Method and system for improving alarm balance of temperature-sensitive fire detector

Abstract

The invention discloses a method and a system for improving alarm balance of a temperature-sensitive fire detector, relates to the technical field of temperature-sensitive detector control, and solves the technical problem of poor fire detection and prevention effects caused by poor adaptability of an existing differential temperature alarm mode. The method comprises the steps of 1, obtaining a plurality of adjacent temperature detection values of temperature detection equipment; step 2, acquiring a set differential temperature discrimination short period, acquiring detection area characteristics corresponding to the detection area, and generating a temperature difference discrimination long period according to the detection area characteristics; 3, generating a short-period temperature difference value according to the temperature difference discrimination short period and the temperature detection value; generating a long-period temperature difference value according to the temperature difference discrimination long period and the temperature detection value; 4, selecting a temperature early warning value according to the short-period temperature difference value and the long-period temperature difference value; 5, acquiring a current temperature detection value of the temperature detection equipment, and generating an alarm signal according to the current temperature detection value and the temperature early warning value; and the fire detection and prevention effects are improved.

Description

Technical Field

[0001] The present application belongs to the field of temperature-sensing detector control technology, and relates to temperature-sensing detector control technology, and specifically to a method and system for improving the alarm balance of temperature-sensing fire detectors. Background Art

[0002] In the automatic fire alarm system, the temperature detector is a common fire trigger device. The basic principle of the temperature detector is: when the burning of the material releases a lot of heat, the ambient temperature rises sharply, and the thermistor in the temperature detector is very sensitive to the temperature change. The physical change is processed by the single-chip microcomputer to generate an electrical signal and perform alarm processing. The temperature detector is mainly divided into two types: fixed temperature alarm and differential temperature alarm. The fixed temperature alarm detector starts the alarm when the temperature rises to the set value. The differential temperature alarm detector can also start the alarm when the temperature rise rate reaches the set value before the temperature rises to the set value.

[0003] At present, temperature detectors mainly use differential temperature alarms, which can realize timely alarms in the case of rapid temperature rise. However, this alarm mode can only identify rapid temperature rise. However, in the case of slow temperature rise, the temperature difference in a short period cannot trigger the alarm, resulting in the inability of the differential temperature alarm to give a timely alarm, missing the best rescue opportunity, and making the fire monitoring and prevention effect poor; therefore, a method and system for improving the alarm balance of temperature-sensitive fire detectors is needed. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a method and system for improving the alarm balance of temperature-sensitive fire detectors, which is used to solve the technical problem that the existing differential temperature alarm method has poor adaptability and leads to poor fire detection and prevention effects.

[0005] To achieve the above-mentioned purpose, the first aspect of the present application provides a method for improving the alarm balance of a heat-sensitive fire detector, comprising the following steps:

[0006] Step 1: Obtain several nearby temperature detection values ​​of the temperature detection device;

[0007] Step 2: Obtain the set short period of temperature difference discrimination, obtain the detection area characteristics corresponding to the detection area, and generate a long period of temperature difference discrimination according to the detection area characteristics;

[0008] Step 3: Generate a short-period temperature difference value based on the temperature difference and the temperature detection value; generate a long-period temperature difference value based on the temperature difference and the temperature detection value;

[0009] Step 4: Select the temperature warning value according to the short-term temperature difference value and the long-term temperature difference value;

[0010] Step 5: Obtain the temperature detection value of the current temperature detection device, and generate an alarm signal according to the current temperature detection value and the temperature warning value.

[0011] The present application obtains the set short period of temperature difference discrimination, obtains the detection area characteristics corresponding to the detection area, generates a long period of temperature difference discrimination according to the detection area characteristics; generates a short period temperature difference value according to the short period of temperature difference discrimination and the temperature detection value; generates a long period temperature difference value according to the long period of temperature difference discrimination and the temperature detection value; selects a temperature warning value according to the short period temperature difference value and the long period temperature difference value; obtains the temperature detection value of the current temperature detection device, and generates an alarm signal according to the current temperature detection value and the temperature warning value; and effectively solves the problem of untimely alarm under slow temperature rise by generating the length of the long period of temperature difference discrimination in a targeted manner according to the regional characteristics of the detection area. At the same time, the temperature warning value is selected by the short period temperature difference value and the long period temperature difference value to reduce the probability of false alarm of the alarm, thereby improving the effect of fire detection and prevention.

[0012] Preferably, the step three specifically includes the following steps:

[0013] S31: Obtaining a short period of temperature difference determination;

[0014] S32: Acquire a number of temperature detection values ​​within a differential temperature determination short period, and number the temperature detection values ​​in the order of acquisition time to form a differential temperature determination short period array;

[0015] S33: Recording the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination short-cycle array as the short-cycle temperature difference;

[0016] S34: Obtaining a long period of temperature difference determination;

[0017] S35: obtaining a plurality of temperature detection values ​​within a differential temperature discrimination long period, and numbering the plurality of temperature detection values ​​in the order of acquisition time, and combining them into a differential temperature discrimination long period array;

[0018] S36: Record the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination long-period array as the long-period temperature difference.

[0019] Preferably, the step of generating a temperature difference discrimination long period according to the detection area characteristics comprises:

[0020] Extracting regional structural features and stored item features from the detection area features, and inputting the regional structural features into a transfer influence coefficient generation model to obtain a corresponding transfer influence coefficient; the transfer influence coefficient generation model is obtained by training an artificial intelligence model;

[0021] Extract the combustion heat release level and the combustion speed score from the characteristics of the stored items; generate the item influence coefficient according to the combustion heat release level and the combustion speed score;

[0022] Generate a long period of temperature difference judgment based on the transfer influence coefficient and the item influence coefficient.

[0023] Preferably, the transfer influence coefficient generation model is obtained by training an artificial intelligence model, including:

[0024] Acquire several regional structural characteristic data and the corresponding transfer influence coefficient; the transfer influence coefficient is the influence coefficient of the regional structure on the temperature transfer during combustion according to the experts, and the greater the resistance to temperature transfer, the greater the corresponding transfer influence score; the structural characteristic data is the material and installation position of the structural components within the detection range of the temperature sensor;

[0025] Integrate several regional structural characteristic data and transfer influence coefficients into training data and test data; use the training data to train the artificial intelligence model, and use the test data to test the trained artificial intelligence model, and finally obtain a transfer influence coefficient generation model with the regional structural characteristic data as input and the corresponding transfer influence coefficient as output, wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0026] Preferably, generating the item influence coefficient according to the combustion release heat level and the combustion speed score includes:

[0027] The combustion release score is marked as RD; the combustion speed score is marked as SD; the combustion release score is a score of the combustion condition of the item according to the heat released by the combustion per unit weight of the corresponding item, and the combustion speed score is a score of the combustion condition of the item according to the time required for combustion per unit volume of the corresponding item;

[0028] The corresponding item influence coefficient is calculated by the formula WY=α1×exp(-β1×RD)+α2×exp(-β2×SD); wherein α1 and α2 are proportional coefficients; β1 and β2 are adjustment coefficients.

[0029] Preferably, the step of generating a long period of temperature difference determination according to the transfer influence coefficient and the item influence coefficient includes:

[0030] Obtain the transfer influence coefficient CY and the item influence coefficient WY;

[0031] The long-period proportionality coefficient K is calculated by the formula K=γ1×CY+γ2×WY; wherein γ1 and γ2 are proportionality coefficients used to limit the range of the long-period proportionality coefficient K;

[0032] The temperature difference judgment long period is calculated by the formula CZQ=K×DZQ, and DZQ is the corresponding temperature difference judgment short period.

[0033] Preferably, the selecting of the temperature warning value according to the short-period temperature difference value and the long-period temperature difference value comprises:

[0034] S41: Obtaining a short-period temperature difference value and a long-period temperature difference value;

[0035] S42: When the short-cycle temperature difference value is greater than the set high temperature rise short-cycle threshold; or the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the set high temperature rise long-cycle threshold; the temperature rise level is set to level one;

[0036] When the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the set high temperature rise long-cycle threshold; or when the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the low temperature rise long-cycle threshold; the heating level is set to level 2;

[0037] When the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the low temperature rise long-cycle threshold, the temperature rise level is set to level three; among which, the high temperature rise long-cycle threshold is greater than the low temperature rise long-cycle threshold, and the high temperature rise short-cycle threshold is greater than the low temperature rise short-cycle threshold;

[0038] S43: When the temperature rise level is level one, the temperature warning value is set to a high temperature rise warning threshold; when the temperature rise level is level two, the temperature warning value is set to a medium temperature rise warning threshold; when the temperature rise level is level three, the temperature warning value is set to a low temperature rise warning threshold.

[0039] Preferably, generating an alarm signal according to the current temperature detection value and the temperature warning value includes:

[0040] Get the temperature detection value and determine whether the temperature detection value is greater than the temperature warning value;

[0041] If yes, then when the temperature warning value is the high temperature rise warning threshold, a rapid fire warning signal is generated; otherwise, when the temperature warning value is the medium temperature rise warning threshold, a conventional fire warning signal is generated; otherwise, when the temperature warning value is the low temperature rise warning threshold, a low-speed fire warning signal is generated;

[0042] No, no alarm signal is generated and the process goes to step three; the alarm signal includes a rapid fire warning signal, a conventional fire warning signal and a low-speed fire warning signal.

[0043] This system can classify fires in real time. When there are multiple fires and there are not enough firefighters, areas with high alarm levels can be dealt with first.

[0044] Another aspect of the present application provides a system for improving the alarm balance of a temperature-sensitive fire detector, comprising: a data acquisition module, a cycle setting module, a threshold generation module, an anomaly detection module, an alarm module, and a database;

[0045] The data acquisition module is used to obtain the temperature detection value in the detection area in real time through the data acquisition device connected to it, and the data acquisition device includes a temperature detection device;

[0046] The cycle setting module is used to obtain the set temperature difference determination short cycle and the detection area characteristics corresponding to the detection area, and generate the temperature difference determination long cycle according to the detection area characteristics;

[0047] The threshold generation module selects a temperature warning value according to a short-period temperature difference value and a long-period temperature difference value;

[0048] The abnormality detection module is used to obtain the temperature detection value of the current temperature detection device and generate an alarm signal according to the current temperature detection value and the temperature warning value;

[0049] The alarm module is configured to generate an alarm according to the alarm signal.

[0050] Compared with the prior art, the beneficial effects of this application are:

[0051] 1. The present application obtains the set short period of temperature difference discrimination, obtains the detection area characteristics corresponding to the detection area, generates a long period of temperature difference discrimination according to the detection area characteristics; generates a short period temperature difference value according to the short period of temperature difference discrimination and the temperature detection value; generates a long period temperature difference value according to the long period of temperature difference discrimination and the temperature detection value; selects a temperature warning value according to the short period temperature difference value and the long period temperature difference value; obtains the temperature detection value of the current temperature detection device, and generates an alarm signal according to the current temperature detection value and the temperature warning value; effectively solves the problem of untimely alarm under slow temperature rise by generating the length of the long period of temperature difference discrimination in a targeted manner according to the regional characteristics of the detection area, and at the same time selects the temperature warning value through the short period temperature difference value and the long period temperature difference value to reduce the probability of false alarm of the alarm, thereby improving the effect of fire detection and prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A schematic diagram of the method steps of the alarm balance method in this application;

[0054] Figure 2 It is a flowchart of the alarm balance method in this application;

[0055] Figure 3 This is a schematic diagram of the principle of the alarm balance system in this application. DETAILED DESCRIPTION

[0056] The technical solution of the present application will be described clearly and completely in conjunction with the embodiments below. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0057] See also Figure 1-Figure 2 The first aspect of the present application provides a method for improving the alarm balance of a temperature-sensitive fire detector, comprising the following steps:

[0058] Step 1: Obtain several adjacent temperature detection values ​​of the temperature detection device; the temperature detection value is the temperature value detected by the temperature detection device; the temperature detection device includes a temperature sensor;

[0059] Step 2: Obtain the set short period of temperature difference discrimination, which is a detection period artificially set for fast-heating fires; obtain the detection area characteristics corresponding to the detection area, which are the regional characteristics of the area that can be detected by the corresponding temperature detection device, including the storage item characteristics of the items stored in the area and the regional structural characteristics of the area, which need to be input in advance; for example, the corresponding temperature detection device is installed in a warehouse with shelves and partitions for storing books or chemicals; its corresponding regional structural characteristics include the position, size and material of the partitions in the detection area; the storage item characteristics are the relevant characteristics of books or chemicals; generate a long period of temperature difference discrimination based on the detection area characteristics; the long period of temperature discrimination is a period for slow temperature rise anomaly detection, which is specifically set according to the characteristics of the stored items and the characteristics of the regional structure affecting heat transfer;

[0060] Step 3: Generate a short-cycle temperature difference value based on the short-cycle temperature difference and the temperature detection value; the short-cycle temperature difference value is the temperature difference before and after the short-cycle temperature difference; generate a long-cycle temperature difference value based on the long-cycle temperature difference and the temperature detection value; the long-cycle temperature difference value is the temperature difference before and after the long-cycle temperature difference; the specific steps are:

[0061] S31: Obtaining a short period of temperature difference determination;

[0062] S32: Acquire a number of temperature detection values ​​within a differential temperature determination short period, and number the temperature detection values ​​in the order of acquisition time to form a differential temperature determination short period array;

[0063] S33: Recording the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination short-cycle array as the short-cycle temperature difference;

[0064] S34: Obtaining a long period of temperature difference determination;

[0065] S35: obtaining a plurality of temperature detection values ​​within a differential temperature discrimination long period, and numbering the plurality of temperature detection values ​​in the order of acquisition time, and combining them into a differential temperature discrimination long period array;

[0066] S36: Record the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination long-period array as the long-period temperature difference.

[0067] Step 4: Select the temperature warning value according to the short-cycle temperature difference value and the long-cycle temperature difference value; the temperature warning value is set according to the temperature rise in the short cycle and the long cycle. For example, when the temperature rises quickly in the short cycle, setting a suitable temperature warning value can not only give an alarm in time, but also reduce the probability of false alarms;

[0068] Step 5: Obtain the temperature detection value of the current temperature detection device, and generate an alarm signal according to the current temperature detection value and the temperature warning value.

[0069] This embodiment obtains the set short period of temperature difference judgment, obtains the detection area characteristics corresponding to the detection area, and generates a long period of temperature difference judgment according to the detection area characteristics; generates a short period temperature difference value according to the short period of temperature difference judgment and the temperature detection value; generates a long period temperature difference value according to the long period of temperature difference judgment and the temperature detection value; selects a temperature warning value according to the short period temperature difference value and the long period temperature difference value; obtains the temperature detection value of the current temperature detection device, and generates an alarm signal according to the current temperature detection value and the temperature warning value; by generating the length of the long period of temperature difference judgment in a targeted manner according to the regional characteristics of the detection area, the problem of untimely alarm under slow temperature rise is effectively solved, and at the same time, the temperature warning value is selected by the short period temperature difference value and the long period temperature difference value to reduce the probability of false alarm of the alarm, thereby improving the effect of fire detection and prevention.

[0070] Generating a temperature difference discrimination long period according to the detection area characteristics, including: extracting the regional structure characteristics and the storage item characteristics from the detection area characteristics, inputting the regional structure characteristics into the transfer influence coefficient generation model to obtain the corresponding transfer influence coefficient; the transfer influence coefficient generation model is obtained by training the artificial intelligence model;

[0071] Extract the combustion heat release level and the combustion speed score from the characteristics of the stored items; generate the item influence coefficient according to the combustion heat release level and the combustion speed score;

[0072] Generate a temperature difference discrimination long period according to the transfer influence coefficient and the item influence coefficient, including: obtaining the transfer influence coefficient CY and the item influence coefficient WY;

[0073] The long-period proportionality coefficient K is calculated by the formula K=γ1×CY+γ2×WY; wherein γ1 and γ2 are proportionality coefficients, which are used to limit the range of the long-period proportionality coefficient K so that K>2;

[0074] The temperature difference judgment long period is calculated by the formula CZQ=K×DZQ, and DZQ is the corresponding temperature difference judgment short period.

[0075] This embodiment generates a long period of temperature difference determination in a targeted manner through the regional structural characteristics of the detection area and the characteristics of the stored items, so that the long period of temperature difference determination is more adapted to the current detection environment, thereby improving the accuracy of subsequent fire warnings.

[0076] The transfer influence coefficient generation model is obtained through artificial intelligence model training, including: obtaining a number of regional structural feature data and the corresponding transfer influence coefficient; the transfer influence coefficient is the influence coefficient of the regional structure on the temperature transfer during combustion according to the expert, and the greater the resistance to temperature transfer, the greater the corresponding transfer influence score; for example, if there are structural objects such as walls set between the fire point and the temperature detector within the detection range of the temperature detector, it will affect the speed at which the temperature of the fire point is transmitted to the temperature detector; or a digital twin model can be constructed to simulate the actual situation of the fire point and the temperature detector, and the temperature transfer time without the influence of regional structural features and the temperature transfer time under the influence of various regional structural features are obtained, and the ratio between the temperature transfer time under the influence of the structure and the temperature transfer time without the influence of the regional structure is used as the transfer influence coefficient; the structural feature data is the material and setting position of the structural components within the detection range of the temperature detector.

[0077] Integrate several regional structural feature data and transfer influence coefficients into training data and test data; use the training data to train the artificial intelligence model, and use the test data to test the trained artificial intelligence model. Specifically, input the regional structural features in the test data into the trained artificial intelligence model to obtain the corresponding transfer influence coefficient output, and judge whether the difference between the output transfer influence coefficient and the transfer influence coefficient recorded in the corresponding test data is within an acceptable range. If so, the test data passes the test and the next set of test data is tested; if not, adjust the relevant parameters of the artificial intelligence model and use the set of test data again for testing; until a set proportion of the test data passes the test; finally, a transfer influence coefficient generation model is obtained with the input of regional structural feature data and the output of the corresponding transfer influence coefficient, wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.

[0078] Generate item impact coefficients based on the level of heat released by combustion and the burning speed score, including:

[0079] The combustion release score is marked as RD; the combustion speed score is marked as SD; the combustion release score is a score for the combustion condition of the item according to the heat released by the combustion per unit weight of the corresponding item. The higher the combustion release score, the more heat is released by the combustion per unit weight; the combustion speed score is a score for the combustion condition of the item according to the time required for the combustion per unit volume of the corresponding item; the higher the combustion speed score, the shorter the time required for the combustion per unit volume;

[0080] The corresponding item influence coefficient is calculated by the formula WY=α1×exp(-β1×RD)+α2×exp(-β2×SD); wherein α1 and α2 are proportional coefficients, which are used to set the range of the item influence coefficient to be the same as the range of the transfer influence coefficient; β1 and β2 are adjustment coefficients.

[0081] In this embodiment, the requirement for the detection period for the burning of objects in the detection environment is calculated through the above formula; when the heat generated by the burning of objects is more and the speed of the burning of objects is faster, the temperature of the detection area rises faster and the time for the temperature to rise to the temperature threshold is shorter. In order to ensure the sensitivity of the fire alarm, the required detection period is shorter and the corresponding object influence coefficient is set smaller; when the heat generated by the burning of objects is less and the speed of the burning of objects is slower, the temperature of the detection area rises slower and the time for the temperature to rise to the temperature threshold is longer, the required detection period is longer and the corresponding object influence coefficient is set larger.

[0082] The temperature warning value is selected according to the short-term temperature difference value and the long-term temperature difference value, including:

[0083] S41: Obtaining a short-period temperature difference value and a long-period temperature difference value;

[0084] S42: When the short-cycle temperature difference value is greater than the set high temperature rise short-cycle threshold; or the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the set high temperature rise long-cycle threshold; the temperature rise level is set to level one;

[0085] When the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the set high temperature rise long-cycle threshold; or when the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the low temperature rise long-cycle threshold; the heating level is set to level 2;

[0086] When the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the low temperature rise long-cycle threshold, the temperature rise level is set to level three; among which, the high temperature rise long-cycle threshold is greater than the low temperature rise long-cycle threshold, and the high temperature rise short-cycle threshold is greater than the low temperature rise short-cycle threshold;

[0087] S43: When the temperature rise level is level one, the temperature warning value is set to a high temperature rise warning threshold; when the temperature rise level is level two, the temperature warning value is set to a medium temperature rise warning threshold; when the temperature rise level is level three, the temperature warning value is set to a low temperature rise warning threshold.

[0088] Generating an alarm signal according to the current temperature detection value and the temperature warning value includes: obtaining the temperature detection value, and determining whether the temperature detection value is greater than the temperature warning value;

[0089] If yes, then when the temperature warning value is the high temperature rise warning threshold, a rapid fire warning signal is generated; otherwise, when the temperature warning value is the medium temperature rise warning threshold, a conventional fire warning signal is generated; otherwise, when the temperature warning value is the low temperature rise warning threshold, a low-speed fire warning signal is generated;

[0090] No, no alarm signal is generated and the process goes to step three; the alarm signal includes a rapid fire warning signal, a conventional fire warning signal and a low-speed fire warning signal.

[0091] This system can classify fires in real time. When there are multiple fires and there are not enough firefighters, areas with high alarm levels can be dealt with first.

[0092] See also Figure 3 , another aspect of the present application provides a system for improving the alarm balance of a temperature-sensitive fire detector, comprising: a data acquisition module, a cycle setting module, a threshold generation module, an anomaly detection module, an alarm module and a database;

[0093] Data acquisition module: acquires the temperature detection value in the detection area in real time through the data acquisition device connected to it, and the data acquisition device includes a temperature detection device;

[0094] Cycle setting module: obtains the set short cycle of temperature difference discrimination and the detection area characteristics corresponding to the detection area, and generates a long cycle of temperature difference discrimination according to the detection area characteristics;

[0095] Threshold generation module: selects temperature warning value according to short-term temperature difference value and long-term temperature difference value;

[0096] Abnormal detection module: obtains the temperature detection value of the current temperature detection device, and generates an alarm signal according to the current temperature detection value and temperature warning value;

[0097] Alarm module: alarm according to the alarm signal;

[0098] The database is used to store each short-period temperature difference value, long-period temperature difference value, temperature difference discrimination short-period array and temperature difference discrimination long-period array; this facilitates subsequent verification of the alarm results.

[0099] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0100] How this application works:

[0101] The present application obtains the set short period of temperature difference discrimination, obtains the detection area characteristics corresponding to the detection area, generates a long period of temperature difference discrimination according to the detection area characteristics; generates a short period temperature difference value according to the short period of temperature difference discrimination and the temperature detection value; generates a long period temperature difference value according to the long period of temperature difference discrimination and the temperature detection value; selects a temperature warning value according to the short period temperature difference value and the long period temperature difference value; obtains the temperature detection value of the current temperature detection device, and generates an alarm signal according to the current temperature detection value and the temperature warning value; and effectively solves the problem of untimely alarm under slow temperature rise by generating the length of the long period of temperature difference discrimination in a targeted manner according to the regional characteristics of the detection area. At the same time, the temperature warning value is selected by the short period temperature difference value and the long period temperature difference value to reduce the probability of false alarm of the alarm, thereby improving the effect of fire detection and prevention.

[0102] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A method for improving the alarm balance of a heat-sensitive fire detector, characterized in that: The following steps are involved: Step 1: Obtain several nearby temperature detection values ​​of the temperature detection device; Step 2: Obtain the set short period of temperature difference discrimination, obtain the detection area characteristics corresponding to the detection area, and generate a long period of temperature difference discrimination according to the detection area characteristics; Step 3: Generate a short-period temperature difference value based on the short-period temperature difference and the temperature detection value; Generate a long-period temperature difference value based on the temperature difference long-period and the temperature detection value; Step 4: Select the temperature warning value according to the short-term temperature difference value and the long-term temperature difference value; Step 5: Obtain the temperature detection value of the current temperature detection device, and generate an alarm signal according to the current temperature detection value and the temperature warning value.

2. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 1, characterized in that: The step three comprises the following steps: S31: Obtaining a short period of temperature difference determination; S32: Acquire a number of temperature detection values ​​within a differential temperature determination short period, and number the temperature detection values ​​in the order of acquisition time to form a differential temperature determination short period array; S33: Recording the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination short-cycle array as the short-cycle temperature difference; S34: Obtaining a long period of temperature difference determination; S35: obtaining a plurality of temperature detection values ​​within a differential temperature discrimination long period, and numbering the plurality of temperature detection values ​​in the order of acquisition time, and combining them into a differential temperature discrimination long period array; S36: Record the difference between the last temperature detection value and the first temperature detection value in the temperature difference discrimination long-period array as the long-period temperature difference.

3. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 1, characterized in that: The method of generating a temperature difference determination long period according to the detection area characteristics includes: Extracting regional structural features and stored item features from the detection area features, and inputting the regional structural features into a transfer influence coefficient generation model to obtain a corresponding transfer influence coefficient; the transfer influence coefficient generation model is obtained by training an artificial intelligence model; Extract the combustion heat release level and the combustion speed score from the characteristics of the stored items; generate the item influence coefficient according to the combustion heat release level and the combustion speed score; Generate a long period of temperature difference judgment based on the transfer influence coefficient and the item influence coefficient.

4. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 3, characterized in that: The transfer influence coefficient generation model is obtained through artificial intelligence model training, including: Acquire several regional structural characteristic data and the corresponding transfer influence coefficients; integrate several regional structural characteristic data and transfer influence coefficients into training data and test data; use the training data to train the artificial intelligence model, and use the test data to test the trained artificial intelligence model, and finally obtain a transfer influence coefficient generation model with the regional structural characteristic data as input and the corresponding transfer influence coefficient as output, wherein the artificial intelligence model includes a BP neural network model and an RBF neural network model.

5. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 3, characterized in that: The item impact coefficient is generated according to the combustion heat release level and the combustion speed score, including: The combustion release score is marked as RD; the combustion speed score is marked as SD; The corresponding item influence coefficient is calculated by the formula WY=α1×exp(-β1×RD)+α2×exp(-β2×SD); wherein α1 and α2 are proportional coefficients; β1 and β2 are adjustment coefficients.

6. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 3, characterized in that: The method of generating a temperature difference determination long period according to the transfer influence coefficient and the item influence coefficient includes: Obtain the transfer influence coefficient CY and the item influence coefficient WY; The long-period proportionality coefficient K is calculated by the formula K = γ1×CY+γ2×WY; where γ1 and γ2 are proportionality coefficients; The temperature difference discrimination long period is calculated by the formula CZQ=K×DZQ, and DZQ is the corresponding temperature difference discrimination short period.

7. A method for improving the alarm balance of a heat-sensitive fire detector according to claim 1, characterized in that: The selecting of the temperature warning value according to the short-period temperature difference value and the long-period temperature difference value comprises: S41: Obtaining a short-period temperature difference value and a long-period temperature difference value; S42: When the short-cycle temperature difference value is greater than the set high temperature rise short-cycle threshold; or the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the set high temperature rise long-cycle threshold; the temperature rise level is set to level one; When the short-cycle temperature difference value is between the set low temperature rise short-cycle threshold and the high temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the set high temperature rise long-cycle threshold; or when the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is greater than the low temperature rise long-cycle threshold; the heating level is set to level 2; When the short-cycle temperature difference value is less than the set low temperature rise short-cycle threshold, and the long-cycle temperature difference value is less than the low temperature rise long-cycle threshold, the temperature rise level is set to level three; among which, the high temperature rise long-cycle threshold is greater than the low temperature rise long-cycle threshold, and the high temperature rise short-cycle threshold is greater than the low temperature rise short-cycle threshold; S43: When the temperature rise level is level one, the temperature warning value is set to a high temperature rise warning threshold; when the temperature rise level is level two, the temperature warning value is set to a medium temperature rise warning threshold; when the temperature rise level is level three, the temperature warning value is set to a low temperature rise warning threshold.

8. The method for improving the alarm balance of a heat-sensitive fire detector according to claim 1, characterized in that: The generating of an alarm signal according to the current temperature detection value and the temperature warning value comprises: Get the temperature detection value and determine whether the temperature detection value is greater than the temperature warning value; If yes, then when the temperature warning value is the high temperature rise warning threshold, a rapid fire warning signal is generated; otherwise, when the temperature warning value is the medium temperature rise warning threshold, a conventional fire warning signal is generated; otherwise, when the temperature warning value is the low temperature rise warning threshold, a low-speed fire warning signal is generated; No, no alarm signal is generated and the process goes to step three; the alarm signal includes a rapid fire warning signal, a conventional fire warning signal and a low-speed fire warning signal.

9. A system for improving the alarm balance of a heat-sensitive fire detector, applied to a method for improving the alarm balance of a heat-sensitive fire detector according to any one of claims 1 to 8; characterized in that: include: Data acquisition module, cycle setting module, threshold generation module, anomaly detection module, alarm module and database; The data acquisition module is used to obtain the temperature detection value in the detection area in real time through the data acquisition device connected thereto; The cycle setting module is used to obtain the set temperature difference determination short cycle and the detection area characteristics corresponding to the detection area, and generate the temperature difference determination long cycle according to the detection area characteristics; The threshold generation module selects a temperature warning value according to a short-period temperature difference value and a long-period temperature difference value; The abnormality detection module is used to obtain the temperature detection value of the current temperature detection device and generate an alarm signal according to the current temperature detection value and the temperature warning value; The alarm module is configured to generate an alarm according to the alarm signal.

Images