AI Predictions in the Energy Simulation Industry
The advent of machine learning in the energy sector has the potential to bring a golden age for building energy modelers. Imagine being able to generate building geometry from a PDF, or calibrate a model as easily as using the solver function in Excel? Imagine running a pre-simulation that approximates results in real time. These things are already possible and being done on small scale. We just need the right people to step up and develop such tools and offer them open-source or at reasonable prices.
But who will develop the AI tools? And why?
Well, several stakeholders are perhaps unknowingly sitting atop AI gold mines. This series of blog posts aims to assist companies in identifying these untapped resources, and provides strategies for individuals to plan for the evolving landscape.
Initially, I reserved these insights for a competitive edge, I've decided to disseminate some thoughts so as to get this party started. (I know factually that many important people will read this)
Understanding Machine Learning's First Wave
Let's delve into how the first round of machine learning works. AI evolves in a stepwise fashion. First, machines require databases filled with existing inputs and outputs. Note that the format is important. The machine maps inputs to outputs and "learns" to identify patterns and inferences in ways that are beyond human comprehension.
This round of AI is characterized by three key factors:
- Data existing in a predefined input and output format.
- Large volumes of data in a consistent format.
- A valuable computation between the input and output, offering creative or predictive potential.
A Simple Example
Consider the trend on social media where users posted a current picture of themselves alongside an image from 10 years prior. This phenomenon met the first two conditions perfectly, offering a vast dataset in a standardized file format of photographs, with clearly defined inputs and outputs.
The result was an AI that could predict what you will look like in a decade. Most people think that the value is for fun little apps that predict what you look like. The real value goes far beyond personal curiosity. Consider the potential for unsolved crimes, missing persons cases, and military uses. Plus, when combined with other data, such as facial recognition technology, this capability becomes even more powerful. For instance, posting an old photo on Facebook might automatically tag and connect you to a childhood friend, enhancing social interactions and potentially influencing the platform's stock value.
A Forgotten Goldmine of Energy Modeling
So why do I mention stock price? What does that have to do with energy modeling?
Don't ever forget that our models have a significant influence on infrastructure. They determine investments. Our work drives building design decisions, which drives data, ultimately accumulating into hundreds of billions of dollars towards various industry trends (Ground source VRF anyone?)
Therefore machine learning provides two advantages 1) make it easier to save energy 2) identify products that save the most energy in various applications (data driven proof that a product worked!)
Who will tap into this goldmine? If you locate large amounts of energy modeling data, the data formatting is often inconsistent. You can’t just dump randomly organized data into a machine. That’s not how AI works.
Data needs to be parsed, organized, and THEN mapped.
If you have pre-organized data, you have a massive headstart. The question is, Where is the machine-ready data?
Finding Large, Consistent Data Sets
Entities with access to vast amounts of consistently formatted data can skip the parsing and pre-processing. They can dump their data directly into a machine.
Who has large amounts of simplified and organized data?
In the context of energy simulation, several organizations come to mind:
- U.S. Green Building Council (USGBC)
- ENERGY STAR Portfolio Manager
- Utility company rebate programs
- Various government initiatives
Example, Turning the G in USGBC from Green to Goldmine?
I've simulated hundreds of LEED buildings, so I'm going to use the USGBC as an example (this is purely speculative, as it's the best example I have personal experience with).
In order to earn various LEED certifications, modelers are required to download and fill out a detailed spreadsheet with various inputs and outputs from their energy modeling files. This spreadsheet alone, when fed into a neural network, could yield predictions beyond our current understanding. Alongside other submitted documents, USGBC could amass a treasure trove of data invaluable to corporations (e.g., "To obtain LEED Platinum, and cut your energy bills by 40%, use one of these three light bulb manufacturers").
How could this change how you do your job?
Beyond benefits to corporations, machine learning, in the USGBC data alone, could pioneer an accurate "pre-simulation" engine. For instance, with as few as ten inputs, a well-trained AI could predict building energy usage with better accuracy than many full energy simulation models. Additionally, the software might recommend commonly used energy strategies that have resulted in LEED certification and highlight the most cost-effective solutions.
Test and improve a simple free tool
In line with this, we took some limited information, combined it with a custom GPT and developed "Optimize Architect," a free chatbot designed to guide you in identifying the ten most effective energy conservation measures. It begins with basic queries and progressively offers more creative suggestions.
Check it out Optimize Architect here
If you're keen to understand energy modeling in ways AI currently cannot, our ecourses offer a comprehensive learning experience that focus on "why" and "how" and other items where you can beat the bots.
Stay tuned
I have been experimenting with various AI tools and plan to share insights on safeguarding your career against the AI revolution in upcoming blog posts. This will ensure you remain invaluable during the golden age of energy simulations, allowing you to focus on the unique contributions only humans can provide.