Analog fire sensor

Fire Sensor

This invention relates to an analog fire detector for use in a fire alarm system, and more particularly to an analog fire detector that detects a fire-like condition such as smoke density, temperature or gas concentration or the like and transmits it to a central signal station and detects a fire from a central signal station.

A conventional fire alarm system uses so-called on-off type fire detectors which are adapted to close their contacts when a fire is detected and to send a fire signal to a central signal station. However, this conventional fire alarm system is not able to successfully perform the two functions required of a fire alarm system, namely early detection of a fire and prevention of false alarms.

To solve this problem, an analog-type fire alarm system has recently been proposed in which an analog quantity detected by detectors, such as temperature or smoke density, is transmitted as such to a central signal station and fire detection is performed by an analog quantity detected at the central signal station.

However, if all the detection data of the analog fire detectors is received and the fire detection processing is performed for each detector, the sampling period of the sensor data with the poll becomes long as the number of detectors increases. In addition, the central signal station has to perform a complex fire determination and the processor of the central signal station becomes busy so that the rotation of other sensors must be interrupted. The processing work of the processor of the central signal station thus becomes too large. As a result, the number of sensors used is limited.

In addition, since the analog information transmitted from the sensors contains variations due to interference, when the data as such is used, an incorrect determination of the occurrence of a fire can be made. Therefore, pretreatment must be performed to remove unwanted interference components contained in the received analog information. This pretreatment further increases the CPU load on the central signal station.

The present invention has been made in order to avoid the above-mentioned problems, and it is an object of the present invention to provide a fire alarm sensor comprising an analog sensor for detecting a state such as temperature, smoke density, etc. in analog form and a reference analog value and a predetermined threshold value. to compare. The invention is characterized by what is stated in claim 1.

According to the present invention, the number of analog fire alarm sensors for which the central signal station performs a fire determination is much smaller and the load on the processor of the central signal station is considerably reduced. In addition, disturbances in the area below the threshold level of the sensor can be eliminated.

Fig. 1 is a block diagram of a first embodiment of the present invention, Fig. 2 is an explanatory view showing data averaging, Fig. 3 is an explanatory view showing the relationship between an analog fire sensor threshold level and a threshold level used by a central signal station for fire determination, Fig. 4 is a flow chart of central signal station processor processing, and 6 are explanatory images, Fig. 7 is an explanatory view of the quadratic function prediction calculation of the central signal station processor, and Fig. 8 is an explanatory view showing the time required to reach the hazard level calculated by the processor of the central signal station processor.

Figure 1 shows a basic embodiment of a first embodiment of the present invention. 1 is a central signal station and includes a processor CPU that performs fire determination processing.

The analog fire detectors 3 are connected to the signal lines 2a to 2n conducted from the central signal station 1. The analog fire alarm sensors 3 detect the state resulting from the fire in an analogous large form, such as temperature, smoke density, CO gas concentration, quantity, and transmit the detection information, for example in electric current, in response to a poll from the central signal station 1.

Each analog fire sensor 3 operates by receiving a power supply from a central signal station, and the sensor has an analog detection section 4, which includes elements for detecting temperature, smoke density, etc. in the form of an analog quantity. 5 is a sampling circuit that samples analog detection signals at a predetermined period. The A / D converter 6 converts the sampling data from the sampling circuit 5 into digital data and it is input to the averaging section 7.

This averaging section 7 calculates a moving average and a simple average of the sampling data. The point. As explained in more detail, as shown in Figure 2, the averages (AVERAGE) of three consecutive sampling data are calculated sequentially, and then simple averages of the data provided by the six moving average calculations are computed to form one data to be transmitted to the central signal station.

This averaging processing, which comprises moving average calculation and simple averaging, acts as a digital low-pass filter to remove the higher harmonic components generated by the fundamental temperature components associated with the fire temperature or smoke contained in the analog detection signals. With this digital low-pass filter, the original signal can be reproduced truthfully. In addition, this averaging section can act as a digital filter by calculating only the moving average.

Because analog detection signals are sampled, the probability of pulse interference being sampled is reduced. In addition, even if the pulse disturbance is taken as sampling data, the average calculation can provide sufficient disturbance attenuation.

is a digital comparator acting as a comparison means for comparing the output data from the averaging section 7 with the sensor threshold represented by the reference voltage source 9, and the comparator generates a higher level signal to issue a data transmission command when the averaging information exceeds the sensor threshold.

As the average set for the comparator 8, for example, the highest normally expected room temperature, e.g. 30 ° C, can be mentioned in the case of a fire temperature indication. In this case, data transmission to the central signaling station is allowed only when detection information of 30 ° C or more is obtained.

is a call detection section which counts clock pulses transmitted from the central signal station 1, for example in voltage form, and detects an incoming call when the clock count value reaches a predetermined number, thereby giving a data transmission signal (higher level signal). The outputs of the call detection section 10 and the comparator 8 are fed to the AND gate 11. The AND gate 11 provides a higher level output to the signal transmitting section 12 when the detected analog level is higher than specified and when the call detection section 10 recognizes its own call, and ⁵ The information provided by the averaging section 7 of the 85629 is sent to the central signal station 1, for example in current mode after D / A conversion.

The central signal station 1 comprises a call control section 13, a processor 14 for performing fire determination processing, an A / D converter for converting analog signals from the sensors 13 into digital signals, and a display section 16.

The call control section 13 repeatedly outputs in voltage form a number of clock pulses corresponding to the number of analog fire detectors 3 connected to the central signal station 1, followed by a long-term reset pulse, to perform a rotation of the detectors. The voltage generated by the detection current transmitted from the sensors 3 over the resistor 17 is fed to the A / D converter 15 and it converts the voltage into a digital signal for supply to the processor 14.

The processor 14 collects analog data corresponding to the sensor addresses determined by the clock pulse counts and performs a fire occurrence determination by predictive calculation according to a function approximation, as will be explained in detail later so that the display section 16 may display a fire message together with the sensor 25 address.

In the following, the fire detection concept performed by the central signal station 1 on the basis of the sensor information of the processor will be explained.

The content of the fire detection process is divided into two parts as follows:

a. security handling for non-fire alarm;

b. predictive fire calculation according to a function approximation.

Figure 3 shows the relationship between the threshold levels used for the fire determinations of a) and b above and the threshold level set for the control of signal transmission to the analog fire detectors 3. For fire determination, the start level is set to start the predictive calculation with a function approximation and the Hazard Level is obtained to obtain the time remaining before it reaches the fire based on the result of the predictive calculation, while the threshold level of analog fire sensors is set to eliminate continuous interference below the start level.

Thus, when the detection levels of the analog fire detectors shown in white dots are lower than the threshold level, signal transmission is not performed even if polling from the central signal station 1 is performed, and only analog signals above the threshold level shown by the black dots 14 are transmitted to the central signal station 1. the amount of information represented by the points.

Fig. 4 is a flowchart of an example of the fire determination processing performed by the processor 14 of the central signal station 1. In this processing, a predictive calculation is performed by a function approximation.

First, in block 20, it is checked by a poll whether the response information is obtained.

If the response information is received, the step proceeds to the next decision block 21, where it is checked whether the last transmitted information, after averaging it and has exceeded the sensor threshold, is greater than the calculation start level as shown in Fig. 3.

The operation of the processor 14 of the central signal station 1 sequentially stores 20 sensor data LD1 to LD20 for calculation processing by function approximation.

If the last received sensor information LD20 exceeds the count start level, the step proceeds to block 22 for non-fire safety processing.

Fig. 5 shows an detection example showing the slope coefficients y1 to y3 as examples. In this case, the slope factor y1 is negative and the slope factors y2 and y3 are positive. For positive slopes, it is examined whether or not they are greater than a predetermined slope, and the number of slopes greater than one is calculated. When the number of slope coefficients larger than the slope factor one is two or more, as shown in Fig. 6, the existence of a fire possibility is determined, and the step proceeds to the next step 23 to start the predictive calculation by function approximation.

On the other hand, when, as shown in Fig. 5, the number of slope coefficients larger than one is less than two, it is found that the change in information is due to tobacco smoke, etc., and predictive calculation by function approximation is not performed.

A predictive calculation is performed in block 23 on the information that has passed the non-fire safety processing performed in block 22.

In this predictive calculation, the time change of temperature or smoke due to fire is approximated by the equation y = ax ^ + bx + c and the values ​​of the coefficients a, b and c of the quadratic function shown in Fig. 7 are obtained from the data LD1 to LD20 obtained by averaging. The coefficients a, b and c are obtained by calculating the simultaneous equations consisting of the determinants according to the least squares method by the Gaussian-Jordan method.

If the coefficients a, b, and c are obtained, the trajectory of future data changes can be determined as shown in Fig. 8.

Thus, in the next block 24, the time tr, which is the time required to reach the hazard level, is obtained as a function of its squares in Fig. 8, and at that time the remaining predicted time Tpu for reaching the hazard level is calculated.

Since the shorter the time remaining to reach the hazard level, the greater the probability of an actual fire, the time in decision block 25 is compared to, for example, a threshold time of 800 s, and if the time is less than 800 s, it is determined as a fire and a fire alarm is given in block 21.

As will be apparent from the above description, the preprocessing calculation of the processor 4 of the central signal station 1 does not have to be performed for all the sensor data. For a signal change within the level range where predictive calculation is not required, signal transmission to the central signal station 1 is blocked by the analogue fire sensor 3 and signal processing is started only when the change reaches a level which obviously requires prediction calculation so that the number of analogue fire sensors 3 to be 1 the fire determination performed by the processor 14 is greatly reduced. The load on the processor 14 of the central signal station 1 is thus greatly reduced, and the processor 14 may have additional processing power, determination capability, so that the number of sensors 3 to be connected to the central signal station 1 can be increased.

In this connection, it should be noted that the sampling circuit 5, the A / D converter 6 and the averaging part 7 of the analog fire sensor 3 can be omitted. In this case, where the analog information indicating the analog fire detector portion is provided directly from the analog fire detector, the content of the processing and determination performed by the processor 14 of the central signal station 1 is as follows:

629

a. elimination of higher harmonic disturbances in my average step 11a,

b. security handling for non-fire alarm;

c. predictive fire calculation by function approximation.

The predictive calculation of the processor 14 of the central signal station 1 by a function approximation can be performed instead of the quadratic function approximation described above alternatively by a linear function y = ax + b or by a combination of a linear function and a quadratic function.

In addition, the fire determination in the central signal section does not always have to be performed on the basis of a function approximation. A fire can be determined directly from analog data if the data has a value to be defined as a fire. Furthermore, in addition to the polling system, other types of systems can also be used as the communication system from each analog fire detector to the central signal station.

Although the fire detection processing is performed in the above example by predictive calculation by function approximation, the present invention is not limited thereto, and the fire detection processing may alternatively be performed under the control of a suitable program.