Industrial pump failures — What causes them and how to predict them
What is an industrial pump?
An industrial pump is generally a heavy-duty process pump which is used to force different types of products, including water, chemicals, fuel, waste water, sludge or slurry. The most common pumps used for industrial purposes are centrifugal pumps and positive displacement pumps.
What causes industrial pump failure?
It is very easy to say nearly all the faults lie with user error. But there are many things that can go wrong. Often, centrifugal and positive-displacement pumps are treated the same, so the fixes will actually cause problems over time. Pumps will at times operate at improper speeds, be misaligned, or simply be the wrong size for the environment and there can always be contamination in the lubricant.
Refer to the flowchart below to see common root causes for industrial pump failure, and how each problem eventually impacts the equipment, environment, and business.
The U.S. industrial pump market size was estimated at $7.62 billion in 2019. It is expected to grow at a CAGR of 1.7% from 2020 to 2027 due to the rise in crude oil production in both offshore and onshore oil fields. That means in America alone, with its roughly 150 large refineries, there are over half a billion dollars invested in pumps in the oil and gas industry. It also lends credence to the idea that the health of the oil and gas industry depends on its equipment to run efficiently. A study revealed that, an average oil and gas company goes through at least 27 days of unplanned downtime each year, resulting in huge loss of approximately $38 million.
The general practice is for most pumps to be left in place until it breaks, maximizing its life value. One of the most commonly used artificial lift methods is the electrical submersible pump (ESP). Using this method, a pump can handle large volumes of hydrocarbons and work under many conditions in both onshore and offshore reservoirs. B. Even though the ESP is extensively used by oil producers, it still suffers from many failures due to mechanical, electrical, and operational problems associated with its downhole assembly. While a pump failure in a refinery may only affect one part of a process, pump failures in an oil field can shut down a well or pipeline. It becomes very important to identify the reasons behind ESP failures and come up with preventive measures to rapidly and effectively avoid and mitigate these failures to avoid further damage and costs, with improved rig-personnel and operational safety.
How does predictive maintenance work for industrial pump failures?
To truly avoid the headaches, you need to predict the failures instead of fixing them. For that, with any great levels of accuracy, a business would need to employ a predictive maintenance solution driven by AI, machine learning, and advanced analytics. And along the digital transformation journey, be sure to adopt a wireless sensor and warning system! After all, these machines are often at a remote location and take time to get to.
Regardless of the solution, you end up going with, you need to have sensors in place to monitor vibrations, seals, cavitations, and any other necessary warning signs (dependent on your pump and infrastructure). Without accurate data points, whether you are running a digital and wireless predictive maintenance solution or a manual one, it’ll always be guesswork. But, with IoT and AI technologies, it becomes feasible to collect and monitor vast quantities of data in real-time. Machine learning algorithms and advanced analytics can help in tracking wear and tear of an equipment to predict potential failures and recommend ways to fix such anomalies with predictive maintenance, helping optimize production and decrease risk of costly breakdowns.
Neal Analytics has extensive expertise in the field of artificial intelligence (AI) and predictive maintenance. Learn how Neal worked with a multi-national oil & gas company to identify and solve issues with pump jack failures using predictive maintenance. Contact us to get started!