Exceptional digital control has been a feature of ABB high-power rectifiers since the 1980s – it is essential, for example, in managing high-performance thyristor-based AC-to-DC current conversion. Today, ABB is responding to customer requests for predictive maintenance of rectifier systems by deploying AI monitoring-and-response capabilities. An intuitive suite of digital offerings helps operators of electrical equipment in critical industries such as metals production to optimally run and troubleshoot their power assets. 

“In workshops with customers,” says Jonas Dänzer, ABB’s Product Manager for Digital Solutions for HPR, “they expressed a need for condition monitoring and predictive analytics, so we began working on a digital platform that would deliver that capability.” On the journey, the team discovered that customers’ needs extended beyond condition monitoring and predictive analytics. In exploring these requirements, they uncovered additional functionality that could address several customer pain points – and many of these turned out to be relatively low‑hanging fruit. 

Electric light-bulb moments

ABB’s first ​​HPR digital offering is a “co-pilot”, in the style of AI language tools such as ChatGPT or Microsoft Copilot, which presents HPR operators and maintenance teams with analysis and assistance much more quickly than they can achieve by combing through systems documentation. That is, if they have an issue with the equipment, they need only prompt the ABB Ability™ ConSys Connect solution, which will present relevant information within seconds. Issues can therefore be resolved much faster, reducing or completely avoiding downtime in sensitive situations – such as ensuring continuous power to an aluminium-smelting pot line. 

In the event of a system alarm, ConSys Connect also sources the information relevant to that aspect of the system – to give operators an instant, enriched view of the potential problem – and provides tailored instructions on how to resolve the issue. 

Dänzer says, “When we first tried out the solution at a plant using high-power rectifiers to make graphite anodes for steel industry arc furnaces, operators found they could respond more quickly to issues, which helped keep production running smoothly.”

Quick, smart response module

Another quick-win capability of the solution is built on the foundation provided by ABB AC 800PEC controller, and pushes high-speed data triggered by a system outage or issue into the cloud so that it is accessible to the customer as well as an ABB remote support team. Response time is dramatically enhanced, to more rapidly restore or stabilize power flow and maintain continuous operations. 

The critical component of this AC 800PEC capability is an ​​​​integrated transient recorder, which is unique to ABB controllers, “It’s not available with any other control platform in the HPR industry,” says Dänzer.​​​​

Although customers can request and quickly construct dashboards from within the solution, the emphasis is on ease of use and speed of response. 

”We are thinking one step ahead,” says Dänzer. “We don’t see a future in building a lot of dashboards, but rather in language recognition.” Using keyboard or voice, the user can simply prompt ConSys Connect for the information they want, and it will bring up a temporary dashboard displaying relevant data.  

A data collection scheme to build clarity

One challenge to the development of the ultimate condition monitoring and predictive maintenance software for HPRs has been that the installed base is small. Dänzer compares the hundreds of installed high-power rectifiers to the billions of cars on the road.  

He says that we understand the concept of condition monitoring and predictive maintenance “mainly from our experience in automotive”. For the past two decades or so cars have alerted us when they need service, instead of car owners having to adhere to a regular service schedule. The data that makes this possible was collected from the performance indicators of vast numbers of vehicles. 

Not only is the installed base of high-power rectifiers small in comparison, but, “HPR systems are all individual to the customer, so they all behave differently, which makes identifying common patterns much harder,” says Dänzer. 

Another barrier has been the scarcity of installed sensors or instrumentation on rectifiers that actually record data. Classic rectifier systems were for many years fitted with local gauges and indicators, which can only be read on site. They may trip a remote connection in an alarm-level situation, but the value on the gauge isn’t recorded. Some time ago, says Dänzer, “we started to introduce sensor instrumentation on our rectifiers, especially on the cooling circuit and other critical components”. The first benefit of remote sensing was that personnel no longer had to go into the sometimes hazardous HPR environment to get a reading – a significant safety improvement. 

These sensors were then also remotely monitored, and the data stored for collective analysis.  

“Now,” says Dänzer, “we have far more infrastructure in place to record more systems data, to extract more insights, and to really push the development of predictive analytics for industrial systems.”

Analytics as a service

ConSys Connect is provided via subscription, because it relies on data processing in the cloud, and because collected data will benefit the whole HPR community.  

Built on lean data, the solution can be adapted to virtually any HPR system in ABB’s existing installed base. “The more instrumentation we have, the more we can do, but we can provide functionality for all our customers,” says Dänzer. “Some sensors can be retrofitted, but for implementing ConSys Connect the most important requirement is the availability of digitized system documentation.” 

Ultimately, even competitor HPRs will be able to be integrated with the ABB solution.  

Says Dänzer, “In building a solution that functions well on fewer signals, we have created something outstanding for our HPR customers.”