Method, system, device and medium for discriminating between true and false alarms of a gas alarm

Abstract

The application provides a method, system, device and medium for distinguishing true and false alarms of a gas alarm, the method comprising: obtaining false alarm sample data, determining false alarm interference factors and false alarm identification features; calculating the probability of false alarms caused by each false alarm interference factor under each false alarm identification feature to construct a false alarm probability list of 'identification feature-interference factor'; determining the weight corresponding to each false alarm identification feature; in true and false alarm discrimination, obtaining a piece of alarm data to be tested, determining the interference factors and identification features corresponding to the alarm data to be tested as target interference factors and target identification features, and extracting a target false alarm probability value; based on the weight and the target false alarm probability value, calculating the expected value of false alarms caused by each interference factor corresponding to the alarm data to be tested; comparing the calculated expected value with a set value, and discriminating whether the alarm data to be tested is a false alarm according to the comparison result, so that false alarms are effectively identified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gas leakage detection, in particular to a method, system, device and medium for discriminating true and false alarms of a gas alarm. BACKGROUND

[0002] With the improvement of urban gas pipeline network supporting facilities, gas as a clean and efficient new energy is widely used in various fields, which brings great convenience to people's daily life, but also exists the safety hazard of gas leakage. Once gas leakage explosion occurs, it is easy to cause casualties and heavy losses.

[0003] Although gas leakage alarm products are also widely used, the quality of alarms produced by various gas alarm manufacturers is uneven, and false alarms often occur, which is a prominent problem. It not only interferes with the normal production and life of users, but also brings operational pressure to gas companies and industry supervision departments.

[0004] Therefore, how to effectively filter the interference of other factors on the alarm has become a problem to be solved.

[0005] In order to solve the above problems, people have been seeking an ideal technical solution. SUMMARY

[0006] The purpose of the present application is to overcome the shortcomings of the prior art, and to provide a method, system, device and medium for discriminating true and false alarms of a gas alarm, which can screen out interference factors affecting the alarm and obtain the performance characteristics of the interference factors, and effectively screen out false alarms according to the interference factors and their performance characteristics, and improve the accuracy of gas alarm.

[0007] In order to achieve the above purpose, the first aspect of the present application provides a method for discriminating true and false alarms of a gas alarm, comprising: obtaining false alarm sample data; determining false alarm interference factors and false alarm identification characteristics based on the false alarm sample data; calculating the probability of false alarm caused by each false alarm interference factor under each false alarm identification characteristic to construct a "identification characteristic-interference factor" false alarm probability list;

[0008] Determine the weight corresponding to each false alarm identification characteristic;

[0009] In the discrimination of true and false alarms, a piece of alarm data to be tested is obtained, and the interference factors and identification characteristics corresponding to the alarm data to be tested are determined as target interference factors and target identification characteristics. Through the target interference factors and the target identification characteristics, the target false alarm probability value is extracted from the "identification characteristic-interference factor" false alarm probability list;

[0010] Based on the weight and the target false alarm probability value, an expected value of false alarm caused by each interference factor corresponding to the to-be-tested alarm data is calculated; the calculated expected value is compared with a set value, and whether the to-be-tested alarm data is a false alarm is determined according to a comparison result.

[0011] To achieve the above-mentioned purpose, the second aspect of the present application provides a system for discriminating true and false alarms of a gas alarm, comprising an alarm data discrimination module and a task generation module, wherein the alarm data discrimination module is used to discriminate to-be-tested alarm data by using the method for discriminating true and false alarms of a gas alarm; the task generation module is used to generate an alarm processing task when it is determined that the to-be-tested alarm data is true alarm data, and not to generate an alarm processing task when it is determined that the to-be-tested alarm data is false alarm data.

[0012] To achieve the above-mentioned purpose, the third aspect of the present application provides a device for discriminating true and false alarms of a gas alarm, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus; the memory is used to store a computer program; the processor is used to execute the program stored on the memory to realize the method for discriminating true and false alarms of a gas alarm as described above.

[0013] To achieve the above-mentioned purpose, the fourth aspect of the present application provides a readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processor to execute the method for discriminating true and false alarms of a gas alarm as described above.

[0014] The present application has the following beneficial effects:

[0015] The present application analyzes false alarm interference factors and false alarm recognition features based on false alarm sample data, and constructs a false alarm probability list of "recognition feature-interference factor"; the weight corresponding to each false alarm recognition feature is determined, and in true and false alarm discrimination, the corresponding false alarm expected value is calculated based on the weight and the interference factor and the recognition feature corresponding to the to-be-tested alarm data, so as to discriminate whether the to-be-tested alarm data is a false alarm, which can effectively screen out false alarm data. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the method for discriminating true and false alarms of a gas alarm of the present application Figure 1 ;

[0017] Figure 2 is a flowchart of the method for discriminating true and false alarms of a gas alarm of the present application Figure 2 ;

[0018] Figure 3 is a block diagram of the system for discriminating true and false alarms of a gas alarm of the present application. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be further described in detail below through specific embodiments.

[0020] To facilitate understanding, the interactive parties and / or terms and / or custom terms involved in this invention will first be explained in conjunction with the technical solution of this invention:

[0021] Alarm devices: These include devices with gas leak detection and alarm functions, such as gas alarms or gas detectors.

[0022] False alarm sample data: refers to historical sample data where alarm data exists but there is actually no gas leak.

[0023] False alarm interference factors: denoted as A, A = {A1, A2, ..., A...} n A1 represents the first type of false alarm interference factor, ..., A n This represents the nth type of false alarm interference factor; specifically including but not limited to oil fumes, water vapor, volatile gases, and the use of self-heating hot pot, etc.

[0024] False alarm detection features: denoted as B, B = {B1, B2, ..., B} m}, B1 represents the first type of false alarm detection feature, ..., B m This represents the m-th false alarm identification feature, where n and m are both integers greater than or equal to 1; specifically including but not limited to the rate of change of gas concentration, alarm duration, and the time period in which the alarm occurs.

[0025] Rate of change of gas concentration: When a false alarm occurs, the gas concentration measured by the alarm first increases and then decreases; when a true alarm occurs, the gas concentration measured by the alarm continues to rise.

[0026] Alarm duration: When a false alarm is triggered, the alarm duration is mainly concentrated in a very short period of time, while when a real alarm is triggered, the alarm time continues until human intervention.

[0027] The timing of alarms: False alarms caused by characteristics such as cooking fumes, water vapor, and volatile gases mainly occur during the daytime, with the frequency increasing from a certain time in the morning and peaking at a certain time in the afternoon, which is basically consistent with people's cooking and other daily life activities. False alarms caused by using self-heating hot pot also mainly occur during the daytime, with two peaks in frequency, one at noon and one in the evening, which coincide with lunch and dinner times. Real alarms do not have the above characteristics.

[0028] Weight: Used to indicate the magnitude of the influence of each false alarm detection feature on false alarms, denoted as It represents the weight corresponding to the jth false alarm identification feature, and the greater the weight, the greater the impact on false alarms; for example, the number of times each false alarm identification feature in the false alarm sample data causes false alarms is taken as the corresponding weight.

[0029] The "identification feature-interference factor" false alarm probability list: used to mark the probability of false alarms caused by each false alarm interference factor under each false alarm identification feature. Taking three false alarm interference factors and three false alarm identification features as an example, the "identification feature-interference factor" false alarm probability list is shown in the following table:

[0030] [p(A1|B1)] [p(A2|B1)] [p(A3|B1)] [p(A1|B2)] [p(A2|B2)] [p(A3|B2)] [p(A1|B3)] [p(A2|B3)] [p(A3|B3)]

[0031] Wherein, p(A1|B1) represents the probability of false alarms caused by the first false alarm interference factor under the first false alarm identification feature, p(A1|B2) represents the probability of false alarms caused by the first false alarm interference factor under the second false alarm identification feature, and p(A1|B3) represents the probability of false alarms caused by the first false alarm interference factor under the third false alarm identification feature.

[0032] p(A2|B1) represents the probability of false alarms caused by the second false alarm interference factor under the first false alarm identification feature, p(A2|B2) represents the probability of false alarms caused by the second false alarm interference factor under the second false alarm identification feature, and p(A2|B3) represents the probability of false alarms caused by the second false alarm interference factor under the third false alarm identification feature.

[0033] p(A3|B1) represents the probability of false alarms caused by the third false alarm interference factor under the first false alarm identification feature, p(A3|B2) represents the probability of false alarms caused by the third false alarm interference factor under the second false alarm identification feature, and p(A3|B3) represents the probability of false alarms caused by the third false alarm interference factor under the third false alarm identification feature.

[0034] Set value: a comparison value set by the user based on experience and can be modified at any time, related to the number of false alarms caused by false alarm interference factors, such as set to one quarter, one third, one half, four-thirds or other of the total number of false alarms caused by the corresponding false alarm interference factor in the false alarm sample data.

[0035] Embodiment 1

[0036] As shown in FIGS. 1-3, a method for discriminating true and false alarms of a gas alarm includes: Figure 1 and 2 As shown in FIGS. 1-3, a method for discriminating true and false alarms of a gas alarm includes:

[0037] Obtaining false alarm sample data; wherein the false alarm sample data is used to record the total number of false alarms, the interference factors that cause false alarms, and the number of false alarms related to each interference factor, the identification features, and the number of false alarms related to each identification feature, etc.

[0038] Based on the false alarm sample data, a false alarm interference factor and a false alarm identification feature are determined, and a probability of false alarm caused by each false alarm interference factor under each false alarm identification feature is calculated to construct a "identification feature-interference factor" false alarm probability list;

[0039] A weight corresponding to each false alarm identification feature is determined.

[0040] In the true or false alarm discrimination, a piece of to-be-tested alarm data is obtained, and a corresponding interference factor and identification feature of the to-be-tested alarm data are determined as a target interference factor and a target identification feature. A target false alarm probability value is extracted from the "identification feature-interference factor" false alarm probability list through the target interference factor and the target identification feature.

[0041] Based on the weight and the target false alarm probability value, an expected value of false alarm caused by each interference factor corresponding to the to-be-tested alarm data is calculated. The calculated expected value is compared with a set value, and whether the to-be-tested alarm data is a false alarm is determined according to the comparison result.

[0042] It should be noted that the embodiment analyzes the interference factors in the false alarm sample data, obtains the identification features of the interference factors, and then constructs the "identification feature-interference factor" false alarm probability list. When it is necessary to discriminate the true or false alarm data, the interference factors and identification features corresponding to the to-be-tested alarm data are determined first, and then the probabilities of false alarm caused by each false alarm interference factor under the false alarm identification features corresponding to the to-be-tested alarm data are determined. Then, the expected values of false alarm corresponding to each false alarm identification feature are calculated in combination with the preset weight. Finally, the true or false alarm of the alarm is discriminated according to the expected values of false alarm, and the true or false alarm discrimination accuracy is high and fast.

[0043] In some embodiments, the calculation formula of the probability of false alarm caused by each false alarm interference factor under each false alarm identification feature is as follows:

[0044]

[0045] wherein, p(A i |B j ) represents the probability of false alarm caused by the i th false alarm interference factor under the j th false alarm identification feature, A i represents the i th false alarm interference factor, B j represents the j th false alarm identification feature; p(A i ) represents the probability of false alarm caused by the i th false alarm interference factor, p(B j ) represents the false alarm probability conforming to the j th false alarm identification feature, and p(B j |A i ) represents the probability of conforming to the j th false alarm identification feature when the i th false alarm interference factor causes false alarm.

[0046] Specifically, i is in the range of [1, n], and j is in the range of [1, m].

[0047] It can be understood that, assuming that the false alarm sample data contains M false alarm data;

[0048] If, in the M false alarm data, the number of data caused by the first false alarm interference factor is a1, the number of data caused by the second false alarm interference factor is a2,..., and the number of data caused by the n-th false alarm interference factor is an, a1+a2+…+an=M; at this time, the probability p(A i ) that the i-th false alarm interference factor causes false alarm is ai / M, such as the probability p(A1) that the first false alarm interference factor causes false alarm is a1 / M,..., and the probability p(A n ) that the n-th false alarm interference factor causes false alarm is an / M;

[0049] If, in the M false alarm data, the number of data meeting the first false alarm identification characteristic is b1, the number of data meeting the second false alarm identification characteristic is b2,..., and the number of data meeting the m-th false alarm identification characteristic is bm, b1+b2+…+bm=M; at this time, the probability p(B j ) that the j-th false alarm identification characteristic is bj / M, such as the probability p(B1) that the first false alarm identification characteristic is b1 / M,..., and the probability p(B m ) that the m-th false alarm identification characteristic is bm / M;

[0050] If, in the a1 data caused by the first false alarm interference factor, the number of data meeting the first false alarm identification characteristic is a11, the number of data meeting the second false alarm identification characteristic is a12,..., and the number of data meeting the m-th false alarm identification characteristic is a1m, a11+a12+…+a1m=a1; at this time, the probability p(B j |A1) that the j-th identification characteristic is a1j / a1, such as the probability p(B1|A1) that the first identification characteristic is a11 / a1,..., and the probability p(B m |A1) that the m-th identification characteristic is a1m / a1;

[0051] If, in the a2 data caused by the second false alarm interference factor, the number of data meeting the first false alarm identification characteristic is a21, the number of data meeting the second false alarm identification characteristic is a22,..., and the number of data meeting the m-th false alarm identification characteristic is a2m, a21+a22+…+a2m=a2; at this time, the probability p(B jp(B1|A2) = a21 ÷ a2, p(B2|A2) = a22 ÷ a2,..., p(Bm|A2) = a2m ÷ a2. m p(B1|A2) = a21 ÷ a2, p(B2|A2) = a22 ÷ a2,..., p(Bm|A2) = a2m ÷ a2.

[0052] If a3 is the number of data caused by the third false alarm interference factor, a31 is the number of data caused by the first false alarm recognition characteristic, a32 is the number of data caused by the second false alarm recognition characteristic,..., a3m is the number of data caused by the mth false alarm recognition characteristic, a31 + a32 +... + a3m = a3; p(Bj|A3) = a3j ÷ a3 is the probability of the first false alarm recognition characteristic when the third false alarm interference factor causes false alarm. j p(B1|A3) = a31 ÷ a3, p(B2|A3) = a32 ÷ a3,..., p(Bm|A3) = a3m ÷ a3. m p(B1|A3) = a31 ÷ a3, p(B2|A3) = a32 ÷ a3,..., p(Bm|A3) = a3m ÷ a3.

[0053] In a specific embodiment, it is assumed that the obtained false alarm sample data is 1000, the number of false alarms caused by the first false alarm interference factor is 300, the number of false alarms caused by the second false alarm interference factor is 300, and the number of false alarms caused by the third false alarm interference factor is 400, as shown in the following table:

[0054] False alarm interference factors Number of bars causing false alarms The probability p(A i ) of false alarms [A1] 300 300 / 1000＝0.3 [A2] 300 300 / 1000＝0.3 [A3] 400 400 / 1000＝0.4

[0055] In 1000 false alarm sample data, the number of false alarms caused by the first false alarm recognition characteristic is 200, the number of false alarms caused by the second false alarm recognition characteristic is 300, and the number of false alarms caused by the third false alarm recognition characteristic is 500, as shown in the following table:

[0056]

[0057]

[0058] 300 of the first kind of false alarm interference factors cause false alarms, the number of false alarms caused by the first false alarm identification feature 150, the second false alarm identification feature causes false alarms 100, the third false alarm identification feature causes false alarms 50; At this time, the probability of conforming to the first identification feature when the first false alarm interference factor causes false alarms is p(B1|A1) = 150 ÷ 300 = 0.5, the probability of conforming to the second identification feature when the first false alarm interference factor causes false alarms is p(B2|A1) = 100 ÷ 300 = 0.33, and the probability of conforming to the third identification feature when the first false alarm interference factor causes false alarms is p(B3|A1) = 50 ÷ 300 = 0.17;

[0059] 300 of the second kind of false alarm interference factors cause false alarms, the number of false alarms caused by the first false alarm identification feature 50, the second false alarm identification feature causes false alarms 100, the third false alarm identification feature causes false alarms 150; At this time, the probability of conforming to the first identification feature when the second false alarm interference factor causes false alarms is p(B1|A2) = 50 ÷ 300 = 0.17, the probability of conforming to the second identification feature when the second false alarm interference factor causes false alarms is p(B2|A2) = 100 ÷ 300 = 0.33, and the probability of conforming to the third identification feature when the second false alarm interference factor causes false alarms is p(B3|A2) = 150 ÷ 300 = 0.5;

[0060] 400 of the third kind of false alarm interference factors cause false alarms, the number of false alarms caused by the first false alarm identification feature 0, the second false alarm identification feature causes false alarms 100, the third false alarm identification feature causes false alarms 300; At this time, the probability of conforming to the first identification feature when the third false alarm interference factor causes false alarms is p(B1|A3) = 0 ÷ 400 = 0, the probability of conforming to the second identification feature when the third false alarm interference factor causes false alarms is p(B2|A3) = 100 ÷ 400 = 0.25, and the probability of conforming to the third identification feature when the third false alarm interference factor causes false alarms is p(B3|A3) = 300 ÷ 400 = 0.75;

[0061] The probability of conforming to a certain identification feature when a certain false alarm interference factor causes false alarms is obtained, as shown in the following table:

[0062] [p(B1|A1) = 0.5] [p(B1|A2) = 0.17] [p(B1|A3) = 0] [p(B2|A1) = 0.33] [p(B2|A2) = 0.33] [p(B2|A3) = 0.25] [p(B3|A1) = 0.17] [p(B3|A2) = 0.5] [p(B3|A3) = 0.75]

[0063] Therefore, the "identification feature - interference factor" false alarm probability list obtained is shown in the following table:

[0064]

[0065]

[0066] In some embodiments, the calculation formula of the expected value of false alarm caused by each interference factor corresponding to the to-be-tested alarm data is:

[0067]

[0068] wherein, E i represents the expected value of false alarm caused by the ith false alarm interference factor calculated based on the to-be-tested alarm data; represents the weight corresponding to the jth identification feature configured based on the false alarm sample data.

[0069] In a specific implementation, based on the above example of 1000 false alarm sample data, the weight corresponding to the 1st identification feature configured is 200, the weight corresponding to the 2nd identification feature is 300, the weight corresponding to the 3rd identification feature is 500; the set value corresponding to the false alarm interference factor A1 is set to 150, the set value corresponding to the false alarm interference factor A2 is set to 150, and the set value corresponding to the false alarm interference factor A3 is set to 200.

[0070] After obtaining a to-be-tested alarm data, the interference factor corresponding to the to-be-tested alarm data and the identification feature corresponding to the to-be-tested alarm data are analyzed, and the expected value of false alarm caused by the false alarm interference factor corresponding to the to-be-tested alarm data is calculated, as shown in the following table:

[0071]

[0072] For example, when the false alarm interference factor corresponding to the to-be-tested alarm data is A1 and there is only one false alarm identification feature, the expected value of false alarm caused by the false alarm interference factor A1 corresponding to the to-be-tested alarm data is: or or

[0073] Therefore, when the false alarm interference factor corresponding to the to-be-tested alarm data is A1 and there are two false alarm identification features, the expected value of false alarm caused by the false alarm interference factor A1 corresponding to the to-be-tested alarm data is:

[0074]

[0075] or

[0076] or

[0077] When the false alarm interference factor corresponding to the to-be-tested alarm data is A1 and there are three false alarm identification features, the expected value of false alarm caused by the false alarm interference factor A1 corresponding to the to-be-tested alarm data is: ​​​

[0078]

[0079] In some embodiments, when comparing the calculated expected value with the set value, and determining whether the to-be-tested alarm data is false alarm according to the comparison result, the following is performed: determining the expected value and the set value corresponding to each false alarm interference factor;

[0080] If the expected value corresponding to each false alarm interference factor does not exceed the set value corresponding to the false alarm interference factor, it is determined that the to-be-tested alarm data is true alarm data; if the expected value corresponding to at least one false alarm interference factor exceeds the set value corresponding to the false alarm interference factor, it is determined that the to-be-tested alarm data is false alarm data.

[0081] Based on the above example of 1000 false alarm sample data, it is assumed that a to-be-tested alarm data is obtained, and it is analyzed that the interference factors corresponding to the to-be-tested alarm data are A1, A2 and A3, the identification features corresponding to the interference factor A1 are B1 and B3, the identification features corresponding to the interference factor A2 are B1, B2 and B3, and the identification features corresponding to the interference factor A3 are B2; at this time,

[0082] The expected value of false alarm caused by the false alarm interference factor A1 corresponding to the to-be-tested alarm data is: 201> The set value corresponding to the false alarm interference factor A1 is 150;

[0083] The expected value of false alarm caused by the false alarm interference factor A2 corresponding to the to-be-tested alarm data is: 300> The set value corresponding to the false alarm interference factor A2 is 150;

[0084] The expected value of false alarm caused by the false alarm interference factor A3 corresponding to the to-be-tested alarm data is: 99< The set value corresponding to the false alarm interference factor A3 is 200;

[0085] Therefore, the expected values of false alarms caused by two false alarm interference factors in the to-be-tested alarm data exceed the set values corresponding to the false alarm interference factors, and therefore it is determined that the to-be-tested alarm data is false alarm data.

[0086] Embodiment 2

[0087] It should be noted that in order to enhance the relevance between the data in the false alarm sample data and the nearest false alarm data, and improve the accuracy of the true / false alarm detection of the gas alarm, the false alarm sample data needs to be updated each time new false alarm data is detected.

[0088] Therefore, the embodiment differs from the embodiment 1 in that the method for distinguishing the true alarm from the false alarm of the gas alarm further comprises: after detecting the new false alarm data, storing the false alarm data into the false alarm sample data, and updating the false alarm probability list of the "recognition feature-disturbing factor", the weight corresponding to each false alarm recognition feature, and the set value corresponding to each false alarm disturbing factor.

[0089] Specifically, the new false alarm data comprises the disturbing factor (target disturbing factor) and the recognition feature (target recognition feature) causing the false alarm.

[0090] It should be further noted that after storing the new false alarm data into the false alarm sample data, the number of false alarms caused by the disturbing factors, the number of false alarms caused by the recognition features, the number of false alarms caused by the disturbing factors, and the number of false alarms caused by various false alarm recognition features will change, thereby causing the probability values related to the target disturbing factor and the target recognition feature in the false alarm probability list of the "recognition feature-disturbing factor" to change; in addition, the update of the false alarm sample data will also cause the set value corresponding to the target disturbing factor to change.

[0091] Therefore, when there is new false alarm data, the false alarm probability list of the "recognition feature-disturbing factor", the weight corresponding to each false alarm recognition feature, and the set value corresponding to the false alarm disturbing factor need to be updated.

[0092] Embodiment 3

[0093] Based on the above embodiments, the embodiment provides a specific implementation of a system for distinguishing the true alarm from the false alarm of a gas alarm.

[0094] As shown in the accompanying drawings, Figure 3 the system for distinguishing the true alarm from the false alarm of the gas alarm comprises an alarm data identification module and a task generation module, wherein the alarm data identification module is configured to distinguish the to-be-tested alarm data by using the method for distinguishing the true alarm from the false alarm of the gas alarm in the embodiments 1 or 2.

[0095] The task generation module is configured to generate an alarm processing task when the to-be-tested alarm data is determined to be true alarm data, and not to generate an alarm processing task when the to-be-tested alarm data is determined to be false alarm data.

[0096] It should be noted that when new alarm data is detected, the gas company, the fire unit, and the property management unit, etc. related units usually need to send staff to the scene for viewing; however, the energy of the staff is limited every day, and if the staff is sent to the scene regardless of true alarm or false alarm, it will inevitably waste the cost of human resources, and may also cause the alarm that really needs to be handled to be not handled in time.

[0097] To solve the problem, the embodiment provides a system for distinguishing true alarms and false alarms of a gas alarm, wherein the system generates an alarm processing task only when it is determined that the alarm data to be measured is true alarm data, so as to effectively filter false alarm data; therefore, in the case of frequent gas explosion accidents, the system can well identify the false alarm problem of the alarm, realize accurate identification of alarm information on the alarm platform side, reduce the frequency of false alarms, and better guide the operation of gas companies and the like, and promote better operation of gas companies and the like.

[0098] The system can also ensure safe, smooth and normal operation of the urban gas pipe network, without increasing labor costs and reducing the work intensity of rescue operation and maintenance of gas companies and the like.

[0099] Embodiment 4

[0100] Based on the above-mentioned embodiments, the embodiment provides a specific implementation of a device for distinguishing true alarms and false alarms of a gas alarm.

[0101] Specifically, the device for distinguishing true alarms and false alarms of a gas alarm comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus; the memory is used for storing a computer program; and the processor is used for executing the program stored on the memory to realize the method for distinguishing true alarms and false alarms of a gas alarm in the embodiment 1 or 2.

[0102] Embodiment 5

[0103] Based on the above-mentioned embodiments, the embodiment also provides a readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processor to execute the method for distinguishing true alarms and false alarms of a gas alarm in the embodiment 1 or 2.

[0104] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer non-transitory readable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer program code.

[0105] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0106] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0107] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0108] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the technical solutions of the present application; although the present application is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the specific implementation ways of the present application can be modified, or some technical features can be replaced by other technical features with the same function; without departing from the spirit of the technical solutions of the present application, all of them should be included in the technical solutions of the present application.

Claims

1. A method of discriminating between true and false alarms of a gas alarm, characterized in that, The method comprises the following steps: acquiring false alarm sample data; based on the false alarm sample data, determining false alarm interference factors and false alarm identification features; calculating the probability of false alarms caused by each false alarm interference factor under each false alarm identification feature to construct a "identification feature-interference factor" false alarm probability list; determining the weight corresponding to each false alarm identification feature; in the true or false alarm discrimination, acquiring a piece of alarm data to be tested and determining the interference factors and identification features corresponding to the alarm data to be tested as target interference factors and target identification features; extracting the target false alarm probability value from the "identification feature-interference factor" false alarm probability list through the target interference factors and the target identification features; based on the weight and the target false alarm probability value, calculating the expected value of false alarms caused by each interference factor corresponding to the alarm data to be tested; comparing the calculated expected value with a set value, and discriminating whether the alarm data to be tested is a false alarm according to the comparison result.

2. The method of claim 1, wherein the method further comprises: The calculation formula of the probability of false alarms caused by each false alarm interference factor under each false alarm identification feature is as follows: wherein p(A i |B j ) represents the probability of false alarm caused by the i-th false alarm interference factor under the j-th false alarm identification feature, A i represents the i-th false alarm interference factor, B j represents the j-th false alarm identification feature; p(A i ) denotes the probability of false alarm caused by the i-th false alarm interference factor, p(B j ) denotes the probability of false alarm conforming to the j-th false alarm identification feature, and p(B j |A i ) denotes the probability of false alarm conforming to the i-th false alarm identification feature when the i-th false alarm interference factor causes false alarm.

3. The method of claim 2, wherein the method further comprises: The calculation formula of the expected value of false alarms caused by each interference factor corresponding to the alarm data to be tested is as follows: Ei = Ei + Ei i Ei represents the expected value of false alarm caused by the ith false alarm interference factor calculated based on the alarm data to be measured. represents the weight corresponding to the jth identification feature configured based on the false alarm sample data.

4. The method of claim 1, wherein: In the comparison of the calculated expected value with the set value and the discrimination of whether the alarm data to be tested is a false alarm according to the comparison result, the following steps are performed: determining the expected value corresponding to each false alarm interference factor and the set value; if the expected value corresponding to each false alarm interference factor does not exceed the set value corresponding to the false alarm interference factor, it is determined that the alarm data to be tested is true alarm data; if the expected value corresponding to at least one false alarm interference factor exceeds the set value corresponding to the false alarm interference factor, it is determined that the alarm data to be tested is false alarm data.

5. The method of claim 4, wherein the method further comprises: Further comprising: after detecting new false alarm data, storing the false alarm data into the false alarm sample data, and updating the "identification feature-interference factor" false alarm probability list, the weight corresponding to each false alarm identification feature, and the set value corresponding to each false alarm interference factor.

6. A system for discriminating between true and false alarms of a gas alarm, characterized by: The alarm data discrimination module is used to discriminate the true or false alarm of the gas alarm by using the method of any one of claims 1 to 5. The task generation module is used to generate an alarm processing task when it is determined that the alarm data to be tested is true alarm data, and not to generate an alarm processing task when it is determined that the alarm data to be tested is false alarm data. The processor, the communication interface, and the memory complete communication with each other through the communication bus.

7. A device for discriminating between true and false alarms of a gas alarm, characterized in that: The memory is used to store computer programs. The processor is used to execute the programs stored on the memory to realize the method of discriminating the true or false alarm of the gas alarm according to any one of claims 1 to 5. The instructions stored thereon, when executed by one or more processors, cause the processors to perform the method of discriminating the true or false alarm of the gas alarm according to any one of claims 1 to 5.

8. A readable storage medium characterized by: ​

Images