One of the most significant trends that could revolutionize fleet management in the power industry is the Industrial Internet of Things. Jonas Berge explains why

Power companies are required to meet ever-increasing demands on energy efficiency and reduced emissions, comply with environmental regulations, and reduce health and safety incidents across a growing and aging fleet of plants.

At plant level, personnel are expected to minimize outages and reduce maintenance costs with fewer people and reduced human error even as experienced personnel retire, taking decades of know-how with them.

Power plant operators can improve fleet management through digital transformation, a new strategy enabled by digital technology in general and the Industrial Internet of Things (IIoT) in particular. Digital transformation can be accomplished at up to four different transformation levels: aftermarket services, business model, operations and automation. These four transformations can be implemented from the bottom up, as each one builds on the next. However, it makes sense to review them from the top down to understand why the underlying digital transformation is required.

Aftermarket services

Power companies want to focus on what they do best – generating power – not performing equipment maintenance. Thus, more companies are employing outcome-focused measures to streamline operations by shifting maintenance services to equipment vendor partners.

Some industry watchers predict a trend of companies ultimately shifting to an outcome-focused service strategy where the equipment manufacturer or an authorized independent service provider is responsible for ensuring their product works.

This transfers the burden of machinery maintenance from the power company to the equipment vendor. This kind of managed service is best suited for complex equipment, from large gas or steam turbines to smaller control valves, that needs frequent attention and is difficult for plant personnel to support themselves.

Such arrangements require equipment vendors to come to the plant and service the equipment when it starts showing early signs of trouble to avoid equipment failure that can result in an unplanned outage. However, the vendor cannot frequently stop critical equipment in a planned outage for inspection, as it must keep running. To enable an outcome-based service strategy, companies can link maintenance contracts to IIoT-based connected services so the vendor can monitor the equipment continuously while it is running, without disruption. This type of digital transformation is an example of how maintenance can be conducted in the plant more efficiently long after the equipment has been sold.

For instance, IIoT-based connected services allow a GC manufacturer’s expert to monitor the equipment from a global monitoring centre instead of through periodic preventive maintenance visits. This can be particularly beneficial for gas-fired power plants that use GCs, as variations in gas composition can negatively affect combustion and emissions, and may even damage the turbine, as the combination of natural gas with different composition and heating values is becoming more common. Power plants can easily adopt this service strategy based on connected services, benefitting from the GC manufacturer’s experienced analytical engineers in the event they don’t have their own. As a result, performance is optimized with minimal burden on plant personnel.

Business model

Despite preventive maintenance and periodic inspection by maintenance and reliability teams, plant equipment still fails and fouls occasionally, causing outages and inefficiencies.

The main problem is that with periodic inspections, early signs of trouble often go undetected, and the equipment runs to failure. There simply isn’t enough time or skilled people to inspect and collect the data as frequently as required. Plants often must rely on external contractors for vibration testing, steam trap surveys, corrosion audits etc. Still, even with contractors, the inspection is not frequent enough.

Therefore, plants are now adopting IIoT-based connected services for maintenance management across their fleet whereby an external third-party service provider is continuously monitoring the equipment in the plant from their global monitoring centre. The service provider generates a weekly report listing equipment in the plant that needs servicing. This report is used by the plant maintenance team for planning of their daily maintenance activities. Depending on the type of equipment and service agreement level chosen, immediate exception reports can be automatically triggered in case of sudden critical events. This represents digital transformation of maintenance management.

To enable the connected services, the service provider installs instrumentation on the equipment to be monitored (pumps, cooling towers, relief valves, steam traps etc), making it all ‘connected equipment’. The number and types of sensors used depends on the type of equipment to be monitored and the failure modes of interest. The data from these sensors are sent securely to a cloud-based server where the analytics software runs 24/7. When combined with sensors and software, a ‘dumb pump’ then becomes a smart pump.

Most importantly, the service provider has a pool of experts with domain expertise in many areas such as turbines, compressors, pumps, control valves, relief valves, steam traps and many other equipment classes. The more asset classes the service provider can support, the fewer service providers and less technology infrastructure will be required.

The experts extract and review the reports from the analytics software before they are sent to the plant maintenance department, and the plant maintenance engineers can also collaborate remotely with the service provider’s experts if needed. This pool of experts is very knowledgeable and experienced, often working with multiple sites and constantly expanding their knowledge base.

To make connected services possible, the plant must permit the required data to leave the site. Keep in mind, this only pertains to equipment data used by external service contractors, not proprietary power generation data. The service provider is not operating the plant, just monitoring the equipment condition. If external service contractors are already collecting vibration data and doing steam trap surveys, this data is already leaving the plant with them on their laptops. IIoT-based connected services just transmit that same data over a secure network without proprietary power generation data ever leaving the plant.

While most plants purchase this instrumentation and network infrastructure as a traditional capital investment, some plants prefer not to purchase the equipment but rather require the service provider to install and support these sensors and infrastructure as part of the service contract.

This is a new business model where the plant pays a monthly fee per piece of equipment, which includes the full service: monitoring, reporting, and the associated hardware and software (although the plant must set aside operational budget for these services). The service provider takes responsibility for maintaining the networks and keeping software up-to-date. A service contract can start with simple assets like steam traps and relief valves, then gradually grow to include monitoring of other assets using the same infrastructure.

Alternatively, the power company can set up their own global monitoring centre to manage the maintenance across their plant fleet anywhere in the world. The infrastructure can use the same underlying products and technologies (digital sensors, networks and analytics apps), but the pool of experts is their own.

Several security options are available to send the data required for condition and performance monitoring services into the cloud:

• De-Militarized Zone (DMZ): The traditional way of connecting the control system to the enterprise network through back-to-back firewalls already existing in some plants and supported by their IT department. The same path can be used to send data to the service provider;

• Historian: Enterprise historian infrastructure may include cloud connectivity, meaning the service provider can have a historian and mirror the required data from the plant across the Internet;

• Industrial Data Diode: A unidirectional connection path whereby data can only leave the plant, meaning the connection cannot be used to send malicious commands or malware to the plant. It is also very easy to administer as there are no certificates or policies involved, and no firmware patches. Selected data is pushed to the service provider;

• Physically Private Network: A 3G/4G router connects through the mobile network to the cloud, totally separate from the control system and the plant’s networks.

Often the most appropriate solution is selected based upon the application and current plant infrastructure:

• By subscribing to an IIoT-based steam trap health monitoring connected service, instead of relying on yearly steam trap surveys, the plant receives a weekly report telling them which steam traps have failed. By replacing them sooner than ever possible in the past, the plant is able to reduce steam losses;

• By establishing their own global monitoring centre with a pool of experts, fleet managers can have control valves in a plant checked by valve experts from elsewhere in the world to support the limited number of staff at sites;

• A coal-fired power plant in Southeast Asia stores coal in silos monitored by multiple 3D solids scanners to accurately map the volume, even though the profile is very uneven due to peaks and valleys formed while drawing and filling the coal. The solids scanner manufacturer supports the power plant by periodically checking the state of the scanner system from the global monitoring centre.

Other plants can easily replicate these solutions, resulting in minimized equipment downtime and optimized performance, with minimal burden on plant personnel who can now complete their tasks with a minimum amount of fatigue, minimum risk of injury and maximum on-time performance.


A current trend in many industries is a gradual shift to data-driven maintenance and operations. Many power plants have already been modernized with an on-premise second layer of automation for performance and condition analytics of equipment beyond only the most critical turbines, and to automate manual maintenance, operations and emissions data collection tasks.

Availability of condition and performance data gives plant personnel the ability to perform tasks through a digital transformation of how daily operations are carried out in the plant. This provides the opportunity to develop new data-driven standard operating procedures (SOP) – which is the whole point of digital transformation. It also means plants must manage a data-driven process of organizational change and inculcate a new mindset of checking the software first before going to the field to inspect or service the equipment.

Training also must be provided to change behaviour so people can follow the new procedures, use the new tools, and see how those tools help them in their daily tasks to establish the new predictive maintenance culture.

The enablers for automating the data collection and interpretation are sensors installed on the equipment and the analytics software. This includes common sensors for pressure, temperature, level and flow, but also more exotic sensors for factors such as corrosion, vibration, acoustics, power consumption and position. Many of these sensors are non-intrusive, meaning they can be installed without cutting, drilling or welding while the plant is still running.

As a result, power plants have been able to reduce outages, reduce maintenance costs, improve equipment efficiency and overall plant heat rate, comply with emissions regulations and increase productivity.


All these solutions require hundreds or thousands of additional sensors around the plant. It is not practical to connect all these sensors using traditional point-to-point 4-20 mA and on-off signal hardwiring, and existing control systems don’t have sufficient spare I/O points or space to install more. In addition, opening junction boxes and cable trays to lay the new cable risks damaging the existing installation.

Power plants can deploy a plant-wide wireless sensor network, which can be a practical way to add sensors in an existing plant. Using digital sensors and digital networks requires a digital transformation of how the I&C department in the plant thinks about the instrumentation to be used in an application. Also, digital automation is the basis of all digital transformation. That is, the power plant must start by deploying a digital network infrastructure for the digital sensors and other digital instrumentation.

The desire for new technology is strong because it enables new solutions to old problems. Outcome-focused strategy, connected services, and digital transformation of operations ultimately boils down to the use of digital sensors, networks and analytics apps. New plants are built with fully digital control systems, including digital networks for the control loops and motor controls.

Wireless sensor developments

Wireless sensors have been used in plants for more than ten years now, starting with sensors for pressure and temperature, followed by an endless parade of additional measurements such as level, position, on-off contact, vibration, corrosion, temperature, acoustic noise, level switch, wireless adapter etc. The past two years have seen the addition of non-intrusive corrosion sensors, wireless pressure gauges, turbine flow sensors, electric power sensors, and integrated sensors with differential pressure and pressure in the same module. Solids level scanners supporting wireless adapters have also been made available.

Reliability engineers and maintenance technicians do not have time to interpret all the data from dozens to hundreds of pieces of equipment manually. Moreover, they are not data scientists or mathematicians, and they cannot use complex general purpose analytics tools.

Predictive analytics software helps in the interpretation of data streaming in from equipment sensors to diagnose signs of trouble early. This analytics software must be easy for reliability engineers and maintenance technicians to use. Simple apps should warn them if there is a potential problem with pumps, gearboxes, cooling towers, compressors, blowers, heat exchangers, fans and air-cooled heat exchangers with plain-text, actionable information.

To get this level of ease-of-use, plants use specialized equipment analytics apps rather than generic data analytics tools. At the highest level, the analytics app displays equipment health as a simple dashboard, allowing the user to drill down into details.

Deep equipment knowledge is contained within the analytics to predict specific equipment problems. The apps perform real-time analytics on multiple variables such as vibration, temperature and pressure as the data is coming in from the sensors.

Analytics pick up developing trends, instability and changes in background noise in real-time in the early stages, thus predicting failure and fouling. This gives maintenance a chance to schedule service in advance, and if the problem is caused by operations, they can make changes to prevent it from getting worse.

In the case of a pump, issues like developing strainer blockage, bearing failure, mechanical seal failure and pre-cavitation are presented as easy-to-understand plain text. Engineers need not have data analytics skills to interpret the data because the diagnostics are descriptive and actionable. ‘Pre-cavitation’ tells personnel there probably is a blockage, so they will check for closed upstream and downstream valves or a plugged strainer.

Moving the needle

IIoT and on-prem automation helps plants remain competitive by improving their performance in many areas including: reduced outages, extended life of equipment, reduced maintenance costs, improved heat-rate, reduced emissions, reduced HS&E incidents, improved response time and enhanced productivity.

This is achieved through digital transformation of how the plant is run and maintained, using outcome-focused service strategies based on external connected services for equipment condition monitoring and maintenance management, which in turn is based on automatic data collection and analysis, and ultimately on a digital ecosystem.

Jonas Berge is Director of Applied Technology at Emerson Automation Solutions