What makes a Commercial Building Smart?

Let me start fresh at MMXX with a teaser from Bill Gates;

"You can completely ignore whoever tries to close their system"

Business Insider November 2019

Bill Gates said that is the only way to hire and keep AI talent is to let them share their research openly. He also mentioned that the US has long benefitted from openly sharing scientific research and that it remains a huge advantage, especially within the field of artificial intelligence. In his view, there’s no going back from that approach. “AI is very hard to put back in the bottle and whoever has the open system will so vastly get ahead,” Gates said.

Just keep that for a few moments in mind and take The Global Perspectives

The energy intensity per square meter (m2 ) of the global buildings sector needs to improve on average by 30% by 2030 (compared to 2015) to be on track to meet global climate ambitions set forth in the Paris Agreement.

Buildings and construction together account for 36% of global final energy use and 39% of energy-related carbon dioxide (CO2 ) emissions when upstream power generation is included. Progress towards sustainable buildings and construction is advancing, but improvements are still not keeping up with a growing building sector and rising demand for energy services.

Active controls could save up to 230 EJ cumulatively to 2040, roughly twice the energy consumed by the entire buildings sector in 2017.

Source: IEA (2017), Digitalization and Energy, IEA/OECD, Paris, www.iea.org/digital/

Now a bit more focused aspect on these subjects with Buildings, Energy with monitoring and controls as the rationale to start to optimize and transform buildings to SMART my favorite focus would be optimizing the machines that deliver the indoor climate. And just to give a bit of global rational, Cooling is the fastest-growing end-use in buildings, as its energy demand more than tripled between 1990 and 2018 to around 2.000 terawatt-hours (TWh) of electricity. Good ENOUGH?

To conclude the headline of my very first post in MMXX "what makes a Commercial Building Smart" -

Optimize building performance and occupant experience

AGREE... or would you rather prefer the subsets "Cutting Edge Technology to Enhance the experience" (Schneider Electric) or "Intelligent Infrastructure, which enables the building to collect and analyze data" (Siemens) what basically is great marketing to sell you more of the same story after the proprietary devices and platforms they have been selling for decades.?

We do know that Open Source is supporting innovation, creating automation innards and brings countless benefits to building owners, property managers, end-users and system integrators. but not for the OEM's "cash cows" of devices with proprietary device protocols and legacy platforms. Open source is nothing new to the software industry. It’s been around for over 40 years. However, for smart buildings, history is far more recent and its uptake could change this industry as we know it.

The medicine is available already for decades

The Equal Marginal Performance Principle was back in 2005 an entirely new way of looking at systems that are composed of multiple power-modulating components such as fans, chillers, and pumps. It’s of particular value because applying variable frequency drives (VFDs) on all motors in an HVAC system and configuring it as an “all-variable speed” system can improve efficiency and performance enormously. The problem has been how to optimize the design and operation of a configuration such that it is simple, stable, and yet achieves the highest possible efficiencies and improves overall system performance. Now we have Machine Learning and even AI toolsets are available the use of relational control algorithms based on the Equal Marginal Performance Principle has taken away the concerns from that time.  

The Optimum energy Systems can be applied to all water-cooled centrifugal chillers and select water-cooled screw chillers supporting single or multiple condenser water loops and multiple chilled water distribution loops. The ML Engines are delivering (fully automated) optimized setpoints for the high impact machines. Many real-world cases are delivering energy savings in the range of 15-20% in Combined Cooling Capacity of Chiller Plant up to 3.000 tons.

• Primary chilled water pump speed
• Chilled water differential pressure setpoint
• Condenser water pump speed
• Condenser water supply setpoint
• Cooling tower fan speed
• Chilled water setpoint 

At larger (> 3.000 Tons) like Data Centres, Industrial and District Cooling plants it could include to optimized Chiller Staging and Thermal Energy Storage with savings raising up 50%. With Machine Learning the advancements are typically Dynamic Sequencing, Predictive Free Cooling, Chiller Diagnostics and real-time asset condition monitoring including predictive failures on the Electrical assets like Air Blowers and Pumps.

The next target is HEATING

ML engines that learn how boilers systems perform and uses this trained data to improve overall plant efficiency

Now we can get Cooling optimized the next target would obviously be to Optimize HEAT that learns how boilers perform in a variety of operating conditions. Demand-based relational control algorithms optimize hot water and steam systems, develop optimal equipment combinations and sequences for maximizing system efficiency, and route these recommendations to the BMS or Building Automation System (BAS). Staging boilers to operate at their “sweet spot” for optimum performance, and prolonging the life of plant equipment. Typical benefits are;

• Automatically, continuously optimizes boiler systems in the most holistic, intelligent manner
• Delivers consistent savings across industries, settings and control systems
• Streamlines operations and lengthens equipment life
• Data-driven HVAC optimization that learns and adapts over time significantly lowers operating expenses while increasing safety and reliably
• Sustainably reduces energy use by up to 50%

Complete the optimization with Ventilation to a full SMART HVAC

Demand-based relational control algorithms automatically adjust airflow to deliver precise output while using the least amount of fan power, chilled water, and heating energy to meet temperature, humidity and airflow requirements. In contrast to conventional methods that operate systems to static temperatures and pressures—or optimize based on “worst-case” zones — Optimizing AIR reads the actually required airflow and holistically regulates energy exchange across the chiller plant, boiler plant, and AHU (Air Handling Unit) system in real-time. The ML relational, systems-approach ensures energy use and savings are always optimal while safeguarding space requirements/occupant experience. Typical benefits are:

• improving occupant comfort
• Extending equipment life
• Lowering operating / maintenance expenses
• Reduction of energy use by up to 40%
• Reduction of water use
• Adapts and responds to real-time building loads and changing ambient and occupancy conditions

Think BIG, START small and scale FAST

No need to implement all ML engines at the same time, start small with one of the three and add-on/upgrade/scale whenever the first savings are delivered. And yes, the EMPP Machine Learning engine is reusing the data from existing Buildings Management Systems BMS / SCADA or MES that can stay in operation as is with NO need for new controls. A secure API or cloud-based human interface with customizable dashboards would report display meticulous, real-time chiller plant performance and operational insights including enterprise views that summarize information from a single or even multiple chiller plants in case of an across building portfolio. This to ensure propper Measurement and Verification will take place from the start the ML engine is delivering its new setpoints to optimize the system.

While this all sounds wonderful, there are some hurdles the industry must overcome to reap the rewards of a SMART Buildings community leveraging ML/AI to its full potential and pretty quickly we can all agree that education is one of the biggest barriers to adoption, Let's do that TOGETHER!

What is your definition of a SMART Commercial Building?