- May 13, 2014
By Nigel Rozier, Raima
From batch processing to sophisticated real-time processes, effective data management is the key to boosting efficiency, increasing productivity and reducing costs in industrial applications. This article explores the role of data management solutions embedded within industrial automation systems.
By Nigel Rozier, Raima
From simple batch processing to the most sophisticated real-time processes, effective data management is the key to boosting efficiency, increasing productivity and reducing costs in industrial applications. This article explores the role of data management solutions embedded within industrial automation systems. Managing data effectively within industrial automation systems has become one of the key challenges of the 21st Century, as companies across all sectors of industry strive to improve efficiency and boost output. This may mean managing a switch from manual data collection and analysis to automated systems; it may mean making better use of existing batch or alarm information; or it may mean managing the ever higher amounts of live data generated within today’s increasingly sophisticated automation products. Common across all of these applications, regardless of their levels of data generation, is the requirement to make better automation and control decisions that will address issues such as the need to reduce downtime, increase availability, boost productivity, eliminate waste, reduce maintenance costs, and ultimately increase profitability. Making improvements in any of these areas requires businesses to look at how they collect and process data, and what they do with that data once it has been collected. At one end of the scale, consider a petroleum processing plant where many automation systems are still manually monitored and sensor-generated alarms are passed on to responders in the form of a phone call or radio communication. These systems may well have been designed at a time when data management and dataflow solutions were still in their infancy. Because of this, alarms can often be delayed in their delivery to responders and many of them can be inaccurate or even false. This could result in frequent and unnecessary maintenance, costly production stoppage or, in rare cases, catastrophic failure. In this environment, the ability to make pro-active decisions is nearly impossible. Because of the lack of a real-time aggregated data system, operators may wait for scheduled batch updates before re-configuration decisions can be made, perhaps postponing tasks that could have increased the overall production. At the other end of the scale, we can look at high throughput discrete manufacturing applications, with automation systems that are generating more and more data, as increasing numbers of sensors and control outputs provide ever more detailed feedback on processes and production lines. That raw data may well hold the key to improved overall equipment effectiveness, reduced energy consumption and vastly improved productivity, but it’s more than traditional data management systems at the higher levels of the enterprise have been designed to handle. In these high data throughput applications, although the need for database systems has been identified, the database solutions that have conventionally been deployed have been slow and bulky, requiring an interface like SQL to access data. The interfaces have been difficult to set up, often representing a bottleneck between the plant floor and the higher level enterprise systems, both in sending data up from the plant floor to be analyzed, and down to the plant processes to be acted upon. Just as the conventional model of managing data in our hypothetical petroleum processing plant was inadequate, so the higher level database model in discrete manufacturing operations fails to meet modern needs, with data being generated at a rate that far exceeds any ability to use it. An alternative data management approach that meets the needs of all levels of industrial automation is the embedded database, which can be tightly integrated with real-time automation processes, and which can manage live real-time data streams. These data management systems can take captured live data, process it (aggregating and simplifying the data as required) and then distribute it to deliver visualization and analytics that will enable meaningful control decisions to be made. The ability to do all of this locally within embedded systems – acting on data that may be of real value only in the moment – has a huge impact on the performance of plant and assets. Complete accuracy in events and alarms allows the operational system to be much less prone to unnecessary maintenance and production stoppage. Real-time decision making capabilities allow system operators to optimize total production and reduce risk by reducing the reliance on human capital. Business managers will have up-to-date information regarding the state of the system and accurate and reliable data for reporting and trend analysis. When evaluating embedded databases, there are some key features to assess. The database technology should already be implemented in numerous industrial automation systems, ranging from simple batch processing systems to complex turnkey power plant systems. Common across many of such systems is the requirement for live real-time data management performance at the controller (device) level, with automated data movement to upstream shop floor management systems and beyond to corporate management systems. Embedded technologies should provide a rugged, scalable and local solution for the handling of large amounts of production-specific data, directly on the plant floor. Platform independence allow them to run on everything from popular OS options such as MS Windows, Linux and iOS to real time systems such as Wind Rivers’s VxWorks, QNX Neutrino and Green Hills INTEGRITY, as well as many others. As well as supporting multiple processor and multi-core architectures, the data storage engine should provide a set of data organisational features that can be use to control in-memory, disk-based or remote storage to provide the best possible performance in an embedded systems application. Importantly, the technology should be able to make data available wherever it is needed. This is often achieved by replicating data between computers on a network and via the Internet to systems outside the embedded network environment. This can be used to improve the speed of processing, data backup security and system-wide data availability. Simple set up is desirable because it increases the visibility of data, and so makes whole production processes more efficient. And if it is embedded within the industrial automation system itself, it eliminates a whole layer of costly PC installations and the associated development and support costs. The conventional embedded database products that typically reside in higher level enterprise systems, often store data in a flat file. This can lead to slow, bulky data handling. As more and more industrial automation systems generate data, there are too many log files for a flat file system to cope with efficiently. Efficient systems not only collects the data in a more structured and meaningful way, but also allows pre-processing of the data actually on the embedded device itself before sending the most relevant data to other systems for further analysis or long-term historical data storage. System developers working with industrial automation systems in all sectors can take full advantage of the advanced features of state of the art technologies, such as support for multi-core processors, in-memory limitation, encryption, shared memory and a host of other high-performance features. It is an ACID compliant database, meaning that the information collected is guaranteed to be accurate. Automatic recovery features ensure that data will never be lost due to a system failure. And often most critical of all is a very high level of availability needed to support modern industrial automation processes. Where once those tasked with collecting, analysing and acting upon data might only have been interested in overall output levels, critical systems alarms, production line quotas and information on plant or asset uptime, today they are increasingly looking at systems that can give far greater transparency into the automation system, accessing and acting on the data generated by the myriad of automated devices in a modern manufacturing or processing plant. Modern embedded database technology addresses the need for fast, reliable data management, flow and analysis that today’s industrial automation systems need. ABOUT THE AUTHOR Nigel Rozier is a Chartered Engineer with a degree in physics and computer science. He began his career as an Embedded Software Developer at Ericsson. This experience led to working as a Field Application Engineer for Telelogic where alongside other duties he taught high level software languages before moving into a commercial role. Nigel was UK Sales Manager at Green Hills Software for nine years before joining Raima in 2010. He is now the EMEA (Europe, Middle East and Asia) Sales Manager and responsible for some of Raima’s largest accounts.. Raima is a provider of high performance, always on database management system technology for both in-memory database usage and persistent storage devices. It delivers database solutions which are cross-platform, small footprint and designed for distributed architecture in resource-constrained environments.
Did you enjoy this great article?
Check out our free e-newsletters to read more great articles..Subscribe