define the problem:
We start the place we start with any problem. We define it:
That wasn’t so hard. With the problem defined as broadly as possible, we are now free to go on to the next step.
Measure as much as possible:
Use tools like stuff like the Powerkuff or TED, and Google’s real-time charts to find typical power usage. Use a Kill a Watt on individual appliances. Use an IR thermometer to look for hot spots around your home. Use punks to check for drafts. Look around for inefficiencies in the specifications of things you can’t measure directly.
Collect all of this data somewhere. You’ll need most of it later.
Analyze and filter:
What uses 80% of the energy in your home? Find the answer to that question and you’ll know what you need to correct. You can also find the real cause of your problem though Root Cause Analysis, Ishikawa (fishbone) diagrams, and process tools like SIPOC. None of these tools are any harder to use than DMAIC, and they’re easy to put into practice.
I find that visualizing your data is the quickest way to make intuitive leaps. Using Google spreadsheet, you can plot the consumption, and in the case of real-time data, you can tag certain spikes in your use to devices around the house. A spike of 45 watts at 15 minute intervals may be your refrigerator, for example.
It’s critical to filter your data for meaningful results. Figure out if your central air system is healthy before replacing all your windows. Check to see if your windows are leaking before increasing your central air system’s capacity. Using the old adages:
Let’s call it “Less loss; Use drops” Start with that razor and cut back the noise.
Improve the system:
The fun part, which everyone skips to without research, is the Improve step. Now that you’ve done your analysis, and you know what your need to improve, you have to find ways of setting about improving the situation.
If you’re talking about a minor appliance or light drawing too much energy, simple replacement or careful use will do. I have some decision tools for changing out an appliance, and a personal example involving my bathroom lights, with calculations.
Once you have an improvement planned, you set your plan to action. You make your purchases, your replacements, and you habit changes. This is not the end of the process.
Control and refine:
You have to continue to Control the situation. Keep taking those measurements, analyzing your powermeter account, checking your TED or PowerKuff screen. Take more physical measurements whether directly or indirectly. Do this regularly, and continue the moving average of your consumption to find trends.
Staying Under Control
The important thing is you take your measurements in the same way as you did before, whether you think they are needed or not. You’d be surprised that sometimes replacing a refrigerator affects your home heating bill. The waste heat has to go somewhere, doesn’t it?
Indirect methods include the built-in analysis tools provided by your power company. For example, I include a side-by-side comparison of last year to this year’s June power bill. I’ve made a number of improvements in the last year, the biggest of which was 8 new windows:
The financial result of controlling one's spending
The net result was a 25% drop in my bill, and 260 less total kWh drawn for the billing period. As I mentioned, this method is indirect. It took half a year to start gathering results, and the drop can be attributed to many improvements made in the interim. That’s only part of the story, however. I created a plot of the past 24 months as well, including a 12 month rolling average:
My electricity use from July 07 to Jun 09
This is the process I try to employ in all improvements at home. I can’t wait to get my hands on Google’s PowerMeter, or Microsoft’s Hohm, but I may end up purchasing a TED since neither of those options will be available for some time.
Related Posts: