What Makes Pumps Critical to Industrial Operations?
Of course, put simply, pumps facilitate the transport of fluids from one point to another. If a sophisticated manufacturing facility were to be broken down into discrete equipment, it would be all the more easy to recognise how pumps and piping play a significant role in making the entire plant operational. Consider, for instance, the benefits of a compressed air distribution network from air compressors, as well as high-pressure reverse osmosis pumps that both purify water and transport the essential raw materials to their common reaction chamber (aka reaction vessel).
A close-up of an industrial pump system. Image Credit: Pumps & Systems.
A malfunctioning pump, on the other hand, can seriously affect an ongoing industrial operation and may lead to an unplanned shutdown in worse scenarios. This is a clear reflection of the importance of IIoT-based online continuous monitoring, which the next sections explore further.
The Evolution from Reactive Maintenance to Predictive Maintenance
Particularly in chemical process industries, the approach to equipment maintenance has evolved over the past few years. Prior to this, most industry practices relied on the reactive (rather than proactive or preventive) maintenance of pumps. This meant that, typically, any troubleshooting used to be achieved only when the equipment stopped functioning entirely. And while this approach will always come with minimal upfront costs, both the operational risk and potential operational losses are often too high for any plant to manage.
To counter such problems with reactive maintenance, the affected industries eventually moved to the traditional approaches of preventive maintenance, or otherwise scheduled maintenance. In either precaution, the maintenance activities are planned on set intervals based on the historical data and equipment manufacturer’s recommendations.
An engineer prepares his tablet to digitally inspect an industrial pump system. Image Credit: Freepik.
Acknowledging the Need for Predictive Maintenance
Despite the fact that the above measures reduced (and will always have the potential to reduce) unplanned downtime, there was still no guarantee that a problem wouldn’t arise in-between the maintenance intervals. Moreover, the repeated start-ups after planned shutdowns would make the pumps highly vulnerable. Any missed SOP (standard operating procedure) could lead to poor lubrication and mispositioned valves, as well as other pitfalls that could result in the immediate loss of the plant’s critical assets.
Now, however, with better access to advanced, digitised infrastructures, industries are shifting to proactive maintenance or predictive maintenance (which were preceded by the periodic monitoring of pumps and other equipment by the area operators). What’s more, thanks to economical sensors now being available commercially, today it is possible to maintain the real-time, online continuous monitoring of several important physical parameters—to altogether evaluate the health of pumps and other plant machinery. (Such parameters relevant to pumps can be vibration analysis, discharge pressure, temperature, and speed.)
While the upfront cost of predictive maintenance may be much higher than its reactive counterpart, the long-term benefits, such as reduced operational losses and improved safety, outweigh the capital investment. Also, unlike the now-outdated periodic monitoring approach, continuous monitoring with IIoT helps engineers to identify and respond to the issue within a matter of hours (if not, minutes).
A close-up of two industrial pumps and their pressure readings. Image Credit: Great Industrial Pumps.
Essential Requirements for IIoT-based Pump Monitoring
Understanding the potential impact of predictive maintenance leads to an important question: how can engineers equip themselves for IIoT-based continuous monitoring for their industrial pumps?
To start with, there are three fundamental resources required:
Sensors to collect physical data about various process parameters, such as pressure, temperature, speed, and vibrational analysis
Cloud computing applications to allow both remote access to the required data (such as those mentioned above) and a secure means of data storage (i.e. historical records)
Data analytics technology that utilises machine learning to gain valuable insights from the data
With these three essentials in mind, it’s notable that the real advantage of an IIoT-based monitoring system lies in its ability to drive more valuable insights—based on combined data from various sources, rather than standalone input. For example, consider the fact that monitoring the vibration pattern of pumps may not be sufficient in detecting a potential issue in itself. Yet, by combining that data with the pump’s speed measurements and comparing the results with the historical record, the overall readings will help engineers to identify operational abnormalities. They may even be quick to reach the root cause.
Ultimately, therefore, the industrial internet of things-based approach to online continuous predictive maintenance is much more than generating alarms in case of malfunctioning: it altogether enables engineers to take timely action when addressing the primary source of the problem.