109166505 | Theerapong28 | Dreamstime
658215cd092d23001eb7a846 Dreamstime M 109166505

Technology solutions for energy monitoring on the plant floor

Dec. 27, 2023
Step-by-step, how data can impact decisions across an entire organization, from floor technicians and supervisors to administration or C-level decision-makers to facility management staff.

As the saying goes, you can’t manage what you don’t measure. But with all the demands that today’s manufacturing professionals face, is monitoring the energy usage of machines worth the time and effort?

Let’s look at a few scenarios:

Plant floor technician: A technician has been replacing one of the parts on a machine every four months, even though that part is supposed to last a year. He suspects surges might be the cause, but he doesn’t see any visibly damaged boards. How can he get to the bottom of the issue?

Plant floor supervisor: A production supervisor is looking at the output numbers for the entire process of a particular line: uptime, production values, and output numbers. Is it more efficient to let a machine go into a lower power mode when a shift isn’t running, or is it better to power machines on and off during those times?

Administration or C-level decision-maker: A plant director is looking at the overall cost of running the operation. How is energy usage impacting the bottom line?

Facility management staff: A building automation team has been charged with reducing the overall energy usage at the facility to help meet corporate sustainability directives. Where do they start?

See also: How manufacturers can cut complexity with integration technology

One key to solving these dilemmas is energy monitoring. By collecting energy usage data from machines and pulling it into a system where it can be analyzed, manufacturers can make more informed decisions. Employees throughout a facility can access the data needed for their specific roles to make decisions that will improve operations and sustainability and increase profitability.

These are the steps to implementing an energy monitoring plan and will highlight some use cases based on Phoenix Contact’s experience with energy monitoring.

Step 1: Monitoring

The first step is to start monitoring. At the beginning of its own energy monitoring plan, Phoenix Contact identified several machines in its U.S. production facility that consume higher amounts of energy: the wave solder machine, the Varioface Professional (VIP) slice machine, and the Surface-Mount Technology (SMT) machine. Collecting the energy data can be simple, but different machines require different approaches.

Basic energy meters can measure and collect system conditions like voltage, current, and power. Energy meters can measure AC power in single-phase and three-phase installations. Other sensors can capture key parameters such as pressure, temperature, and flow. While the energy meters and sensors can’t perform the analytics themselves, more advanced meters can send the data directly to the cloud, so it’s accessible for other uses.

The wave solder and VIP slice machines only require measuring power at a single point of delivery. Here, Phoenix Contact uses a single channel energy meter. The energy meter directly measures voltage and uses a current transformer to measure how much current is flowing through the machine. Based on these measurements, it calculates how much power is being consumed. The energy meter is then polled by an internal microarchitecture application via a REST API.

See also: Making the business case for PLM to SaaS deployment

For more complex machines or full production lines, a programmable logic controller (PLC) offers a higher level of data collection. The PLC acts as the “brain” of the system. Rather than just gathering the data, the PLC can preprocess the data with scalers, multipliers, and more. It can then send the data to an internal server or external cloud service, such as Amazon Web Services or Microsoft Azure, for use later.

The SMT machine is more complicated than the other two machines. Phoenix Contact needs to measure and merge multiple circuits at different voltage levels. An open-source, Linux-based controller and power monitoring modules collect data for this machine. Because of its open nature, this edge PLC can work with hardware and software from many other vendors.

Step 2: Analyze the data

Collected data doesn’t have much impact on its own. Monitoring and collection efforts are only valuable if someone studies that data to identify insights into operations or faults. The information must be easily available to different people across the organization so that they can take the necessary actions to meet their department’s objectives.

Getting a clear picture of a machine’s energy usage requires processing that data, via human or automated means. Looking at the machine’s historical consumption, utilization, and throughput/yield will establish a baseline. From here, you can define performance metrics or identify anomalies.

To this end, Phoenix Contact has set up databases and an internal microarchitecture (a series of servers running on its own network). The historical database can store years of data samples. To help analyze the data, Phoenix Contact uses an open-source building automation software based on the Niagara Framework. The software helps create a “digital twin,” a digital representation of what’s happening on-premises. A graphical interface makes it easier for the human brain to process the data.

See also: Industrial Applications of Generative AI: A Smart Industry eBook

Within the software, the team can create different user profiles to enable different levels of access. People in different roles can look at the visualizations most relevant to their jobs. They can calculate important information like power, electrical characteristics, power factor, and reactance values.

In the near future, an automated software program, utilizing machine learning (ML), neural networks, or other artificial intelligence (AI), will pore over and analyze the data. The machine will have access to years’ worth of data, helping to train it in the proper data sets. From there it could analyze the data, notify key people about anomalies, and determine the next logical step to improve functionality and cost-benefit ratios.

While Phoenix Contact prepares to implement ML or AI, it uses an MQTT broker for important notifications. Any software on the network can subscribe to the energy monitoring system. An internal application can pull data from the machines to notify team members about anomalies so that they can take action.

Step 3: Take action

Once you’ve ensured that collected data is being monitored and analyzed, what can you do with it? Let’s go back to the scenarios discussed earlier.

Plant-floor technician: From a diagnostics standpoint, the technician is who suspects a power surge but doesn’t see a visibly damaged board is trying to solve the problem directly at the machine. But without energy data, any theories he might form are just pure speculation. In the absence of a visibly damaged board, the problem could be due to any number of things—a surge or something else.

With data in hand, the technician can analyze historic voltage values and compare them to current data to pinpoint the root cause. He can take the actions necessary to repair the machine, and from there, measure how that part impacts the machine’s shelf life. This information can further be used to implement a predictive or preventive maintenance plan that reduces downtime.

See also: How manufacturers can break data silos through DOP-enabled technology roadmaps

Plant floor supervisor: Without data, the supervisor trying to determine whether it would be better to shift a machine into a lower power mode or shut it off between shifts has no useful basis for making his decision. In one such case, a manufacturing supervisor at Phoenix Contact analyzed these data points and found that the energy monitoring clearly shows when a new shift starts: The power consumption shoots up and stays consistent throughout the shift.

Reviewing the data, the supervisor can see that the baseline power usage, or standby current power, for a particular machine is high. Even when that machine is not producing anything, it consumes about 5 kilowatts of power. This data makes it possible to evaluate whether it would be more energy- and cost-efficient to put the machine into sleep mode or to shut the machine down entirely between shifts.

Administration or C-level decision-maker: With data to analyze, the plant director trying to determine how energy usage is impacting the bottom line can now compare the energy usage data of an existing machine to the projected energy usage of a new machine. This information can clarify whether the new machine will save enough money on energy costs to justify its purchase.

Such data could get even more granular in the future, perhaps to the product pricing level. How many kilowatt hours of energy does it take to make a single terminal block, and how does that impact where the price is set?

Facility management staff: The building automation engineer or facility manager charged with reducing energy usage to meet sustainability objectives is not interested in the fine details of any machine but in energy usage across the entire campus. The facilities team can use collected plant floor data to decrease overall energy demand and maximize efficiency.

For example, Phoenix Contact generates a significant part of the energy to meet its own electrical needs through an internal solar power system and a combined cooling, heating, and power (CCHP) system.

Energy monitoring could enable the company’s facility manager to partner with the production team in determining the best times to run specific machines to optimize for green energy usage—which could reduce the company’s electric bills. In the future, the facilities department could even use this data to bill different departments for their individual energy usage.

Walking the talk

While Phoenix Contact USA is still in the early stages of its energy monitoring journey, the company has already implemented a digital factory in Bad Pyrmont, Germany. The American team is working closely with the German team to implement the best practices already learned.

See also: How to harness operational data to meet sustainability targets

As Phoenix Contact continues to refine its own energy monitoring, employees across the entire organization will soon be able to make data-driven decisions. These decisions could have a significant impact on the bottom line, by helping to reduce downtime, plan predictive maintenance, improve resource utilization, and even determine product pricing.

But just as important as the financial impact, energy monitoring will also help Phoenix Contact reach its goal to become a carbon-neutral company by 2030. Having a clear understanding of energy usage can reduce overall energy consumption, allocate resources for cleaner energy usage, reduce emissions, and allow the company to become a better steward of the environment.

About the Author

Russ Kolacek

Russ Kolacek has been the building automation engineer at Phoenix Contact USA in Middletown, Pennsylvania, since 2019. Phoenix Contact is a manufacturer of industrial automation, interconnection, and interface solutions.