Will the rise of AI help ease the worries of an energy manager or give them a bigger headache?
I’ve read an excellent book by Professor Steve Peters. called ‘The Chimp Paradox’. It uses scientific facts to help us understand the human mind and its thought process.
The Chimp Paradox uses a simple analogy to explain functional brain types, dubbing one the ‘Human’ and the other ‘Chimp’. We all have them both, the Human side is logical, calm and emotionally assured brain, whereas the Chimp brain can be fiery, spontaneous and alert to help you avoid trouble.
Both the Human and Chimp are essential keeping you in control of your actions and away from danger.
The Data Chimp
One of the challenges I’ve faced in Energy Management has been vast and growing array of data feeds from energy meters; whether half hourly, sub meters or BMS systems. Whilst data is useful it can also be hugely distracting and create emotional unease should it highlight unexpected results.
I recall an instance when a data feed suggested a remotely located property, that I’d recently invested energy saving technology in, was suddenly using excessively move electricity than anticipated. With my employers focused on protecting margins and me keen to justify my investment choices, my ‘data chimp’ ignited a flurry of activity intended to resolve the problem and protect my reputation.
After checking the data quality and running numerous comparison reports the consumption was indeed high but the cause uncertain. Only after I’d spoken to colleagues on site did I hear that overnight repair work was ongoing requiring the extra energy temporarily. Had I known this earlier my ‘data chimp’ could have remained calm I’d have invested my time better in other activity.
Another way?
Recently I’ve been testing an Artificial Intelligence (AI) tool for energy data. AI systems bring in extra contributors to data analysis process, that being remote verification and memory.
When abnormal energy consumptions occur, the system alerts an end user and asks him/her to validate the result. From this feedback the AI system will learn the appropriate reasons and corrective actions.
Over time the AI generates more appropriate alerts and suggested actions to the end user, thus removing erroneous alerts.
AI then becomes a massive database of actions and replies. Each time the end user gives appropriate feedback and this loop enhances the AI’s knowledge base, so continuously improving its response time and action.
A lookout?
In the future Energy Managers with the support of AI, are less likely to have a ‘data chimp’ worrying about irregular energy.
They’ll be kept calm knowing that the memory bank of past occurrences are constantly reviewing and assessing data irregularities in the AI system. AI could become a lookout, keeping a watching eye on your energy usage and only raising concerns when consumption fails to match previous patterns.
Should AI achieve this, the life of Energy Managers could be much more productive!
