Intelligent fire-fighting supervision platform system for old-age care institutions

Through the smart fire supervision platform system, automated detection and cloud data analysis are used to solve the problem of low fire supervision and management efficiency, efficient monitoring and responsibility management of fire equipment are achieved, and fire safety response capabilities are improved.

CN119048314APending Publication Date: 2024-11-29NANJING JIURUN SECURITY TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411527988.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-30
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing fire supervision technology has problems such as large workload and insufficient staff in large buildings and nursing homes, and cannot achieve timely and accurate supervision and management, especially in terms of fire safety.

Method used

A smart fire supervision platform system for nursing homes was designed, including basic equipment units, Internet of Things units, data processing units and remote reporting units. Through automated detection and data analysis, real-time monitoring and management of fire-fighting equipment is realized, and cloud computing is used. Perform data processing and analysis, detect the operating status of the equipment regularly, and issue a one-click alarm when a fire occurs.

Benefits of technology

Improve the supervision and management efficiency of fire-fighting equipment, reduce personnel investment, realize joint responsibility management of fire-fighting equipment, ensure the accuracy and safety of data information, improve the speed of fire detection and alarm response, and enhance fire safety management ability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048314A_ABST
    Figure CN119048314A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent fire-fighting supervision platform system for an old-age care institution, and the system comprises an infrastructure unit which comprises fire-fighting equipment and supervision equipment. The Internet of Things unit comprises a communication device and a cloud calculator; the data processing unit comprises a preprocessing module and an analysis module; the management unit comprises an equipment responsibility module, an equipment list and personnel management; the remote reporting unit comprises a data reporting module and a one-key alarm module, the fire station can improve the inspection efficiency and reduce the personnel investment through the responsibility associated management of fire fighting equipment, and the collected data information can be quickly calculated, analyzed and processed; the accuracy and the analysis capability of data information are improved, the data information is reported, the inspection efficiency is improved, the personnel investment is reduced, and one-key alarm can be realized when a fire disaster occurs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of fire supervision, and in particular to a smart fire supervision platform system for a nursing home. Background Art

[0002] Elderly care institutions are a special term for social elderly care, which refers to institutions that provide comprehensive services such as food, daily living, cleaning, life care, health management, and cultural and sports entertainment activities for the elderly. It can be an independent legal entity, or a department or branch affiliated to a medical institution, enterprise, institution, social group or organization, or a comprehensive social welfare institution; the main target of elderly care institutions is the elderly, but some elderly care institutions (such as rural nursing homes) also accept orphans or disabled children or people with disabilities in their jurisdiction.

[0003] Nursing homes are the homes of the elderly, so fire monitoring is particularly important. Therefore, in order to effectively improve the fire prevention capabilities of nursing homes and create a safer and more comfortable living environment for the elderly, relevant departments should carry out efficient supervision and management of fire protection facilities in buildings, not only to ensure the quality and quantity of fire protection facilities, but also to establish a sound management mechanism to give full play to the protective role of fire protection facilities and equipment on buildings.

[0004] However, the existing fire supervision has a large workload and too few supervisory personnel, making it impossible to achieve timely and accurate supervision and management. For some large building groups or people with disabilities, fire safety is more important. In order to improve safety supervision and management, it is necessary to supervise and manage fire-fighting equipment in building groups, thereby improving safety protection and supervision and management capabilities.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In response to the problems in the related technology, the present invention proposes a smart fire supervision platform system for nursing homes to overcome the above-mentioned technical problems existing in the existing related technology.

[0007] To this end, the specific technical solution adopted by the present invention is as follows: A smart fire supervision platform system for nursing homes, comprising: A basic equipment unit, wherein the basic equipment unit includes fire-fighting equipment and monitoring equipment. The fire-fighting equipment is used to detect the environment of the nursing home to determine whether a fire has occurred, and to automatically extinguish the fire when a fire occurs. The monitoring equipment is used to monitor and supervise the fire-fighting equipment, and collect liquid level data, water pressure data, voltage data, and trigger signal data to monitor whether the fire-fighting equipment can be used normally or whether the fire-fighting equipment is damaged. An Internet of Things unit, wherein the Internet of Things unit is used to upload the detection data of the monitoring equipment and the fire-fighting equipment in the basic equipment unit, and the fire-fighting equipment is used to detect the fire situation and firefighting situation of the nursing home, and the monitoring equipment is used to regularly detect whether the fire-fighting equipment can operate normally; and the Internet of Things unit includes a cloud-based computer, and the cloud-based computer is used for the intelligent fire-fighting monitoring platform to calculate, analyze and process data information; A data processing unit, wherein the data processing unit includes a preprocessing module and an analysis module, wherein the analysis module is operated through the cloud computer, and the preprocessing module is used to preprocess the data information collected by the basic equipment unit to receive, convert, filter, amplify and clean the data information; A management unit, the management unit is used to manage the fire-fighting equipment and the supervision equipment, register the management personnel information, and clearly record the responsible management personnel of the fire-fighting equipment and the supervision equipment, and register the daily management, maintenance and replacement equipment information of the fire-fighting equipment and the supervision equipment. The management unit includes an equipment responsibility module, an equipment list and personnel management; A remote reporting unit, which is used to transmit data information to a fire station and to sound an alarm when a fire occurs in a nursing home. The remote reporting unit includes a data reporting module and a one-button alarm module. The data reporting module uploads the data information of the smart fire supervision platform to the fire station through the Internet of Things unit, and the one-button alarm module sounds an alarm through the Internet of Things unit.

[0008] As a preferred embodiment, the preprocessing module includes a receiving component for acquiring data information, a conversion component for performing analog-to-digital conversion on the data information, a filtering component for filtering out noise from the data information, a gain component for amplifying the data information, and a cleaning component for clearing abnormal data from the data information.

[0009] As a preferred embodiment, the filtering component is used to smooth the data information and remove noise. The filtering component uses the following calculation formula to process the data information: , in, Represented as the original data corresponding to time point t in the data information, Represented as a time point in the data information The corresponding smoothing value is is a smoothing factor used to assign weights to new data points and historical data points, and Indicates data information in scale and location The transformation structure on is the conjugate of the wavelet transform, is the scale parameter, is the translation parameter.

[0010] As a preferred embodiment, the cleaning component is detected by the following algorithm: , in, It is expressed as the calculation value of data information cleaning. Indicates data information. It is expressed as the mean of the data group in the data information, is expressed as the standard deviation of the data set in the data information, and is expressed as the number of data in the data set, and It is represented as the position of the data in the data group, and the judgment of the anomaly is >2, the data point is an outlier. For the data set Data information.

[0011] As a preferred embodiment, when the cleaning component performs abnormality detection on the data information, it first normalizes the data information and maps the data information between [0, 1], and the normalization formula is as follows: ,in, Represented as the minimum value in the data information, Represented as the maximum value in the data information; For the replacement of abnormal data, the average value is used for replacement: ,in, is the amount of non-anomalous data.

[0012] As a preferred embodiment, the analysis module uses a decision tree to calculate and process the fire fault assessment, realizes the assessment and prediction of the basic equipment unit in the smart fire supervision platform, and determines whether there is damaged equipment in the basic equipment unit, and the calculation formula is as follows: , where T is the number of trees, and It is the prediction result of the tth tree. Through multiple predictions of the equipment, it is determined whether there are faulty or damaged equipment in the basic equipment unit in the smart fire supervision platform.

[0013] As a preferred embodiment, when the analysis module performs fire fault assessment through a decision tree, it selects the optimal features of the processed data information, and the processed data information includes liquid level data, water pressure data, voltage data and trigger signal data. Perform information gain processing on the data information of liquid level data, water pressure data, voltage data and trigger signal data, and convert the collected data information into a data set, and: , in, It is a dataset The entropy of is the optimal feature mean Dataset under conditions The entropy of is the information gain The entropy of is the adjustment factor, is the complexity of the feature.

[0014] As a preferred embodiment, the data set The entropy is used to calculate the uncertainty of the data set, and the calculation formula is as follows: , in, represents the number of categories, It is Probability of class; and For the dataset Divide and split the data set according to information gain to form subsets; Compare the information gains of all features and select the feature with the largest information gain as the split feature of the current node; divide the data set into subsets according to the selected features, and each subset corresponds to a value of the feature; repeat the above process for each subset until the stopping condition is met; and the stopping condition is that the number of samples in the node is less than 10 data information.

[0015] As a preferred embodiment, after the decision tree is constructed, the decision tree is pruned by complexity penalty: ; is the error rate of the decision tree; is the number of leaf nodes of the decision tree; is the maximum depth of the decision tree; Penalty coefficient for controlling tree size; is the penalty coefficient for controlling the depth of the tree.

[0016] As a preferred embodiment, the liquid level data, the water pressure data, the voltage data and the trigger signal data respectively establish a decision tree, and the number of trees T is 4, namely, the liquid level decision tree, the water pressure decision tree, the voltage decision tree and the trigger signal decision tree. Fire supervision evaluation is performed on the prediction results of the four decision trees, and the evaluation levels are as follows: One decision tree predicts a malfunction or damaged equipment, which is considered a first-level fire accident; The second decision tree predicts a faulty or damaged equipment, which is considered a Level 2 fire incident; The three decision trees predict that if the equipment is faulty or damaged, it will be considered a serious fire accident; If the four decision trees predict a malfunction or damaged equipment, it is considered an emergency fire incident.

[0017] The beneficial effects of the present invention are: 1. The present invention realizes the registration of fire-fighting equipment through the basic equipment unit, and selects manual detection and automatic detection according to the characteristics of the fire-fighting equipment, so as to facilitate the supervision and management of all fire-fighting equipment, and records all fire-fighting equipment through the management unit, and realizes the joint responsibility processing of the equipment in the basic equipment unit through the equipment responsibility module in the management unit, so as to facilitate the tracking of the management of the basic equipment unit and realize a clear management system; 2. When the present invention is in use, data transmission is performed through the Internet of Things unit, and functions such as long-distance, short-distance and direct call alarms are realized, and the analysis module in the data processing performs calculation and analysis processing through a cloud computer, thereby improving the calculation efficiency of the platform system and reducing the calculation pressure of the platform system; 3. When the present invention is used, the data information collected by the basic equipment unit is pre-processed and analyzed by the data processing unit, and the pre-processing module receives, converts, filters, amplifies and cleans the data information, effectively improving the accuracy and security of the data information, especially the filtering process, improving the smoothing and denoising of the data information, improving the processing efficiency, and cleaning the data information, which can clear the abnormal data in the collected data information, thereby improving the accuracy of the data and preventing the abnormality of supervision caused by the error data; 4. When the present invention is in use, the analysis module performs calculation and analysis based on the cloud computer, thereby improving the calculation efficiency of the platform system and reducing the calculation pressure of the platform system. The analysis module determines the supervision results of the basic equipment unit by means of a decision tree calculation method, calculates whether the basic equipment unit has a fault, and performs maintenance processing through fire management personnel, and records the maintenance information and fault information in the equipment list or the equipment responsibility module, so as to facilitate uploading of data information, so that the fire station can improve the inspection efficiency, reduce personnel input, coordinate the management of various supervision and inspection departments, solve the problem of unclear responsibility chain of fire management in nursing homes, and realize normalization and precision update of fire database; 5. When the present invention is in use, the collected data information can also be reported to the fire station through the remote reporting unit, and when a fire occurs, the alarm can be processed through the one-key alarm module, so that the fire station can improve the inspection efficiency, reduce the personnel input, and coordinate the management of various supervision and inspection departments; The present invention has the advantages of joint responsibility management of fire-fighting equipment, so that fire-fighting equipment can be easily registered, including equipment information, maintenance information and fault information, so that fire stations can improve inspection efficiency, reduce personnel input, and coordinate the management of various supervision and inspection departments, as well as realize rapid calculation and analysis processing of collected data information, improve the accuracy and analysis ability of data information, and report data information, so that fire stations can improve inspection efficiency, reduce personnel input, and in the event of a fire, can realize one-click alarm, thereby improving the efficiency of fire station dispatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required to be used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 It is a structural diagram of a smart fire supervision platform system for a nursing home according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in the field should be able to understand other possible implementations and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0021] According to an embodiment of the present invention, a smart fire supervision platform system for a nursing home is provided.

[0022] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, a smart fire monitoring platform system for a nursing home according to an embodiment of the present invention includes: A basic equipment unit, wherein the basic equipment unit includes fire-fighting equipment and monitoring equipment. The fire-fighting equipment is used to detect the environment of the nursing home to determine whether a fire has occurred, and to automatically extinguish the fire when a fire occurs. The monitoring equipment is used to monitor the fire-fighting equipment to determine whether the fire-fighting equipment can be used normally or whether the fire-fighting equipment is damaged. The firefighting equipment includes manually detected fire extinguishers, fire axes, fire extinguisher boxes for storing fire extinguishers and fire axes, fire hoses, fire doors, fire suits and breathing apparatus, including automatically detected fire hydrants, smoke detectors, temperature detectors, fire alarm hosts, automatic sprinkler systems and emergency lighting; Classify firefighting equipment according to its characteristics, and implement detection of different types of firefighting equipment. Use automated detection for intelligent equipment to reduce personnel input. Use manual detection for equipment that cannot be intelligently detected, and store the detection result information in the equipment list or equipment responsibility module to achieve clear equipment management. The fire hydrant and the automatic sprinkler fire extinguishing system are automatically detected by arranging a liquid level detection sensor and a pressure sensor at the end, that is, the water level is detected by the liquid level detection sensor, the water pressure is detected by the pressure sensor, the smoke detector, the temperature detector, the fire alarm host and the emergency lighting are detected by the voltage acquisition circuit to see whether they are powered on, and by actively triggering the smoke detector and the temperature detector, it is detected whether the fire alarm host will sound an alarm to prompt personnel to take timely action or evacuate; It should be noted that different equipment is tested by different detection methods. Fire hydrants and automatic sprinkler systems need to be filled with water at all times. Automated detection is completed through liquid level and water pressure, and real-time detection can be achieved, which is convenient for reporting liquid level data and water pressure data. For the detection of electrical equipment, the voltage is detected and voltage data is obtained to determine whether the power supply is normal. In addition, real-time detection can be performed by actively triggering or actively sending pulse information to obtain trigger signal data, thereby improving the accuracy of detection. An Internet of Things unit, wherein the Internet of Things unit is used to upload the detection data of the monitoring equipment and the fire-fighting equipment in the basic equipment unit, and the fire-fighting equipment is used to detect the fire situation and firefighting situation of the nursing home, and the monitoring equipment is used to regularly detect whether the fire-fighting equipment can operate normally; and the Internet of Things unit includes a cloud-based computer, and the cloud-based computer is used for the intelligent fire-fighting monitoring platform to calculate, analyze and process data information; A data processing unit, wherein the data processing unit includes a preprocessing module and an analysis module, wherein the analysis module is operated through the cloud computer, and the preprocessing module is used to preprocess the data information collected by the basic equipment unit to receive, convert, filter, amplify and clean the data information; A management unit, the management unit is used to manage the fire-fighting equipment and the supervision equipment, register the management personnel information, and clearly record the responsible management personnel of the fire-fighting equipment and the supervision equipment, and register the daily management, maintenance and replacement equipment information of the fire-fighting equipment and the supervision equipment. The management unit includes an equipment responsibility module, an equipment list and personnel management; The personnel management is used to register fire management personnel, the equipment list is used to register and manage the basic equipment units, and the equipment responsibility module is used to establish the connection between fire management personnel and basic equipment, that is, to assign fire management personnel to each basic equipment for management and maintenance, and establish responsibility management of basic equipment; The equipment list includes the type, model, installation location, manufacturer, installation date and maintenance record of the equipment, as well as the inspection record of the equipment, and the fire management personnel are connected with the equipment in the equipment list to form the equipment responsibility module, and an identification code is affixed to each device for easy tracking and management, and the identification code includes an NFC tag or a QR code, and the type, model, installation location, manufacturer, installation date and maintenance record of the equipment are entered in the NFC tag or QR code, as well as the information of the fire management personnel; It should be noted that by establishing a connection between management personnel and fire-fighting equipment, the joint responsibility between management personnel and fire-fighting equipment can be achieved, which is convenient for tracking fire management and can improve the clarity of fire management. In addition, the information of the equipment can be labeled or converted into a QR code, which is convenient for spot checks and inspections of the equipment and improves the efficiency of inspections. A remote reporting unit, which is used to transmit data information to a fire station and to give an alarm when a fire occurs in a nursing home. The remote reporting unit includes a data reporting module and a one-key alarm module. The data reporting module uploads the data information of the smart fire monitoring platform to the fire station through the Internet of Things unit, and the one-key alarm module gives an alarm through the Internet of Things unit. That is, when a fire occurs in a nursing home, the one-key alarm module contacts the fire station through the Internet of Things unit to realize an automatic fire alarm; The Internet of Things unit includes a communication device, and the communication device includes long-distance transmission, short-distance transmission and telephone communication. The long-distance transmission uses an Ethernet device to transmit the collected and processed data information to the fire station, that is, the data reporting module is connected to the fire station through the Ethernet device, and the collected and processed data information is uploaded, and the upload time interval is one month. The short-distance transmission uses a wireless network or Bluetooth communication to transmit the collected data information to the data processing unit. The telephone communication establishes a virtual phone through the Internet or uses a SIM card for calls, that is, the one-button alarm module contacts the fire station through the telephone communication; It should be noted that the communication equipment in the Internet of Things unit facilitates the transmission of data information, which makes it easy to realize close-range data collection and data processing, and can report the collected and analyzed data information to the fire station, which is convenient for the fire station to supervise and manage. In the event of a fire, the fire station can be contacted with one click, and the address information can be carried, which is convenient for the fire station to respond to emergencies in a timely manner.

[0023] As a preferred embodiment, the preprocessing module includes a receiving component for acquiring data information, a conversion component for performing analog-to-digital conversion on the data information, a filtering component for filtering out clutter on the data information, a gain component for amplifying the data information, and a cleaning component for clearing abnormal data from the data information; It should be noted that preprocessing of data information can improve the accuracy of data information, facilitate subsequent calculation and analysis based on the collected data information, and facilitate rapid processing of data information.

[0024] As a preferred embodiment, the filtering component is used to smooth the data information and remove noise. The filtering component uses the following calculation formula to process the data information: , in, Represented as the original data corresponding to time point t in the data information, It is represented as the smoothing value corresponding to the time point t-1 in the data information, is a smoothing factor used to assign weights to new data points and historical data points, and Represents the transformation structure of data information at scale a and position b, is the conjugate of wavelet transform, a is the scale parameter, i.e. the research and expansion degree of data information, b is the translation parameter, i.e. the moving position on the time axis; It should be noted that for the collected data information, there is a lot of noise in the data information. These noises are likely to cause errors when analyzing and processing the data. The noise in the data information needs to be removed. In addition, the data information will fluctuate when it is transmitted, resulting in large data differences. Smoothing the data information can maintain the stability of the data. By combining data smoothing and denoising, the impact caused during data transmission can be reduced.

[0025] As a preferred embodiment, the cleaning component is detected by the following algorithm: , Among them, Z represents the calculated value of data information cleaning, X represents data information, It is expressed as the mean of the data group in the data information, is represented by the standard deviation of the data set in the data information, and n is represented by the number of data in the data set, and i is represented by the position of the data in the data set, and the judgment of the abnormality is >2, the data point is an outlier; It should be noted that the abnormal data in the data information will affect the analysis and processing of the data information. During data preprocessing, the abnormal data in the data information needs to be cleared to maintain the accuracy of the data information and facilitate subsequent calculation, analysis and processing of the data.

[0026] As a preferred embodiment, when the cleaning component performs abnormality detection on the data information, it first normalizes the data information and maps the data information between [0, 1], and the normalization formula is as follows: ,in, Represented as the minimum value in the data information, Represented as the maximum value in the data information; For the replacement of abnormal data, the average value is used for replacement: , where m is the amount of non-abnormal data; It should be noted that the collected data information is large. When it is directly calculated and processed, it is easy to cause high pressure on data calculation, affect the efficiency of calculation, put high pressure on the computer, and after cleaning the abnormal data, the missing positions should be supplemented to maintain the integrity of the data.

[0027] As a preferred embodiment, the analysis module uses a decision tree to calculate and process the fire fault assessment, realizes the assessment and prediction of the basic equipment unit in the smart fire supervision platform, and determines whether there is damaged equipment in the basic equipment unit, and the calculation formula is as follows: , where T is the number of trees, and is the prediction result of the t-th tree. Through multiple predictions of the equipment, it is determined whether there is any faulty or damaged equipment in the basic equipment unit in the smart fire monitoring platform; It should be noted that after the detection of fire-fighting equipment, the collected data information is analyzed and processed through the decision tree to predict the possibility of fire-fighting failures. Through multiple predictions of the equipment, it is determined whether the basic equipment unit in the smart fire-fighting supervision platform has faulty or damaged equipment, and it is convenient to repair the faulty or damaged equipment in a timely manner; As a preferred embodiment, the liquid level data, the water pressure data, the voltage data and the trigger signal data respectively establish a decision tree, and the number of trees T is 4, namely, the liquid level decision tree, the water pressure decision tree, the voltage decision tree and the trigger signal decision tree. Fire supervision evaluation is performed on the prediction results of the four decision trees, and the evaluation levels are as follows: One decision tree predicts a malfunction or damaged equipment, which is considered a first-level fire accident; The second decision tree predicts a faulty or damaged equipment, which is considered a Level 2 fire incident; The three decision trees predict that if the equipment is faulty or damaged, it will be considered a serious fire accident; The four decision trees predict that if the equipment is faulty or damaged, it will be considered an emergency fire incident; It should be noted that, through the above-mentioned classification, it is possible to determine fire failures and to facilitate maintenance of fire equipment according to the levels.

[0028] As a preferred embodiment, when the analysis module performs fire fault assessment through a decision tree, it selects the optimal features of the processed data information, and the processed data information includes liquid level data, water pressure data, voltage data and trigger signal data. Perform information gain processing on the data information of liquid level data, water pressure data, voltage data and trigger signal data, and convert the collected data information into a data set, and: , in, It is a dataset The entropy of is the optimal feature mean Dataset under conditions The entropy of is the information gain The entropy of the data set Including liquid level data, water pressure data, voltage data and trigger signal data, as well as the optimal feature mean They are the mean of liquid level data, water pressure data, voltage data and trigger signal data respectively. is the complexity of the feature, and , is the adjustment factor, and , d is the number of data in the dataset N.

[0029] It should be noted that by performing information gain on the data information, it is convenient to calculate the information gain of each feature, measure the reduction in uncertainty of information after splitting using the feature, and when calculating and processing, combine the introduction of information gain rate and feature bias to avoid information gain biased towards high-cardinality features, and adjust the information gain to suppress the preference for complex features.

[0030] As a preferred embodiment, the entropy of the data set is used to calculate the uncertainty of the data set, and the calculation formula is as follows: , in, represents the number of categories, It is Probability of class; and For the dataset Divide and split the data set according to information gain to form subsets; Compare the information gains of all features, and select the feature with the largest information gain as the split feature of the current node; divide the data set into subsets according to the selected features, and each subset corresponds to a value of the feature; repeat the above process for each subset until the stopping condition is met; and the stopping condition is that the number of samples in the node is less than 10 data information; It should be noted that the data set is split through the entropy of the data set, which facilitates the determination of a subset of the decision tree, realizes the construction of the decision tree, and determines the stop information for the recursive construction of the data information through the number of samples.

[0031] As a preferred embodiment, after the decision tree is constructed, the decision tree is pruned by complexity penalty: ; is the error rate of the decision tree; is the number of leaf nodes of the decision tree; is the maximum depth of the decision tree; Penalty coefficient for controlling tree size; Penalty coefficient for controlling tree depth; It should be noted that in order to reduce overfitting, the decision tree is pruned, and after construction, unimportant nodes are removed. Pruning the decision tree through complexity penalty can take into account both the size and depth of the tree, control the complexity more comprehensively, and more effectively suppress trees with excessive depth, thereby enhancing generalization ability, so that the decision tree can better adapt to the characteristics of different data sets and avoid excessive complexity.

[0032] To sum up, the present invention has the function of jointly managing the fire-fighting equipment, so that the fire-fighting equipment can be easily registered, including equipment information, maintenance information and fault information, so that the fire station can improve the inspection efficiency, reduce personnel input, and coordinate the management of various supervision and inspection departments, as well as realize rapid calculation and analysis of the collected data information, improve the accuracy and analysis ability of the data information, and report the data information, so that the fire station can improve the inspection efficiency, reduce personnel input, and in the event of a fire, can realize one-click alarm, thereby improving the efficiency of the fire station's dispatch.

[0033] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A smart fire supervision platform system for nursing homes, characterized in that: include: A basic equipment unit, wherein the basic equipment unit includes fire-fighting equipment and monitoring equipment. The fire-fighting equipment is used to detect the environment of the nursing home to determine whether a fire has occurred, and to automatically extinguish the fire when a fire occurs. The monitoring equipment is used to monitor and supervise the fire-fighting equipment, and collect liquid level data, water pressure data, voltage data, and trigger signal data to monitor whether the fire-fighting equipment can be used normally or whether the fire-fighting equipment is damaged. An Internet of Things unit, wherein the Internet of Things unit is used to upload the detection data of the monitoring device and the fire-fighting equipment in the basic equipment unit, and the fire-fighting equipment is used to detect the fire situation and fire-fighting situation of the nursing home, and the monitoring device is used to regularly detect whether the fire-fighting equipment can operate normally; And the Internet of Things unit includes a cloud computer, which is used for the smart fire monitoring platform to calculate, analyze and process data information; A data processing unit, wherein the data processing unit includes a preprocessing module and an analysis module, wherein the analysis module is operated through the cloud computer, and the preprocessing module is used to preprocess the data information collected by the basic equipment unit to receive, convert, filter, amplify and clean the data information; A management unit, the management unit is used to manage the fire-fighting equipment and the supervision equipment, register the management personnel information, and clearly record the responsible management personnel of the fire-fighting equipment and the supervision equipment, and register the daily management, maintenance and replacement equipment information of the fire-fighting equipment and the supervision equipment. The management unit includes an equipment responsibility module, an equipment list and personnel management; A remote reporting unit, which is used to transmit data information to a fire station and to sound an alarm when a fire occurs in a nursing home. The remote reporting unit includes a data reporting module and a one-button alarm module. The data reporting module uploads the data information of the smart fire supervision platform to the fire station through the Internet of Things unit, and the one-button alarm module sounds an alarm through the Internet of Things unit.

2. According to claim 1, a smart fire supervision platform system for nursing homes is characterized in that: The preprocessing module includes a receiving component for acquiring data information, a conversion component for performing analog-to-digital conversion on the data information, a filtering component for filtering out clutter on the data information, a gain component for amplifying the data information, and a cleaning component for clearing abnormal data from the data information.

3. According to claim 2, a smart fire supervision platform system for nursing homes is characterized in that: The filter component is used to smooth the data information and remove noise. The filter component uses the following calculation formula to process the data information: , in, Represented as the original data corresponding to time point t in the data information, Represented as a time point in the data information The corresponding smoothing value is is a smoothing factor used to assign weights to new data points and historical data points, and Indicates data information in scale and location The transformation structure on is the conjugate of the wavelet transform, is the scale parameter, is the translation parameter.

4. According to claim 3, a smart fire supervision platform system for nursing homes is characterized in that: The cleaning component is detected by the following algorithm: , in, It is expressed as the calculation value of data information cleaning. Indicates data information. Represented as the mean of the data group in the data information, is expressed as the standard deviation of the data set in the data information, and is expressed as the number of data in the data set, and It is represented as the position of the data in the data group, and the judgment of the anomaly is >2, the data point is an outlier. For the data set Data information.

5. According to claim 4, a smart fire supervision platform system for nursing homes is characterized in that: When the cleaning component performs anomaly detection on the data information, it first normalizes the data information and maps the data information between [0,1], and the normalization formula is as follows: ,in, Represented as the minimum value in the data information, Represented as the maximum value in the data information; For the replacement of abnormal data, the average is used for replacement: ,in, is the amount of non-anomalous data.

6. The intelligent fire supervision platform system for nursing homes according to claim 1 is characterized in that: The analysis module uses a decision tree to calculate and process the fire fault assessment, realizes the assessment and prediction of the basic equipment unit in the smart fire supervision platform, and determines whether there is damaged equipment in the basic equipment unit, and the calculation formula is as follows: , where T is the number of trees, and It is the prediction result of the tth tree. Through multiple predictions of the equipment, it is determined whether there are faulty or damaged equipment in the basic equipment unit in the smart fire supervision platform.

7. The intelligent fire supervision platform system for nursing homes according to claim 6 is characterized in that: When the analysis module performs fire fault assessment through the decision tree, the optimal feature selection is performed on the processed data information, and the processed data information includes liquid level data, water pressure data, voltage data and trigger signal data. Perform information gain processing on the data information of liquid level data, water pressure data, voltage data and trigger signal data, and convert the collected data information into a data set, and: , in, It is a dataset The entropy of is the optimal feature mean Dataset under conditions The entropy of is the information gain The entropy of is the adjustment factor, is the complexity of the feature.

8. The intelligent fire supervision platform system for nursing homes according to claim 7 is characterized in that: The dataset The entropy is used to calculate the uncertainty of the data set, and the calculation formula is as follows: , in, represents the number of categories, It is Probability of class; and For the dataset Divide and split the data set according to information gain to form subsets; Compare the information gains of all features and select the feature with the largest information gain as the split feature of the current node; divide the data set into subsets according to the selected features, and each subset corresponds to a value of the feature; repeat the above process for each subset until the stopping condition is met; and the stopping condition is that the number of samples in the node is less than 10 data information.

9. The intelligent fire supervision platform system for nursing homes according to claim 8 is characterized in that: After the decision tree is constructed, it is pruned by using complexity penalty: ; is the error rate of the decision tree; is the number of leaf nodes of the decision tree; is the maximum depth of the decision tree; Penalty coefficient for controlling tree size; is the penalty coefficient for controlling the depth of the tree.

10. The intelligent fire supervision platform system for nursing homes according to claim 9 is characterized in that: The liquid level data, the water pressure data, the voltage data and the trigger signal data respectively establish a decision tree, and the number of trees T is 4, namely the liquid level decision tree, the water pressure decision tree, the voltage decision tree and the trigger signal decision tree. Fire supervision evaluation is performed on the prediction results of the four decision trees, and the evaluation levels are as follows: One decision tree predicts a malfunction or damaged equipment, which is considered a first-level fire accident; The second decision tree predicts a faulty or damaged equipment, which is considered a Level 2 fire incident; The three decision trees predict that if the equipment is faulty or damaged, it will be considered a serious fire accident; If the four decision trees predict a malfunction or damaged equipment, it is considered an emergency fire incident.

Citation Information

Patent Citations

  • Fan gear lubrication system fault diagnosis method based on improved decision tree

    CN112697424A

  • Intelligent fire-fighting centralized and unified supervision platform based on AI, big data and Internet of Things

    CN116976832A

  • Fan blade damage area positioning method in strong noise environment, electronic equipment and storage medium

    CN118014940A

  • Intelligent equipment monitoring and maintenance system

    CN118134458A