The global site of the UK's leading magazine for automation, motion engineering and power transmission
7 December, 2023

Twitter link

Taking the guesswork out of VSD maintenance

15 October, 2021

Cloud computing allows us to track and analyse VSD data accurately, and to schedule maintenance exactly when it is needed. Liam Blackshaw, product manager for LV drives at ABB Motion UK, believes this is “a game-changer”.

Today’s digital variable-speed drive (VSD) typically generates around 200GB of data every year from a series of thermal, voltage and power sensors that track ambient temperatures and load variations. However, until recently, collecting this amount of data and putting it to good use has been impractical. With the advent of cloud computing, the opportunity arises to track and analyse this data accurately. For instance, the output can determine the impact that temperature has on the ageing of a VSD’s three main components – the cooling fan, semiconductor converter module and DC link capacitors.

This is a game-changer as it has the potential to take the guesswork out of scheduling drive repair and replacement activities based on time-based, preventive maintenance. It moves maintenance to a data-driven condition-based approach that provides clarity on how components are ageing. This means that VSDs which are overloaded and at risk of failing before the scheduled replacement date, can be identified, triggering maintenance shutdowns more efficiently to minimise downtime. By contrast, those VSDs which appear to be at the end of their normal working lives might, in fact, be under very little stress and therefore have many years of useful work left.

The secret to unlocking this data lies in a combination of cloud computing, machine learning and component manufacturers’ life prediction curves, based on L10 probability. The L10 life calculates how long 90% of a group of apparently identical components will last when subjected to the same operating conditions. The L10 curves can be used to develop cloud-based algorithms that predict when a key component is likely to fail. Details from these individual components can then be used to build an algorithm to predict the life of the VSD.

A dashboard can be created that shows the typical 25-year life span of a VSD, overlaid with the actual trajectory of the VSD under its present stress conditions. It can show alarms points – such as a yellow line indicating when component replacement is due, and a red line warning when a component must be replaced to avoid failure.

This data-driven approach can predict failures up to two years ahead, giving reliability managers the confidence to book maintenance for a particular time and avoiding costly emergency maintenance in case of failure. If a component is changed, the manager can insert the time of maintenance into the portal. The lifetime plan is then re-calculated. The data can also be shown as a daily lifetime impact chart, which gives an insight into whether plant setting optimisation is needed to extend VSD component lifetime

  • To view a digital copy of the latest issue of Drives & Controls, click here.

    To visit the digital library of past issues, click here

    To subscribe to the magazine, click here



"Do you think that robots create or destroy jobs?"



Most Read Articles