Application of Linear Optical Fiber Temperature Detector in Roof Buildings

Most of the historical buildings and ancient sites with wooden structures have roofs made of plant materials such as thatch, bark, and thin wood. In the hot summer, such buildings are more likely to catch fire from the roof surface. Once a fire occurs in a building, the flames will develop rapidly, which may not only cause the ancient buildings and ancient sites to burn down, but also spread to adjacent buildings. Traditional fire detectors are mainly used for early detection of fires inside buildings, and the early detection effect of fires on the roof surface is not very obvious.

This paper sets up a building model with a plant material roof outdoors, sets a linear fiber optic temperature detector on the roof surface to collect data, and verifies the detection performance of the linear fiber optic temperature detector using a fire judgment algorithm through combustion tests of various materials. The test proves that by setting a linear fiber optic temperature detector on the roof surface, it is possible to effectively realize the early detection of fires in ancient buildings and ancient sites with plant material roofs.

1 Fire judgment algorithm

The flow chart of the fire judgment algorithm is shown in Figure 1. The linear fiber optic temperature detector is set on the outer and inner surfaces of the roof, and outputs analog values ​​as the ambient temperature changes. The analog value is converted into a detection temperature value tn after data processing. tn is compared with a preset fixed temperature value tsl in the fixed temperature comparison part. When it is greater than tsl, a fire alarm signal is issued.

Figure 1 Flowchart of fire judgment algorithm

The detected temperature value tn is input to the temperature difference detection part and compared with the reference temperature setting value tc at the same time, and the temperature difference value is Δt. The reference temperature value tc is the corrected temperature value calculated each time in the reference temperature correction part according to the following formula.

tc＝tc'+0.16(tn-tc)

tn: the detected temperature value of this time
tc: the reference temperature value of this time
tc': the reference temperature value of the next time

The temperature difference detection part inputs the calculated temperature difference value Δt to the fire judgment part. When Δt is greater than the pre-set temperature value ts in the fire judgment part, a fire alarm signal is issued.

Figure 2 is a curve of the reference temperature value tc and the temperature difference value Δt changing with time when the detected temperature value tn of the linear optical fiber temperature detector rises linearly. As shown in Figure 2, when the detected temperature value tn represented by the solid line rises linearly according to a certain trend, the reference temperature value tc calculated according to the formula, as shown by the dotted line, also rises according to the same trend as time changes. Therefore, the temperature difference value Δt tends to a constant value after a period of time.

Figure 2 Detection temperature value tn and reference temperature value tc

Figure 3 is a time variation curve of the temperature difference Δt between the detection temperature tn and the reference temperature tc, with the temperature rise rate of the detection temperature tn as a parameter. It can be seen from the figure that the greater the temperature rise rate of the detection temperature, the higher the temperature difference tends to be; the smaller the temperature rise rate of the detection temperature, the lower the temperature difference tends to be.

Figure 3 Temperature difference time variation with the detection temperature rise rate as parameter

2 Data collection test

The temperature of thatched roofs is high in summer, so the temperature data of the thatched roof surface is collected through the test, the causes and characteristics of false alarms are analyzed, and real fires are effectively distinguished.

As shown in Figure 4, the roof material of the building model is thatch. The roof size is 240cm long, 130cm wide, 35cm thick, with an inclination angle of 45 degrees and placed east-west. The thatched part is 220cm long and 110cm wide.

Figure 4 Building model

As shown in Figure 5, linear fiber optic temperature detectors are set vertically along two lines on the roof surface, with an interval of 50 cm. Thermocouples and illuminance meters are also set on the roof surface. The data collection time interval is 16 seconds.

Figure 5 Fiber Optic Setup Diagram

The test results are shown in Figure 6. The detected temperature of the linear optical fiber temperature detector increases with the rise of the sun, and decreases when the weather changes from sunny to cloudy. The test data proves that the detected temperature of the linear optical fiber temperature detector has a great relationship with the illumination of sunlight.

3 Combustion test

Linear fiber optic temperature detectors, point infrared flame detectors, and point ultraviolet flame detectors were set on the outer surface of the model roof, and linear fiber optic temperature detectors and point smoke detectors were set on the inner surface of the roof. The detection temperature value of the linear fiber optic temperature detector is tn, the fixed temperature judgment value tsl is set to 60ºC, and the temperature difference judgment value ts is set to 7ºC. The roof materials are thatch, cypress bark, and thin wood for combustion tests.

3.1 Thatched roof combustion test

The test conditions are set as a fire on the roof surface and wind blowing over the roof. The test results are shown in Figure 7. The linear fiber optic temperature detector sends out a fire alarm signal almost at the same time as the flame is emitted. Point ultraviolet and infrared flame detectors also alarm quickly for open flames on thatched roofs. The point smoke detector set on the inner surface of the roof has a late alarm time, mainly because the wind blows over the roof and it is difficult for smoke to enter the building.

Figure 7 Thatched roof combustion test

3.2 Cypress bark roof combustion test

The test conditions were set as a fire on the roof surface, and the wind blew directly to the roof, making it easy for smoke to enter the room. The test results are shown in Figure 8. The linear fiber optic temperature detector issued a fire alarm signal in the smoldering stage. When the cypress bark is burning, the smoldering fire characteristics are more obvious. When the smoke in the room reaches a certain concentration, the point-type smoke detector sends a fire alarm signal; while the point-type ultraviolet and infrared flame detectors are mainly based on the detection principle of flames, so the alarm time is relatively late.

Figure 8 Cypress bark roof combustion test

3.3 Thin wood roof combustion test

The test conditions are the same as those of the cypress bark roof combustion test. The test results are shown in Figure 9. The linear fiber optic temperature detector sends out a fire alarm signal in the smoldering stage. When the thin wood burns, the flame is extremely small and is almost in a smoldering state. Therefore, similar to the cypress bark roof combustion test, the point-type smoke detector has a faster alarm time, while the point-type ultraviolet and infrared flame detectors have a slower alarm time.

Figure 9 Thin wood roof combustion test

试验结果表明，采用该火灾判断算法的线型光纤感温探测器可以有效地防止误报，同时和设置在屋顶外表面的点型红外、紫外火焰探测器及设置在屋顶内表面的点型感烟火灾探测器的报警时间相比较在火灾探测方面具有显著的有效性。

4 结束语

通过在室外设置植物性材料屋顶的建筑模型的燃烧试验，证明采用本文火灾判断算法的线型光纤感温探测器应用在采用植物性材料屋顶的古建筑和古代遗址中，可以有效地实现火灾早期探测，具有较强的实用性和有效性。

参考文献：
[1] 山下邦博等．光纤探测器在古建筑中的应用．日本报知机株式会社技术论文集，1998年总第14期．