CN110807905B

CN110807905B - Community fire monitoring system based on end-edge-cloud architecture

Info

Publication number: CN110807905B
Application number: CN201911013218.XA
Authority: CN
Inventors: 丁辉; 史运涛; 王力; 董广亮; 董哲; 雷振伍; 孙德辉; 刘大千; 李超
Original assignee: North China University of Technology
Current assignee: North China University of Technology
Priority date: 2019-10-23
Filing date: 2019-10-23
Publication date: 2020-12-18
Anticipated expiration: 2039-10-23
Also published as: CN110807905A

Abstract

The embodiment of the invention provides a community fire monitoring system based on a terminal-edge-cloud architecture, which comprises: the system comprises a 6LoWPAN wireless sensing network, a cloud gateway and a cloud platform; the 6LoWPAN wireless sensing network has a fire risk sensing function; the cloud gateway is connected with the 6LoWPAN wireless sensing network and is accessed to the cloud platform through the NB-IOT network; the cloud gateway transmits data by using an MQTT protocol; the cloud platform utilizes the fuzzy neural network early warning model of the fire hazard of the community and edge calculation of the cloud gateway to realize real-time early warning and control of the fire hazard risk in a cooperative mode. The embodiment of the invention realizes the low-cost arrangement of the fire monitoring network, improves the signal quality and enhances the reliability; data transmission of a wireless sensor network and a cloud platform is realized through a single MQTT protocol, and the data transmission efficiency and the timeliness of fire early warning are improved; the sensing, transmission, analysis, early warning and control of the fire risk of the community are realized in an all-round manner.

Description

Community fire monitoring system based on end-edge-cloud architecture

Technical Field

The invention relates to the technical field of communication, in particular to a community fire monitoring system based on an end-edge-cloud architecture.

Background

Fire is one of the most serious disasters, and directly threatens property and life safety. With the development of economy in China, large-scale high-rise and super high-rise buildings are more and more, and if a fire disaster happens, the loss is heavy. Therefore, effective measures are taken before the fire disaster happens to carry out comprehensive monitoring, and the method has great significance for preventing the fire disaster in the past.

The existing community fire monitoring system adopts a wired or wireless mode to transmit data. Among them, the wired installation method is high in cost and difficult to install. For a wireless installation mode, the existing technologies directly adopt communication technologies of wireless wide area networks such as GPRS and 4G to carry out networking on fire sensing equipment, but most of community fire fighting facilities are deployed in buildings with complex structures, and the requirement on the signal quality of the wireless sensing equipment is high, so that the reliability of the networking mode system is difficult to guarantee. In addition, for massive fire-fighting wireless sensing equipment, wireless telecommunication network transmission is directly adopted, the cost is very high, and the existing telecommunication network is difficult to ensure simultaneous online communication of massive wireless sensing equipment in a complex community building. And the wireless sensing equipment is very close in distance and does not need to directly adopt remote communication in the community fire scene, thereby causing the waste of resources.

With the development of wireless communication technology and internet of things technology, a wireless sensor network with a potentially great application value attracts attention and research of people. The advantages of low power consumption and low cost enable the Zigbee protocol to be widely applied in the field of wireless sensing, and particularly the application of the Zigbee protocol based on IEEE 802.15.4 is widely applied to various industries such as industrial control, smart cities, medical treatment, home furnishing and the like. However, the Zigbee protocol and its terminal can only implement interaction with the Zigbee terminal, and lack support for IPV6, and cannot implement IP addressing, so it is impractical to directly access the WSN to the internet due to the limitations of its hardware performance and its working environment. The research on the 6LoWPAN wireless sensor network only stays at the wireless personal area network level, and the wide area network interconnection problem is not considered. The fire-fighting network system is generally applied in a local area, the problem of interconnection and intercommunication between a personal area network and a wide area network is not solved, and a certain distance is kept from the aim of really realizing the interconnection of everything in the community fire-fighting system.

Therefore, how to acquire fire monitoring information by using a wireless sensor network, and quickly send data to a cloud platform, and realize remote monitoring or risk early warning is a problem to be solved urgently.

Disclosure of Invention

In order to solve the problems in the prior art, embodiments of the present invention provide a community fire monitoring system based on an end-edge-cloud architecture, where the system includes a 6LoWPAN wireless sensor network on an end side, a cloud gateway with edge computing capability, and a cloud platform on a cloud side; wherein: the wireless sensing nodes in the 6LoWPAN wireless sensing network have a fire risk sensing function, so that the fire risk of the community can be sensed in an all-round manner; the OT layer of the cloud gateway is connected with the 6LoWPAN wireless sensing network, and the IT layer of the cloud gateway is accessed to the cloud platform through an NB-IOT network, so that interconnection and intercommunication between a wireless personal area network and a wide area network are realized; for 6LoWPAN and NB-IOT heterogeneous networks, the cloud gateway transmits data by using an MQTT protocol to realize the MQTT network to the end; moreover, the cloud gateway realizes the preliminary judgment of the fire risk through the edge computing capability of the cloud gateway; the cloud platform has the functions of community fire monitoring data storage, big data calculation and early warning, a community fire fuzzy neural network early warning model is stored in the cloud platform in advance, and real-time early warning and control of fire risks are achieved through cooperation of the community fire fuzzy neural network early warning model and edge calculation of a cloud gateway. .

Further, the wireless sensor nodes in the 6LoWPAN wireless sensor network include a wireless sensing device and a wireless execution device, wherein: in data uplink, the wireless sensing equipment is used for acquiring the community fire monitoring data, converting the community fire monitoring data into an MQTT protocol data format, and then converging the community fire monitoring data to a cloud gateway through the 6LoWPAN wireless sensing network, wherein the cloud gateway transmits the community fire monitoring data in the MQTT protocol data format to a cloud platform through an NB-IOT wireless wide area network; in data downlink, if the cloud platform judges that the fire risk exists, the cloud platform sends a control instruction in an MQTT protocol data format to the cloud gateway through an NB-IOT wireless wide area network, the control instruction comprises address information of target wireless execution equipment, and the cloud gateway sends the control instruction in the MQTT protocol data format to the target wireless execution equipment according to the address information so that the target wireless execution equipment executes corresponding actions.

Further, the cloud gateway comprises a radio frequency unit, a microprocessor unit and a cloud transmission unit; the microprocessor unit comprises an MQTT proxy server, an MQTT client and a threshold early warning module; wherein: in the data uplink, the radio frequency unit is used for receiving the community fire monitoring data converted into an MQTT protocol data format and converging the community fire monitoring data to the MQTT proxy server; the MQTT proxy server is used for sending the community fire monitoring data to the MQTT client; the MQTT client is used for sending the community fire monitoring data to the threshold early warning module and sending the community fire monitoring data to the cloud platform through the cloud transmission unit; the threshold early warning module is connected with the MQTT client and used for receiving the community fire monitoring data acquired by the MQTT client, comprehensively analyzing the community fire monitoring data of each wireless sensing device and judging whether the abnormality occurs or not; if the situation that the abnormality occurs is judged and known, the abnormal information is sent to the MQTT client; the MQTT client sends the abnormal information to the cloud platform through the cloud transmission unit; in data downlink, the cloud platform sends the control instruction in the MQTT protocol data format to the cloud transmission unit through an NB-IOT wireless wide area network, and the cloud transmission unit sends the control instruction to the target wireless execution device sequentially through the MQTT client, the MQTT proxy server and the radio frequency unit.

Further, the cloud platform is further configured to compare the community fire monitoring data with a preset threshold value when it is judged that there is a fire risk, so as to determine a target wireless sensing device of which the community fire monitoring data exceeds the preset threshold value, and obtain the address information of the target wireless execution device in the same area according to the number information of the target wireless sensing device.

Further, the radio frequency unit is further configured to receive execution result data of the target wireless execution device converted into an MQTT protocol data format, and send the execution result data to the cloud platform through the cloud gateway.

Further, the abnormal information is false alarm information, fault information or fire alarm information, wherein: the threshold early warning module is used for comprehensively analyzing the community fire monitoring data of each wireless sensing device, and if the data of a single wireless sensing device is abnormal occasionally, the abnormal information is known to be false alarm information; if the data of a single wireless sensing device is continuously abnormal and the data of the rest wireless sensing devices are normal, acquiring that the abnormal information is fault information; and if the data of the plurality of wireless sensing devices in the same area are abnormal, acquiring that the abnormal information is fire alarm information.

Furthermore, the threshold early warning module is further configured to send sampling period adjustment information to the wireless sensing device with data abnormality through the MQTT client, the MQTT proxy server, and the radio frequency unit, and the wireless sensing device increases sampling frequency according to the sampling period adjustment information.

Further, the threshold early warning module is further configured to, when the abnormal information is the fire alarm information, obtain a second probability of occurrence of a fire according to a ratio of abnormal data in an area where the fire alarm information is generated, and send information of the second probability to the cloud platform; when the cloud platform is used for realizing real-time early warning and control of fire risks by utilizing the community fire fuzzy neural network early warning model and the edge calculation of the cloud gateway in a coordinated manner, the cloud platform is specifically used for: inputting the community fire monitoring data into a pre-established community fire fuzzy neural network early warning model, and obtaining a first probability of fire according to an output result of the community fire fuzzy neural network early warning model; carrying out weighted summation on the first probability and the second probability through a preset weight value to obtain a third probability of fire occurrence, and judging whether the fire risk exists according to the third probability; and if the fire risk is judged and known, sending a control instruction to the target wireless execution equipment through the cloud gateway.

Furthermore, the wireless sensing equipment comprises a temperature and humidity sensor, a smoke sensor, a flame sensor, a door magnetic switch sensor, a combustible gas sensor and a fire alarm; the wireless execution equipment comprises a spraying device, a ventilation valve and a rolling door.

Furthermore, the cloud gateway and the 6LoWPAN wireless sensing network are multiple and in one-to-one correspondence.

According to the community fire monitoring system based on the end-edge-cloud architecture, disclosed by the embodiment of the invention, the 6LoWPAN wireless sensing network is arranged for carrying out fire monitoring, so that the low-cost arrangement of the fire monitoring network is realized, the signal quality is improved, and the reliability is enhanced; the 6LoWPAN wireless sensing network and the cloud platform are connected through the cloud gateway, so that the cloud of the community fire monitoring data is realized; data transmission is carried out by adopting a single MQTT protocol, so that one MQTT network is completed, the data transmission efficiency is improved, and the timeliness of fire early warning is improved; the sensing, transmission, analysis, early warning and control of the fire risk of the community are realized in an all-round manner.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic structural diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating the operation of a community fire monitoring system based on a peer-to-peer cloud architecture according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a wireless sensor node in a 6LoWPAN wireless sensor network of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention;

FIG. 7 is a schematic structural diagram of a community fire monitoring system based on a peer-to-peer-cloud architecture according to an embodiment of the present invention;

fig. 8 is a schematic diagram of a data uplink workflow of a community fire monitoring system based on a terminal-edge-cloud architecture according to an embodiment of the present invention;

fig. 9 is a schematic data downlink workflow diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a schematic structural diagram of a community fire monitoring system based on an end-edge-cloud architecture according to an embodiment of the present invention. As shown in fig. 1, the system includes an end-side 6LoWPAN wireless sensor network 1, a cloud gateway 2 with edge computing capability, and a cloud platform 3 in a cloud; wherein: the wireless sensing nodes in the 6LoWPAN wireless sensing network 1 have a fire risk sensing function, and all-round sensing of the fire risk of the community is achieved; the cloud gateway 2 is connected with the 6LoWPAN wireless sensing network 1 and is accessed to the cloud platform 3 through an NB-IOT network, so that interconnection and intercommunication between a wireless personal area network and a wide area network are realized; for 6LoWPAN and NB-IOT heterogeneous networks, the cloud gateway 2 utilizes an MQTT protocol to transmit data, and MQTT one-network-to-all is realized; moreover, the cloud gateway 2 realizes the preliminary judgment of the fire risk through the edge computing capability; the cloud platform 3 has the functions of community fire monitoring data storage, big data calculation and early warning, the community fire fuzzy neural network early warning model is stored in the cloud platform 3 in advance, and real-time early warning and control of fire risks are achieved through cooperation of the community fire fuzzy neural network early warning model and edge calculation of the cloud gateway 2.

The 6LoWPAN wireless sensor network 1 refers to a 6LoWPAN based wireless sensor network. Among them, 6LoWPAN is a low-speed wireless personal area network standard based on IPv6, i.e. IPv6 over IEEE 802.15.4. The wireless sensing nodes in the 6LoWPAN wireless sensing network 1 have a fire risk sensing function, and all-around sensing of the fire risk of the community is achieved through the arranged multiple wireless sensing devices. According to the embodiment of the invention, the cloud gateway 2 is connected with the 6LoWPAN wireless personal area network with low power consumption and high reliability, so that the communication quality under the scene of the community complex building is ensured, and the defect of poor wireless wide area network signal quality under the scene of the community complex building is effectively overcome. The cloud gateway 2 is connected with the 6LoWPAN wireless sensing network 1(OT layer), the IT layer of the cloud gateway 2 is accessed to the cloud platform 3(IT layer) through an NB-IOT network, and interconnection and intercommunication between a wireless personal area network and a wide area network are achieved.

The 6 LoWPAN-based community fire monitoring system provided by the embodiment of the invention realizes the communication between the community wireless fire monitoring network 1 and the cloud platform 3 in a low-cost and high-efficiency mode, namely, the application layer uniformly adopts a single MQTT protocol to establish the communication between the 6LoWPAN wireless sensing network 1 and the cloud platform 3; aiming at 6LoWPAN and NB-IOT heterogeneous networks, an MQTT over XNET (XNET refers to 6LoWPAN and NB-IOT) heterogeneous network communication method is provided, and MQTT one-network-to-one-end is realized.

The cloud gateway provided by The embodiment of The invention solves The problem that The traditional fire monitoring network is isolated and can not be on The cloud, and realizes a cloud + end mode of IOT (The Internet of Things). The OT layer, namely the 6LoWPAN community fire wireless sensing network, is communicated downwards, and the personal area network interconnection of community fire monitoring data can be realized; the IT layer is communicated upwards, and the interconnection with the cloud platform is realized through an NB-IOT (Narrow Band Internet of Things, NB-IoT) wide area network.

The cloud gateway has edge computing capability, primary judgment of fire risks is achieved through the edge computing capability, and real-time early warning and control of the fire risks can be achieved in cooperation with the cloud end. The cloud platform 3 is stored with a community fire fuzzy neural network early warning model in advance, and utilizes the community fire fuzzy neural network early warning model and the edge calculation of the cloud gateway 2 to realize real-time early warning and control of fire risks in a synergetic mode.

The cloud platform 3 establishes a fuzzy neural network early warning model of the community fire through historical data of the community fire monitoring, trains the neural network by using an error back propagation method, modifies the weight of the neural network, and obtains an accurate fuzzy control rule suitable for the field environment. And after the modified membership function and the fuzzy rule are extracted from the neural network, the modified membership function and the fuzzy rule are used as a fuzzy inference system in a field environment. When the cloud fuzzy inference system receives real-time data input from the community fire wireless sensor network, the system can calculate according to a pre-trained community fire fuzzy neural network early warning model, and therefore the probability of fire occurrence in the corresponding community scene at the current moment can be obtained.

The specific design steps of the fuzzy neural network early warning model for the community fire are as follows:

(1) and establishing a membership function library and fuzzy rules according to different scenes (community types, geographical positions, floor heights, building types and sensor types) in the community.

(2) Fuzzy rules and membership functions are represented by neural networks.

(3) And (3) modifying parameters of the membership function by training a neural network through the historical data of the community fire monitoring to obtain an accurate fuzzy rule.

(4) And extracting the modified membership functions and the fuzzy rules from the neural network and storing for later use.

(5) Real-time data monitored by the community fire wireless sensor network are input into a community fire fuzzy inference system constructed by a fuzzy neural network, and the probability of a comprehensive fire is obtained through signals of all paths through a fusion center.

The cloud platform 3 is connected with the cloud gateway 2 through an NB-IOT network, receives real-time data issued by the cloud gateway 2 and edge calculation result data of the threshold early warning module 223 of the cloud gateway 2, synthesizes the real-time data and the edge calculation result data to carry out comprehensive early warning on fire risks, and if a fire alarm is found, the cloud platform 3 issues a corresponding control instruction to carry out control response on the fire alarm. Cloud platform 3 can be with conflagration risk recognition result propelling movement to cloud platform 3's high in the clouds control terminal, and high in the clouds control terminal can issue control command to cloud gateway 2 according to conflagration recognition result, and cloud gateway 2 sends corresponding instruction to wireless execution unit (like fire prevention rolling slats door, spray set, ventilation unit) and carries out the long-range processing of conflagration police situation. (for example, if fire information is continuously sent out by fire sense in a certain area of a building, the fire state of the sensor in the same area also sends out the fire information, and the data of the temperature and smoke information is abnormal, the fire alarm condition in the area corresponding to the building of the community is judged, the cloud platform 3 carries out rapid comprehensive reasoning according to the real-time data and the characteristics of the fire occurrence place, and sends out an instruction to a corresponding execution node (such as a fire-proof rolling door, a spraying device and a ventilating device) according to the reasoning result to control the fire so as to realize the comprehensive early warning of the community fire).

In the community fire monitoring system with the end-edge-cloud architecture provided by the embodiment of the invention, the end refers to the 6LoWPAN wireless sensing network, the edge refers to the cloud gateway with edge computing capability, and the cloud refers to the cloud platform. Wherein:

end: the fire risk sensing function is realized, namely the 6lowpan wireless sensing network nodes of the community realize the all-round sensing of the fire risk of the community through the 6lowpan wireless sensing network nodes;

edge, refers to a cloud gateway with edge computing capability, which has the following features:

(1) the OT layer is composed of 6LoWPAN wireless sensing nodes, the OT layer data is accessed to the cloud platform through the NB-IOT network, data interaction with the IT layer (cloud platform) is achieved, and interconnection and intercommunication between a wireless personal area network and a wide area network are achieved; aiming at 6LoWPAN and NB-IOT heterogeneous networks, an MQTT over XNET (XNET refers to 6LoWPAN and NB-IOT) heterogeneous network communication method is provided, and MQTT one-network-to-one-end is realized.

(2) The system has an edge calculation function, and can realize real-time early warning and control of fire risks in cooperation with a cloud end;

cloud: the fire disaster early warning system has the functions of equipment fire disaster data storage, big data calculation and early warning. The cloud gateway is designed with a community fire fuzzy neural network early warning model, and real-time early warning and control of fire risks are achieved through cooperation with edge computing of the cloud gateway.

The terminal-edge-cloud-architecture community fire monitoring system provided by the invention realizes sensing, transmission, analysis, early warning and control of the community fire risk in an all-around manner.

According to the embodiment of the invention, the 6LoWPAN wireless sensing network is arranged for fire monitoring, so that the low-cost arrangement of the fire monitoring network is realized, the signal quality is improved, and the reliability is enhanced; the 6LoWPAN wireless sensing network and the cloud platform are connected through the cloud gateway, so that the cloud of the community fire monitoring data is realized; data transmission is carried out by adopting a single MQTT protocol, so that the data transmission efficiency is improved, and the timeliness of fire early warning is improved; the sensing, transmission, analysis, early warning and control of the fire risk of the community are realized in an all-round manner.

Further, based on the above embodiment, the wireless sensor node in the 6LoWPAN wireless sensor network 1 includes a wireless sensing device and a wireless execution device, where:

in data uplink, the wireless sensing equipment is used for acquiring the community fire monitoring data, converting the community fire monitoring data into an MQTT protocol data format, then converging the community fire monitoring data to a cloud gateway 2 through the 6LoWPAN wireless sensing network 1, and transmitting the community fire monitoring data in the MQTT protocol data format to a cloud platform 3 through an NB-IOT wireless wide area network by the cloud gateway 2;

in data uplink, the wireless sensing equipment is used for collecting community fire monitoring data, after the community fire monitoring data are converted into an MQTT protocol data format, the community fire monitoring data are gathered to the cloud gateway 2 through the 6LoWPAN wireless sensing network 1, the cloud gateway 2 does not need to carry out protocol conversion, and the community fire monitoring data in the MQTT protocol data format are transmitted to the cloud platform 3 through an NB-IOT wireless wide area network. In an application layer, data transmission is carried out by using a single MQTT protocol data format, and the data transmission (to the cloud platform 3) of the 6LoWPAN wireless sensing network 1 through the single MQTT protocol is realized.

In the data downlink, if the cloud platform 3 judges that the fire risk exists, the control instruction in the MQTT protocol data format is sent to the cloud gateway 2 through the NB-IOT wireless wide area network, the control instruction includes address information of a target wireless execution device, and the cloud gateway 2 sends the control instruction in the MQTT protocol data format to the target wireless execution device according to the address information, so that the target wireless execution device executes a corresponding action.

The cloud platform 3 judges whether a fire risk exists or not, when the fire risk is judged and known, a control instruction in an MQTT protocol data format is sent to the cloud gateway 2 through an NB-IOT wireless wide area network, the control instruction comprises address information of target wireless execution equipment, and the cloud gateway 2 sends the control instruction in the MQTT protocol data format to the target wireless execution equipment according to the address information, so that the target wireless execution equipment executes corresponding actions, and fire prevention and control are performed. In an application layer, data transmission is carried out by using a single MQTT protocol data format, and the control instruction of the cloud platform 3 is transmitted to the wireless execution equipment of the 6LoWPAN wireless sensor network 1 through the single MQTT protocol.

According to the embodiment of the invention, the wireless sensing equipment senses the environmental information, the sensed community fire monitoring data is converted into MQTT protocol data, the data is converged to the cloud gateway 2 through the 6LoWPAN wireless sensing network, and the cloud gateway 2 transmits the MQTT data of the 6LoWPAN wireless sensing network 1 to the cloud platform 3 through the NB-IOT wireless wide area network. When an abnormal state occurs, the cloud platform 3 can directly send a control instruction to an execution node with a given IPV6 address through the cloud gateway 2, and perform cloud emergency handling. The implementation method completes the interconnection and intercommunication of the 6LoWPAN wireless personal area sensor network (6LoWPAN wireless sensor network) and the cloud platform 3 through a single MQTT protocol. In the system, the cloud gateway 2 directly communicates the channel between the 6LoWPAN wireless sensing node and the cloud platform 3 through a single MQTT protocol, so that the cloud platform 3 directly supervises the wireless sensing equipment of the IPV6 personal area network, the cloud and cloud instruction issuing efficiency of sensing data is improved, and the timeliness of fire early warning is improved.

The 6 LoWPAN-based community fire monitoring system provided by the embodiment of the invention realizes communication between a community wireless fire monitoring network and a cloud platform in a low-cost and high-efficiency mode, namely, an application layer uniformly adopts an MQTT protocol to establish communication between a 6LoWPAN personal area network and the cloud platform, and fusion of IT (meaning cloud platform) and OT (meaning 6LoWPAN wireless sensing network) is realized through a single MQTT protocol.

On the basis of the embodiment, after the community fire monitoring data are converted into an MQTT protocol data format through the wireless sensing equipment in the data uplink, the community fire monitoring data are converged to a cloud gateway through the 6LoWPAN wireless sensing network, and then the cloud gateway transmits the community fire monitoring data to a cloud platform through an NB-IOT wireless wide area network; in data downlink, the cloud platform sends the control instruction of the MQTT protocol data format to the target wireless execution equipment through the cloud gateway, and transmission of single MQTT protocol data in the data cloud-up and cloud-down processes is achieved.

Fig. 2 is a schematic structural diagram of a 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 2, the wireless sensor node transmits monitoring data or control instructions of data in MQTT protocol format between the 6LoWPAN wireless sensor network and the cloud gateway, and allocates addresses by using IPV 6; the cloud gateway transmits monitoring data or control instructions of MQTT protocol format data between the NB-IOT wireless wide area network and the cloud platform, and performs address allocation through IPv6 or IP 4. The cloud gateway transmits data through the uniform MQTT protocol without data protocol conversion, and MQTT over XNET (XNET refers to 6LoWPAN and NB-IOT) is realized. The cloud platform and the 6LoWPAN network are directly communicated through the MQTT protocol, namely the MQTT is communicated to the end.

Fig. 3 is a schematic structural diagram of a 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 3, the cloud gateway 2 includes a radio frequency unit 21, a microprocessor unit 22, and a cloud transmission unit 23; the microprocessor unit 22 comprises an MQTT proxy server 221, an MQTT client 222 and a threshold early warning module 223; wherein: in the data uplink, the radio frequency unit 21 is configured to receive the community fire monitoring data converted into an MQTT protocol data format, and converge the community fire monitoring data to the MQTT proxy server 221; the MQTT proxy server is configured to send the community fire monitoring data to the MQTT client 222; the MQTT client 222 is configured to send the community fire monitoring data to the threshold early warning module 223 and to the cloud platform 3 through the cloud transmission unit 23; the threshold early warning module 223 is connected to the MQTT client 222, and is configured to receive the community fire monitoring data acquired by the MQTT client 222, perform comprehensive analysis on the community fire monitoring data of each wireless sensing device, and determine whether an abnormality occurs; if the situation that the abnormal situation occurs is judged and known, the abnormal information is sent to the MQTT client 222; the MQTT client 222 sends the abnormal information to the cloud platform 3 through the cloud transmission unit 23; in the data downlink, the cloud platform 3 sends the control instruction in the MQTT protocol data format to the cloud transmission unit 23 through an NB-IOT wireless wide area network, and the cloud transmission unit 23 sends the control instruction to the target wireless execution device sequentially through the MQTT client 222, the MQTT proxy server 221, and the radio frequency unit 21.

After the system is started, the radio frequency unit 21 of the cloud gateway 2 automatically acquires an IPv6 address automatically generated by a network prefix, and a tree network using a border router as a root node is constructed through an IPv 6-based Low-power-consumption lossy wireless local area network (IPv6R outgoing Protocol for Low power and loss networks, RPL) routing Protocol. The border router is composed of a radio frequency unit 21 and a microprocessor unit 22. The radio frequency unit 21 communicates with wireless sensing nodes in the 6LoWPAN wireless sensing network through a 6LoWPAN wireless communication protocol. The radio frequency unit 21 has a routing function of a wireless sensing network, and realizes community fire wireless sensing node access based on a 6LoWPAN personal area network. The cloud transmission unit 23 establishes a channel between the cloud gateway 2 and the cloud platform 3 through the NB-IOT wireless wide area network.

Firstly, the cloud gateway realizes the fusion of the 6LoWPAN and the NB-IOT heterogeneous network, namely the gateway establishes the communication between the 6LoWPAN personal area network and the NB-IOT wide area network and establishes the communication with the cloud platform through the NB-IOT network. The cloud gateway realizes IT and OT fusion by uniformly adopting an MQTT protocol at an application layer, namely, communication between the wireless fire sensing node and a cloud platform is realized without protocol conversion in a heterogeneous network. An MQTT proxy server (MQTT Broker) and an MQTT client on the cloud gateway 2 subscribe the information published by the wireless sensing nodes through a 6LoWPAN wireless sensing network, wherein the information comprises the community fire monitoring data and is published to the cloud platform 3 by the identity of the information publisher, so that the cloud transmission of the fire sensing data is realized, and the community fire monitoring and monitoring are realized; in addition, the MQTT proxy server and the MQTT client on the cloud gateway 2 can also subscribe the control instruction issued by the cloud (cloud platform 3) by the identity of the subscriber, and then issue the control instruction to the corresponding wireless sensing node by the identity of the publisher, so that the fire alarm execution node under the 6LoWPAN network of the community can be controlled.

The microprocessor unit 22 of the embodiment of the present invention performs data communication with the cloud transmission module 23 through a soft serial communication interface (software-implemented serial port), and the cloud transmission module 23 is connected with an NB-IOT wireless communication module and is connected with the cloud platform 3 for data interaction.

Specifically, in the data uplink, the radio frequency unit 21 is configured to receive the community fire monitoring data converted into an MQTT protocol data format, and converge the community fire monitoring data to the MQTT proxy server 221; the MQTT proxy server 221 is configured to send the community fire monitoring data to the MQTT client 222; the MQTT client 222 sends the community fire monitoring data to a cloud transmission unit 23, and the cloud transmission unit 23 sends the community fire monitoring data to the cloud platform 3 through an NB-IOT wireless wide area network; the MQTT client 222 further sends the community fire monitoring data to the threshold early warning module 223, and the threshold early warning module 23 has edge computing capability, performs threshold analysis and calculation on the real-time monitoring data, and sends abnormal information to a cloud platform for comprehensive early warning when an abnormal value is found. The threshold early warning module 223 is connected to the MQTT client 222, and is configured to receive the community fire monitoring data acquired by the MQTT client 222, perform comprehensive analysis on the community fire monitoring data of each wireless sensing device, and determine whether an abnormality occurs; specifically, the community fire monitoring data may be compared with a preset threshold value, so as to determine whether an abnormality occurs. If the situation that the abnormal situation occurs is judged and known, the abnormal information is sent to the MQTT client 222; the MQTT client 222 sends the abnormal information to the cloud platform 3 through the cloud transmission unit 23. The cloud platform 3 can perform comprehensive early warning on the processing result of the real-time data by combining the received abnormal information and the processing result of the real-time data.

In order to solve the problems of low response speed, high false alarm rate and more false alarms of the traditional community fire monitoring system, the cloud gateway 2 supports the edge computing function, and the cloud gateway 2 and the cloud platform 3 perform collaborative computing to obtain fire alarm information. When the fire real-time monitoring data is clouded, the threshold early warning processing unit 223 of the cloud gateway 2 performs threshold analysis calculation on the real-time monitoring data, and sends abnormal information to the cloud platform 3 for cooperative calculation when an abnormal value is found. The edge calculation function has the advantages that data of a large number of fire monitoring nodes are analyzed and calculated in real time and are transferred to the edge side, abnormal warning information can be found in advance, comprehensive calculation of the cloud is reduced, the fire emergency processing speed is improved, and false alarms are reduced.

In the data downlink, the cloud platform 3 sends the control instruction in the MQTT protocol data format to the cloud transmission unit 23 through an NB-IOT wireless wide area network, and the cloud transmission unit 23 sends the control instruction to the target wireless execution device sequentially through the MQTT client 222, the MQTT proxy server 221, and the radio frequency unit 21.

On the basis of the above embodiment, the embodiment of the invention ensures timeliness and reliability of data transmission and early warning by arranging the radio frequency unit, the cloud transmission unit and the microprocessor unit including the MQTT proxy server, the MQTT client and the threshold early warning module in the cloud gateway.

Further, based on the above embodiment, the cloud platform 3 is further configured to compare the community fire monitoring data with a preset threshold value when it is determined that there is a fire risk, so as to determine a target wireless sensing device of which the community fire monitoring data exceeds the preset threshold value, and obtain the address information of the target wireless execution device in the same area according to the number information of the target wireless sensing device.

When judging that the fire risk exists, the cloud platform 3 compares the community fire monitoring data of each wireless sensing device with a preset threshold value, so as to determine a target wireless sensing device of which the community fire monitoring data exceeds the preset threshold value, and can determine an area where a fire alarm occurs according to the number information of the target wireless sensing device. And obtaining the address information of the target wireless execution equipment in the same region according to the region where the fire alarm occurs, and further sending a control instruction to the target wireless execution equipment through a cloud gateway according to the address information.

On the basis of the above embodiment, the embodiment of the invention obtains the address information of the target wireless execution device in the same area according to the number information of the target wireless sensing device by determining the target wireless sensing device with the community fire monitoring data exceeding the preset threshold value, thereby improving the accuracy of sending the control instruction.

Further, based on the above embodiment, the radio frequency unit 21 is further configured to receive execution result data of the target wireless execution device converted into an MQTT protocol data format, and send the execution result data to the cloud platform 3 through the cloud gateway 2.

The MQTT proxy server and the MQTT client on the cloud gateway 2 subscribe the information published by the wireless sensing nodes through the 6LoWPAN wireless sensing network 1, the information also comprises execution result data of target wireless execution equipment, and the information is published to the cloud platform 3 by the identity of an information publisher, so that cloud transmission of the execution result data is realized, the execution condition of the terminal is known by the cloud, and the reliability of fire early warning is further improved.

On the basis of the embodiment, the embodiment of the invention further improves the reliability of fire early warning by sending the execution result data of the target wireless execution device to the cloud platform through the cloud gateway.

Further, based on the above embodiment, the abnormal information is false alarm information, fault information or fire alarm information, wherein: when the threshold early warning module 223 is used for performing comprehensive analysis on the community fire monitoring data of each wireless sensing device, if the data of a single wireless sensing device is abnormal occasionally, it is known that the abnormal information is false alarm information; if the data of a single wireless sensing device is continuously abnormal and the data of the rest wireless sensing devices are normal, acquiring that the abnormal information is fault information; and if the data of the plurality of wireless sensing devices in the same area are abnormal, acquiring that the abnormal information is fire alarm information.

The threshold early warning module 223 may not only obtain whether an abnormality occurs according to a comprehensive analysis of the community fire monitoring data, but also classify the abnormal information and inform the cloud platform 3 of the classification result, so as to perform a comprehensive process using the cloud platform 3. The abnormal information is false alarm information, fault information or fire alarm information. If it is monitored that the data of a single sensing node in the community building region is abnormal accidentally, the false alarm is judged, and the false alarm information is sent to the cloud platform 3, wherein the false alarm information comprises: abnormal equipment number, abnormal equipment type, abnormal occurrence time and abnormal value. If it is monitored that data of a single sensing node in a community building region is abnormal continuously, but data information of other fire monitoring nodes in the same region is normal, the wireless sensing node is judged to be in fault, equipment fault information is sent to the cloud platform 3, and the fault information comprises: fault equipment number, fault equipment type, fault occurrence time and abnormal value. (for example, if the fire sense in a certain area of the building continuously gives fire information, but the other fire sense in the same area as the sensor is normal and the temperature and smoke sense information is in the normal range, the fire sense is judged to have equipment failure). If data abnormality of a plurality of sensing nodes in the same area of the community building is monitored, the occurrence of fire alarm is judged, and fire alarm information is transmitted to the cloud platform 3. And the cloud end carries out comprehensive early warning according to the real-time data and the fire alarm information.

On the basis of the above embodiment, the embodiment of the invention further improves the reliability of fire early warning by distinguishing the abnormal information into the false alarm information, the fault information and the fire alarm information.

Further, based on the above embodiment, the threshold early warning module 223 is further configured to send sampling period adjustment information to the wireless sensing device with data abnormality through the MQTT client 222, the MQTT proxy server 221, and the radio frequency unit 21, where the wireless sensing device increases the sampling frequency according to the sampling period adjustment information.

The MQTT proxy server 221 and the MQTT client 222 of the cloud gateway 2 receive the wireless sensing node data, transmit the wireless sensing node data to the cloud platform 3 through the cloud transmission unit 23, and simultaneously transmit the data to the threshold early warning module 223, and the threshold early warning module 223 determines the wireless sensing node safety threshold according to the relevant standards. When the data of the wireless sensing node received by the threshold early warning module 223 exceeds the safety threshold, the abnormal information is sent to the cloud platform 3 and a control instruction is automatically sent to the abnormal wireless sensing device, so that the sampling period of the corresponding wireless sensing device is shortened to 1/2 or other set proportions of the original sampling period.

On the basis of the embodiment, the embodiment of the invention further improves the timeliness of fire early warning by increasing the acquisition frequency of the wireless sensing equipment with abnormal data.

Further, based on the above embodiment, the threshold early warning module 223 is further configured to, when the abnormal information is the fire alarm information, obtain a second probability of fire occurrence according to a ratio of abnormal data in an area where the fire alarm information is generated, and send information of the second probability to the cloud platform 3; when the cloud platform 3 is used for realizing real-time early warning and control of fire risks by utilizing the community fire fuzzy neural network early warning model and the edge calculation of the cloud gateway in a coordinated manner, the cloud platform is specifically used for: inputting the community fire monitoring data into a pre-established community fire fuzzy neural network early warning model, and obtaining a first probability of fire according to an output result of the community fire fuzzy neural network early warning model; carrying out weighted summation on the first probability and the second probability through a preset weight value to obtain a third probability of fire occurrence, and judging whether the fire risk exists according to the third probability; and if the fire risk is judged and known, sending a control instruction to the target wireless execution equipment through the cloud gateway 2.

And the cloud platform 3 inputs the community fire monitoring data into a pre-established community fire fuzzy neural network early warning model, and obtains a first probability of fire according to an output result of the community fire fuzzy neural network early warning model.

The threshold early warning module 223 is further configured to, when the abnormal information is the fire alarm information, obtain a second probability of occurrence of a fire according to a ratio of abnormal data in an area where the fire alarm information is generated, and send information of the second probability to the cloud platform 3.

After receiving community fire monitoring data (real-time data), the cloud platform 3 inputs the community fire monitoring data into a community fire fuzzy neural network early warning model established by using community wireless sensor network monitoring historical data, and obtains a first probability of fire occurrence according to an output result of the community fire fuzzy neural network early warning model; and the cloud platform 3 receives the second probability for fire risk judgment sent by the threshold early warning module 223. And the cloud platform 3 performs weighted summation on the first probability and the second probability through a preset weight value to obtain a third probability of fire occurrence, and judges whether the fire risk exists according to the third probability. When weighted summation is performed, the weight of the first probability and the weight of the second probability may be the same or different.

And if the third probability is greater than a preset probability threshold value, judging that the fire risk exists. The cloud platform 3 further acquires the address information of the target wireless execution device in the fire risk area, and the cloud platform 3 may acquire the address information by acquiring the number information of the wireless sensing device corresponding to abnormal data in the real-time data, and may acquire more comprehensive address information of the target wireless execution device by combining with the information of the wireless sensing device in the abnormal information, and further send a control instruction to the target wireless execution device based on the address information.

On the basis of the embodiment, whether the fire risk exists is judged by integrating the processing result of the threshold early warning module and the processing result of the cloud to the real-time data, and the reliability of fire risk early warning is further improved.

The community fire wireless sensing network and the cloud gateway provided by the embodiment of the invention realize uplink and downlink transmission of data, realize cloud collaborative early warning of a community fire scene by using the cloud gateway threshold processing unit and the cloud fuzzy inference system on the basis, and issue a control instruction for cloud emergency disposal according to an early warning result and a community fire actual scene.

Fig. 4 is a flowchart illustrating an operation of a 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 4, the 6LoWPAN wireless sensor network collects community data to the cloud gateway, the cloud gateway receives the data and sends real-time data to the threshold processing unit for processing, and the threshold processing unit judges whether the threshold is exceeded; if the data of a single sensor in the same area accidentally exceeds a threshold value and other sensors are normal, the sensor is considered to be a false alarm, and the false alarm information is transmitted to the cloud platform; if the data of a single sensor in the same area continuously exceed the threshold value and other sensors are normal, the sensor is considered to be in fault, and fault information is transmitted to the cloud platform; if the data of the sensors in the same area continuously exceed the threshold value, the fire hazard is considered to occur, the data are uploaded to a cloud platform fuzzy inference system, the cloud platform synthesizes early warning information and real-time data to judge whether the fire hazard occurs or not, and the fire hazard is processed according to the actual fire hazard.

Further, based on the above embodiment, the wireless sensing device includes a temperature and humidity sensor, a smoke sensor, a flame sensor, a door magnetic switch sensor, a combustible gas sensor, and a fire alarm; the wireless execution equipment comprises a spraying device (a spraying execution valve), a ventilation valve and a rolling door.

Fig. 5 is a schematic structural diagram of a 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 5, each wireless sensing device sends sensed community fire monitoring data to the cloud gateway, the cloud gateway forwards real-time data to the cloud platform, the cloud gateway can perform edge computing to confirm abnormal information, and if abnormal information exists, the abnormal information is sent to the cloud platform. And the cloud platform synthesizes the edge calculation result and the data processing result thereof to carry out comprehensive study and judgment and intelligent decision, confirms whether fire risk exists or not, and sends a control instruction to corresponding wireless execution equipment if the fire risk exists, so as to realize remote control.

Fig. 6 is a schematic structural diagram of a wireless sensing node in a 6LoWPAN wireless sensing network of the 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 6, the wireless sensing node (fire monitoring sub-node) includes a microprocessor, a radio frequency module, a wireless sensing device and a wireless execution device. The communication interfaces of the microprocessor integration standard comprise serial communication, 485 communication interfaces, analog quantity communication interfaces, 0/1 digital quantity communication interfaces and the like. The microprocessor is connected with the radio frequency module through a serial bus interface (uart), and the wireless sensing network based on the 6LoWPAN is connected with the boundary router of the cloud gateway 2. After the connection is successful, the border router allocates an IPV6 address to the fire monitoring sub-node to realize the ad hoc network of the wireless sensor network. The 485 communication interface, the analog quantity communication interface and the 0/1 digital quantity communication interface are in butt joint with equipment and meters related to fire fighting and fire fighting, and comprise smoke sensing equipment, fire sensing equipment, gas monitoring equipment, fire alarm equipment and the like. The data storage unit is connected with the microprocessor.

On the basis of the above embodiments, the embodiments of the present invention further improve the reliability of risk early warning by providing a plurality of wireless sensing devices and wireless execution devices.

Fig. 7 is a schematic structural diagram of a 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. The cloud gateway with 6LoWPAN wireless sensing network is a plurality of, and one-to-one. That is, the cloud platform may receive monitoring information of different wireless sensor networks sent by cloud gateways of multiple areas, and perform corresponding control.

On the basis of the embodiment, the embodiment of the invention realizes the extension of the wireless personal area communication network to the Internet of things, can meet the 6LoWPAN wireless fire-fighting sensing network between different buildings and different floors of a community, and realizes the exchange, communication and control of fire-fighting alarm information of different subnets by connecting the cloud gateway and the cloud platform together.

To further illustrate the functions of the 6 LoWPAN-based community fire monitoring system provided in the embodiment of the present invention, the workflow of the 6 LoWPAN-based community fire monitoring system provided in the embodiment of the present invention is further described below by using a specific embodiment.

The 6 LoWPAN-based community fire monitoring system provided by the embodiment of the invention comprises a 6LoWPAN wireless sensing network for community fire sensing, a cloud gateway and a cloud platform. The wireless sensing network comprises a plurality of wireless sensing nodes (including wireless sensing equipment and line execution equipment) deployed in a community building, the wireless sensing nodes gather data to cloud gateway nodes in a multi-hop ad hoc network mode through the 6LoWPAN wireless sensing network, and the cloud gateway nodes transmit fire sensing data to a cloud platform through the NB-IOT network.

Fig. 8 is a schematic diagram of a data uplink workflow of the 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 8, the data uplink process is as follows:

after the system is started, the radio frequency unit of the cloud gateway automatically acquires an IPv6 address automatically generated by a network prefix, and a tree network with a border router as a root node is constructed through an IPv 6-based Low-power-consumption lossy wireless local area network router (IPv6R outgoing Protocol for Low power and loss networks, RPL). The method comprises the steps that a radio frequency unit of the cloud gateway establishes a network channel with a wireless sensing node through a 6LoWPAN wireless personal area network, data transmission is carried out on an application layer through an MQTT protocol, and data messages of the wireless sensing node are issued to an MQTT browser of the cloud gateway to achieve data aggregation. On one hand, an MQTT client of the cloud gateway transmits data converged by the cloud gateway to a cloud platform through a cloud transmission unit; on the other hand, data gathered by the cloud gateway are transmitted to the threshold early warning module, the threshold early warning module conducts pre-analysis on the data gathered by the wireless sensing nodes, if the analysis result of the corresponding data exceeds the normal threshold range, corresponding early warning information is generated and automatically sent to the cloud platform through the cloud gateway MQTT client, the cloud platform combines real-time monitoring data issued by the cloud gateway and conducts comprehensive early warning through the fuzzy neural network, and risk early warning information of the community fire monitoring system is obtained.

Fig. 9 is a schematic diagram of a data downlink work flow of the 6 LoWPAN-based community fire monitoring system according to an embodiment of the present invention. As shown in fig. 9, the data downlink process is as follows:

when a control instruction needs to be sent to the wireless execution node, the cloud platform issues the control instruction to the cloud gateway MQTT Broker according to a comprehensive fire early warning result of the community or other control requirements, the cloud gateway sends a corresponding instruction to the corresponding wireless sensing actuator device under the 6LoWPAN network through the MQTT client, and then the corresponding actuator device executes the related instruction. After the instruction execution is finished, the wireless sensing execution node feeds the execution result back to the cloud platform: the execution result is firstly issued to the MQTT Broker of the cloud gateway, and then the MQTT client of the cloud gateway feeds the execution result back to the cloud platform through the cloud transmission unit.

The community fire monitoring system based on the 6LoWPAN provided by the embodiment of the invention has the following advantages:

(1) the problem that the reliability is difficult to guarantee when wireless community fire fighting equipment networking of wireless wide area networks such as GPRS and 4G are directly deployed in a building with a complex structure is solved.

(2) For massive fire-fighting wireless sensing equipment, if the wireless telecommunication network is directly adopted for transmission, the cost is very high, and the existing telecommunication network is difficult to ensure that the massive wireless sensing equipment in a complex community building can be simultaneously communicated on line. The community fire monitoring system based on the 6LoWPAN provided by the embodiment of the invention utilizes the 6LoWPAN wireless personal area network to realize the self-networking of the mass fire monitoring sensing nodes, not only is the networking convenient, but also the problem of poor signal of the wide area network in a building is solved.

(3) Some researches on the 6LoWPAN wireless sensor network only stay at the wireless personal area network level, and the problem of wide area network interconnection is not considered. They are generally applied in local areas, and the networks do not generally communicate with each other, which is a certain distance away from the goal of really realizing the interconnection of everything in the community fire-fighting system. For example, since the nodes have to have extremely low power consumption and limited computation and storage resources, the communication bandwidth, transmission distance and coverage area must be reduced. In addition, the technologies of different communication protocols cannot be interconnected. In order to solve the problems, the cloud gateway in the 6 LoWPAN-based community fire monitoring system provided by the embodiment of the invention has the following advantages: the wireless sensor network device is widely accessed to the 6LoWPAN of the community fire fighting system, and one network is achieved through the MQTT protocol.

(4) The 6 LoWPAN-based community fire monitoring system utilizes 6LoWPAN and MQTT technologies, and achieves the cloud reaching of environmental parameter information of each sensor node in the 6LoWPAN wireless sensing network and the remote control of wireless node equipment in the community fire detection system. The interconnection and intercommunication between the wireless personal area network and the wireless wide area network are realized, and the cloud platform and the 6LoWPAN network are directly communicated through the MQTT protocol, namely the MQTT network is completed. The edge gateway realizes the function of community fire prediction and early warning, firstly, the gateway realizes the function of threshold early warning, and when the numerical value of the wireless sensing node exceeds the threshold range, the gateway equipment sends warning information to the cloud platform for reminding. The cloud platform can directly utilize the early warning information sent by the gateway to carry out comprehensive early warning of fire risks, and the scheme not only avoids the complexity of developing an alarm algorithm on the cloud platform or a server, but also improves the real-time performance of the early warning information.

The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A community fire monitoring system based on an end-edge-cloud architecture is characterized by comprising a 6LoWPAN wireless sensing network on an end side, a cloud gateway with edge computing capability and a cloud platform of a cloud side; wherein:

the wireless sensing nodes in the 6LoWPAN wireless sensing network have a fire risk sensing function, so that the fire risk of the community can be sensed in an all-round manner;

the cloud gateway is connected with the 6LoWPAN wireless sensing network and is accessed to the cloud platform through an NB-IOT network, so that interconnection and intercommunication between a wireless personal area network and a wide area network are realized; for 6LoWPAN and NB-IOT heterogeneous networks, the cloud gateway transmits data by using an MQTT protocol, and an application layer establishes communication between the 6LoWPAN wireless sensing network and a cloud platform by uniformly adopting a single MQTT protocol so as to realize one-network-to-one-network MQTT; moreover, the cloud gateway realizes the preliminary judgment of the fire risk through the edge computing capability of the cloud gateway;

the cloud platform has the functions of community fire monitoring data storage, big data calculation and early warning, a community fire fuzzy neural network early warning model is stored in the cloud platform in advance, and real-time early warning and control of fire risks are achieved through cooperation of the community fire fuzzy neural network early warning model and edge calculation of a cloud gateway;

the wireless sensing node in the 6LoWPAN wireless sensing network comprises a wireless sensing device and a wireless execution device, wherein:

in data uplink, the wireless sensing equipment is used for acquiring the community fire monitoring data, converting the community fire monitoring data into an MQTT protocol data format, and then converging the community fire monitoring data to a cloud gateway through the 6LoWPAN wireless sensing network, wherein the cloud gateway transmits the community fire monitoring data in the MQTT protocol data format to a cloud platform through an NB-IOT wireless wide area network;

in data downlink, if the cloud platform judges that the fire risk exists, sending a control instruction in an MQTT protocol data format to the cloud gateway through an NB-IOT wireless wide area network, wherein the control instruction comprises address information of target wireless execution equipment, and the cloud gateway sends the control instruction in the MQTT protocol data format to the target wireless execution equipment according to the address information so as to enable the target wireless execution equipment to execute corresponding actions;

the cloud gateway comprises a radio frequency unit, a microprocessor unit and a cloud transmission unit; the microprocessor unit comprises an MQTT proxy server, an MQTT client and a threshold early warning module; wherein:

in the data uplink, the radio frequency unit is used for receiving the community fire monitoring data converted into an MQTT protocol data format and converging the community fire monitoring data to the MQTT proxy server; the MQTT proxy server is used for sending the community fire monitoring data to the MQTT client; the MQTT client is used for sending the community fire monitoring data to the threshold early warning module and sending the community fire monitoring data to the cloud platform through the cloud transmission unit; the threshold early warning module is connected with the MQTT client and used for receiving the community fire monitoring data acquired by the MQTT client, comprehensively analyzing the community fire monitoring data of each wireless sensing device and judging whether the abnormality occurs or not; if the situation that the abnormality occurs is judged and known, the abnormal information is sent to the MQTT client; the MQTT client sends the abnormal information to the cloud platform through the cloud transmission unit;

in data downlink, the cloud platform sends the control instruction in the MQTT protocol data format to the cloud transmission unit through an NB-IOT wireless wide area network, and the cloud transmission unit sends the control instruction to the target wireless execution device through the MQTT client, the MQTT proxy server and the radio frequency unit in sequence;

the abnormal information is false alarm information, fault information or fire alarm information, wherein:

the threshold early warning module is used for comprehensively analyzing the community fire monitoring data of each wireless sensing device, and if the data of a single wireless sensing device is abnormal occasionally, the abnormal information is known to be false alarm information; if the data of a single wireless sensing device is continuously abnormal and the data of the rest wireless sensing devices are normal, acquiring that the abnormal information is fault information; if the data of the wireless sensing devices in the same area are abnormal, the abnormal information is acquired as fire alarm information;

the threshold early warning module is further used for sending sampling period adjustment information to the wireless sensing equipment with data abnormality through the MQTT client, the MQTT proxy server and the radio frequency unit, and the wireless sensing equipment improves sampling frequency according to the sampling period adjustment information;

the threshold early warning module is further used for obtaining a second probability of fire occurrence according to the proportion of abnormal data in an area generating the fire alarm information when the abnormal information is the fire alarm information, and sending the information of the second probability to the cloud platform;

when the cloud platform is used for realizing real-time early warning and control of fire risks by utilizing the community fire fuzzy neural network early warning model and the edge calculation of the cloud gateway in a coordinated manner, the cloud platform is specifically used for: inputting the community fire monitoring data into a pre-established community fire fuzzy neural network early warning model, and obtaining a first probability of fire according to an output result of the community fire fuzzy neural network early warning model; carrying out weighted summation on the first probability and the second probability through a preset weight value to obtain a third probability of fire occurrence, and judging whether the fire risk exists according to the third probability; and if the fire risk is judged and known, sending a control instruction to the target wireless execution equipment through the cloud gateway.

2. The end-edge-cloud architecture-based community fire monitoring system according to claim 1, wherein the cloud platform is further configured to compare the community fire monitoring data with a preset threshold value when it is determined that there is a fire risk, so as to determine a target wireless sensing device of which the community fire monitoring data exceeds the preset threshold value, and obtain the address information of the target wireless execution device in the same area according to the number information of the target wireless sensing device.

3. The end-edge-cloud architecture-based community fire monitoring system according to claim 1, wherein the radio frequency unit is further configured to receive execution result data of the target wireless execution device converted into MQTT protocol data format, and send the execution result data to the cloud platform through the cloud gateway.

4. The end-edge-cloud architecture-based community fire monitoring system according to claim 1, wherein the wireless sensing equipment comprises a temperature and humidity sensor, a smoke sensor, a flame sensor, a door magnetic switch sensor, a combustible gas sensor and a fire alarm; the wireless execution equipment comprises a spraying device, a ventilation valve and a rolling door.

5. The end-edge-cloud architecture-based community fire monitoring system of claim 1, wherein the cloud gateway and the 6LoWPAN wireless sensor network are multiple and in one-to-one correspondence.