Anindya Chatterjee, Global Head of Value Engineering and Data Science for ABB's Process Automation Digital Business Area, discusses how the latest Asset Performance Management (APM) tools, such as digital twins, can help manufacturers ensure reliability of assets and optimise energy consumption.
What is asset performance management (APM)? This ubiquitous term will be familiar to most people working in the manufacturing sphere; however, the advent of industry 4.0 solutions such as artificial intelligence (AI), the industrial internet of things (IIoT) and big data have transformed APM from its traditional condition monitoring role in a single industrial setting into something more advanced.
From a domain side, ABB has identified two major themes in APM. Firstly, how can we improve the reliability, availability and maintainability of key assets in an industrial setting? In other words, use digital and automation tools to ensure production targets are met with a high degree of confidence.
The second theme is this: how can we utilise advance analytics tools and practices like AI, machine learning and digital twins to really assess the performance of equipment, with a view to helping operators optimise energy consumption, reduce costs and emissions, and hit sustainability targets?
The equipment can range from pumps and compressors to pulp and paper machines; every industry has a high-level value chain, and APM is an important tool that can be deployed in partnership with a technology vendor such as ABB to run complex, heavy machinery in a more energy-efficient manner.
The power of digital twins
Let’s take digital twins as an example. At ABB, we currently work with three kinds, the first being a physics-based solution, which may utilise traditional methods like thermodynamics and hydraulics. Second, we also employ code engineering to build the actual models, which is a ‘real’ digital twin.
Then, we have a ‘hybrid’ model, whereby we extract information from the measured equipment or instruments and create the digital twin based on those. We employ thermodynamics to estimate some of the parameters required, and then apply diagnostic models to identify faults or bottlenecks.
ABB is currently using this hybrid digital twin model during a project at a refinery in Turkey, where we are using operating data from the refinery’s compressors to create a digital twin, which we can then use to perform diagnostics and identify improvements in performance management.
In another project, we are taking the concept still further by creating a neural network-based mode of our digital twin, more of an AI-driven solution, to identify the optimal operational set point for an entire data centre, everything from halls and chillers to cooling towers, across 4,000-plus variables.
This is crucial to allow the client to satisfy their service level agreements from a temperature and humidity perspective and assess whether equipment such as compressors are operating sustainably and in the most energy-efficient fashion. This is what we could call next-level APM – or APM 4.0.
Unlock the value of contextualised data
We should also point out that there’s a big difference between having data and making it work. Less than 20% of data generated by industrial companies is actually used. Even less is analysed. ABB is changing that using AI and industrial analytics tools such as ABB Ability Genix, which unlocks the value of contextualised data through IIoT, extracting actionable insights from massive amounts of industrial data with the ultimate goal of improving both productivity and operational excellence.
Not only is it more effective than any traditional software because of its ability to solve complex scenarios, it is a completely self-service-based configuration application that has matured to a no-code configuration of all AI/ML-based algorithms for training and deployment. In other words, you don’t need to be a data scientist to operate and benefit from these systems. Simplifying the complex is a huge part of what ABB is trying to achieve with platforms such as ABB Ability Genix APM.
In addition, ABB Ability Genix also provides a flexible application to capture the learnings of the field in the model, which is called “codifying knowledge”.
By bringing together data from all sources (IT, OT and ET) and contextualising it with additional parameters and inputs, a regular engineer or operations manager can derive value/insight – and leverage those insights to make more informed, data-driven decisions that drive real business value.
To read more on the benefits of Digital Transformation, click here.
About the author
Anindya Chatterjee is the Global Head of Value Engineering and Data Science for ABB’s Process Automation Digital Business Area where his specialty portfolio covers predictive maintenance and digital transformation.
He has over 30 years of experience across operation and maintenance of oil refineries and petro-checmical manufacturing industries. With a background in Mechanical Engineering and a certified Reliability Engineer, he started his career under different roles in Digital solution and product development for Asset Management and predictive maintenance, and has played significant roles in helping customers implement asset management solutions, asset integrity and change management.
Anindya, with his deep technical knowledge in the digital portfolios coupled with innovative solution mindset, has been supporting industries around the world to embark on their digital transformation journey while boosting their performance, operational and maintenance efficiency.