[
Large atria require creative approaches to fire safety. Here we profile an imaginative use of video smoke detection and beam detection at a hotel and apartment development in London, the design of which was the winner in the residential category of the FSE Design Awards 2006.
The development is a new build project known as Galliards Apart Hotel, which lies adjacent to the south side of Waterloo Station, fronting Addington Street in London. It comprises 18 levels: basement, ground, mezzanine, floors one through to 14, and a roof level. With around 400 hotel accommodation suites, there are also restaurant and bar areas, meeting rooms and leisure facilities, with a small retail annex to the ground floor.
The building is constructed mainly of steel and reinforced concrete, and has a glazed front aspect and an atrium spanning from the ground floor to the fourteenth floor. It has automatic sprinkler protection throughout and an automatic fire detection and alarm system to BS 5839 pt.1: 2002 category L1. Detection in the atrium consists of two beam detector sets located on the fifth, tenth and fourteenth floors, providing six in all.
Because the included a restaurant at the mezzanine level which is open to the atrium, the building control authority insisted upon additional protection to the base of the atrium by way of automatic water drenching. But water supply and pressure limitations meant that the protected area (effectively the underside of the 2nd floor to ground floor) needed to be divided into five protection zones, enabling a maximum of two zones of drenchers to operate, via electronic metron valves, at any given time. Detection and valve selection would be paramount, not only to avoid problems with the water supply, but also to avoid the possibility of extensive water damage on false activations. Cost, operational and aesthetic considerations ruled out point type detection, flame detection, sole optical beam and aspirating detection.
Video smoke detection
In order to pinpoint the exact area of any fire to enable the correct metron valve selection, a video smoke detection (VSD) option was chosen. Each of the detection zones will be covered by positioning two static overlapping CCTV cameras, so that they can detect and confirm the area of smoke (providing confirmation on the ‘double knock’ principle). The UK manufacturer of the system, however, was uncomfortable in offering its equipment as the sole means of drencher activation in cases of false activation. So to was decided to use it solely as the means for valve selection, and to introduce optical beam detection to confirm a fire within the wider area, then allowing the selected metron valve(s) to be opened.
This method suited the client’s operational needs while satisfying the requirements of the building control authority and the fire safety strategy. Activation of any stage of the VSD system would operate the building fire alarm system and its cause and effects. Upon activation of the video system, the affected area can be viewed on a monitor to assist staff in assessing the situation.
Video smoke detection is based on sophisticated computer analysis of images captured by a CCTV camera (sensor). Using advanced image processing technology, detection algorithms, and known false alarm phenomena, the system can automatically identify the distinct characteristics of smoke patterns. The fire detection industry has an abundance of known smoke signatures, and all of these are built into the system to give an accurate decision on whether or not smoke is present. It is so accurate in its analysis that it can even differentiate between steam and smoke.
The system uses standard CCTV equipment linked to a self-contained processing system, which is capable of recognising small amounts of smoke within the video image and alerting the system operator. It employs highly complex algorithms to process video information from up to eight cameras simultaneously. Under normal conditions with all eight cameras connected, the system achieves a 5Hz-frame rate for each channel.
Real-time information
The video hardware is designed to allow simultaneous, real time digitising of all eight images. This means that no information is lost or delayed, as the system does not multiplex or timelapse images. All alarm condition images are logged, time and date stamped, and stored within the system’s memory.
The VSD system detects smoke rapidly by looking for small areas of change within the image at the digitisation stage, and only passing these pixel changes to the main processor for further filtering. The video information is passed through a series of filters, which seek particular characteristics that can be associated with smoke behaviour. Further analysis is then carried out on the relationships between the filtered characteristics, to determine whether all the conditions have been met for the system to confidently predict the presence of smoke.
The system installer has the ability to vary the amount of smoke and the length of time that the smoke exists, before an alarm condition is raised, to cater for situations where there may be background smoke present. The installer may also divide the video image into zones, and programme the system to alarm only if smoke is present in two or more zones.