The rise of the Internet of Things brings many benefits as it means a more connected world. On the flipside, the uptake of a hyper-connected automated workforce may be a cause for concern for workers who fear the rise of the machines.
It’s a concern that’s not entirely unmerited: at Amazon, the number of robot workers brought online during 2016 was up 50 percent compared to a 46 percent growth in human workers.
The Industrial Internet of Things (IIoT) is on the rise too, with connected tech implemented across multiple areas of work. One of the biggest rollouts has been the application of predictive maintenance regimes.
As the name suggests, predictive maintenance regimes work by making forecasts based on algorithms. The cleaner the provided data, the more accurate the projections. Predictive maintenance is most useful in industrial settings, where it can lead to cost savings in upwards of the millions – though there is a caveat: the quality of the data is key.
This is where connected machinery truly comes to the fore – with no threat to the human workforce. Sensors are one of the most important parts of any predictive maintenance regime and while most machines manufactured in the last five years will come with sensors as standard, older machinery isn’t usually equipped with requisite sensor technology.
In most cases, sensors can be retro-fitted to the machinery to track the often minute changes that occur when components are degrading. Adding sensors doesn’t have to be a prohibitively expensive outlay for most businesses (it’s certainly less expensive than replacing the machinery). The process of installing the sensors can usually be completed within a week or so with minimum interruption to the production schedule.
But what exactly will the sensors be monitoring – and why are they so useful? Industrial predictive maintenance has four main types of sensors, which we’re explaining below.
Vibration sensors are used on machinery with moving parts like motors, fans, pumps, engines, turbines, conveyors and tools with rotational elements. The rotating elements of these machines emit vibrations at a specific frequency. Sensors can pick up on any anomalies in this frequency and alert the system, which can assess the danger. Among the faults that can be picked up by vibration sensors include:
In Sweden, the national railway system has been modernised with the rollout of a maintenance scheme built around vibrating sensors.
SFK, the company who oversaw the upgrade, installed a powerful blend of bearing technology and cutting edge wireless communications. As well as bearing data, the system analyses information on speed and positional data using GPS.
The data is stored in the cloud and gives rail operators the information they need to run a cost-efficient predictive maintenance programme.
As the name suggests, temperature sensors monitor the heat of the machinery. Heat monitoring is extremely prevalent in food production facilities, but it’s also very common in industrial settings.
Temperature sensors can apply to cooling tanks, kilns, industrial ovens and even conveyor lines. Again, like vibration sensors, the data collected by temperature sensors will flag any overheating or overcooling components so they can be assessed by the predictive maintenance system.
While the measurements of temperature sensors assess whether an element is at the right temperature, it can also alert the existence of:
While many of the applications are in the industrial world, vineyards are at the forefront of a pilot around the benefits of temperature sensors.
Winemaker Henry of Pelham has a mobile weather station and two sub-stations onsite at its 300-acre vineyard. The climate can vary widely within a single vineyard, making constant monitoring a must.
Henry of Pelham intends to use sensors to increase the accuracy, volume, and timeliness of its weather data – which is vital to harvesting grapes for premium wine products. Each IoT sensor is equipped with GPS and is super-portable so it can be used to fill the gaps that can’t be covered by the weather station and sub-stations.
The sensors can be rolled out year-round – which means they have serious potential to bear fruit for the winemaker.
Perhaps the latest technology in the world of sensors, acoustic sensors can detect anomalies based on how the machinery sounds. While the uptake of the technology is relatively new, Amnon Shenfield of 3D Signals says that technicians have always used sound to monitor the health of their machines.
"All the various technicians understand what is going on with their machines when they walk into a production floor and just listen,” he says. “All the pumps and electrical motors and valves and turbines, generators and compressors and so on, they are all very acoustic. When you walk around them, you get a sense, as with your car, whether they sound normal or not."
Acoustic sensors are particularly valuable when monitoring:
Again, the potential applications of the tech are vast: optical internet cables combined with acoustic sensing could be used to detect earthquakes while the Scripps Institution of Oceanography deployed acoustic monitoring equipment to study how marine life reacts to ocean noise.
60 to 70 percent of the electrical energy used in industrial settings is consumed by running electric motors. Because of this, there is huge value in current sensors – which can monitor energy usage as well as energy surges.
A machine using more energy than normal can be doing so for many reasons including:
Regardless of what you want to measure, sensors are a core component of any predictive maintenance regime. While relatively inexpensive and easily installed, sensors pack serious power in an industrial environment.
Whatever sensor your machines need, Statwolf's expertise can help you find the best solution for you and your business.
Our team of data scientists have worked on predictive maintenance programmes around the world so can help you with any queries you might have.
Want to make sense of your data? Download our comprehensive guide: The Predictive Maintenance Cookbook.