RU2646204C2 - Method of automatic detection of initial stages of fires in premises of hazard locations containing heat-generating equipment - Google Patents

Abstract

FIELD: fire safety.

SUBSTANCE: invention relates to methods for providing fire safety in premises of hazard locations containing heat-generating equipment and can be used in shipbuilding in ship fire and temperature alarm systems to detect fires at the initial stages of their development. A lot of controlled physical parameters on the object are grouped together, each control parameter in the group is assigned a serial number; on each control cycle, the numerical values are formed from the obtained parameter values in accordance with their sequence numbers, which are stored and in the subsequent control cycles are compared with the newly generated numerical series, which, in case of a mismatch, is also remembered for subsequent comparison, in this case, the memorization of new numerical ranges in accordance with a number of rules is performed during the period of training and self-adjustment, after the end of which, when a numerical range different from the memorized ones appears, signal about the occurrence of the initial stage of a room fire.

EFFECT: technical result consists in the expansion of the functional capabilities of fire and temperature alarm systems in terms of detecting the initial stages of the formation and development of fires by implementing training and self-adjustment functions with automatic adaptation of the systems to the features of the hazard locations and changes in operating conditions.

1 cl, 2 dwg, 2 tbl

Description

The present invention relates to methods for ensuring fire safety in the premises of fire hazardous facilities containing fuel equipment, and can be used in shipbuilding, in particular in ship fire and temperature alarm systems for detecting fires at the initial stages of their development.

There are various methods for detecting the initial stages of fires:

- a method for detecting fire danger and fire in the ship’s premises (RF patent No. 2179470) and a method of detecting fire danger in the compartment of a submarine (RF patent No. 2573305), which extends the scope of its application for pressurized compartments, which consists in monitoring the percentage of oxygen in the air of the submarine compartment and with an increase in the percentage of oxygen above the set value, zones are determined where contact between the combustible substance and the ignition source is possible, the temperature of which is sufficient full-time to start ignition of a combustible substance at a current increased percentage of oxygen, and signal this as a fire hazard;

- a fire detection method and an intelligent control station for implementing the method (RF patent No. 2344859). The method consists in constantly isolating fire hazard factors of the environment of the controlled object, converting the selected factors into an array of digitized data, comparing this data array with an a priori data array, classifying the obtained comparison results in accordance with the extrema and generating, depending on the hazard class, a control signal. The technical result is achieved by continuously comparing the set of current informative parameters with an array of sets of specified informative parameters, and the control signal for extinguishing is generated only if the combination of the current informative parameters matches one of the sets of specified informative parameters. Comparison with an array of predetermined populations is carried out by calculating the dependence (function) of the current values of informative parameters, the development of which leads to the emergence of a certain type of fire, and which is obtained by modeling different types of fires and entered into the device's memory in advance. The number of sets of preset informative parameters stored in the memory of an intelligent control station depends on the technical capabilities and facilities for which the station is intended to be protected.

As a significant drawback of these methods, it should be pointed out that it is necessary to carry out a large complex of preliminary work on the analysis of each protected object, modeling the processes of formation and development of fires in it, and determining the sets of informative parameters.

Since the proposed method can be implemented in modern fire and temperature alarm systems with analog addressable fire detectors, the closest analogue is the well-known indoor fire detection method used in most modern fire alarm systems and described in many published sources, including including in the book of Stumpf E.P., Shtellinga V.N. "Setting up ship's fire alarm systems", Leningrad, "Shipbuilding", 1988, p. 5. The method consists in receiving and generating a signal about the condition of the premises in terms of fire hazard by controlling hazardous fire factors; collecting this information; processing the information received and making a generalized decision on the presence or absence of a fire by comparing the numerical values of the controlled parameters of the fire factors with the established limit values; generating a fire signal and indicating the location of the fire; the inclusion of optical and sound alarms. The above method has the following drawback: the initial stages of the fire development go unnoticed, the decision to turn on the alarm means is delayed only when the monitored parameters of the dangerous fire factors have reached the set limit values, which leads to the untimely start of fire fighting.

The technical result is the expansion of the functionality of fire and temperature alarm systems in terms of detecting the initial stages of the formation and development of fires by implementing training and self-tuning functions with automatic adaptation of the systems to the characteristics of the objects of protection and changes in operating conditions.

The method consists in cyclic determination of the values of a set of physical parameters controlled at the facility (including hazardous fire factors), which are grouped into groups, each of which characterizes the processes of occurrence and development of a fire in a particular room, and each serial parameter in the group is assigned a serial number. The block diagram shown in FIG. 1, reflects a possible implementation of the proposed method and explains the sequence of actions.

, так как

. Полученные по каждому помещению численные ряды запоминают в том случае, если контролируемые параметры не превысили предельно допустимых значений x_i<x^max,

, и на последующих циклах контроля сравнивают с вновь сформированным численным рядом, который, в случае несовпадения, также запоминают для последующего сравнения.On each control cycle, from the obtained parameter values (x ₁ , x ₂ , ..., x _N ) ∈X, numerical series X → { a ₁ , a ₂ , ... a _N ) are formed in accordance with their serial numbers. Each numerical series, in fact, reflects the situation A prevailing in the room and expressed through the numerical values of the controlled physical parameters x ₁ , x ₂ , ... x _N. For example, for the control cycle s, the situation A ^s in a room can be represented as A ^s ≡ X ^s , where

, as

. The numerical series obtained for each room are remembered if the monitored parameters do not exceed the maximum permissible values x _i <x ^max ,

, and in subsequent control cycles, they are compared with the newly formed numerical series, which, in case of a mismatch, is also remembered for subsequent comparison.

, указанного на фиг. 1, автоматически формирует и пополняет базу данных, содержащую данные о всех произошедших в каждом помещении ситуациях. Запомненные численные ряды для каждого помещения образовывают прямоугольную матрицу размерностью z×ν, где z - количество запомненных численных рядов, a ν - количество точек контроля в помещении.Remembering new numbers, for example, numbers

indicated in FIG. 1, automatically generates and replenishes a database containing data on all situations that occurred in each room. The stored numerical series for each room form a rectangular matrix of dimension z × ν, where z is the number of stored numerical series, and ν is the number of control points in the room.

The process of memorizing new numbers is a learning process and actually self-tuning the system. It is advisable to memorize the numerical series in the period immediately after the installation of fire and temperature-alarm systems at the protected facility for a period of time during which changes in the monitored parameters reached the maximum possible values, but did not exceed the maximum permissible values. Additional replenishment or adjustment of stored numbers can also be carried out periodically during operation of the systems in the event of significant changes in the operating conditions of the protected facility.

The memorization of new numerical series and their subsequent comparison with the newly formed numerical series can be carried out separately for each room, as shown in FIG. 2. This allows you to speed up the calculation process by making it parallel.

The algorithm shown in FIG. 1 in the frame of the computation unit, can also be used when processing information in multisensor fire detectors.

After the period of training and self-tuning of the system, the appearance of a numerical series that differs from the already remembered numerical series will mean that an abnormal situation has developed in the room, which is characteristic of the initial stage of a fire.

A change in the steady-state temperature in a controlled ship’s room can occur when weather conditions change, the operating mode of regular heat sources or the occurrence of abnormal heat sources. The fact of the occurrence of an abnormal heat source means that a fire hazard situation has developed in the vessel's premises, the further development of which can lead to a fire. The classification and characteristics of the initial stages (sequentially: fire hazard situation, pre-fire situation, fire, fire) of the formation and development of fires on ships is presented in the article "Ways to solve the problem of automation of fire hazard control in ship premises" published in the journal "Shipbuilding" 2. 2005 ( 759). In the table. 1 presents the main characteristics of the stages of development of fires in ship premises.

To detect an abnormal heat source in the ship’s room in which the heat-generating equipment is installed, during the period of the system’s self-adjustment, for each violation of the established thermal regime in the room, the sequence of the temperature rise at the control points is remembered. The sequence of numbers of control points form a distinctive digital series that allows you to uniquely identify the current situation. If at any control points the parameter has not changed, then the digit zero is assigned to it. That is, the identification digital series characterizing the steady state will contain only zeros. For each situation, determine: the duration of the time between the start of the temperature increase in each subsequent pair of control points and remember the minimum values achieved; the maximum achieved values of the temperature difference at the control points; the maximum achieved values of the rate of change of the temperature difference at the control points; the maximum achieved temperature growth rate. Subsequently, during the fire control period, after identifying the situation, the length of time between the start of the temperature increase at each next control point and the temperature difference and its growth rate for each pair of control points are determined and the results are compared with the corresponding stored values. In cases where the situation has not been identified or the duration of any period of time between the beginning of the temperature increase at the next control point turned out to be less than the stored value, and also if an excess of the stored numerical values of the indicators over the current values was detected, then they signal the presence of an abnormal heat source.

To clarify the foregoing in FIG. Figure 3 shows in graphical form the hypothetical results of measuring temperature x ₁ , x ₂ , x ₃ , x ₄ , x ₅ at five control points in one room. On the graph along the x-axis, the time t, measured in seconds, is plotted. Instead of time, you can use the number of control cycles. The control cycle in the presented example is repeated every 5 s. The ordinate shows the temperature T, measured in degrees Celsius. Let us agree that a steady-state temperature regime occurs in a room when the temperature has not changed at all five control points during a period of 20 seconds. The graph shows that the steady-state temperature in the room, when the temperature at all five control points remained unchanged x ₁ = const, x ₂ = const, x ₃ = const, x ₄ = const, x ₅ = const, was up to time t ₁ . From time t ₁ to t ₄ there were changes at control points 4, 3 and 5. At control points 1 and 2, the temperature did not change. From time t ₆ we can assume that the steady state has again come. A distinctive digital series identifying the situation in the room for the case under consideration will be 43500.

The duration of the time intervals between the beginning of the temperature increase in each subsequent pair of control points will be determined as:

Δt ₂₁ = t ₂ -t ₁ = 5c for control points 4-3;

Δt ₄₂ = t ₄ -t ₂ = 10c for control points 3-5.

The values of the temperature difference at the control points at different points in time will be defined as:

at time t ₁ :

x ₅ -x ₁ = 27-25 = 2; x ₄ x ₁ = 29.5-25 = 4.5; x _3- x ₁ = 26-25 = 1; x _5- x ₂ = 27-30 = -3; x ₄ x ₂ = 29.5-30 = -0.5; x _3- x ₂ = 26-30 = -4; x ₅ x ₃ = 27-26 = 1; x ₄ x ₃ = 29.5-26 = 3.5; x _1- x ₂ = 25-30 = -5; x ₅ -x ₄ = 27-29.5 = -2.5;

at time t ₂ :

x ₅ x ₁ = 27-25 = 2; x ₄ x ₁ = 29.5-25 = 4.5; x _3- x ₁ = 26.5-25 = 1.5; x _5- x ₂ = 27-30 = -3; x ₄ x ₂ = 29.5-30 = -0.5; x _3- x ₂ = 26.5-30 = -3.5; x ₅ x ₃ = 27-26.5 = 0.5; x ₄ x ₃ = 29.5-26.5 = 3; x _1- x ₂ = 25-30 = -5; x _5- x ₄ = 27-29.5 = -2.5;

at time t ₃ :

x ₅ -x ₁ = 27-25 = 2; x ₄ -x ₁ = 30.5-25 = 5.5; x ₃ -x ₁ = 26.5-25 = 1.5; x ₅ -x ₂ = 27-30 = -3; x ₄ -x ₂ = 30.5-30 = 0.5; x ₃ -x ₂ = 26.5-30 = -3.5; x ₅ x ₃ = 27-26.5 = 0.5; x ₄ -x ₃ = 30.5-26.5 = 4; x ₁ -x ₂ = 25-30 = -5; x ₅ -x ₄ = 27-30.5 = -3.5;

at time t ₄ :

x ₅ -x ₁ = 28.5-25-3.5; x ₄ -x ₁ = 30.5-25-5.5; x ₃ -x ₁ = 28-25 = 3; x ₅ -x ₂ = 28.5-30 = -1.5; x ₄ -x ₂ = 30.5-30 = 0.5; x _3- x ₂ = 28-30 = -2; x ₅ -x ₃ = 28.5-28 = 0.5; x ₄ x ₃ = 30.5-28 = 2.5; x ₁ -x ₂ = 25-30 = -5; x ₅ -x ₄ = 28.5-30.5 = -2.

The maximum achieved values of the temperature difference at the control points for a period of time from t ₁ to t ₄ for the situation under consideration with an identifying digital series 43500 will be as follows:

| x ₅ -x ₁ | = 3.5; | x ₄ -x ₁ | = 5.5; | x ₃ -x ₁ | = 3; | x ₅ -x ₂ | = 3; | x _4- x ₂ | = 0.5; | x _3- x ₂ | = 4; | x ₅ -x ₃ | = 1; | x ₄ -x ₃ | = 4; | x ₁ -x ₂ | = 5; | x ₅ -x ₄ | = 3.5.

The maximum achieved values of the rate of change of the temperature difference at the control points are determined for a specified period of time that is a multiple of the periods between control cycles, for example, for 10 s. Then for control points 4-3 in the time interval Δt ₁₃ = t ₃ -t ₁ = 10c temperature differences will be defined as:

;

;

in the time interval Δt ₂₄ = t ₄ -t ₂ = 10c:

;

;

in the time interval Δt ₃₅ = t ₅ -t ₃ = 10c:

;

;

in the time interval Δt ₄₆ = t ₆ -t ₄ = 10c:

.

.

In the table. Figure 2 shows the values of the rate of change of the temperature difference at the control points 4-3 and 3-5.

.The achieved numerical values of the temperature growth rate for control points 1 and 2 over a period of time from t ₁ to t ₄ , which can be denoted as Δt ₁₄ , will be 0, that is:

.

.For control point 3:

.

.For control point 4:

.

.For control point 5:

.

It can be seen that at control point 5, the most intense increase in temperature was observed during the time period Δt ₁₄ .

After the end of the training and self-tuning period, during fire control, the following actions are sequentially performed: identify the situation; determine the duration of the time intervals between the beginning of the temperature increase at each next control point and the value of the difference in temperature and its growth rate for each pair of control points; comparing the obtained data with the corresponding stored values; signal the occurrence of the initial fire stage in the room at the stage of a fire hazard situation in cases where the situation has not been identified or the length of any period of time between the start of the temperature rise at the next control point was less than the stored value or an excess of the stored numerical values of the indicators (temperature difference at control points, the rate of change in temperature difference at control points, the rate of temperature rise at control points) over current values.

The essential features of the claimed invention are:

- combining the set of physical parameters controlled at the facility into groups, each of which characterizes the processes of the onset and development of a fire in a particular room, with the assignment of a serial number to each controlled parameter in the group;

- the formation on each control cycle for each group of the obtained values of the parameters of the numerical series in accordance with the assigned serial numbers of the controlled parameters;

- memorization during the training and self-adjustment for each group of controlled parameters of the first formed numerical series for subsequent comparison;

- Comparison of the numerical series generated at each control cycle with the parameters that are remembered for each group of controlled parameters;

- memorization during the training and self-adjustment for each group of controlled parameters of the generated number series, provided that it does not coincide with those already remembered;

- determination of the duration of the training period and self-tuning by the time during which the changes in the controlled parameters reached the maximum possible values, but did not exceed the maximum permissible values;

- signaling about the occurrence of the initial stage of a fire in a room when it is detected as a result of a comparison of the appearance of a numerical series other than remembered;

- identification of the current situation by using a distinctive digital series formed by the numbers that are assigned to the control points, depending on the sequence of the beginning of the temperature increase in them;

- remembering the sequence of the onset of temperature increase at the control points, the numbers of which form a distinctive digital series, which allows to identify each prevailing situation;

- determination for each current situation of the length of time between the onset of temperature increase for all consecutively formed pairs of control points;

- memorization for each current situation of the minimum achieved values of the length of time intervals between the onset of temperature increase for all consecutively formed pairs of control points;

- remembering for each current situation the maximum values of the temperature difference and its growth rate for all successively formed pairs of control points;

- alarm about the occurrence of a fire hazard situation in the room if the duration of any length of time between the start of the temperature increase at the next control point was less than the stored value;

- alarm about the occurrence of a fire hazard situation in the room when it detects that the current values exceed the stored maximum values of the temperature difference and its growth rate;

- alarm about the occurrence of a fire hazard situation in the room, if the situation has not been identified.

An additional technical result is the possibility of detecting a fire at the very early stage of its development in the event of a fire hazard, when only the conditions for its occurrence develop, and the dangerous factors of the fire do not yet manifest themselves. Immediate adoption of appropriate measures to eliminate the causes that contributed to the occurrence of a fire hazard situation, avoids its further development, that is, to prevent a fire.

Claims (2)

1. The method of automatic detection of the initial stages of a fire in the premises of fire hazardous facilities containing heat-generating equipment, which consists in cyclically determining the values of the monitored parameters of hazardous fire factors, comparing the results with specified maximum permissible values and signaling a fire, if these values are achieved, characterized in that that in order to detect a fire at the initial stage of its development, the whole set of physical parameters controlled at the facility are combined into groups, each of which characterizes the processes of occurrence and development of a fire in a particular room, while each monitored parameter in the group is assigned a serial number and on each control cycle, from the obtained parameter values, numerical series are formed in accordance with their serial numbers, which are remembered for subsequent cycles controls are compared with the newly formed numerical series, which in case of mismatch is also remembered for subsequent comparison, while remembering new numerical series during the initial period of training and self-tuning, or periodically during the time during which the changes in the controlled parameters reached the maximum possible values, but did not exceed the maximum permissible values, and after the end of the training and self-tuning period, when a numerical series other than remembered appears, they signal the occurrence the initial stage of a fire indoors. 2. The method according to p. 1, characterized in that during the period of training and self-adjustment for each violation of the established thermal conditions in the room, remember the sequence of the onset of temperature increase at the control points, the numbers of which form a distinctive digital series identifying each current situation, for which, including, at its subsequent occurrences, the duration of the time intervals between the onset of temperature rise in each subsequent pair of control points is determined and the minimum values achieved are stored, also for Each pair of control points remembers the maximum achieved values of the temperature difference and its growth rate, and subsequently, during fire control, they identify the situation, determine the length of time between the start of the temperature increase at each next control point and the temperature difference and its growth rate for each pairs of control points and compare with the corresponding stored values and, in cases where the situation has not been identified, or the duration of any length of time dy start temperature increase control in the following point was less than the stored value, or the excess was found stored numerical values of the indices of the current values, signal the occurrence of indoors initial stage of fire.

Images