Fire-fighting inspection system based on big data

Abstract

The invention discloses a fire-fighting inspection system based on big data, and relates to the technical field of fire-fighting inspection. The method comprises the following steps: acquiring inspection features corresponding to inspection contents through an inspection feature acquisition module, and obtaining a historical inspection result library corresponding to the inspection features; processing the historical inspection result library to obtain an abnormal inspection feature probability, and predicting an inspection period corresponding to each inspection feature according to the abnormal inspection feature probability; a training time period is set through a polling period prediction module, polling training is carried out on each polling feature in the training time period according to a predicted polling period, and abnormal frequencies corresponding to the polling features are obtained; correcting the inspection period according to the abnormal frequency, and obtaining a normal predicted inspection period corresponding to the inspection feature; obtaining a standard inspection mode according to the normal prediction inspection period; and the accuracy of the inspection period is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fire inspection, and in particular to a fire inspection system based on big data. Background Art

[0002] Fire inspection is an important work of regular inspection and maintenance of fire protection facilities, equipment and related systems, aiming to ensure that they can operate normally and play a role in time when a fire occurs, so as to protect the lives and property of personnel. The purpose of fire inspection is to ensure the integrity of facilities, improve emergency response capabilities and prevent fire accidents. In the current life, people attach more importance to national security, among which the use of fire protection systems in national security is extremely important; but due to the existing technology, regular inspections can be carried out to timely discover damage, failure or aging of fire protection facilities and equipment, so as to carry out maintenance or replacement to ensure that they are in good standby status; however, there are diversity in the inspection cycle, inspection content and inspection method of regular inspections; the inspection cycle cannot be updated in real time according to the status of the inspection content, which may cause abnormalities in the inspection content in advance, and then cause delayed warning; for this reason, the present invention provides a fire inspection system based on big data. Summary of the invention

[0003] In order to solve the above technical problems, the present invention provides a fire inspection system based on big data; The object of the present invention can be achieved by the following technical solutions: A fire inspection system based on big data, comprising a central processing module, wherein the central processing module is connected to an inspection feature acquisition module, a forecast inspection cycle module and an inspection mode decision module; The inspection feature collection module is used to collect inspection features corresponding to each inspection content and obtain a historical inspection result library corresponding to each inspection feature; The central processing module is used to process the historical inspection result library to obtain the abnormal inspection feature probability, and predict the inspection cycle corresponding to each inspection feature according to the abnormal inspection feature probability; The predicted inspection cycle module is used to set a training time period, perform inspection training on each inspection feature in the training time period according to the predicted inspection cycle, and obtain the abnormal frequency corresponding to the inspection feature; correct the inspection cycle according to the abnormal frequency, and obtain the normal predicted inspection cycle corresponding to the inspection feature; The inspection mode decision module is used to obtain a standard inspection mode according to a normal predicted inspection cycle.

[0004] Furthermore, the process of the inspection feature collection module acquiring the historical inspection result library includes: The inspection features corresponding to each inspection content are integrated to generate an inspection feature data set corresponding to the inspection content. The inspection features in each inspection feature data set are numbered and recorded as iu, respectively, where i=1, 2, ..., n, and n is a positive integer; u represents the inspection content; Set the inspection characteristic value model, where T iu It is represented by the inspection characteristic value corresponding to the number iu.

[0005] Further, a historical collection cycle set is set, the historical collection cycle set includes a number of different historical collection cycles, and each historical collection cycle is numbered, recorded as r, r=1, 2, ..., u, and u is a positive integer; based on the big data collection, each inspection content of each historical collection period corresponds to each inspection feature of the inspection feature data set corresponding to the historical inspection data corresponding to each inspection feature, and the corresponding inspection data standard is set according to the inspection feature, and compared with the corresponding historical inspection data, if the historical inspection data does not meet the inspection data standard, the historical inspection data is marked as abnormal inspection data, and the corresponding inspection feature is marked as abnormal inspection feature; otherwise, the historical inspection data is marked as normal inspection data, and the corresponding inspection feature is marked as normal inspection feature; and then the inspection feature value corresponding to the historical inspection data is obtained based on the inspection feature value model; The inspection feature values ​​corresponding to each inspection feature in each historical collection cycle are integrated to generate the corresponding historical inspection feature value data set, and the historical inspection feature value data sets corresponding to all inspection contents are integrated to generate the historical inspection result library.

[0006] Furthermore, the central processing module processes the historical inspection result library to obtain the abnormal inspection feature probability: Obtain the total number of abnormal inspection feature values ​​in each historical inspection feature value data set in the historical inspection result library, and obtain the proportion of the total number of inspection feature values ​​in the historical inspection feature value data set. According to the historical collection cycle and the corresponding proportion, obtain the probability of abnormal inspection features. The specific formula is: ; in, It is represented by the inspection feature corresponding number iu in the historical collection period T r The probability of abnormal inspection features corresponding to the abnormal inspection features; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of abnormal inspection features in the corresponding inspection feature value dataset; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of inspection feature values ​​in the corresponding historical inspection feature value data set; T r Indicates the historical collection period.

[0007] Furthermore, the process of predicting the inspection cycle corresponding to each inspection feature according to the probability of abnormal inspection features includes: Set the deviation value corresponding to each historical collection period of the historical collection period set corresponding to the inspection feature, and obtain the predicted inspection period corresponding to the inspection feature according to the deviation value and the probability of abnormal inspection features corresponding to each historical collection period; That is, the specific formula is: ; in, It is represented by the predicted inspection cycle corresponding to the inspection feature numbered iu; It is expressed as the deviation value of the inspection feature corresponding to the historical collection period.

[0008] Furthermore, the process of the predicted inspection cycle module obtaining the abnormal frequency corresponding to the inspection feature includes: The training time period and each predicted inspection cycle are formed into a training instruction and sent to the monitoring device of the corresponding inspection feature; the monitoring device obtains a number of inspection time nodes corresponding to the inspection feature in the training time period through the training instruction, and collects the inspection data corresponding to the inspection time nodes in real time and sends it to the inspection feature value model, and then compares it with the corresponding inspection data standard to obtain the corresponding inspection feature value, obtains the abnormal inspection time point according to the inspection feature value, and then obtains the abnormal frequency of the corresponding inspection feature according to each abnormal inspection time point; The specific formula is: ; in, It is represented by the abnormal frequency of the inspection feature corresponding to the number iu; It is represented by the total number of abnormal inspection time points corresponding to the inspection feature value of 0 during the training period; It is represented by the total number of inspection time points corresponding to the inspection features in the training time period.

[0009] Furthermore, the process of correcting the inspection cycle according to the abnormal frequency and obtaining the inspection cycle instruction corresponding to the inspection feature includes: Set the abnormal frequency threshold corresponding to each inspection feature in the training time period and compare it with the corresponding abnormal frequency. If the abnormal frequency is greater than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a shortened predicted inspection cycle; if the abnormal frequency is less than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as an extended predicted inspection cycle; if the abnormal frequency is equal to the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a normal predicted inspection cycle; The predicted inspection cycle corresponding to the shortened predicted inspection cycle is shortened, the predicted shortened inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained; similarly, the predicted inspection cycle corresponding to the extended predicted inspection cycle is extended, the predicted extended inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained.

[0010] Furthermore, the process of the inspection mode decision module acquiring the inspection mode decision data includes: The normal predicted inspection cycle corresponding to each inspection feature in the inspection feature data set of each inspection content is obtained and mapped with each inspection feature in the inspection feature data set to form a standard inspection mode corresponding to the inspection content.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention collects inspection features corresponding to each inspection content through the inspection feature collection module, and obtains the historical inspection result library corresponding to each inspection feature; processes the historical inspection result library to obtain the probability of abnormal inspection features, and predicts the inspection cycle corresponding to each inspection feature according to the probability of abnormal inspection features; and sets the training time period through the predicted inspection cycle module, performs inspection training on each inspection feature in the training time period according to the predicted inspection cycle, and obtains the abnormal frequency corresponding to the inspection feature; corrects the inspection cycle according to the abnormal frequency, and obtains the normal predicted inspection cycle corresponding to the inspection feature; obtains the standard inspection method according to the normal predicted inspection cycle; effectively predicts the normal predicted inspection cycle for different inspection features, which can not only avoid the mismatch of the same cycle for different inspection features; but also effectively improve the accuracy of the training cycle corresponding to the inspection feature, thereby avoiding the occurrence of abnormalities or delays in advance, and better improving the accuracy and efficiency of fire inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0013] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0015] like Figure 1 As shown, a fire inspection system based on big data includes a central processing module, wherein the central processing module is connected to an inspection feature acquisition module, an inspection cycle prediction module, and an inspection mode decision module; The inspection feature collection module is used to collect inspection features corresponding to each inspection content and obtain a historical inspection result library corresponding to each inspection feature; The central processing module is used to process the historical inspection result library to obtain the abnormal inspection feature probability, and predict the inspection cycle corresponding to each inspection feature according to the abnormal inspection feature probability; The predicted inspection cycle module is used to set a training time period, perform inspection training on each inspection feature in the training time period according to the predicted inspection cycle, and obtain the abnormal frequency corresponding to the inspection feature; correct the inspection cycle according to the abnormal frequency, and obtain the normal predicted inspection cycle corresponding to the inspection feature; The inspection mode decision module is used to obtain a standard inspection mode according to a normal predicted inspection cycle.

[0016] The process of the inspection feature collection module acquiring the historical inspection result library includes: The inspection contents include fire extinguishing equipment, alarm system, smoke exhaust system, fire water source inspection and fire passage; the inspection features corresponding to each inspection content are integrated to generate an inspection feature data set corresponding to the inspection content, and the inspection features in each inspection feature data set are numbered and respectively recorded as iu, where i=1, 2, ..., n, and n is a positive integer; u represents the inspection content; through Set the inspection characteristic value model, where T iu It is represented by the inspection characteristic value corresponding to the number iu; when T iu If it is 0, it means that the inspection feature is abnormal. iu If it is 1, it means the inspection characteristics are normal; In the above embodiments, it should be further explained that the inspection features are used to represent the features that need to be inspected corresponding to the inspection content; for example, the inspection features corresponding to the fire extinguishing equipment include but are not limited to fire extinguisher pressure, fire extinguishing agent validity period, fire hydrant valve, water hose, water gun head, etc.; the inspection features corresponding to the alarm system include but are not limited to detectors, alarms, buttons, signal transmission, etc., the inspection features corresponding to the exhaust and smoke protection system include but are not limited to fans, air ducts, fire dampers, smoke exhaust valves, etc., the inspection features corresponding to the fire water source inspection include but are not limited to fire water pools, water replenishment facilities, fire water tank water level, fire water tank water quality, etc., and the inspection features corresponding to the fire passage include but are not limited to fire passages that are unobstructed, no debris piled up, no vehicles occupied, safe exits, etc.

[0017] A historical collection cycle set is set, wherein the historical collection cycle set includes a number of different historical collection cycles, and each historical collection cycle is numbered, recorded as r, r=1, 2, ..., u, and u is a positive integer; based on the big data collection, each historical collection cycle corresponds to each inspection content of the inspection feature data set corresponding to each inspection feature of the historical collection cycle. The corresponding inspection data standard is set according to the inspection feature, and compared with the corresponding historical inspection data. If the historical inspection data does not meet the inspection data standard, the historical inspection data is marked as abnormal inspection data, and the corresponding inspection feature is marked as an abnormal inspection feature; otherwise, the historical inspection data is marked as normal inspection data, and the corresponding inspection feature is marked as a normal inspection feature; and then the inspection feature value corresponding to the historical inspection data is obtained based on the inspection feature value model; The inspection feature values ​​corresponding to each inspection feature in each historical collection cycle are integrated to generate the corresponding historical inspection feature value data set, and the historical inspection feature value data sets corresponding to all inspection contents are integrated to generate the historical inspection result library.

[0018] In the above embodiment, it needs to be further explained that the historical collection period is set according to the weight corresponding to the importance of the inspection content corresponding to the inspection feature, so as to better focus on a certain inspection feature.

[0019] Get the probability of abnormal inspection features: Obtain the total number of abnormal inspection feature values ​​in each historical inspection feature value data set in the historical inspection result library, and obtain the proportion of the total number of inspection feature values ​​in the historical inspection feature value data set. According to the historical collection cycle and the corresponding proportion, obtain the probability of abnormal inspection features. The specific formula is: ; in, It is represented by the inspection feature corresponding number iu in the historical collection period T r The probability of abnormal inspection features corresponding to the abnormal inspection features; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of abnormal inspection features in the corresponding inspection feature value dataset; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of inspection feature values ​​in the corresponding historical inspection feature value data set; T r It is expressed as a historical collection period; The process of the central processing module obtaining the inspection cycle corresponding to the inspection feature includes: Set the deviation value corresponding to each historical collection period of the historical collection period set corresponding to the inspection feature, and obtain the predicted inspection period corresponding to the inspection feature according to the deviation value and the probability of abnormal inspection features corresponding to each historical collection period; That is, the specific formula is: ; in, It is represented by the predicted inspection cycle corresponding to the inspection feature numbered iu; It is expressed as the deviation value of the inspection feature corresponding to the historical collection period; In the above embodiment, it needs to be further explained that the historical collection periods in the historical collection period set are not equal, and the unequal historical collection periods have abnormal inspection feature deviations, and setting corresponding deviation values ​​is beneficial to improving the accuracy.

[0020] The process of the predicted inspection cycle module obtaining the abnormal frequency corresponding to the inspection feature includes: The training time period and each predicted inspection cycle are formed into a training instruction and sent to the monitoring device of the corresponding inspection feature; the monitoring device obtains a number of inspection time nodes corresponding to the inspection feature in the training time period through the training instruction, and collects the inspection data corresponding to the inspection time nodes in real time and sends it to the inspection feature value model, and then compares it with the corresponding inspection data standard to obtain the corresponding inspection feature value, obtains the abnormal inspection time point according to the inspection feature value, and then obtains the abnormal frequency of the corresponding inspection feature according to each abnormal inspection time point; The specific formula is: ; in, It is represented by the abnormal frequency of the inspection feature corresponding to the number iu; It is represented by the total number of abnormal inspection time points corresponding to the inspection feature value of 0 during the training period; It is represented by the total number of inspection time points corresponding to the inspection features in the training time period.

[0021] In the above embodiment, it needs to be further explained that, for the same training time period, the training time points corresponding to the training features are obtained for different prediction training cycles, the corresponding inspection data are collected in real time according to the training time points, the inspection feature values ​​of the inspection data are determined, and then the corresponding abnormal frequencies are obtained, and the corresponding prediction inspection cycles are trained to improve the accuracy of the prediction training cycles.

[0022] The process of correcting the inspection cycle according to the abnormal frequency and obtaining the inspection cycle instruction corresponding to the inspection feature includes: Set the abnormal frequency threshold corresponding to each inspection feature in the training time period and compare it with the corresponding abnormal frequency. If the abnormal frequency is greater than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a shortened predicted inspection cycle; if the abnormal frequency is less than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as an extended predicted inspection cycle; if the abnormal frequency is equal to the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a normal predicted inspection cycle; The predicted inspection cycle corresponding to the shortened predicted inspection cycle is shortened, the predicted shortened inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained; similarly, the predicted inspection cycle corresponding to the extended predicted inspection cycle is extended, the predicted extended inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained.

[0023] In the above embodiment, it needs to be further explained that each predicted inspection cycle is trained through a training time period to obtain a normal predicted inspection cycle, so as to better strictly notify the fire inspection cycle, avoid abnormalities in advance, and avoid waste caused by frequent monitoring of the monitoring device.

[0024] The process of the inspection mode decision module acquiring the inspection mode decision data includes: The normal predicted inspection cycle corresponding to each inspection feature in the inspection feature data set of each inspection content is obtained and mapped with each inspection feature in the inspection feature data set to form a standard inspection mode corresponding to the inspection content. Example

[0025] Another method is proposed for the inspection mode decision module: The inspection feature values ​​corresponding to the inspection features of the standard inspection method are collected in real time, and the corresponding abnormal frequency of each inspection feature after each normal predicted inspection cycle is obtained. The abnormal frequency threshold is set through the predicted inspection cycle module, and compared with the abnormal frequency to obtain the corresponding abnormal frequency, and then the normal predicted inspection cycle is corrected to obtain the kth normal predicted inspection cycle, and then the inspection features are inspected according to the kth normal predicted inspection cycle, and the cycle is repeated.

[0026] The features and exemplary embodiments of various aspects of the present application are described in detail above. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail in combination with the drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application rather than to limit the present application. For those skilled in the art, the present application can be implemented without some of these specific details. The above description of the embodiments is merely to provide a better understanding of the present application by showing examples of the present application.

[0027] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A fire inspection system based on big data, including a central processing module, characterized in that: The central processing module is connected to the inspection feature acquisition module, the inspection cycle prediction module and the inspection mode decision module; The inspection feature collection module is used to collect inspection features corresponding to each inspection content and obtain a historical inspection result library corresponding to each inspection feature; The central processing module is used to process the historical inspection result library to obtain the abnormal inspection feature probability, and predict the inspection cycle corresponding to each inspection feature according to the abnormal inspection feature probability; The predicted inspection cycle module is used to set a training time period, perform inspection training on each inspection feature in the training time period according to the predicted inspection cycle, and obtain the abnormal frequency corresponding to the inspection feature; correct the pre-inspection cycle according to the abnormal frequency, and obtain the normal predicted inspection cycle corresponding to the inspection feature; The inspection mode decision module is used to obtain a standard inspection mode according to a normal predicted inspection cycle.

2. According to the big data-based fire inspection system of claim 1, it is characterized in that: The process of the inspection feature collection module acquiring the historical inspection result library includes: The inspection features corresponding to each inspection content are integrated to generate an inspection feature data set corresponding to the inspection content. The inspection features in each inspection feature data set are numbered and recorded as iu, respectively, where i=1, 2, ..., n, and n is a positive integer; u represents the inspection content; Set the inspection eigenvalue model, where T iu It is represented by the inspection characteristic value corresponding to the number iu.

3. A fire inspection system based on big data according to claim 2, characterized in that: A historical collection cycle set is set, wherein the historical collection cycle set includes a number of different historical collection cycles, and each historical collection cycle is numbered, denoted as r, where r=1, 2, ..., u, and u is a positive integer; based on big data, historical inspection data corresponding to each inspection feature of each inspection content corresponding to the inspection feature data set of each historical collection cycle is collected, and a corresponding inspection data standard is set according to the inspection feature, and compared with the corresponding historical inspection data; if the historical inspection data does not meet the inspection data standard, the historical inspection data is marked as abnormal inspection data, and the corresponding inspection feature is marked as an abnormal inspection feature; Otherwise, the historical inspection data is marked as normal inspection data, and the corresponding inspection features are marked as normal inspection features; then the inspection feature values ​​corresponding to the historical inspection data are obtained based on the inspection feature value model; The inspection feature values ​​corresponding to each inspection feature in each historical collection cycle are integrated to generate the corresponding historical inspection feature value data set, and the historical inspection feature value data sets corresponding to all inspection contents are integrated to generate the historical inspection result library.

4. A fire inspection system based on big data according to claim 3, characterized in that: The central processing module processes the historical inspection result library to obtain the probability of abnormal inspection features: Obtain the total number of abnormal inspection feature values ​​in each historical inspection feature value data set in the historical inspection result library, and obtain the proportion of the total number of inspection feature values ​​in the historical inspection feature value data set. According to the historical collection cycle and the corresponding proportion, obtain the probability of abnormal inspection features. The specific formula is: ； in, It is represented by the inspection feature corresponding number iu in the historical collection period T r The probability of abnormal inspection features corresponding to the abnormal inspection features; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of abnormal inspection features in the corresponding inspection feature value dataset; It is represented by the inspection feature corresponding number iu in the historical collection period T r The total number of inspection characteristic values ​​in the corresponding historical inspection characteristic value data set; T r Represents the historical collection period.

5. A fire inspection system based on big data according to claim 4, characterized in that: The process of predicting the inspection cycle corresponding to each inspection feature according to the probability of abnormal inspection features includes: Set the deviation value corresponding to each historical collection period of the historical collection period set corresponding to the inspection feature, and obtain the predicted inspection period corresponding to the inspection feature according to the deviation value and the probability of abnormal inspection features corresponding to each historical collection period; That is, the specific formula is: ； in, It is represented by the predicted inspection cycle corresponding to the inspection feature numbered iu; It is expressed as the deviation value of the inspection feature corresponding to the historical collection period.

6. A fire inspection system based on big data according to claim 5, characterized in that: The process of the predicted inspection cycle module obtaining the abnormal frequency corresponding to the inspection feature includes: The training time period and each predicted inspection cycle are formed into a training instruction and sent to the monitoring device of the corresponding inspection feature; the monitoring device obtains a number of inspection time nodes corresponding to the inspection feature in the training time period through the training instruction, and collects the inspection data corresponding to the inspection time nodes in real time and sends it to the inspection feature value model, and then compares it with the corresponding inspection data standard to obtain the corresponding inspection feature value, obtains the abnormal inspection time point according to the inspection feature value, and then obtains the abnormal frequency of the corresponding inspection feature according to each abnormal inspection time point; The specific formula is: ； in, It is represented by the abnormal frequency of the inspection feature corresponding to the number iu; It is represented by the total number of abnormal inspection time points corresponding to the inspection feature value of 0 during the training period; It is represented by the total number of inspection time points corresponding to the inspection features in the training time period.

7. A fire inspection system based on big data according to claim 6, characterized in that: The process of correcting the inspection cycle according to the abnormal frequency and obtaining the inspection cycle instruction corresponding to the inspection feature includes: Set the abnormal frequency threshold corresponding to each inspection feature in the training time period and compare it with the corresponding abnormal frequency. If the abnormal frequency is greater than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a shortened predicted inspection cycle; if the abnormal frequency is less than the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as an extended predicted inspection cycle; if the abnormal frequency is equal to the abnormal frequency threshold, the corresponding predicted inspection cycle is marked as a normal predicted inspection cycle; The predicted inspection cycle corresponding to the shortened predicted inspection cycle is shortened, the predicted shortened inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained; similarly, the predicted inspection cycle corresponding to the extended predicted inspection cycle is extended, the predicted extended inspection cycle is obtained, and inspection training is performed again until a normal predicted inspection cycle is obtained.

8. The fire inspection system based on big data according to claim 7 is characterized in that: The process of the inspection mode decision module acquiring the inspection mode decision data includes: The normal predicted inspection cycle corresponding to each inspection feature in the inspection feature data set of each inspection content is obtained and mapped with each inspection feature in the inspection feature data set to form a standard inspection mode corresponding to the inspection content.

Images