Wedoany.com Report-Dec 21, Asset performance management solutions that incorporate artificial intelligence can now analyse vast datasets from sensors and control systems to predict water and dam asset failures, optimise maintenance schedules and improve decision-making, as Stacey Jones, Global APM Portfolio Leader Energy Industries, Digital from ABB, explains.
Asset performance management (APM). The term is now so ubiquitous across multiple industries, particularly those with a high-level value chain like the global water and dam sector, it is easy to forget what it is, why it is critical to modern industrial operations – and how rapidly it is evolving.
Despite the shift towards more proactive maintenance strategies, businesses still face unforeseen errors and mishaps that impact operations and safety. Current volatile economic conditions have limited funding and reduced onsite staff, increasing the need for remote access, automation and efficient risk management. There is also the need to generate more value from existing assets.
For water utility managers, dam operators, infrastructure engineers and asset managers, this means avoiding costly downtime and environmental risks, as well as managing increased maintenance and repair costs due to ageing infrastructure; growing regulatory pressures; environmental factors like water scarcity and climate change; and higher energy costs.
In this article, we will discuss how AI technologies such as machine learning (ML), predictive analytics and digital twins are being integrated into APM to address these issues, improving the reliability and longevity of assets like turbines, pumps and control systems, and reducing the environmental impact of water and dam operations through optimized resource management and energy consumption.
What is asset performance management?
Put simply, asset performance management is a strategic approach to managing and optimizing assets, helping companies from all industry sectors to predict process failures in real time. This improves the reliability, availability and maintainability of their critical equipment through predictive rather than reactive maintenance, enabling them to hit production, safety and sustainability KPIs with a higher degree of confidence.
Fine. However, the current industry standard for small-to-medium-sized rotating equipment is little to no condition monitoring (CM). Once the values are considered to be within ‘normal’ range, what little information has been gathered is often discarded – meaning industrial operators are getting rid of their most precious asset, data, which can be used to drive uptime, production and profitability.
Things are changing, however. The fall in the cost of wireless technology using connection protocols like Bluetooth or WirelessHart means CAPEX is no longer prohibitively high, meaning the ability to monitor assets across the value chain 24/7 is a cost-effective alternative to manual, infrequent CM.
What is more, with data-driven APM for rotating, electrical and instrumentation assets, integrating remote and wireless CM on the Edge into the existing OT landscape is seamless. Once connected, the data can then be analysed, with outputs/alarms made available via email or common dashboards, on-premises or in the cloud.
AI and digital twins: the advent of APM 4.0
Hard to believe now, but just a few decades ago the majority of asset performance management solutions were ‘run-to-failure reactive’. Equipment had to be shut down at short notice for unplanned maintenance, which is four to ten times more expensive than using quantitative risk analysis and the current state of machine health to prioritize maintenance, and also avoids both unscheduled downtime and safety incidents.
The next evolution was usage or time-based maintenance, in which a schedule was used to assess when equipment was about to fail, so it could be fixed ahead of time. A drawback of this approach was that it treated/treats all assets as the same, rather than identifying those critical to production and created unmanageable maintenance backlogs.
That’s when APM took a major step forward into risk-based maintenance, using failure modes and effects analysis and reliability-centred maintenance to prioritize critical assets. This constituted a vast improvement; however, maintenance was (and still is) based on how an asset has behaved (i.e. failed) in the past, whereas, in reality, 82% of asset failures happen at random intervals. The goal of APM 4.0, then, is to leverage real-time data available to drastically reduce unplanned equipment failures.
Let’s illustrate the point using the analogy of a physical exam. If a physician only asks a patient how you are feeling as opposed to taking their temperature, doing blood work, etc, then they will only get half the picture. It stands to reason that more quantitative information equates to much better decision-making. In an industrial context, maintaining assets efficiently (i.e. only when needed) has been shown to decrease maintenance costs by 20–30%, and machine downtime by 20–50%.
Digital twins – digital replicas of a physical assets – are also evolving, enabling industries to test and refine multiple critical processes and scenarios without impacting real-world operations. ABB uses a hybrid modelling approach. We begin with a physics-based solution that may utilise traditional methods such as thermodynamics and hydraulics. Second, we deploy data driven models using embedded AI and ML to detect equipment anomalies not being monitored by physics based models alone.
Then, in combining rules based, physics based, and data driven models, we create a ‘hybrid’ model that creates a digital twin based on information extracted from the measured equipment or instruments. The ABB team then employs thermodynamics to estimate some of the parameters required, and applies diagnostic models to run “What-If” scenarios that can assist in addressing faults or bottlenecks.
APM in the water and dam industry
As we have seen, AI technologies including machine learning, predictive analytics and digital twins are being successfully integrated into APM tools, analysing vast datasets from sensors and control systems to predict asset failures, optimise maintenance schedules and improve decision-making.
For water and dam operators, the benefits include improved reliability and longevity of critical assets like turbines, pumps, and control systems; the ability to more accurately predict asset degradation, reducing unplanned downtime and costly repairs; increased operational efficiency via data-driven decision-making for more proactive maintenance and optimized resource allocation; and reduced environmental impact through optimized water resource management and energy consumption.
Key features of ABB’s APM solution are predictive maintenance (potential of reducing maintenance costs by more than 15%); real-time monitoring, resource optimisation (potential to increase overall productivity by 8%+), extended asset lifespan (25%+) and environmental and regulatory compliance.
Let’s look at how this works in a real-world project. In Italy, ABB is delivering predictive maintenance solutions that will enable a major operator’s hydroelectric plants – across 33 sites, comprised of around 100 units – to move from hours-based maintenance to predictive and condition-based maintenance11.
The contract includes digital software solutions and services that will provide analysis of more than 190,000 signals and the deployment of around 800 digital asset models. The overarching aim of the project to improve plant operational performance, reduce unplanned failures and enable more efficient planned maintenance practices through predictive maintenance. The integration of the ABB system is expected to yield savings in both fleet maintenance costs and increase plant productivity.
The ABB Ability™ Collaborative Operations Center for power generation and water will help bring wider benefits of digitalization and engagement, supporting informed decision-making, real-time solutions and cost savings. The centre provides similar digital solutions and advanced applications for more than 700 power plants, water facilities and electric vehicle charging stations globally.
Future-proof operations using APM
In conclusion, then, the integration of digital innovations such as AL, ML, digital twins and big data into more traditional APM represents a step change in how industrial assets along the value chain are managed. This gives water utility managers, dam operators, infrastructure engineers and asset managers unprecedented visibility across their operations, enabling them to make the transition to predictive maintenance, reducing downtime while optimizing production, safety and sustainability.
New-generation APM systems are designed to meet increasing demand for enhanced APM systems to handle the complexities of modern water infrastructure, such as aging assets, stringent regulatory requirements and environmental concerns. Deployed strategically, in partnership with a trusted technology provider, these smart solutions have the potential to transform how critical assets are monitored and maintained, future-proofing operation and ensuring that the water keeps flowing.
