Project Title
Measurement and improvement of energy management for manufacturing processes.
2019 - Current
Verdigris' AI-based submeters at Katerra's Factory in Tracy, CA: 1) Automatically labeled the factory's load types with 98.9% accuracy 2) Detected and alerted facility teams to energy usage anomalies 3) Detected unbalanced phases and compressor short-cycling

Verdigris’ smart energy meters showcased ease of installation, instant and seamless data access, auto-classification of end uses, and valuable insights on equipment health and energy efficiency in Katerra’s factory in Tracy, California.

What does advanced manufacturing look like? Smart robots, brilliant people, highly automated assembly lines, self-guided vehicles? All these are on display at Katerra’s factory located in Tracy, CA, along with an AI-enhanced smart energy metering system.

Large manufacturing facilities can consume significant amounts of energy–often thousands of megawatt-hours (MWH) every month– and have untapped opportunities to reduce energy use through the application of smart metering technology. For most facilities, the best insight they have into their operations is a bill from utility companies which can make for a difficult time assessing the efficiency of specific factory equipment.


Katerra, a technology company optimizing every aspect of building design, materials supply, and construction, has a number of domestic and international offices, factories and building projects. Founded in 2015, Katerra was recently ranked by NMHC as the fifth largest builder in the U.S.

Verdigris installed its AI-enhanced submeters at Katerra’s factory in Tracy, CA. At 577,000 square feet, the factory is an advanced manufacturing facility with automated production lines for wood frame walls, floors, roof truss lines, robotics-enabled cabinet, finish and window lines, and cold-formed steel. Katerra’s Tracy factory began operations last fall and when it reaches full production with three shifts, the factory will employ more than 500 workers who will operate some of the most advanced manufacturing equipment available in the industry. At full capacity, the factory will be able to produce, on an annual basis, the equivalent of 12,500 multifamily units. The factory also features a recently installed 1.117 megawatt (MW) solar array to offset a majority of its energy use. With clean onsite generation and energy submetering, Katerra’s Tracy factory is positioned to deliver high-quality, low-carbon building components with even greater operational excellence.


Customer Challenge

With a goal to deliver building products with improved environmental performance, a reduced carbon footprint, and do so without adding costs, Katerra sought to find metering solutions to help them measure the energy impact and carbon footprint across all their construction and manufacturing processes. The features of an energy management system that were paramount to Katerra included seamless data and easy installation and integration into their existing systems.

The Verdigris Advantage

Verdigris, an industry leading AI-enhanced submetering solution provider, installed a total of six Verdigris meters and 216 smart sensors on six electrical panels and 72 circuit breakers at Katerra’s Tracy factory.

Fast Data Access

Data was live on Verdigris’ dashboard within 24 hours after installation, including highly granular data of energy demand down to each phase of each circuit.


Ease of Installation

For the six panels at Katerra’s Tracy site, the total installation time was 8.7 hours, averaging under 1.5 hours for each panel. For the same number of panels and data points, a typical submetering installation time could be up to five times higher.

Smart End-Use Classification

Of the validation set, Verdigris AI automatically “blind labeled” of all the load types. 98.9 percent of energy was labeled accurately.

One of Verdigris’ most advanced AI features is end-use classification. With this feature, Verdigris automatically recognizes the different types of equipment to which Verdigris’ sensors are connected, based on the highly granular energy data they collect.

At Katerra’s factory, Verdigris’ AI identified several key end-use categories of interest: lighting, HVAC, compressor, dust collector, large automation, plug loads, and other process equipment by analyzing the unique waveform for each load type.

An example of how Verdigris AI distinguishes between compressors and dust collectors is shown below. Verdigris samples energy data at very high frequency (8,000 times per second), enabling Verdigris to tell the two equipment types apart by their waveforms.

To develop the Verdigris AI, a total of 21 days of energy data plus external inputs (like weather data) were fed into Verdigris’ algorithm. Fourteen days of energy data acted as the training set, and the remaining seven days as the validation set.


Verdigris Intelligent Alerts: Energy Usage Anomalies

Verdigris’ technology is able to identify energy anomalies automatically. By looking at the past three months of historical energy usage for every circuit in a building, Verdigris’ AI can construct an expected region of energy usage with a 95 percent degree of confidence. Customers will be alerted if energy usage falls outside of this expected region. This feature offers our customer an immediate payback. In one instance, Verdigris detected abnormally high demand at night at Katerra’s factory. It was traced back to some of the dust collectors running overnight. The cost for this was estimated to be a couple hundred dollars per incident. If this was left undetected every night, the energy costs for this equipment alone would be significantly higher. With the Verdigris intelligent alert, the facility team was able to take quick action to correct otherwise wasted energy usage. 


Verdigris Intelligent Alerts: Safeguarding Equipment

Phase imbalance

For motor loads, phase imbalance can cause reduced output horsepower, overheating, and even shut down of the equipment. So, for facility managers at sites with large energy loads, being able to proactively identify this problem is critical to avoiding equipment downtime, which could lead to financial losses and energy waste. Verdigris AI technology also detected phase-imbalance at several breakers for Katerra’s Tracy factory, which the team then fixed before problems arose. 


Compressor short-cycling

Short cycling occurs when a mechanical system that operates in cycles turns on and off more frequently than expected. Short cycling can occur in air compressors as well as other equipment such as chillers, AC units, fans, and pumps. This rapid state-change adds unnecessary stress on the equipment, reducing its lifespan and efficiency. For any efficient manufacturing operation, being able to identify short-cycling has a profound impact on the total productivity and throughput of the factory. Issues like short-cycling are not easily identified, especially among millions of pieces of equipment running different processes at a big manufacturing factory. Identifying short-cycling was only possible with Verdigris high-granular energy data. Verdigris is like the “ECG” (Electrocardiography) scan, listening to the heartbeat of each equipment, and providing a comprehensive diagnosis at a spectacular granularity. For Katerra, Verdigris’ AI identified a potential short-cycling compressor, which, as confirmed by the facility team, uncovered a number of minor leakages.

Contact us today to make your buildings smart:

Have questions? Check out our FAQs in the Help Center.