Emerging Trends in Instrumentation & Controls
Published on : Tuesday 17-12-2019
Jasbir Singh dwells upon the factors that influence the selection of instrumentation and systems in the IIoT era.
Industrial Internet of Things (IIoT) is mainly focused on industrial operation, by digital transformation to the device network for reliable, sturdy and instant field information. The Fourth Industrial Revolution is the era of Industry 4.0, where available data from machines improve manufacturing by digitally exchanging information on single platform from available technologies and current automation. The main aim is moving from reactive to predictive maintenance and develops better asset management strategies to improve reliable, trouble free plant operation on good asset management platform. The goal is to use digitally available real time information from devices and existing data in the system for configuration, calibration, commissioning, diagnosis and again shelving data at designated place for future use.
The IIoT is convergence of operation technologies (OT) and information technology (IT) brought together to a wider platform, which gives the operator access to large fragmented information in precise, digitally understandable data used for operation. It automatically translates into required output to correct the deviation in operation at micro level before it affects the quality of the product. This also gives the operator information to make right decisions, mitigate risk in terms of quality and improve efficiency.
The evolution in sensor, software and control technologies leaves the existing system obsolete, if not upgraded with time. This leads to limited data collection and restricts making the best use of ERP. Today speed is highly important in collecting data from devices, its segregation into useful information, comparison with required standard and translation into understandable format for operator to take necessary action. The selection of instrumentation and systems in IIoT era left us with no choice other than selection of devices capable of fast exchange of digital data, software having instant calculation capability and produce output in understandable visual information at HMI to correct output and tract the results.
On the one hand most of the applications require traffic scaling due to dynamic change in demand, while on the other, the available personalised hardware; it’s becoming difficult to scale the resources for meeting the new requirements. Here it makes sense to use cloud based system to meet the dynamic changes in demand. No single cloud architecture is available that guarantees complete performance for every user application, which forces users to add several services and practices, which can boost cloud performance to achieve the requirement. Cloud providers offer myriad instance types, each with a unique mix of vCPUs, memory, storage and networking. These resources can be further tailored to execute for specific tasks. Public cloud computing mostly available is dynamic in nature, where it offers the potential to add or remove instances and their related resources on demand.
The User/Vendor must implement the rules to decide when and what to scale when demand changes to enhance the cloud based performance. The monitoring services track load
characteristics, and when the workload exceeds a defined utilisation threshold of vCPU, memory, network traffic the system alert triggers the autoscaling service, which follows a predefined plan to add/withdraw resources and set load-balancing preferences for optimisation. The application can execute tasks involving data much faster with cached information. Microservice architecture is adopted, i.e., an approach of application development in which a large application is built as a suite of modular components or services. This platform break applications into a series of interrelated programs that are individually deployed, operated and scaled. These independent services work together through Application Programming Interface (APIs) to meet the full features and functionality. The moment one service reaches its performance limit, only that gets scaled to required size. This is faster and resource-efficient way to optimise application.
Artificial Intelligence (AI) enables people to integrate information, analyse data, and use the resulting insights to improve decision making process in machine. AI is a process of machine learning and adopting as it makes decisions. It uses data available in various digital forms from network or wirelessly and by using advance algorithms developed in software, which behave like highly intelligent programmed human being to act instantly for getting the precise output as desired. The results are consistent over the period of production without rejection. All machines respond to action taken by the human to modify the set parameters based on judgment and intention to obtain require result; whereas AI is a software based system that makes decision by itself, which normally requires human level of expertise to anticipate problems and deal with the issues and operate in an intelligent and adoptive manner. These are basically written algorithms to make decisions using real time data during machine operation. The process uses sensors, discrete inputs, digital data from interface network from a variety of sources, analyse instantly and act on the output control to correct the deviation. AI facilitates high speed processing having large storage capacity and use of analytical capabilities improves sophistication in decision making. AI deals with large volumes of data and figures out efficient output for responding to predictive deviation in automation and control. This is the reason for its growing role in machine learning and using for control while in operation, which improves productivity and economic gain. Wireless technology is gaining importance in industry due to its reliable and safe encrypted data in wide range of machine/plants operation.
Wireless instrumentation is a competitive technology in cases where there is need to remotely monitor instrument condition, reconfigure instruments and gather process data to optimise performance of machine. The reliability of wireless instrumentation is critical as any automation system relies only on accurate data for control and safe operations. Security always remains a concern in any digital network, wireless networks are considered to pose these distinctive challenges. Both transmitter and receiver are authenticated with the network control system while on operation. Transmissions are encrypted using a 128-bit NIST-certified algorithm and verified for completeness and accuracy upon reception. Keys are managed by the gateway and rotated automatically. This combination of authentication, encryption, verification, and key management makes a wireless network as secure as a wired system.
As technology evolves it gives better possibilities to vendors to stand firm with modular and good product/solution to offer and position themselves ahead of market demand. The adoption of private clouds, either within existing contracts, internal infrastructure, or with fresh installs, has meant a slash in running costs and possibilities of better offering by vendors to their customers. That change needs to start with a clear roadmap in existing environment where you will have clear overview what to improve.
Jasbir Singh, Director, ECPR Technologies has over 32 years of business experience with extensive international exposure, both from business and cultural points of view. He is one of the experts in Foundation Fieldbus technology and recognised for continuous improvement in technological development.
