Batteries, Coal

SaskPower Uses Wireless Measurements to Improve Process Visibility at Coal-Fired Generating Station

Issue 9 and Volume 116.

By Scott McLeod, P.Eng, Supervisor – Performance, SaskPower and Mark Menezes, P.Eng, Canada Manager – Measurement, Emerson Process

SaskPower is publicly owned, and generates and distributes electricity to 473,000 customers in the province of Saskatchewan, Canada, with exports to adjacent provinces and North Dakota. Total generating capacity of 4 GW is obtained from a mix of coal, gas, hydro and wind. Their engineers use advanced software to optimize their operations and these expert systems require a lot of real-time data, which drives the need for many new process measurements. For their 267 MW coal-fired Shand station, adding new measurements using traditional field wiring was deemed prohibitively expensive in cost, time and physical space. Instead, SaskPower used Foundation fieldbus and WirelessHART transmitters. One hundred sixty-five fieldbus and 67 wireless measurements were added at less than half the cost of traditional wired, and the system has proven extremely reliable since startup in 2009. SaskPower has since added wireless systems to other facilities, and currently operate more than 200 WirelessHART transmitters in four coal-fired plants with total generating capacity of 1.1 GW.

Need for New Measurements

SaskPower’s internal “Performance” team is tasked with optimizing all of the company’s assets, with the biggest opportunities in coal-fired plants. The team has installed Scientech “FAMOS” software, which continuously analyzes and quantifies opportunities for improving plant performance (heat rate), and detects problems before they can cause shutdown. The system can utilize manual inputs, but the best performance is achieved with continuous, real-time measurements. Although the plant has an automatic control system and associated measurements, the Performance team determined in 2007 that they required additional measurements – 26 pressure, 10 flow and 197 temperature – for maximum benefit.

1 A Screen Capture Of The Famos Screen At The Shand Power Plant From April, 2012. Courtesy Of Saskpower.
Fig. 1

The majority of the temperature measurements were packed into a few locations, so the team decided to utilize eight-input temperature transmitters multi-dropped via Foundation fieldbus. The other measurements were widely distributed over the unit. Adding the new measurements using a traditional wired approach would have incurred significant capital costs in addition to the purchase of the transmitters themselves – new junction boxes, cable trays, marshaling cabinets. The control system, which is itself nearing obsolescence, would require new input cards, racks and power supplies. Physical space for this new equipment would be difficult to find, and existing cable trays were full of aging, brittle cables, prone to failure if disturbed.

Fig. 2

Engineering and labor costs for the new wiring were also expected to be significant, including terminations, checkout and drawings. Project execution was scheduled for 2008, which looked to be a busy year for local contractors, given booming oil, potash and uranium prices. Engineering contractors require accurate information, and of course charge for changes made mid-project. The SaskPower team wanted to ensure maximum flexibility, since the exact number and locations of all of the measurements was not known in advance. Easy expansion was needed, since they expected that as some measurements came online, they would point the need and provide justification for additional measurements. Finally, operations saw advantages in having new measurements integrated without interruption of the existing process control system.

WirelessHART

Wireless technology is used today in a wide variety of power plant applications – to connect a few remote points to the central control system, to connect one control system to another, to track personnel and equipment, for specialized applications like video, and to support the ‘mobile worker’. Various wireless technologies continue to be adopted in these applications, and one of the most applicable to transmitters is the WirelessHART self-organizing network. Each transmitter contains a smart RF radio. While RF is a “line of sight” technology, like a typical RF walkie-talkie it can work through some walls and gratings, and around smaller pipes and pumps. Devices which are within about 300 meters, and not blocked by large, solid vessels or thick, reinforced walls and floors, can communicate directly with the gateway. In a “mesh” topology, a device which cannot see the gateway needs only to see another device, so a data packet can “multi-hop” between as many devices as necessary until it gets back to the gateway.

The mesh architecture allows devices to see around obstacles without the need for tall, costly antennas or repeaters. Once the system reaches sufficient density, the user can usually assume any new device will be able to communicate with several other devices, so failure of any one device will not affect network reliability. While the radios will dynamically switch to unused channels in the RF spectrum, in a typical plant environment much of the interference is generated not by radios but by devices that generate spurious, random emissions, like welding torches, variable-speed drives, etc. In a mesh network, the radios talk ‘around’ rather than ‘through’ electrical or physical obstacles or interference. The network is ‘self-organizing’. Each device automatically forms multiple connections to the gateway, which re-form dynamically as new devices and obstacles appear. This ensures high reliability with minimal engineering, and avoids costly site surveys.

3 WirelessHART Pressure & Temperature Transmitters Simplify Instoallation
fig. 3

Installation benefits are maximized when the self-organizing network devices are powered by internal batteries, with hazardous area approvals for the transmitter and battery system. The SaskPower optimization system expects new inputs every minute – at this update rate, the supplied batteries are expected to last five years. Since the system has only been operating for two years, none of the batteries have yet been replaced. The batteries are user-replaceable in the field, without jeopardizing area approvals or requiring hot work permits.

Users prefer open standards, which ensure access to competitive pricing and best-in-class technology, and minimize the risk of obsolescence. The HART protocol is used in 75 percent of smart wired transmitters shipped today, with devices available from more than 200 suppliers. Not surprisingly, since it leverages existing supplier and user experience, the dominant open protocol for wireless transmitters is WirelessHART, which is also a full international (IEC) standard.

Security and NERC CIP

A significant barrier to the adoption of WirelessHART at SaskPower was concern about security. Cyber security is important because wireless transmissions can be accessed outside the plant fence, which bypasses plant security. A secure system ensures that any data that gets through is valid data, and only allows access to those who should have access to the devices and data.

WirelessHART provides multiple levels of protection against interference and attack:

  • IEC 62591 (WirelessHART) is not a routable protocol = no “IP to the edge” (“serial” is often used in NERC literature as a euphemism for non-routable)
  • Secure device provisioning – wired connection used to insert “shared secret” (Join Key)
  • Robust device join process, including device access control list (whitelisting) when unique join keys employed (Outsiders kept out)
  • AES-128 encryption of individual end-to-end sessions within the sensor web = a cryptographic perimeter
  • Secure browser front-end supports role-based access control
  • Emerson Gateway has an internal firewall with multiple SSL-enabled secure ‘upstream’ communication options (e.g. AMS secure)

SaskPower aims to voluntarily comply with new and emerging NERC CIP security standards since they represent ‘best practice’. Despite the security built into WirelessHART, to minimize risk of non-compliance with existing and future NERC CIP standards, SaskPower elected in 2009 to segregate the wireless system from the control system. The wireless measurements are fed directly to a separate PLC, then on to the historian and optimization system. The wireless system is not currently connected to the control system at any level. This is clearly not the optimal approach for maximizing process visibility, so SaskPower has engaged a third party to test the security of WirelessHART. Discussions with NERC personnel indicate that, with careful planning and documentation, WirelessHART can be utilized in a NERC CIP-compliant system.

To maximize the flexibility of wireless usage, the gateway device should be designated a Cyber Critical Asset and documented as an Electronic Security Perimeter (ESP) Access Point/edge device. The sensors would be designated as outside the ESP and would not need to be Cyber Critical Assets even if they are providing essential information due to the non-routing nature of WirelessHART. Non-NERC facilities would of course have more flexibility in their network design choices (See Fig. 4).

fig. 4

Benefits Achieved and Lessons Learned

The wireless system was installed with ‘no surprises’. Once the gateways were installed and integrated with the PLC, the devices were added online one by one over a period of months, during maintenance ‘spare time’. Adding a device is simple – connect the device to the process, tell the device its network ID and join key (password), and let the device find its optimum path(s) through the network to the gateway. Since December 2009, the system has provided very high data reliability – comparable to a traditional wired system. When compared with estimates provided by local engineering companies for a traditional wired system, the WirelessHART installation was implemented for less than half the cost.

The optimization software has been able to use the new process data to help SaskPower quickly identify and quantify gaps between potential and actual heat rate on a unit-by-unit basis, with the objective of improving average heat rate by 0.5 percent. An additional benefit is the ability to quickly identify potential problems, particularly equipment that is degrading or is not correctly returned to operation after a turn-around. Some examples include:

  • De-aerator steam valves not re-opened after being shut in for maintenance, which if left closed, could have led to a heat rate loss of 10 kJ/kWh, valued at $4,000/month.
  • Extraction line off steam turbine to feedwater heater develops a slowly growing leak inside the condenser. Very slow leak of steam not recognized by the operator, leading to slowly decreasing efficiency, and eventually a more serious rupture causing unscheduled downtime. The slow leak is recognized relatively quickly by the optimization software, and remedied at the next turn-around.
  • Shut-off steam valves are manually closed after a turn-around when cold, but develop slow leaks as they heat up during normal operation. A slow steam leak wastes energy, and eventually wears out the valve, requiring expensive maintenance. This can be detected by a temperature sensor downstream of the steam line. While not available in time for the project, the clamp-on temperature sensor integrated to a wireless transmitter shown on Fig. 5 promises the lowest installed cost.
5 Clamp-on Wireless Temperature Transmitter For Lowest Installed Cost
Fig. 5

While the system has surpassed expectations for cost-effectiveness and reliability, there have been some ‘lessons learned’:

  • Integrated wireless flowmeters, shown in Fig. 6, significantly reduced installed cost. However, the flowmeters were set to read flow every one minute, consistent with the other measurements and the scan rate of the optimization system. In practice, it was found that, unlike other measurements like pressure or temperature, flows are noisy. So, even if the average flowrate is stable, a given flow reading can vary by 2 percent or more between scans taken even a few seconds apart. The optimization software can interpret and flag this random variation as a ‘significant’ change in operation. The solution is to calculate a moving average in the PLC, and send only that averaged value to the optimization software. If comparable response time is desired, the flowmeter can be configured to update more frequently, about every 15 seconds. Faster readings will of course impact battery life, so the flowmeter batteries may have to be changed more frequently than once every five years.
  • Many of the wireless transmitters were purchased with optional local displays, which only turn on for the one second when the transmitter itself turns on and takes its reading. Although the displays are not expensive, SaskPower will probably not be including displays on future wireless transmitters.
  • This was one of the earliest projects to use wireless transmitters in Saskatchewan. Local contractors were not familiar with wireless, and not surprisingly, their quoted prices for installation of the wireless devices were very close to their quoted prices for installation of equivalent wired devices. Hopefully, as contractors gain experience with wireless, their prices will better reflect the available cost savings.

fig. 6

System Expansion and Future Plans

In 2010, SaskPower used the same technology for three of the six units at the 828 MW Boundary Dam Power Station. Approximately 200 WirelessHART transmitters were used for the Boundary Dam system, again installed and commissioned by SaskPower internal maintenance personnel. Like the first, this second installation has provided data reliability comparable to traditional wired.

As new WirelessHART devices become available from different suppliers, SaskPower will be able to easily integrate them into their existing wireless network, which has considerable spare capacity.

The system has succeeded in its primary objective of providing reliable real-time measures and historical trends of actual and potential performance of each unit, allowing SaskPower management to make better long-term decisions. In the future, this will allow the company to better decide where to invest capital to capture potential capacity, and to quickly quantify the return on each investment as it’s made.


Temporary Cooling Solutions to Meet Environmental Needs

By Billy Childers, Cooling Tower Services Manager, Aggreko

New environmental regulations will affect power generation in the U.S. to an unprecedented degree. Two new regulations are aimed at reducing air emissions, while the third focuses on reducing the impingement impact on aquatic life at power stations that utilize once-through cooling water. These regulations are commonly referred to as Maximum Achievable Control Technology (MACT) and 316b – Phase II.

Aggreko's temporary solution to reduce discharge water temperatures at one plant involved 67 modular cooling towers and 27 vertical turbine pumps
Aggreko’s temporary solution to reduce discharge water temperatures at one plant involved 67 modular cooling towers and 27 vertical turbine pumps

In December 2011, Power Engineering reported that MACT and the recently vacated Cross State Air Pollution Rule will push between 10 GW and 80 GW of fossil-fuel fired generation into retirement by 2016. The added pressure this puts on the remaining capacity will be intensified by the impact of 316b – Phase II.

316b of the EPA’s Clean Water Act defines the requirements for facilities that use once-through cooling water. By July 2012, it will require plants that use more than two million gallons of once-through cooling water per day to reduce the impact on aquatic life, in particular fish larvae.

The EPA estimates that the rule will cover 1,260 existing facilities, of which 670 are power generation plants. Plants will have up to eight years (until 2020) to comply. However, most will be required to comply before then when their permits come up for renewal.

Decision time

These imminent changes will mean some tough decisions, regardless of whether the preferred option is to comply or to shut down. In both cases, providers of temporary cooling solutions are in a unique position to help industries manage the transition. Temporary cooling towers, capable of cooling millions of gallons of water per minute, provide reliable and effective engineered solutions whatever the scale of the problem.

For a plant suffering water shortage due to drought, 76 Aggreko mobile cooling towers like these provided the solution. They were fed by 48 pumps, cooling 228,000 gallons per minute
For a plant suffering water shortage due to drought, 76 Aggreko mobile cooling towers like these provided the solution. They were fed by 48 pumps, cooling 228,000 gallons per minute

Where plants are scheduled for closure or have an uncertain future, instead of companies making significant capital investment in existing cooling towers, temporary cooling towers can be substituted at far lower cost.

Where power plants cannot be adapted in time for the new 316b deadline, temporary cooling towers can enable operations to continue at full capacity until long-term solutions can be implemented.

A plan for all seasons

In most cases, the cooling problem is seasonal, peaking in summer when higher intake water temperature changes the volume of water needed to cool plant components.

A typical solution for such a problem was recently provided by temporary cooling specialist Aggreko for a power plant in the southern U.S. A complete system of Aggreko modular cooling towers, vertical circulating pumps and piping delivered 180,000 gallons per minute of cooled water to ensure temperatures complied with regulations. The plant avoided scaling back its power output, reducing the required run time of less efficient peaking plants, saving an estimated $45 million in fuel costs over the next five years.

At another plant in Georgia, when a drought-stricken river limited the plant’s once-through cooling system, Aggreko’s cooling towers delivered 228,000 gallons per minute of cooled water to ensure discharge water remained within acceptable levels.

The complete turnkey temporary installations provided fast and economical solutions with complete on-site technical support. Modular units are scalable and flexible, and companies need only deploy the exact capacity required when needed. Most important, a temporary solution relieves the immediate pressure quickly and gives operators time to engineer the right permanent solution.

Indirect effect

Even facilities not directly affected by the upcoming regulatory changes will feel an indirect effect. With the amount of capacity being retired from operation, demand on the remaining plants is expected to be much higher than before, challenging the capability of existing cooling systems.

Whether a facility uses cooling towers or once-through cooling, temporary cooling tower systems can be used to provide supplemental cooling. Supplemental cooling of cooling towers results in less turbine backpressure and increased efficiency for the whole system.

Whether directly or indirectly affected by the new changes to environmental regulations, those planning the future of their facilities will do well to consider how much of a difference temporary cooling tower systems can make to their business.


Natural Ventilation for use in Power Plants

By John Moffitt, President, Moffitt Corporation

Over the years a variety of industries have discovered the benefits of natural ventilation. While the steel industry has utilized Natural Ventilation principals since its inception, industries such as paper production and glass plants to aluminum smelters and gypsum plants have all seen the advantages of natural ventilation as well. Power Plants, however, have been slow to adopt this technology.

Oftentimes powered fans and equipment were used out of habit or “common sense.” Natural ventilation was dismissed as unpredictable and the potential savings in the electric bill seemed inconsequential. However, in the past decade power plants have been adopting Natural Ventilation, and at an increasing rate. Many of the reasons for power plants to dismiss natural ventilation have been falling away over the years.

In fact in 2010, Alabama Power’s Greene County Electric Generating Plant, a Steam Electric plant under the Southern Company umbrella, embraced the concepts of natural ventilation. Located in west Alabama, this Power Plant would get very hot during the middle of the day. By installing weatherproof roof vents and wall louvers the hot stagnant air was able to more freely move about the facility. Natural ventilation increased internal airflow, leading to a cooler work floor and more comfortable, more productive plant personnel.

Natural ventilation works on the principals of gravity and natural air flow. Warm air has a natural propensity to rise and placing a natural ventilator on the roof allows this warm air to escape. This process is aided by placing wall louvers at ground level to push fresh, clean air into the building. This creates an airflow cycle that then pushes out the hot air at an increased rate. A cycle of air, flowing through the building helps keep the interior environment cool.

Maintaining a consistently cool temperature, and more importantly adapting to changes in temperature are crucial for keeping a building’s interior comfortable. As the day goes on and the temperature increases a natural ventilation system adapts to these changes. By design, a natural ventilation system is self-compensating with the cycle of air moving through the building increasing as the temperatures rise, keeping the interior space cool.

Of course not every environment is ideal for Natural Ventilation. There are certain requirements to ensure that the air will flow at the proper rate to make natural ventilation effective. For instance, a building needs to be of a certain height, to allow for a proper stack effect for air movement. Additionally, natural ventilation works best in a facility that is already producing its own heat. Of course Power Plants often meet these requirements, making them good candidates for natural ventilation.

Alabama Power's Greene County Electric Generating Plant.
Alabama Power’s Greene County Electric Generating Plant.

Advancements in technology, such as Computational Fluid Dynamics (CFD), help better illustrate the effectiveness of natural ventilation. A CFD analysis is an effective tool for determining how to best ventilate a building. By inputting the building dimensions, heat sources, and surrounding areas, a CFD model can simulate the transfers of energy through the building. Utilizing thermal imaging technology CFD modeling helps determine the best design for a ventilation system and depicts the results as easy to interpret 3D models.

CFD modeling allows for a more effective ventilation solution to be made for a given building. Various factors such as placement of vents and louvers, temperature changes, and weather conditions throughout the year, can all be tested before construction ever takes place. This ability to better determine the necessary vent size and placement allows for natural ventilation systems that are more effective than ever before, greatly reducing hot-spots and dead zones.

Utilizing the natural flow of air and heat means a ventilation system with no fans and in turn, no electricity consumption. Of course many plant managers will argue that electricity costs are hardly an issue and that the energy it takes to power fans is insignificant. However with the push from customers and local governments to be go green and be more efficient, many facilities are seeing the benefits of implementing green practices.

Energy costs are not the only issue when it comes to a mechanical ventilation system, however. Any fan or powered ventilation system has moving parts and therefore needs maintenance and repair more often. Whereas the failure of a few fans can lead to the work floor quickly becoming unbearable or difficult to operate a natural ventilation system is continuously operating, with no moving parts to break or wear down. Wall louvers, roof vents, and other natural ventilation products last for a long time before replacement is needed.

Of course for all of its benefits, natural ventilation does not work in all power plants. If the building is not high enough to allow for proper air flow, it will be a bad candidate for natural ventilation. A certain height is needed to ensure that the air will rise through the building at the proper rate. Additionally, a plant where the turbine hall is enclosed would also be unable to utilize natural ventilation. When there is no direct access to the outside through a roof vent, natural ventilation becomes untenable. Other issues such as environments with persistent bad weather or plants located in certain areas, will make natural ventilation a poor option.

However, with advancements in design technology, the increasing need to reduce power consumption, and the long-lasting life of the equipment, more and more power plants are implementing natural ventilation techniques.