Nuclear power station fires can have serious financial implications. FireVu, a specialist in the power generation sector, tells us about the issues and solutions for detecting nuclear power station fires early in April’s edition of Nuclear Engineering International magazine.
Hartlepool Nuclear Power Station’s two reactors produce 1180 MW of electricity, enough to power 2.5 million households across the country.
On 20 April 2013 a fire broke out in the turbine hall of unit 2. The fire, caused by an oil leak, ignited lagging surrounding part of the turbine. Reactor 2 was shut down and cooled, mitigating any threat of nuclear contamination.
The quick actions of staff and the emergency services highlight the importance of early detection. While the cost was relatively small, the fire at Hartlepool still caused material damage and disruption to power generation.
While I don’t have the figures for Hartlepool, we know the estimated losses for coal fired Ferrybridge, which suffered a fire in July 2014. RBC Capital analysts estimated a loss in earnings of around £35 million in energy production.
Transactional costs of buying back forward contracted power can also be factored in as they will now need to be sourced from elsewhere, in addition to the cost of repairing the facilities. Any fire incident adds up to a considerable sum, however “minor.”
Fire detection power plants – the challenges
Nuclear power stations face similar challenges to conventional power plants, including dust, debris, oil and heat generation triggering alarms. There is also the voluminous areas of plants that make fire detection challenging. Fire detection solutions have to be sensitive to danger, raise alerts early, but not be prone to costly false alarms.
The sheer size of power plants makes fire detection challenging. The huge ground areas they occupy and high ceilings (up to 20m) create large volumes of space, such as turbine halls, which must be monitored.
This presents difficulties in terms of the time it takes for smoke to reach some detector systems. In some cases smoke detection can prove ineffective owing to smoke stratification. When smoke cools it reaches a height that ties in with the temperature of the air and remains at this height, which can be below detectors fixed to the ceiling that require smoke to trigger them.
Moreover, we must consider the need to provide a system that can cover large areas effectively – no easy task. Many areas are unmanned such as generator rooms, turbine halls, switch and store rooms and there are also many remote buildings.
Turbines also present a very industry specific issue. Turbine fires can initially be hard to detect as the shell can obscure and contain flames for some time.
In addition the cost of false alarms in turbine halls can be significant. If suppressants are released for false alarms, the cost for big voluminous areas can be substantial, well in excess of £100,000 for many sites.
The attendant environmental impact such as the wash off of foam and liquid based suppressants running into water supplies also needs to be considered.
Fire detection technology
Visual Smoke Detection (VSD), Infra-red (IR) and Aspirating Smoke Detectors (ASD) are the principal options for power station operators.
Each has their advantages and drawbacks. Let’s begin with an overview of the VSD solution.
Power stations are of particular interest to me as our Visual Smoke Detection (VSD) technology was first used to protect power station turbine halls.
The voluminous spaces of these halls and the issue of smoke stratification were key challenges.
VSD uses intelligent video algorithms to identify and analyse the behaviour of smoke patterns and also flame behaviour. Visual monitoring at the very point of fire danger is particularly suited to voluminous areas.
Footage can offer visual verification (usually done on site), large area monitoring and can direct the use of fire suppressants and emergency services.
Temperature sensing (thermopile) capabilities have been incorporated within one system for the first time by FireVu. Thermopile finds its own in situations where fires are not immediately obvious or have not broken out such as overheating turbines. Operation managers can monitor temperatures on the display, without the need to go into the turbine hall, and take appropriate action.
Triple-IR flame detectors compare three specific wavelength bands within the IR spectral region and their ratio to each other allowing the detector to distinguish between non-flame IR sources and actual flames which emit hot CO2 in the combustion process. This reduces detection range and the immunity to false alarms can be significantly increased.
Detecting IR energy emitted by objects takes away reliance on visible light and so obscured conditions should not affect its effectiveness although it still needs a direct line of sight.
IR gives much of the VSD solution, however the latter offers accompanying video, which provides better situational awareness in the event of a fire. It also helps determine the most appropriate action that should be taken, such as triggering an overall suppression system.
ASD is a highly sensitive technology.
It can detect smoke early, which is particularly valuable where a fire develops in obscured or difficult to access locations, such as a turbine, or perhaps in environments that have dangerous and toxic substances. Yet, the sensitivity to distinguish between smoke and dust in early stage fires can be problematic although systems are improving. Moreover, it requires that smoke hits detectors, which can be challenging if smoke stratification is a possibility.
Safety professionals have to analyse their requirements against the solutions available, be it VSD, ASD, IR or another system. Detecting nuclear power station fires requires planning, frequent maintenance and continual assessment. The dangers of a serious incident make this effort essential. Effectiveness should be the driver for investment in fire detection and greater protection of the these high value assets.