Air quality is an issue that manufacturers can’t afford to ignore. Poor indoor air quality (IAQ) in manufacturing facilities is linked to lower productivity, poor morale, increased turnover and negative health impacts for employees. And, of course, manufacturers who do not meet OSHA requirements for IAQ are subject to serious fines and penalties.
However, selection of air quality equipment is rarely given the same attention as the selection of production equipment. Manufacturers typically spend a great deal of time evaluating production equipment on metrics such as product quality, speed and efficiency, reliability and total cost of ownership. When it comes to selecting air quality equipment, often the decision comes down to simple cost.
Many manufacturers make the mistake of thinking “a dust collector is a dust collector.” If they are essentially just big vacuum machines, why invest in extra bells and whistles?
Today’s best dust collectors are more sophisticated than one might think – and the differences matter. Smart control systems that enable dust collectors to sense and respond to their external environment and internal conditions are revolutionizing the air quality industry.
These intelligent systems can help manufacturers get more out of their air quality equipment by improving energy efficiency, extending filter life, optimizing maintenance schedules and reducing total cost of ownership.
Automated systems can turn a dust collector on and off, so that they only run when equipment, such as a laser cutter, is on.
Simple forms of dust collector automation have been around for some time now. These include:
- Systems that turn the dust collector on and off in conjunction with production equipment, such as a laser cutter or weld torch, so it only runs when the equipment is on.
- Safety systems that automatically stop the collector if smoke or excess heat are detected inside the filter cabinet.
- Filter pulsing systems that pulse dust off the surface of the filters at set intervals during operation or at the end of a shift when the collector is powered down.
Intelligent control systems like RoboVent’s eTell take this kind of automation several steps further. These systems actively monitor system performance and environmental conditions and make decisions based on machine learning algorithms. These algorithms detect patterns and adjust system performance to optimize energy use, filter life and maintenance requirements.
Air quality control for manufacturing is getting smarter and more efficient – thanks to machine learning and the Internet of Things.
Smart control systems
Traditionally, dust collectors have been simple machines: turn them on and they suck in dirty air through filters. Turn them off and they stop. Operators are in control of whether they are on or off, and the machines themselves do not need to know anything about their environment or their internal operations.
Smart control systems for dust collectors, however, bring together the Internet of Things (IoT) and artificial intelligence (AI) technologies to create systems that can detect – and respond to – their environments. A smart control system has three main elements:
The first element is a sensor system that enables the machine to detect aspects of its environment. A dust collector may have sensors that tell it when a weld torch or laser is on or off; air quality monitors that detect particulate levels in a hood or enclosure or in the ambient air; and internal sensors that monitor internal pressure drop, temperature, internal particulate levels, energy use or other factors.
The second element is a way for the dust collector to connect and communicate with other devices and with the operators. These communication technologies can be hard-wired or based on WiFi or Bluetooth technologies. Collectors may be connected to an internal network or through a secure cloud application that enables anywhere/anytime visibility and control for users.
The third element is an analytical system that acts as the brains of the dust collector, allowing it to make decisions based on user input or sensor data. These actions may include turning the collector on and off, sounding an alarm or adjusting blower speed in response to real-time conditions.
Sensors, such as this air quality monitor, feed data into smart air quality control systems.
Smart learning
These smart control systems are part of a wider trend in manufacturing: The Industrial Internet of Things (IIoT), or Industry 4.0. This concept brings together two trends that are also transforming the consumer market.
First, the IoT, epitomized by consumer products such as the Nest suite of home electronics and home control systems like Amazon Alexa or Google Home, connects devices such as lighting systems, thermostats, televisions and sound systems directly to the internet. This enables cloud-based communication and control through a mobile app or a computer dashboard. The IIoT extends this technology into the manufacturing environment.
Second, machine learning and predictive analytics are forms of AI that allow software programs to learn from user or sensor inputs and respond to changing needs or environmental conditions. Nest’s intelligent thermostat, for example, learns from user behavior over time so that it is able to anticipate user preferences and make independent decisions to save energy or increase user comfort.
IIoT technologies can be categorized into two broad categories:
- Those that augment human capabilities by providing operators with information and decision-making support.
- Those that automate systems by making decisions on their own.
Smart dust collector control systems may use both types of technologies to reduce the burden of air quality control for the humans in the manufacturing facility.
Automated controls
Smart control systems are responsive to real-time conditions. When connected to internal and external sensors, a smart dust collector can automate a variety of air quality decisions to optimize system performance. For example, it may be able to:
- Cycle on and off in response to real-time airborne particulate monitoring sensors.
- Dynamically adapt blower speeds to compensate for filter loading (as measured by internal pressure drop).
- Adapt dynamic filter pulsing systems to particulate levels to optimize system performance.
Using predictive analytics and machine learning, smart control systems can make predictions about future conditions and adjust system performance based on those predictions. For example, the dust collector could calculate energy use and adjust system settings to maximize energy efficiency based on actual use behaviors. Using machine learning, smart dust collectors could:
- Predict times during the shift when production is heaviest and ramp up the system in anticipation to avoid a spike in airborne particulates.
- Control system settings for multiple dust collectors to adapt when one collector is down for maintenance or repair.
- Optimize the pulsing schedule based on actual patterns of use.
These predictive algorithms save money by reducing energy consumption, maximizing filter life and reducing the risk of system downtime for emergency maintenance. They also shift the decision-making burden from operators to the machines themselves, saving operator time while maximizing system performance.
Preventative vs. predictive
Maintenance is another area that will be transformed by machine learning. Smart dust collector control systems can predict maintenance needs based on actual use patterns, allowing companies to move from standardized preventative maintenance schedules to predictive maintenance schedules.
That means instead of simply changing the filter because “it’s time” or waiting until a sensor indicates a filter has failed, operators will know exactly how much life is left in the filter and be able to schedule filter changes and other predicted maintenance needs at the most convenient time.
Smart dust collectors can even diagnose performance problems and notify maintenance technicians if attention is needed. Predictive maintenance allows companies to:
- Build maintenance schedules based on actual equipment use, performance and needs instead of general rules of thumb.
- Find the right balance between extending the use of each filter and avoiding high energy costs associated with trying to blow air through a heavily loaded filter.
- Optimize maintenance schedules to reduce downtime.
- Avoid unexpected surprises by responding to maintenance needs identified by system sensors and predictive analytics before they become repair emergencies.
Next stages
As both production and environmental control systems become smarter and more automated, the next stage of smart factory evolution will likely involve further integration between systems so they can work together to improve productivity, comfort and operating costs. Soon these scenarios may be possible:
- The dust collector and HVAC system work together to optimize air movement patterns in the facility to increase air quality system performance, reduce overall energy use and maximize comfort levels.
- The HVAC system notices that the post-lunch productivity slump can be combatted with a brief blast of cooler air and programs itself accordingly.
- The lighting system automatically adjusts itself to optimize light levels for certain tasks in response to patterns of employee activity or equipment use.
These scenarios will be real possibilities in the future as sensor data, analytics and control systems continue to advance. While important challenges remain to be solved for system interoperability and security, the core technologies needed to make these possibilities a reality already exist.
In the meantime, Industry 4.0 technologies are already delivering big benefits for the companies that take advantage of them. Manufacturers investing in new air quality equipment should look for smart technologies that will move their facilities into the future of environmental controls.
