A Multidomain Pump Degradation Index Integrating Corrosion, Vibration and Hydraulic Performance in Saline-Service Centrifugal Pumps

In the vast, salt-sprayed world of offshore oil and gas, seawater is both a necessity and a relentless enemy. It cools machinery, fights fires, and moves ballast, but its high salt content acts as a slow, corrosive acid that eats away at metal. For the massive centrifugal pumps that keep these operations running, this environment creates a triple threat. First, the metal itself slowly degrades, losing mass and strength through corrosion. Second, as the metal wears, the pump's moving parts can become unbalanced, causing the machine to shake and vibrate more violently. Third, these physical changes disrupt the smooth flow of water, making the pump less efficient at its job. For decades, engineers have monitored these problems separately, checking vibration levels to spot mechanical trouble and measuring flow rates to gauge performance. However, looking at just one of these signs is like trying to understand a storm by only watching the rain, ignoring the wind and the lightning. To truly know the health of a pump, one needs a way to see all three threats at once.

A researcher has developed a new method to do exactly that, creating a single, simple number that tells the whole story of a pump's decline. This approach, called the Pump Degradation Index, was tested using data from a controlled experiment where a stainless-steel pump was run for 200 hours in synthetic seawater. During this time, the pump was not just watched; it was measured in three distinct ways. Scientists tracked how much metal was lost from test coupons placed in the water, measured the speed of the pump's vibrations, and calculated how much its efficiency dropped as it struggled to push water. The pump started in perfect condition, but as the hours ticked by, the metal loss increased, the shaking grew from a gentle hum to a significant shudder, and the efficiency fell from a healthy 70.6 percent down to 58.0 percent.

The challenge was that these three measurements speak different languages. Metal loss is measured in milligrams, vibration in millimeters per second, and efficiency as a percentage. To combine them, the researcher converted each of these changing values into a common, unitless scale that runs from zero to one. Zero represented the pump's brand-new state, while one represented the worst condition observed during the experiment. By giving equal importance to the metal loss, the vibration, and the efficiency loss, the researcher blended these three separate streams of data into a single, smooth line. This line, the Pump Degradation Index, started at zero and climbed steadily upward, reaching 0.301 after 50 hours, 0.542 after 100 hours, and finally hitting 1.000 at the 200-hour mark.

What makes this finding particularly useful is its stability. The researcher tested whether changing the importance of each factor—perhaps caring more about the metal corrosion than the vibration, or vice versa—would drastically change the result. They ran thousands of different combinations of weights to see if the overall picture would shift. The result was surprisingly consistent. No matter how the importance of the three factors was adjusted, the index followed the same steady path of decline. This suggests that the method is robust; it does not rely on a guess about which problem is the most critical, but rather captures the general, undeniable trend of the machine getting worse over time.

However, the researcher is careful to define what this number is not. The value of 1.000 at the end of the test does not mean the pump has failed or is about to break. It simply means the pump has reached the worst condition that was observed in this specific 200-hour laboratory window. Because the test was conducted in a controlled lab setting with a single pump, the results cannot yet be used to predict exactly when a pump in the real world will fail or to calculate the probability of a breakdown. The method is a tool for spotting the direction of health, not a crystal ball for predicting the future. It serves as a clear, transparent warning system that can tell maintenance teams when a pump is moving from a healthy state into a state of advanced wear, prompting them to look closer with more detailed tools.

This approach offers a new way to think about machinery health, moving away from complex, data-heavy models that require massive amounts of historical failure records. Instead, it provides a straightforward, easy-to-understand score that combines the physical wear of the metal, the dynamic shaking of the machine, and the loss of its pumping power. While it needs further testing on many different pumps in real offshore environments to prove it works everywhere, it lays the groundwork for a simpler, more integrated way to keep vital equipment running safely. It turns a confusing mix of metal loss, shaking, and slowing flow into a single, clear story of degradation, allowing engineers to see the full picture of a pump's life in a single glance.