JPH02271879A - Fire judgement device - Google Patents

Abstract

PURPOSE:To sharply improve the reliability of fire judgement by providing the sensors detecting the physical phenomena accompanying a fire, a primary calculating part of fire information based on the detected informations, and a fire judgement part judging a fire from the changing amount of the fire source information. CONSTITUTION:The outputs detected by a temp. sensor 10, a smoke concn. sensor 12 and a gaseous CO concn. sensor 14 are connected with the signal line of a sampling circuit part 16 which is connected with a fire source information calculating part 18 where a calculation model of fire informations is set in advance. By carrying out the reverse calculation of the calculation model, the heat quantity, the smoke quantity and the gas quantity generated in a fire source are calculated from temp. theta, smoke concn. CS, and gaseous CO concn. G. These calculated fire source informations are given to the fire judgement part 22, and a fire is judged when the changing amount exceeds the starting level set in advance or by carrying out a forecast calculation when it exceeds the starting level. An alarm display part 24 displays the fire alarm according to the obtained results for fire judgement.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention introduces a mathematical model for fire simulation to calculate fire source information such as calorific value, smoke amount, and amount of gas generated to judge a fire. This invention relates to a fire detection device.

[Prior Art] Conventional fire determination is based on the determination of fire based on temporal changes in secondary temperature, smoke concentration, and CO gas concentration accompanying a fire detected by a sensor. In the simplest method, a fire is determined when the sensor detection value exceeds a set threshold. Furthermore, when the sensor detection value is differentiated and the drying time change rate exceeds a predetermined value, it is determined that there is a fire. Furthermore, a method has been proposed in which future changes are predicted based on changes in sensor detection values up to the present time using linear function approximation or quadratic function approximation to determine a fire.

[Problem to be solved by the invention] However, such conventional fire detection is based on detecting heat (temperature), smoke, gas, etc. generated secondarily by a fire using sensors. It is said that However, the process of fire occurrence is diverse, and the combustion materials and surrounding environmental conditions interact in a complex manner during the fire expansion process, so the properties of the secondary generated heat (temperature), smoke, gas, etc. are also diverse. There are many difficulties in determining fire accurately and quickly from such diverse sensor information.
We continue to strive to get closer to our ideal.

The present invention was made in view of the above situation, and it is possible to detect fires such as calorific value, smoke generation amount, gas generation amount, etc. from sensor information such as heat, smoke concentration, CO gas concentration, etc. generated secondarily due to fire. The purpose of this invention is to provide a completely new fire determination device that determines a fire by calculating information on the source itself (secondary fire source information) by introducing a mathematical model for fire simulation.

[Means for Solving the Problem] In order to achieve this object, the present invention includes a sensor that detects physical phenomena accompanying a fire, such as temperature, smoke concentration, and CO gas concentration, which is installed in a fire monitoring area. : A fire source information calculation unit that calculates primary fire source information such as the calorific value, smoke emission amount, and gas generation amount of the fire source based on the detection information of the sensor; and the fire source information calculated by the fire source information calculation unit. The system includes a fire determination section that determines a fire based on the amount of change in the original information.

[Function] In the fire judgment device of the present invention having such a configuration, the temperature, smoke concentration, and The calorific value, smoke amount, and gas generation amount of the fire source itself can be calculated from the CO gas concentration, and a fire can be determined from the amount of change in the primary information of the fire source.

The primary information on the source of a fire, i.e., the amount of heat generated, the amount of smoke generated, and the amount of gas generated, are originally uniquely determined without being affected by the combustion materials or surrounding environmental conditions, and the accuracy of calculating the fire source information depends on the accuracy. By improving this, the reliability of fire judgment can be greatly improved.

[Embodiment] FIG. 1 is a block diagram showing an embodiment of the present invention.

In FIG. 1, 10 is a hot water temperature sensor, 12 is a smoke concentration sensor, and 14 is a CO gas concentration sensor. Each sensor is installed on the ceiling surface of the fire monitoring area, and has temperature θ, smoke concentration Cs, and CO gas concentration G. is detected in an analog manner and a detection signal is output. The detection outputs of the temperature sensor 10, smoke concentration sensor 12, and CO gas concentration sensor 14 are connected to a sampling circuit 16 by signal lines, sampled at predetermined intervals by the sampling circuit 16, and then converted into digital signals by an AD converter. is output.

In this embodiment, one fire monitoring section includes a temperature sensor 10, a smoke concentration sensor 12, and a CO gas concentration sensor 14.
Although one sensor is installed at a time, multiple sensors of the same type may be installed if necessary. In addition, as for the signal transmission method between each sensor and the sampling circuit section 16, in addition to the direct connection method that sends the analog signal as it is, there is also a polling method that polls the sensor side from the sampling circuit section 16 side and sends out a detection signal, etc. A signal transmission method can be applied.

Following the sampling circuit section 16 is the fire source information calculation section 1
8 is provided. A calculation model for fire simulation is preset in the fire source information calculation unit 18, and by executing the inverse calculation of this calculation model, the calorific value, smoke emission amount, and Calculate the amount of gas generated.

The fire source information calculation unit 18 is given various initial values for executing the fire simulation calculation model from the initial value setting unit 20, and is set according to the conditions of the fire alert area where the sensor is installed. After receiving the initial value settings, the next fire origin information is calculated from the sensor detection information.

The fire source information calculated by the fire source information calculation section 18, that is, the calorific value, the smoke amount, and the gas generation amount, is given to the fire determining section 22, and the fire determining section 22 calculates the amount of change in each fire source information. A fire is determined to occur when a preset activation level is exceeded, or fire prediction calculations are performed according to a linear or quadratic function using the fire source information obtained up to that point when the activation level is exceeded. to determine if there is a fire. When the fire determination section 22 obtains a fire determination result based on the fire origin information, a fire determination output is given to the alarm display section 24, and a fire alarm is displayed.

Next, the principle of calculating fire source information performed by the fire source information calculation section 18 of FIG. 1 will be explained in detail.

Currently, mathematical models for analyzing the characteristics of indoor fires can be broadly classified into field models and zone models. These mathematical models calculate the flow of indoor temperature or smoke concentration from the calorific value or smoke amount of a fire source by solving differential equations.

Here, the field model is based on a single indoor closed space in which all entrances, doors, and windows are closed, and the room is divided into several hundred or more units of several tens of centimeters or less, and the mass conservation equation is calculated for each divided space. It calculates the flow of indoor temperature or smoke concentration by applying the momentum conservation equation, energy conservation equation, state equation, and boundary conditions. A feature of this field model is that it performs detailed calculations on the concentration in subdivided rooms, making it possible to accurately grasp phenomena during fires such as temperature distribution and smoke concentration distribution. However, with field models, calculations are performed for each of hundreds of divided spaces, which requires an enormous amount of calculation time, which poses problems in terms of real-time processing, and the inconvenience of not being able to easily change the values of calculation parameters. Yes, there is.

On the other hand, the basis of the zone model is to consider one indoor closed space as a reference and to divide the indoor closed space into several layers in the vertical direction, that is, into two or more layers. The characteristic of the thin model is that it calculates the average temperature or average smoke density in the upper layer, and since it is a simple model, it requires less calculation time and can be processed in real time on a personal computer or the like. In addition, elements such as room size (ceiling area and ceiling height), ambient temperature, heat loss rate, or heat generation amount per unit time can be freely changed, and conversion is easy and boundary layer It has the advantage of being able to determine the height up to the point, and being able to get a rough idea of the dangerous layer inside the room. However, the accuracy of the zone model is not very good because it uses a difference method for calculations and the number of terms used in calculations is truncated to improve calculation time.

Therefore, in the fire source information calculation unit 18 of the present invention, a field model is used when detailed and accurate calculation is required, and a zone model is used when real-time processing is required. good.

In the explanation of the example below, we will use real-time support even if the accuracy is somewhat low to determine the initial stage of a fire due to the fact that the information obtained at the time of fire detection is only from the output of sensors installed indoors. A zone model that allows for

The zone model has been put into practical use in various ways, but
It is not yet an established theory. Therefore, in the present invention, the two-layer model A S E T is one of the mathematical model programs analyzed by L. Y. Coopc and developed by W. D. Walton based on the theory. (A vailable S al
By applying eE Bets Time, )-B and performing inverse calculations based on sensor detection data, changes in calorific value, smoke generation amount, and further gas generation amount are calculated,
A fire is determined based on these calculation results.

FIG. 2 shows an outline of a zone model as a fire simulation calculation model calculated by the fire source information calculation unit 18 of the present invention.

The zone model in FIG. 2 is a two-layer zone model, and since it is a simple model for determining the average temperature θh or average smoke concentration Ca1l of the upper layer 28, the following conditions are set.

It is assumed that all entrances, doors, and windows in the room are closed except for slight leakage from the floor, the pressure in the room is constant, and the increase in pressure is not due to leakage from the floor. are doing.

A fire shall occur at a fire point set on the floor.
The heat and warm smoke generated from the fire point rises due to buoyancy and reaches the ceiling surface. Plume 26 formed at this time
rises, drawing in the cold air around it. The hot airflow reaching the ceiling is diffused and reaches the walls, forming a warm layer or upper layer 28. Then, a boundary 32 is created between the lower layer 30 and the air layer. Upper management 2 as the fire progresses
The boundary 32 with 8 gradually descends to the floor surface with time.

In such a two-layer zone model, the warm upper layer 2
8. It is assumed that the temperature in the lower layer 30, which is the ambient temperature, as well as the smoke concentration are uniform within each layer, and that heat exchange between the layers is performed through the plume 26.

The simulation is based on the calorific value per unit time of the combustion material, which is known in advance, and the temperature θh of the upper layer 28 and the boundary 32.
Find the height Z up to.

That is, the height Z from the fire point to the boundary 32 of the upper layer 28,
The average temperature θh and smoke concentration Csh of the upper layer 28 are given by solving the following differential equation. In addition, the initial conditions set by the initial value setting section 20 are also shown together with the differential equation. Furthermore, the same relational expression as for the smoke concentration Csh is applied to the gas concentration Gh of CO gas.

Initial condition (-〇) Δ0f-d△Oo/d t , ΔC5r-d△Cso/
dt(t-0)ΔQ: Calorific value per unit time, ΔC3
: Production amount per unit time ΔQO: Initial calorific value, ΔC
SO: Initial punishment i S ceiling area, H1 ceiling height, F
: Height of fire point, CP: Specific heat, LR1 heat emissivity, LC
: heat loss rate, θ0: ambient temperature, 9: gravitational acceleration,
ρ: Density of air In such a two-layer zone model, changes in temperature or smoke density are captured as changes in calorific value per unit time or changes in the amount of smoke generated per unit time, and the boundary Regarding the height 32, the change in calorific value per unit time is captured as a change per unit area. Further, the differential equation can be solved by an improved Euler's method.

In the present invention, the temperature sensor 10. The detected temperature θ and the smoke concentration Cs of the smoke concentration sensor 12 are each set to 2.
The average temperature θh and average smoke concentration Cxh of the upper layer 28 in the layer model are treated as the average temperature θh and the average smoke concentration Cxh, and changes in the amount of heat generated and changes in the amount of smoke generated are calculated.

Next, the processing operation of the embodiment shown in FIG. 1 will be explained with reference to the flowchart shown in FIG.

In FIG. 3, when the device is started, first, in step S1, the initial value setting section 20 sets the fire source information calculation section 18 to
Initial value setting is performed. Setting this initial value is as described in 2.
All other values except CI, C2, ΔQf, ΔCsf, amount of heat generated per unit time ΔQ1 and amount of smoke generated per unit time ΔCs shown in the proviso to the initial conditions of the differential equation of the layer zone model.

When the initial values are set in step S1, the process proceeds to step S2, where the sampling circuit unit 16 receives the temperature θ from the temperature sensor 10, and the smoke concentrations C8 and C from the smoke concentration sensor 12.
Data sampling of the gas concentration G from the O gas concentration sensor 14 is performed.

Next, through the processes of steps 83 to S6, the change ΔQ in the amount of heat generated is first determined. That is, in step S3, ΔG is initialized, and in step S4, the average temperature θh and the height zh of the smoke layer are calculated by calculating ASET-B. continue,
Average temperature θh calculated by ASET-B in step S5
The process returns to step S3 and gradually increases the calorific value ΔQ until the absolute value of the difference between the temperature and the detected temperature θ detected by the temperature sensor 10 becomes less than a predetermined value, for example, less than 0.001, and the condition of step S5 is satisfied. In step S6, ΔG at the time when the change in the amount of heat generated is determined as a change in the amount of heat generated.

Next, proceeding to step S7, the smoke generation amount ΔCs and the gas concentration ΔG are set, and the temperature θ that has already been determined at this point is
Using h, smoke concentration Csh and gas concentration Gh are calculated by the calculation of ASET-B in step S8. Subsequently, in step S9, the absolute value of the difference between Cs and Csh and the absolute value of the difference between G and Gh are determined.
Return to step S7 and gradually increase ΔG and ΔCs until the absolute value of the difference becomes less than a predetermined value, for example, 0°001, and when the condition of step S9 is satisfied, ΔG and ΔCs are
Cs is determined as a change in the amount of smoke generated, and ΔG is determined as a change in the amount of gas generated (step 10).

Next, the process proceeds to step Sll, where it is determined whether or not the change ΔQ in the calorific value obtained in step S6, the change ΔCs in the amount of smoke generated, and the change ΔG in the amount of generated gas obtained in step SIO are equal to or higher than a preset fire judgment calculation activation level. Determine whether If step Sll is equal to or higher than the operation activation level, step 8
Proceed to step 12 Calorific value 601 Smoke amount ΔC obtained so far
A predictive calculation is performed using s and the generated gas amount ΔG.

As a predictive calculation, for example, Newton's backward interpolation formula is used to determine a fire. As the calculation for fire determination in step S12, in addition to the prediction calculation, changes in the primary difference and/or secondary difference from the current time point when the activation level is exceeded to a predetermined sampling point before may be calculated.

Figure 4 shows a ceiling area of 28.81m2 and a ceiling height of 2.5m.
Calorific value 601 Smoke amount ΔCs and gas generation amount ΔG time according to the fire source information calculation unit 18 of the present invention for a fire experiment when a chair (material: cloth, urethane foam, wood, etc.) was burned in the center of a room. It is a graph showing changes in temperature θ, upper boundary height L1, smoke concentration C81, and gas concentration G.

Further, FIG. 5 is a graph when the present invention is applied to a non-fire experiment in which nine fish were successively grilled in the same room as in FIG. 4, as an example of cooking in a kitchen.

As is clear from comparing the results in Figure 4 for fire and Figure 5 for non-fire, the change in calorific value ΔQ shown in Figure 4 (a) during a fire changes as the fire progresses. A large peak is shown when the temperature θ suddenly rises. On the other hand, in the calorific value ΔQ during non-fire conditions shown in FIG. 5(a), no peaks like those observed during fire conditions are observed. Therefore,
A fire can be determined based on the correlation between the temperature θ and the calorific value ΔQ shown in FIG. 4(a), which both rise linearly.

Furthermore, in the event of a fire in Figure 4, the difference between the calorific value ΔQ shown in Figure 4 (a), the smoke generation amount ΔCs shown in Figure 4 (b), and the gas generation amount ΔG shown in Figure 4 (C) is There is a correlation in which the amount of change increases at a peak, and such calorific value 6
01 By looking at the correlation between the three types of smoke generation amount ΔCs and gas generation amount ΔG, a more accurate fire judgment can be made.

On the other hand, when there is no fire as shown in FIG. 5, there is no correlation between the amount of smoke generation ΔCs and the amount of gas generation ΔG with respect to the amount of heat generation ΔQ, and from this it is possible to reliably determine that there is no fire. In addition, the amount of smoke generated ΔCs and the amount of gas generated ΔG when there is no fire
In the case of fire, the change pattern is similar to that during fire in Fig. 4, but the amount of change itself is small, so it is possible to distinguish between non-fire and fire.

Figure 6 shows the experimental results of the present invention showing the time change of the calorific value ΔQ with respect to the temperature θ when the room size is changed. The changes in ΔQ were in good agreement, and it was confirmed that according to the present invention, the same change in calorific value ΔQ can be obtained for the same fire regardless of the size of the room. The same applies to the amount of smoke generation ΔCs and the amount of gas generation ΔG.

Next, a specific example of the fire determining section 22 shown in FIG. 1 will be described.

In this embodiment of the fire determination unit 22, a fire is determined by a correlation calculation using two of the calorific value 601 smoke generation amount ΔCs and the gas generation amount ΔG obtained from the fire source information calculation unit 18. It is characterized by

First, the correlation coefficient R is defined by the following equation.

R=Sx7/F-Ichigo (1) Here, each of SIL Sr and Sr is expressed by the following formula, however,
X, Y: any value of ΔQ, ΔCs, ΔG Weighting is performed by multiplying the determined composite vector by the absolute value ID+, and the weighted correlation coefficient Ro
seek.

In other words, the absolute value of the composite vector used for weighting is
It is determined by the following formula.

DI=lUi+Vjl (3) However,
U, V; ΔQ1 ΔCs, ΔG are selected and scaled independently for each value i, j + unit vector of each dimension. Therefore, the weighted correlation coefficient R9 is the correlation coefficient R and Since the composite vector D changes with time, it is expressed as a function of time by the following equation.

R, (t) = R(t) X I D (t) l
(4) For this reason, the correlation value R calculated at a certain point in time
is weighted depending on the absolute value IDI of the composite vector of the two detection quantities U and V at that time, and the correlation value R8 is obtained by weighting the correlation value so that it becomes larger as U and V become larger.

Figure 7 (a) shows the above (1) to (4) using the heat generation amount ΔQ and smoke generation amount ΔCs during a fire shown in Figure 4 (a) and (b).
The weighting obtained by the formula indicates the time change of the correlation value RD,
Correlation value R8 shows a large peak change, and threshold value R5
It can be determined that there is a fire when the

Further, FIG. 7(b) is data obtained by differentiating the correlation value RD shown in FIG. 7(a), and this differential data also shows a remarkable change that can be determined to be a fire.

Figure 8 (a) shows the heat generation amount ΔQ and smoke generation amount ΔCs in non-fire conditions shown in Figures 5 (a) and (b) above (1) to (4).
It shows the correlation coefficient R9 obtained according to the formula, and the correlation coefficient R8 in this case is a value lower than the threshold value RL,
It can be determined that there is no fire. Note that FIG. 8(b) shows the differentiated values of FIG. 8(a).

In the above embodiment, the calorific value 601, the amount of smoke generated ΔCs, and the amount of CO gas generated ΔG are calculated as secondary fire source information, but other primary fire source information includes the fire Since ions are generated by flames, an ion sensor is installed in the fire monitoring area, and the amount of ions generated from the fire source is calculated as primary source fire information from the detection information of the ion sensor to make a fire judgment. It's okay.

[Effects of the Invention] As explained above, according to the present invention, the temperature, smoke concentration, and CO
It is possible to calculate the amount of heat generated, the amount of smoke, and the amount of gas generated from the fire source itself from secondary phenomena accompanying a fire, such as gas concentration, and to judge a fire based on the amount of change in the primary information of the fire source. Therefore, it is expected that the reliability of fire judgments will be greatly improved by minimizing erroneous judgments in which non-fires are judged as fires, without being affected by combustible substances or surrounding environmental conditions.

[Brief explanation of drawings]

Fig. 1 is a configuration diagram of an embodiment of the present invention; Fig. 2 is an explanatory diagram of a two-layer zone model used in the present invention; Fig. 3 is a flowchart showing the operation processing of the present invention; Fig. 4 is a diagram during a fire experiment. An explanatory diagram showing the change in fire source information over time according to the present invention; Figure 5 is an explanatory diagram showing the time change in fire source information according to the present invention during a non-fire experiment; It is an explanatory view showing a time change of heat generation amount by the present invention when changing. Figure 7 is an explanatory diagram showing the temporal changes in the correlation coefficient and its differential value obtained from the calorific value and smoke amount obtained in the fire experiment in Figure 4; Figure 8 is the non-fire experiment in Figure 5. It is an explanatory diagram showing a time change of a correlation coefficient calculated from calorific value and smoke generation amount obtained in , and its differential value. 10: 12 = 14 = 16 = 18: 20: 24: Temperature sensor Smoke concentration sensor CO gas concentration sensor Sampling circuit section Fire source information calculation section Initial value setting section Fire judgment section Alarm display section

Claims (1)

[Claims] 1. A sensor installed in a fire monitoring area to detect physical phenomena associated with fire, such as temperature, smoke concentration, and CO gas concentration; a fire source information calculation unit that calculates primary fire source information such as the amount of smoke emitted and the amount of gas generated; a fire determination unit that determines a fire from the amount of change in the fire source information calculated by the fire source information calculation unit; A fire determination device characterized by comprising;