The Move towards Autonomous Systems and Operations with Digital Twins
We explain what digital twins are and what the four fundamental components of their platforms are.
One of the strengths of our Cloud AVEVA Insight, along with the Advanced Analytics module, is the ease of adopting and implementing digital twin technology for our engineering systems or industrial operations.
But What is a Digital Twin?
A digital twin is a real-time virtual representation of a physical object, system, or process. These virtual replicas can reflect a process within a manufacturing environment, a field object or system, or a complete industrial operation across multiple sites. From a jet engine to wind farms to buildings or the entire earth.
These digital representations are used to create analytical models with which to predict the effects and behaviors of that physical object, system, or process in the face of possible changes, and provides information to facilitate decision-making autonomously. The main objective is to maximize productivity, facilitate innovation and continuous improvement with minimal human intervention.
To create a digital twin, a lot of data must be collected, both from the object or process and from what is around it. The representation of the twin is based on historical data and real-time data captured from the object or process, which provide the basis for generating information and activating actions through the monitoring and analysis of these data. The quantity and quality of data collected for the model will also determine how accurately the digital model represents the physical version.
What are the Key Components of a Digital Twin?
Digital twin platforms have four fundamental components:
- Connectivity and data collection
- Data contextualization
- Analytics
- Operationalization
1. Connect and serve data
The first phase of any digital twin initiative includes the automation of data collection in an object or field system or in a manufacturing production line or plant.
Typically, we use a data server to quickly connect to a data source, capture all the necessary data, and serve that data to a digital twin architecture. Each object, system, or process is likely to have multiple data sources, so we must ensure that we capture information at the scale and frequency necessary to provide a complete picture of what is being modeled.
Any dysfunction in the data can create gaps in the analysis and that could significantly hinder optimization efforts. Data integrity counts.
2. Contextualize
The basis of any digital twin strategy and plan involves the contextualization of data. Contextualized data is a set of data processed so that it can be visualized, monitored, analyzed, and used. Contextualized data is captured from raw historical and time series and event data representing physical systems and processes. Part of this process includes capturing data such as time, location, speed, operating states, and other similar identifiers.
These contextualized data sets increase accuracy and actionability, feeding Machine Learning engines and models for predictive and advanced analytics applications.
3. Analyze
As normalized and contextualized data is captured, a detailed analysis can be performed with a high degree of accuracy and knowledge. A robust digital twin platform will generally include a library of pre-configured applications, each designed to predict and shed light on greater efficiency and effectiveness of an asset or process.
Pre-designed applications may include quality prediction, equipment performance, uptime, OEE, utilization and energy optimization, and other on-demand applications defined by users….
4. Operationalize
As the analysis step translates contextualized data into actionable information, the digital twin will provide specific prescriptive actions to operations personnel. As more data is collected and analyzed by the digital twin, the recommended steps become increasingly accurate, even helping to avoid problems before they occur. This is perhaps the greatest strength of the digital twin model: it moves the entire organization from a reactive to proactive and predictive operating state.
World-class organizations will even integrate their digital twin directly into the control infrastructure. In this capacity, the twin goes from recommending the appropriate action to taking the appropriate action.
This move towards autonomous operations represents the next evolution of automation and optimization efforts, so choosing the digital twin platform you make today is critical to the correct evolution of your business.
I add this link in case you want to expand the information regarding how, from Wonderware, we respond to the challenge of digital twins in your sector:
Need more Information?, our specialists will answer all your questions.
