Hi Guys:
Just would like your suggestions into the following article, any comments are welcome.
Thank you
_____________________________________________________________________
1 Title: Future Concept of Fire Detection Technology Based on Mass Spectrometer & Artificial Intelligence incorporating
(Material Fire Code) MFC
2 AbstractMass Spectrometry technology may be the alternative and effective future method of fire smoke detection. Since it can precisely distinguish the compound’s molecular of fire aerosol to the nanoscale, from any other sort of particulates not genuinely produced by fire signature. Also the introduction of the Material Fire Code MFC as the materials unique ID, and scanned by the fire alarm system, this will add substantial sophistication to addressable fire alarm system, by not only locating the room actually in fire condition, but even the burning material in question within the same room.
3 Key words: Mass Spectrometry, fire detection technology, false alarms, signal processing and algorithms, artificial intelligence, Nanotechnology, material fire code MFC.
4 Introduction:Advancement in microelectronics, information technology and fire science have given substantial progress in fire detection technologies over the last decades. Unwanted alarms rate has been quietly reduced according to statistics but relatively, not enough yet. Since the real world false alarms rate is far serious than the one recorded and compiled by statisticians in the field.
Various new technologies have been proposed such as the video fire detection system, meanwhile the existing ones have been upgraded, namely in regards to multi criteria fire detectors based on advanced signal processing algorithms. Yet, any survey of-the-state-of-the-art fire detection technology will be out of date in few years, if it’s still struggling between the following major challenges:
1. False alarms reduction rate.
2. High sensitivity.
3. Better understanding of fire physics.
4. Cost effective.
In more precise words the existing traditional smoke detection technologies, in regards to ionisation and optical methods have been stretched to their limits, its development is logarithmically evolving, it may end up soon in a situation where the more money spent the less substantial outcomes revealed.
Above those challenges, fire detection main dilemma is that, it covers a wide range of fire scenarios from smouldering to flaming combustions, from small lift motor room to very large multiple occupancies buildings, and unfortunately there is no single standard solution that can be applied to all those wide range of fire scenarios. Basically solutions are formulated to a particular simple fire scenario, of a given space, while neglecting some times some environmental parameters and variables of quite minor impacts… likewise, there is very less probability, that particular predicted fire scenario will occur, in case of a genuine fire… etc
5 Existing technology of sensing particulates: Smoke detectors senses the aerosol produced by flaming or smouldering combustions, either sensed by a spot detectors or drawn through ducts into the sensing unit (aspirated system), the sensed particulates from the air could usually be any thing such as: smoke, dust, powder, steam, fog…etc. The traditional detection process is mostly based on particulate sensing either by light beam scattering (optical) or ionised air (ionisation) type detectors…etc, the principle is based on either ‘increase in percentage of obscuration’ or ‘reduction in current/voltage’ [1]…, those variations are linearly proportional to the total amount of particulates entering the sensing chamber, regardless of what sort of particulates?
The fire detector in this case works then just like a basic ‘lift monitoring system’ when it sense an equivalent weight of say 16 person it generates an overweight warning signal, regardless of what sort of weight is? Are they humans, goods or animals…etc?
There are yet number of reasons and signs proving these existing sensing particulates methods will be out of date soon:
1. Ionisation detector usually the one suitable to detect smaller particulates produced by rapid growing flaming fires, it has been banned from some countiers due to its radioactive source Americium 241, likewise there are severe restrictions into their stores and transportations, which make its use declining rapidly every where. [2]
2. Optical detector is more suitable for large particles of slow smouldering fires. Therefore, they have difficulties in detecting flaming fire with small smoke particles. In addition, non fire-generated particles, such as dust and other nuisance aerosols may cause false alarms.
3. 95% of deaths in fire victims were caused during smouldering phase of fire, and while the victims where sleeping 4/5 have been toxically died [3]. And both fire detection methods are unable to detect the poisonous gases such as CO detector.
4. Both current fire detections technologies are unable to detect the elements and species delivered by a genuine fire signature.
Certainly, signal processing algorithms have advanced the intelligence of these traditional methods, by incorporating the parameter of heat into the same smoke detector (multi sensor), While in fact the parameter of ‘heat’ comes a bit late in the majority of flaming and smouldering fires in terms of early sensing, which then put the sensitivity at risk, if using the parameter of fixed temperature combined with optical, this paradox problem might be overcome for a mid-term solution, if adopting the solution of combined CO/optical with rate of rise heat parameter ...etc, this whole combination build with a further advanced signal processing algorithms, would give better efficiency then the combined fixed heat/optical detector, along with multi mode operations to be configured according to the environment to be used within. But for long-term vision, a radical solution has to take place by brain storming from the beginning the whole analogy of fire detection’s methods, towards other alternatives based on detecting the compounds species at molecular scale, produced particularly from fire signature and not from any other malicious source.
6 From particulate sensing to compounds sensing: Mass spectrometry is a well proven and reliable tool, it’s used to deal with particulates to the nanoscale, it has been applied successfully during the last few decades into other advanced applications and research of both industrial and academia, just to state some; satellites and spacecraft, biotechnology, biochemistry, geology, medical equipment…etc, and incorporating Nanotechnology within the future fire detection systems is really paramount to success.
Mass spectrometer is a sensitive and accurate detector to the species molecular scale or nanoscale; it can even weight the molecular mass of any element with an accuracy of 0.01%, also with advanced software such the one based on Fourier Transform Infra-Red (FTIR) [4, 5] and Chromatography process, it can even trace quantities of contaminants and toxins which might be harmful to human beings, if their threshold level is exceeded into the surrounding environment, such as: CO, CO2, HCN, cyanide…and many other hydrocarbons. The only troublesome with this approach is that, it’s likely to be effective only in a very early stage of fire combustion, so once the fire has reached a serious event, where lots of smoke and heat surrounds the area, it would be practically impossible to decipher the mixture of numerous compounds in relation to their original burned material.
7 How Mass Spectrometer works?In a very brief description, the mass spectrometer has three essential functions or stages: The ion source, mass analyzer and the detector. See, fig: 7-1.
The ion source: In order to measure the characteristics of individual molecules of air sample, a mass spectrometer converts them to ions usualy to cations (positive ions, which are molecules or atoms that lost one electron), so that they can be moved about and manipulated by external electromagnetic field.
The mass analyzer: The ions are sorted and separated according to their masses and charges. By means Lighter ions are deflected more than heavier ions. Thus, the separation of ion’s species from each other
The detector: The positions of separated ions on the detector are a function of their masses; the results are displayed in a specter graph style, ready to be analyzed.
See, fig: 7-2.
8 How can Mass Spectrometer be applied to fire alarm systems?Firstable, mass spectrometer based fire alarm system has to be designed and manufactured such an aspirated system architecture style, while the processing unit is extended by ‘sniffers’ or ‘perforated ducts’. And artificial intelligence software should be in charge of comparing the spectral of the sensed compounds against its vast spectral database. The aspirated style fire alarm system has been well proven, in terms of high sensitivity, by drawing the air samples to the processing unit, instead of passively waiting for the air samples to reach the 'spot detector'.
Secondly, every liquid and solid material has a unique chemical representation, however once the combustible material gets in contact with enough quantity of heat and oxygen, ignition takes place, during the first few seconds of combustion, the first quantity of its molecules gets the first chemical reaction to get transformed to gases, with new chemical forms or representations, by means the same material just took a new unique chemical representation, with a particular compounds and proportions.
Therefore, if the same experiment is repeated several times with the same sort of combustible material or furniture, in quite similar environmental conditions and in a given space and time…etc it will give to some extend, during its first seconds of combustion, the same proportions and compounds which can characterise the burned material, this analogy enables us to set out a unique (Material Fire Code) MFC for each material or furniture in a building. The MFC code could be considered as the DNA of that combustible material or furniture, in terms of species and proportions produced during combustion.
Therefore, if each manufactured material or furniture is labelled with MFC code, which should take the form of a bar code, and once deciphered, it gives normally compounds produced from that material in its combustion stage.
As a result, if in the future before any sort of manufactured combustible material and furniture is launched into the market either solid or liquid ..., a sample is sent to the (Fire Testing Centre) FTC, to get it burnt, tested, analysed and then assigned an MFC code, even the wall paint should be labelled with MFC code, the MFC code can be scanned by the fire alarm system once the material is purchased, placed or fitted; the fire alarm system should then be able to recognise its chemical or gas compounds in its first few seconds of combustion, the fire alarm system must be then of sufficient memory space, to record a full spectral database of all furniture and material’s MFCs in the building. The artificial intelligence embedded fire alarm system should be in charge of sensing that particular species of gases related to the burned furniture or material, and generating ‘analogue values’ instead of one analogue value, stating not only the fire threshold level is reached but also, additional information about the burned material in question, the names of toxic gases, and the location of the combustion within the room…etc
9 Conclusion:As a result, all manufacturers of combustible materials and furniture will have to collaborate indirectly to fire industry. To achieve that, the legislation and standard has to be deeply involved into the matter. This will encourage the manufacturers to drop down the unit price, to its lowest affordable rate, so long as a large scale of customers will purchase the equipment.
Certainly, the introduction of MFC code may look long and tedious requirement, or probably unnecessary process for the majority of public applications, Therfore, further deep researches, testing and analysis have to be carried out to prove the existence and reliability of the MFC code and then to apply this sophistication to high tech applications at a first stage, in fact it may prove beneficial even for future fire investigations, in addition to reducing considerably false alarms rate caused by traditional smoke detectors (optical and ionisations).
Finally, after the first generation fire alarm systems, which are standalone and batteries powered, then the second generation which are conventional fire alarm systems that detect fires by zones only, also the third generation which are analogue addressable fire alarm systems, with their various versions, the fourth generation could be the mass spectrometer and artificial intelligence based fire alarm systems that could even record MFCs codes and detects fires by materials and species within the same room, it would be a Nanotechnology based fire alarm system.
Author:M C Benzerari
Elec.Eng TMIET
Commissioning Engineer
Software Developer
Drax UK LtdIt's now downloadable from FSE Fire Safety Engineering magazine of 08-02-09. click below;
http://www.fseonline.co.uk/articles.asp?article_id=8429&viewcomment=1
