Evolution of turbomachinery controls: Playing catchup with tech innovation

Published on: 
Turbomachinery Magazine, September/October 2020,

Turbomachinery controls and instrumentation have been relatively slow to evolve. But an unprecedented wave of technological innovation over the past decade is bringing about drastic change.

The Industrial Internet of Things (IIoT), digitalization, microprocessors, new software platforms, the cloud, advanced analytics, machine learning, wireless technology, mobility, connectivity, augmented reality, virtualization, artificial intelligence, cybersecurity, big data, Software-as-a-Service (SaaS) and many other breakthroughs have advanced the field like never before. The advantages include remote monitoring and operation, digital twins, remote field service, maintenance automation, real-time plant and fleet management, tighter inventory control, fewer unscheduled outages, and longer plant and equipment lifespans.

Companies like Compressor Controls Corp., Emerson Automation, Baker Hughes, Mitsubishi Hitachi Power Systems, GE Steam Power, Honeywell Process Solutions, Regal Beloit and Veros Systems are in the forefront of this trend. They are harnessing these technologies to make their control systems and instrumentation smarter, more responsive and more user friendly.

Tim Shea, Senior Analyst, ARC Advisory Group, predicts that there will be a period of sustained growth for turbomachinery controls over the next several years.

“This market has tremendous opportunities with steady growth driven by replacement of legacy turbine monitoring and controls with new features such as the growing demand for smart connected products based on the Industrial Internet of Things (IIoT),” said Shea. “The drive for process manufacturing productivity growth is pushing companies to invest in new turbine monitoring and controls for various process industries applications driven by trends such as digitalization and IIoT.”

MHPS-Tomoni

A digitalization initiative from Mitsubishi Hitachi Power Systems (MHPS) is called MHPS-Tomoni This encompasses controls, instrumentation, data analytics, artificial intelligence (AI) and more. It is aimed at making power plants smarter. The average plant has nearly 10,000 sensors that can generate more than a million points of data every minute. MHPS-Tomoni takes this confusion of data and puts it into a useable form.

“Digitalization is becoming the norm,” said Ron Thomas, Senior Manager, Control Systems Integration, Mitsubishi Hitachi Power Systems Americas. “It is crucial to understand what is happening within the turbines and entire plant as operators need to be able to analyze copious amounts of data in real time to detect correlations between different components to improve efficiency, reliability and flexibility.”

MHPS-Tomoni is a suite of digital solutions that can help to create an increasingly smart power plant that will ultimately become capable of various levels of autonomous operation. Increased digitalization of interconnected devices and systems assists control systems to do more, and interface more effectively with advanced analytics. In parallel, steady advances in CPUs, memory and similar digital building blocks provide the opportunity to increase the capabilities of control systems, while boosting speed and cost-effectiveness.

“We are focusing on the continued development of real-time adaptive control and actionable knowledge to optimize power plant performance and improve reliability,” said Thomas. “These technologies hold great potential for making plants more productive, profitable and environmentally friendly.”

The IIoT, for example, offers increased availability of smart sensors and smart devices that can provide expanded data for more comprehensive control and protection strategies, as well as analytics that are derived from control system data. MHPS is investigating how to securely use IIoT sensor inputs for critical low latency control system inputs with jamming immunity.

Modern data management and connection protocols and platforms such as the OSISoft PI System have also proven useful in facilitating communication, data sharing and synergistic total plant control strategies among different control systems and programmable logic controllers (PLCs) in the power plant. Similarly, the steady evolution of computer operating systems adds new capabilities to control system software. This can create a challenge, though, in terms of keeping software updated and incorporating the latest security patches for effective cybersecurity.

The cloud, too, facilitates greater use of advanced analytics and machine learning to provide benefits to power plant operations & maintenance (O&M). Cloud-based analytics, diagnostics algorithms and artificial intelligence are providing more powerful computation capabilities. This offers the ability to make fleet-wide correlations that incorporate the operational experience of many similar power plants and support new concepts such as autonomous logistics systems.

All of these advances are taking place at a time when mobility and wireless functionality are facilitating new O&M support applications such as remote support using augmented reality (AR), auxiliary equipment monitoring and autonomous logistics systems to make operational support more productive. Additionally, virtualization has radically changed product lifecycle costs, removed hardware dependencies, and made it easier to provide cyber security and feature enhancements without costly hardware upgrades.

“Greater connectivity with smart devices in the plant and centralized analytics that perform monitoring, diagnostics and logistics management enable total plant O&M optimization strategies,” said Thomas. “New approaches to O&M support such as remote combustion tuning, controls settings changes and remote balancing support offer a channel for expert advice and support to on-site staff.”

He added that customers are asking about cybersecurity, life extension of existing control systems that are using older components and operating systems, virtualization, and automation of O&M workloads.

What has this meant specifically for MHPS users? The MHPS DIASYS Netmation (Digital Intelligent Automation SYStem) controls platform was initially introduced in the early 1980s. It has undergone several major upgrades as digital technologies advanced in the ensuing years. The latest version, Netmation 4S being introduced to combined cycle plants takes advantage of recent advances to increase operational reliability and reduce maintenance costs. It achieves this by leveraging machine learning, analytics and autonomous systems. The MHPS Virtualized Netmation and Netmation Protect Pack products provide cybersecurity protection and automation of most of the recurring update and maintenance activities related to keeping control system software current.

“The trend toward increased digitalization of power plants will only accelerate over the next several years to improve their flexibility and profitability, making them more intelligent using advanced analytics and cloud computing,” said Thomas. “Plant control systems lie at the heart of the trend and provide the foundation for increased digitalization, connectivity and cybersecurity.”

Further, control systems are advancing to keep pace with changing turbomachinery needs. For example, gas turbines are being developed to run on hydrogen and control systems will play a major role in enabling these new applications.

With MHPS launching the T-Point 2 validation power plant in Japan, the company is in the latest phase of autonomous plant development. Many MHPS-Tomoni digital solutions were validated at the original T-Point 1. T-Point 2 entered commercial operation with a Netmation 4S control system and more than 20 digital solutions that support increased autonomy of critical systems and soon the entire plant.

GE Steam Power

Olivier Jamart, GE Steam Power’s Leader of Automation and Controls, said that current software and controls have changed dramatically from the original analog card versions that still exists in many industrial and nuclear power plants across the world. “The pressure on operating budget and on higher flexibility can be lessened by embracing the new software and controls,” he said.

GE’s Mark VIe control system offers a flexible platform for multiple applications. It features high-speed, networked I/O for simplex, dual, and triple redundant systems, and Ethernet for I/O, controllers, and supervisory interface to operator and maintenance stations and third-party systems. The GE Predix platform offers access to many digital apps and cybersecurity. Plant-wide applications such as Asset Performance Management use historical data and real-time data analysis to predict potential equipment failures and diagnose related issues to help reduce unplanned downtime and improve reliability. Asset-specific applications, such as Generator Health Monitoring, continuously monitor the rotor and stator conditions.

Jamart said GE offers upgrades to the latest industry standard with minimal disruption by easing the migration process. For example, the ALSPA installed base can be switched to the Mark VIe platform. This requires no modifications to I/O modules and related process wiring, reusing of electronic cabinets and automatic migration on the software application (control schemes and control libraries) to the new engineering workstation. The latest cybersecurity practices are incorporated.

“As operators embrace digital technologies to modernize their control and software assets, they are more vulnerable to cyberattacks,” said Jamart. “GE Steam Power offers a proven cybersecurity solution on legacy and new operational technology (OT) to help power companies and utilities plan, design and implement operational resilience into people, processes and technology.”

CCC

For Compressor Controls Corp. (CCC), software changes are mostly focused on cybersecurity, broader integration of high-speed machinery data and virtualization. Collaboration with SEEQ has made it possible to more easily analyze operating data to quantify the potential economic benefits that controls could have on turbomachinery and process performance. Open and standard programming methods have allowed the company to port our algorithms to other platforms in a manageable and supportable way. Broadened access to high-speed control data is also allowing users to do better analysis of data. A new high-speed data historization algorithm allows CCC to send data to the historian while minimizing bandwidth, storage, and loss of data due to data compression.

Cybersecurity and Distributed Control System (DCS) support are the two technologies most mentioned by CCC customers.

“Cloud and wireless have had little impact on our offerings to customers,” said Richard Hall, Vice President of Product Management & Marketing at CCC. “IIoT is something we are watching, but we haven’t had a strong signal from customers during our Advisory Board meetings that it’s something they see as a priority.”

CCC’s Prodigy system is growing well, mostly driven by reliability and control capability. CCC control applications such as Antisurge, Performance, Master/Loadsharing, Speed/Extraction, Fuel and others are at the core of its offerings. But the company has been developing new technology. Process Optimization Studies, Security Update Management, and Guardian Pro ODS (Overspeed Detection System), address areas such as the use of data analytics, improved cybersecurity, and expanding control capabilities to other platforms.

Hall expects market segmentation to continue. Some want automation standardization while others want better performance and security. His company’s approach is to offer solutions for either customer type. For those that want great performance and a focused experience, hardware based solutions are available. For those that want good performance and alignment with broader automation systems, CCC systems are available.

In the aftermarket, users are suffering due to downsizing and retirements. They need support from vendors with long-term turbomachinery experience. As a result, CCC is signing more service and support contracts, as well as doing more retrofits and replacement of controls from other providers.

“The controls that get put in during a project may not be supported by the small companies that specialize in the algorithms, or the mega-firms that focus on large-scale automation systems,” said Hall.

Emerson

Emerson designed its Ovation platform to be able to evolve to take advantage of the latest technologies. This includes analytics, digital twins, mobile and remote work, and cloud-based services. All are embedded in the Ovation automation and software platform. Wireless technology, cloud storage, and mobile devices transmit data from the process and asset level and make it available to the operator, maintenance engineer or plant manager wherever they may be located.

Emerson analytical solutions go beyond basic diagnostics to identify impending process upsets or equipment anomalies and trigger mitigating action to avoid downtime or damage. This is accomplished via modeling, AI based on advanced pattern recognition (APR) and machine learning. Ovation also supports integrated vibration monitoring, generator excitation control, safety instrumented systems, distributed applications, virtualization and digital twin embedded simulation. Subscription services for Ovation include cloud-hosted digital twin and engineering services to supplement onsite resources and reduce the maintenance burden. Cybersecurity features help to bridge the gap between operational technology (OT) and information technology (IT) to mitigate risk.

“Our customers are seeking to digitally transform their operations and our focus is on using technologies and services to accelerate their efforts,” said Jason Blackburn, Director Global Power Marketing, Emerson’s Power and Water Solutions. “There are operational performance and financial results to be gained by leveraging IIoT data across platforms from wireless and other sensors to control systems to asset performance platforms and making this data available across the enterprise.”

Advertisement

He noted that customers are asking how they can keep their plants running safely, even when there are fewer personnel onsite due to limited travel and site-level access. With mobile applications, they can remotely monitor control systems and processes offsite. The Ovation Machinery Health Monitor also provides alerts and insights for operators and maintenance technicians on the health and performance of rotating machinery. Vibration events can develop slowly over time and escalate to failure with little warning.

Blackburn believes that the industry will continue to leverage IIoT technologies, edge computing, AI, machine learning, analytics and expert system technologies. By expanding their digital footprint, users can transition toward semi-autonomous operation.

By installing Emerson’s Ovation technology at its Redbud combined-cycle power plant, Oklahoma Gas and Electric (OG&E) gained the ability to cold start each unit 25 to 35 minutes faster. Located in Luther, Oklahoma, the Redbud plant comprises four identical 1 X 1 units generating a combined output of 1,230 megawatts. The plant began commercial operation in 2003 outfitted with controls from nine different vendors. This patchwork of control systems presented numerous challenges.

“We recognized the inefficiencies inherent in utilizing multiple different control systems. We wanted to consolidate to a standard automation platform to improve reliability and consistency across all units,” said OG&E’s Khoa Le, project manager for the controls upgrade.

Emerson replaced the steam turbine controls as well as static excitation and vibration monitoring systems with an integrated Ovation solution. In addition, Emerson performed steam turbine mechanical retrofits to eliminate single points of failure and provide faster response. With the upgraded and tuned steam turbine controls, along with the redundant mechanical design, the turbines can reach setpoint quicker, reducing cold start times by 25-35 minutes per unit.

Emerson also replaced heat recovery steam generator and balance-of-plant controls at each unit and integrated the auxiliary boiler, water treatment, effluent de-ionization and duct burner controls into the Ovation system. Tied together on a single network, each unit’s Ovation system interfaces with OEM combustion turbine controls for supervisory monitoring and control.

Previously, plant personnel had to monitor multiple process and alarm screens from different vendors. By paring down thousands of control sheets to hundreds and implementing consistent logic, graphics and alarm philosophies across the entire plant, greater visibility was gained into plant processes. Work for all four units was completed during a 16-week outage.

Honeywell Process Solutions

Industrial process automation is an early form of IIoT that has been in existence for many decades. Control systems at large process plants connect thousands of measurement devices, equipment, and valves. Communications with these devices happens in a coordinated way using control methods to operate a complex process in a way that is safe, reliable and efficient.

“We are scratching the surface in helping users achieve better performance during operations,” said Jason Urso, Vice President & CTO, Honeywell Process Solutions. “With more pervasive sensing, new sensor types, faster data collection, and new analytics methods, we will see an inflection point where technology can help users achieve greater performance.”

He considers digital transformation to be invaluable in areas such as converting data to knowledge and knowledge to action. Project execution, for example, typically involves sending large numbers of people to remote locations and working on the physical equipment until a project is complete. By representing the physical equipment as a digital twin, project engineers can work remotely and on a virtual version of equipment. This has been implemented at Honeywell to complete the development of controls, displays, and safety logic, as well as subsequent testing of those systems.

That same digital twin technology can also be used for the training of operations staff. Further, it can be used to monitor process and equipment reliability. With a digital representation of a process and equipment, actual performance can be compared against optimal performance represented by the digital twin.

The cloud is another area of innovation. Honeywell Forge Enterprise Performance Management (Forge) is a software platform that includes applications tailored to improving throughput, yield, reliability and efficiency. It can operate in the cloud and on premise. Those selecting the cloud, receive the set of application capabilities without needing to maintain the software or underlying hardware platforms.

In light of the current crisis, many customers are seeking solutions that allow them to work remotely. Remote operations, remote service, and remote project engineering methods have become essential to maintain business continuity in an environment where the number of personnel onsite must be minimized. Honeywell has evolved its offerings to support these new methods of remote work. Experion PKS Augmented Remote Operations software is a new Honeywell offering that allows process operators to both monitor and control a process from remote locations. An example of this would be a case where a local control operations center must reduce staffing to comply with HSE or local government requirements associated with social distancing. Experion PKS remote operations can quickly and securely establish a remote operations center in an alternate facility either onsite or remote to the site.

“Technologies like digital twins, remote operations, virtualization and wireless are a means to assure every day is the best day of production and every person is an expert,” said Urso.

Further products include:

• Honeywell Forge Enterprise Process Management that combines applications with digital twin technology to achieve better levels of throughput, yield, reliability and efficiency.

• Experion PKS Augmented Remote Operations to extend operations to remote locations.

• Experion PKS HIVE (highly integrated virtual environment) – using Honeywell LEAP project execution principles, software and networking to unchain control applications from physical equipment, and controllers from physical IO.

As for the aftermarket, there is a need to address the many aging controls deployed in the 1980’s and 1990’s. Part obsolescence presents a lifecycle management challenge. Replacing the legacy system can be time consuming and expensive as the built-in intellectual property (IP) has to be recreated. Honeywell’s ELCN offering utilizes virtualization to evolve legacy technology without the need for IP recreation. It does this by taking the legacy hardware and emulating it in software. This enables users to operate existing controls as-is within a software environment. The old hardware is eliminated.

Baker Hughes

The Internet of Things has forced many to take another look at the way users want to operate their turbomachinery units. They are no longer looking to only access data and view data aggregation such as historians. Instead, they are demanding outcomes or recommendations to improve operations. This trend led Baker Hughes to evolve the Nexus OnCore Control System to support edge devices including IIoT. This provides a means of conducting more in-depth analysis as well as remote monitoring for those that want to consolidate expertise in different locations. The Nexus OnCore Control System can integrate with many wireless technologies. It can also support secure remote management, as well as, a Managed Services offering to securely support remote working (work from home) environments and situations.

In addition, the Nexus OnCore platform has a simplified Human Machine Interface (HMI) to make it more intuitive and easier to operate by younger, lesser experienced staff. This helps them to learn the product rapidly and become productive in less time than would be possible using older generation control systems.

Terry Knight, Vice President for Control & Safety Systems at Nexus Controls, Digital Solutions of Baker Hughes, believes that we are not yet in a position where control of equipment can or will be directly from the cloud. However, he said some are becoming more comfortable with sharing information to an external server via their own intranet, or to the cloud. This is especially the case if they have multiple plants in multiple locations to enable them to conduct fleet analytics, for example.

Over the years, he’s witnessed many stages of evolution in controls. It began with mechanical to electrical/electronic drives, then moved from analog to digital technology. Over time, digital technology has become more compact, more reliable and faster.

“The next phase will be with us for some time as we figure out what should remain a typical control system, what can change to edge computing, what can be conducted in a cloud environment, what do we do with all the data available to us to drive outcomes, whether to transmit to the cloud and conduct analytics there, or conduct more analytics at the sensor level,” said Knight.

Regal Beloit

The IIoT megatrend has created more need to efficiently integrate technologies and data sources into one place, while allowing access to different users, said Dan Phillips, Technical Director, Regal Beloit America. His company has been integrating IIoT into its controls, new application technologies to leverage pattern recognition techniques, the cloud for secure remote monitoring, and wireless connectivity.

“One challenge that we see frequently with wireless technologies is that sensors may not be applied correctly to the asset and do not provide advanced warning of faults or degradation,” said Phillips.

The company is also utilizing augmented reality to instruct and train personnel on the assembly and repair of electromechanical systems. It allows users to look at equipment on a tablet or smartphone to monitor equipment conditions. They can peel back equipment layers to look at critical components, read bills of materials and gain a better understanding.

Phillips does not believe in one-size-fits-all systems. Every scenario is a little different. Prior to deploying a technical solution at a customer site, it is important to spend time identifying the value proposition and understanding how success is measured as each asset may have a different critical component. In some applications, the gearbox might be the most critical. In others, it may be the motor, a bearing, or multiple gearboxes or motors.

“Having a scalable system allows for the flexibility to monitor all the components within an asset,” said Phillips. “Additionally, a scalable system allows a user to expand the system to adjacent power transmission applications instead of having to invest in different technology.”

Veros Systems

Veros Systems (Austin, TX) is a startup with backing from Shell Ventures and Chevron Technology Ventures. Veros uses data-processing techniques to decipher the current and voltage waveforms associated with motors to gain insight into the performance, mechanical and electrical condition, and likelihood of failure of rotating equipment. Electrical waveform data are sampled at the existing distribution infrastructure (i.e. VFDs, starters, breakers).

“We collect and analyze just under 400,000 data points each second per motor and only use electrical signals in our analyses,” said Jim Dechman, President & CEO of Veros Systems. “We utilize an edge device in the electrical cabinet to sample, digitize and compress the electrical waveforms that feed a motor. This data is sent to the cloud for machine learning.”

This allows a motor to communicate its condition, the condition of the machine it is driving (compressor, pump, fan) and the overall performance of the drivetrain. Tapping into the electrical waveform signals is non-intrusive and fast. Installation of external sensors (such as accelerometers and proximity probes) and associated cabling is not required.

A massive dataset has been accumulated with readings that show how a motor performs over time. Should the waveform readings deviate from the accumulated dataset, alerts are transmitted. With more than 50 million large motors in factories and plants around the world consuming about half the world’s electricity, Veros provides a means of tapping into their electrical power lines to detect impending issues or failures. The company has licensing or distribution relationships in place with Fluke and Siemens with more in the pipeline wanting to integrate these measurements into electrical power distribution systems and devices.

“Tapping into an electrical conductor and providing warnings a month before a compressor fails helps the operator to conduct repairs or schedule replacement of the machine,” said Dechman. “Veros can also analyze the motion of the motor’s rotor to show how effectively it is operating.”

The technology achieves this by creating a model of motor health. It tracks the motion of the motor rotor for any developing mechanical issues are developing. Flutters in rotor movement are transmitted through the air gap and can be detected in the electrical waveforms. A bearing issue in a compressor, for example, exchanges energy that impacts the rotor. Machine learning is used to filter out process variations, electrical line variations and other noise. ■