Cloud-Edge-End Cooperative Urban Underground Resistivity Sensing System and Data Collection Method

Abstract

This invention is a cloud-based edge-end cooperation-based urban underground space resistivity sensing system and We will disclose the data collection method and develop a cloud-based edge-end architecture based on the advanced cloud-based edge-end architecture design. In this way, data collection operations are submerged in the distributed edge nodes and sensing nodes, and computation tasks are Cloud-intensive data processing and data mining operations are managed on a central cloud computing platform. This was deployed on a data platform, meeting the real-time and efficient requirements for data collection. At the same time, well-ground linkage is adopted to build a three-dimensional spatial randomly distributed sensing network, and both sides of the road Taking full advantage of the advantageous conditions of burying horizontal cables and arranging vertical boreholes in the In order to flexibly arrange the three-dimensional resistivity sensing net across the street, it is necessary to make up for the shortcomings of single surface survey. Achieve precise imaging of targets below city streets.

Description

[Technical field]

[0001] The present invention belongs to the field of electrical method exploration technology, specifically, cloud-edge-end cooperative based This paper describes a resistivity sensing system for urban underground spaces and a data collection method. [Background technology]

[0002] Urban underground construction is the foundation and an important component of urban construction, and in itself, it is also hidden. With the accelerated progress of urbanization, the health status and safety of urban underground space structures are becoming increasingly important. The security of urban areas is directly related to the safety of life and property of urban residents, but how to deal with it quickly, effectively and non-destructively is also important. The task of assessing and evaluating the health of underground structures remains the responsibility of urban management. This is a difficult task ahead of us. Currently, government departments are responsible for oversight during the design, construction, and other stages. They often strengthen supervision and ensure that construction quality of underground works is up to standards. However, the service life and safety of underground construction are closely related to the structural design and construction quality, and It is also closely related to the environmental changes surrounding underground construction during the period. The influence of underground structure and the surrounding environment is mutual. The deformation of the strata and stress changes around the underground structure affect the underground structure and thus This may cause structural destruction, but the destruction of the underground structure may also cause the flow changes of the surrounding soil media. This will accelerate the destruction of the underground structure by promoting the movement of groundwater and the erosion of the soil. From the beginning, the underground space structure, the surrounding geological environment, the underground pipeline network structure, the human and traffic environment were all integrated into one It is necessary to consider the whole picture and study it systematically. Using the drilling surveys, material surveys and other data accumulated in the project, we will build a "transparent city." This is a crucial step and an important foundation for the construction of "smart cities." On the other hand, using emerging technologies to build a four-dimensional dynamic urban sensing network is It is a necessary route to building a "smart city" and is a key component in modernizing cities and creating a livable environment. This has important significance for the

[0003] Changes in the underground structure and the surrounding geological environment will affect the physical parameters of the underground medium ( These changes in the electrical properties of the material can be seen in the following images: Dynamic monitoring of changes in physical parameters of the city will bring dynamic "health" to the city's "body." This is equivalent to installing a "diagnosis" sensor, and can monitor dynamic changes in the underground pipe network and underground space structure of a city. Real-time visualization, detection, and monitoring, and early warning of abnormalities when a set critical value is reached Timely triggering of warning information and distributed multi-sensor monitoring network Rapid identification of the location of abnormalities, timely action, and protection of life and property safety. However, current urban underground sensing systems mainly measure temperature, groundwater level, stress, displacement, etc. In-situ measurements are performed using contact sensors, while sensors with through-beam imaging capabilities are used There is a lack of long-term, remote, non-contact smart sensing.

[0004] The high density resistivity method is a multi-channel, array exploration technique developed on the basis of conventional electrical exploration. The high-density electrical method can collect a large amount of detection data by simply laying cables and electrodes at once. This not only saves manpower and material resources, but also improves data collection efficiency, and The imaging results are intuitive and easy to interpret. However, the traditional high-density resistivity method is not suitable for urban underground. For long-term monitoring (smart sensing) of targets (e.g. road cavity collapses) However, there are still some difficulties and obstacles.

[0005] 1. The surrounding environment of urban streets is complex. (1) Under city streets, the electricity, water, sewerage, This is an area where various pipe networks, including those for communications, pass through intensively, and the underground environment is extremely complex. (2 ) The difference in ground conditions on both sides of city streets is large, and electromagnetic interference is serious. (3) Long and narrow city streets It is only possible to lay electrodes along both sides of the path, which is not possible with the three-dimensional high-density resistivity method. The required vertical expansion space is lacking, making it difficult to lay out a regular measurement network on the surface. (4) Wellbore resistivity imaging has higher resolution but is only vertically sensitive. The pitch of measurements across wells is limited. Currently, resistivity images of the surface and wellbore are Although the advantages of each method have not been fully utilized, the well field survey is This is more advantageous for fully utilizing the advantages of piping.

[0006] 2. Existing resistivity sensing system designs are highly dependent on city facilities and are not compatible with the actual city facilities. The existing design is also limited in its location, making it difficult to effectively utilize its flexibility. At the same time, due to the constraint of the small number of electrode channels of the sensing node, the combination of power supply and potential measurement is required. The combination of the two types is very limited, which seriously affects the sensing imaging effect. Since the combination of the fixed electrodes may belong to different sensing nodes, the existing serial The time delay of the line transmission network has a significant effect on the synchronization of power supply and potential measurement, but the response Adding response waits to the extension also has a profound effect on data collection efficiency.

[0007] 3. The resistivity sensing system, which is remotely managed through a central console, has an inherent There are flaws: (1) All sensing nodes are centrally managed and controlled through a single central console. Scheduled, command and data transmission failure or congestion due to network busy This leads to system startup, time delays, bit errors and character string loss, making it difficult to ensure real-time and reliability. (2) Large-scale data is collected in a central console for centralized storage and Centralized processing, high demands on central console software and hardware, common Public computing resources are not fully utilized, there is duplication of construction and resource waste, and the system itself is This will also increase the operating and maintenance costs of the system in the later stages.

[0008] 4. The main features of the smart sensing system are automatic collection of unmanned on-duty and automatic transmission of remote data. Highly automated remote smart with automatic storage, automatic data processing, automatic analysis prediction and alarm. There are still large differences in some resistivity sensing systems, It is biased towards collection only, and lacks the ability to extract and mine data information, resulting in a lack of a sense of intelligence. It is far from reaching the level of intelligence required for AI to perform predictive analysis.

[0009] Therefore, we have designed a completely new resistivity sensing system that is less dependent on city facilities. General-purpose wireless Internet of Things system, edge cloud (edge ​​storage and edge computing) A platform that can fully utilize common public resources such as cloud computing and central clouds. We will form a platform and rely on cutting-edge technologies such as big data and artificial intelligence to It is necessary to realize a smart sensing system with risk prediction and assessment capabilities. Summary of the Invention [Problem to be solved by the invention]

[0010] The present invention addresses the shortcomings of conventional technologies by providing cloud-edge-end cooperative urban Provide an underspace resistivity sensing system and data collection method, the specific technical proposal is as follows: It is. [Means for solving the problem]

[0011] A cloud-edge-end cooperation-based urban underground space resistivity sensing system, The system adopts a cloud edge-end architecture design and said central cloud computing platform; Multiple edge servers connected to a distributed network, and each edge server has a distributed network. a plurality of resistivity sensing nodes connected to the network; The central cloud computing platform manages the entire resistivity sensing system. Manage, perform global data processing and model inversion, and detect data anomalies that exceed the threshold. and to manage the entire resistivity sensing system. This involves setting up and configuring distributed edge servers, and distributing all resistance sensing through the edge servers. The above-mentioned global data processing and model inversion includes managing the knowledge nodes. Compare real-time data with historical data, mine it, and send the model results to the edge server. and provide guidance on basic data analysis. The edge server may be configured to allow multiple resistivity sensing nodes in a domain to work in a coordinated manner. An edge node for partition-specific control, which controls power supply and power supply within a domain. The selection of the position measurement electrode pair and the collection process are controlled in a coordinated manner, and the data obtained by the data collection are The system selects, organizes, and stores the collected data in a designed format, and Upload and back up data to a central cloud computing platform Once data collection is complete, real-time and historical data can be shared with a central cloud. Cloud computing platform feeds edge nodes based on historical data The results of the area model calculations are compared and analyzed to determine whether there are any abnormalities and to When there is a constant change, abnormal information is sent to the central cloud computing platform. and informing the user of the The resistivity sensing node is an end node, and the resistivity sensing nodes are and / or vertical wellbore, each resistivity sensing node being independently A resistivity sensor unit is provided in a collection station and connected to the collection station. A multi-channel electrode conversion switch, a multi-core high-density electrical cable, and a multi-core high-density electrical and a ground electrode connected to the ground cable, and the resistivity sensing node is In response to a command request from a node, the power supply or potential measurement task is executed, and the measurement data is collected. The data is then uploaded to the corresponding edge node.

[0012] Furthermore, when the resistivity sensing nodes are arranged horizontally along a city road, the resistivity sensing The cable at the node is a multi-core split cascaded high density electric cable, Split cascading cable is a single cable through a cascading electrode conversion switch. The entire cable is connected in series, and the collection station is connected to the end of one entire cable. R, When resistivity sensing nodes are placed along a vertical wellbore, The cable is a single-concentration type high-density electric well cable, and this cable has multiple The electrode junctions are spaced equally apart, each electrode junction is a ground electrode, and the top of the cable is connected to a collection station via a centralized electrode switch; Resistivity sensing nodes are placed horizontally along urban roads and in conjunction with well bores. In this case, the single concentrated high-density electric cable placed in the well hole is first connected to the concentrated electrode conversion switch. It is connected to one end of the multi-core split cascade-connected high-density electric cable on the ground via a switch, and The collection station is connected to the other end of the split cascaded high density electrical cable; and In a concentrated high-density electric cable, multiple electrode junctions are installed at equal intervals, and each electrode The joint is one ground electrode.

[0013] The collection station further includes a control module, a power supply module, and a potential measurement module. a GPS module; The control module, under the direction of the edge node to which it belongs, controls other Controls several modules to manage the operation and self-test of the collection station's system , Communication with edge nodes and mutual exchange of power supply / potential measurement functions under collection command control, It allows you to select channels, execute collection processes, save data, and upload measurement data. After receiving a power supply command, the power supply module controls the corresponding power supply via the control module. Select a pole channel and feed power underground through the cable channel and electrode connected to it At the same time, the magnitude of the power supply current is measured, and after the power supply is completed, the node and its power supply channel are Upload the channel number, measurement start time and power supply current value. The potential measurement module receives a potential measurement command, and then controls the corresponding Select the electrode channel to which you want to apply the potential via the cable channel and electrode connected to it. The measurement is performed and the magnitude of the potential difference is measured. After the measurement is completed, the node and its potential are Upload the measurement channel number, measurement start time and potential difference value. The GPS module is used for precise time signaling and coordination of each node.

[0014] Furthermore, the edge node and the end node communicate with each other via a mobile communication network. data transmission between the edge node and the central cloud computing platform. A data center is a remote data transmission system that transmits data over a wired network.

[0015] A cloud-edge-end cooperative based urban underground space resistivity data collection method, comprising: This method is realized based on the above system, and specifically includes the following steps: Including step, (1) Based on the actual condition of the target street, the maximum search depth, and the resolution of the underground detection target Based on the above, a layout method and collection parameters of the resistivity sensing nodes are determined; (2) Deploy resistivity sensing nodes on target streets and use central cloud computing The platform assigns a unique system number to each edge node, and the edge node Each resistivity sensing node in the domain is assigned a unique system number. A unique system number is assigned to each electrode point in the system, and the three-dimensional location of each electrode point is calculated. Collect the physical coordinates, (3) The central cloud computing platform sequentially distributes data to different edge nodes. Select the edge node to perform block-by-block measurement, and the selected edge node is the resistivity sensing node number. Select one sensing node as a power supply node in the order of The pole combination is the power supply electrode pair AB, and the edge node domain to which this sensing node belongs The electrode combination is selected as a potential measurement electrode pair MN, and the potential measurement electrode pair MN is The pitch of the measurement electrode pair MN and AB is within the effective measurement radius r of AB. If so, power it on and measure the potential. If not, run the next AB. Move to the position of the MN combination and perform a new measurement condition judgment, and JPEG2024534779000002.jpg925, where n is the effective radius coefficient, n=6 to 14, and a is the AB pitch. All power supply electrode pairs and a plurality of potential measurement electrode pairs paired with the power supply electrode pairs in the sensing node When the traversal of the combinations is completed, the power supply and The potential measurement process is complete. (4) Move to the next resistivity sensing node in sequence and perform the power supply and potential measurement process, and finally When all the power supply electrode combinations of the sensing nodes are completed, the power supply and power supply of the current edge node are The entire position measurement process is completed. (5) Then, move on to the next edge node until all edge nodes have been traversed. Go ahead and perform the same power supply and potential measurement process. (6) After the collection operation is completed, the edge node transmits the collected data and its own status information to the edge node. The edge node then forwards the area data to each sensing node. The mat is organized and downloaded from a central cloud computing platform. The edge node performs a rudimentary processing and provides a processing analysis result by quickly comparing the results of the domain model. The analytical results are reported to the central cloud computing platform. The computing platform is capable of storing historical and other multi-source data. Based on the model results of the mart analysis, the model is fed back to each edge node and distributed for subsequent It guides each edge node to perform rapid anomaly analysis and risk identification.

[0016] Furthermore, when AB is a pair of power electrodes, there are two power electrodes between different sensing nodes that satisfy the conditions. The position measurement electrodes of the MN are coordinated by the GPS module time signal, i.e., between different nodes. The multiple potential measurement electrode pairs MN operate in parallel with one power supply electrode pair AB. Realize the measurement.

[0017] In addition, when selecting the power supply electrode pair, the electrode number is selected in ascending order, and the collection station is The electrode point closest to the collection station is designated as electrode A. Then, select the electrode point where the interval between the A and B items is equal to 1 as electrode B and supply power to it. Then, Maintaining the interval of item number B, move A and B to the next electrode point in order, and point B is the current sensing node. When the last electrode point is reached, the power supply process is completed with all AB item intervals equal to 1. , Then, from the starting point, select a measurement point that maintains an interval of two item numbers between A and B and supply power. Then, moving from A to B, when point B arrives at the last electrode point, the gap between A and B becomes two terms. The power supply process is completed for the 1st interval. By repeatedly changing the AB interval, when the maximum isolation factor is reached, the supply of this sensing node is stopped. The charging process is complete.

[0018] Furthermore, the placement of the resistivity sensing nodes is completed at one time, and the position of each electrode point is fixed; and After having the accurate position coordinates, the resistivity sensing node is powered by the edge node corresponding to the resistivity sensing node. The potential measurement collection table is calculated in advance and the sensing node number, potential, The pole numbers and the corresponding sensing node numbers and electrode numbers of the multiple potential measurement points MN are arranged in order, During actual collection, this table is followed in sequence to complete the entire data collection process. Effect of the Invention

[0019] The present invention has the following advantageous effects compared to the conventional techniques:

[0020] 1. Hierarchical storage of sensing data using a "cloud edge-end" architecture By using edge servers at the edge of the network, It realizes edge neighborhood, area division, distributed data collection and data storage, and reduces network congestion. Avoid impacting the collection process and improve the real-time nature of data collection and response efficiency. On the other hand, we will utilize the computing power of the "Center Cloud" to process large amounts of data, Responsible for data mining and risk prediction. "Cloud edge-end" distributed and centralized Adopting the advanced architecture design of medium-sized division of labor collaboration, the capacity and efficiency of the resistivity sensing system is improved. Significantly increase.

[0021] 2. Multi-channel, randomly distributed resistivity sensing node design with unlimited loading capacity By adopting the above, the efficiency of the connection between the electrode channels and the high density power supply / potential measurement combination type To guarantee the type of the system, while reducing the number of sensing nodes and remote transmission devices to the maximum extent possible, Reduces the construction cost of the system.

[0022] 3. Use well-ground linkage to build a three-dimensional spatial randomly distributed sensing network. Both sides of the road Taking full advantage of the advantageous conditions of burying horizontal cables and arranging vertical boreholes in the In order to flexibly arrange the three-dimensional resistivity sensing net across the street, it is necessary to make up for the shortcomings of single surface survey. Provides precise imaging of targets down the street.

[0023] 4. Remote smart communication by combining wireless mobile communication networks and wired public networks Realize area and partition detection in a city of mobile communication network By fully utilizing the characteristics of the resistivity sensing net, the partition and hierarchical management are automatically performed. Due to the large capacity of mobile communication networks, the number of sensing nodes is limited. The scale of the sensing system is flexible and adjustable. The ability to automatically connect to fast backbone networks makes cloud edge-to-end design feasible. Simplify and improve functionality.

[0024] 5. Combining cloud computing platforms with artificial intelligence It realizes automated, intelligent processing and mining of resistivity sensing data, and predicts and alerts. The traditional artificial intelligence Internet of Things (AIoT) technology is now being used to remotely It can be fully utilized as a carrier for information distribution and data mining to smarten resistivity sensing information. It realizes data analysis and predictive warning.

[0025] 6. Make full use of public communication networks and public computing resource platforms to This avoids duplication of system construction and resource waste, while reducing later maintenance costs. The construction of the system is based on the construction of the front-end sensing node, data collection method and system Focus on system architecture design and build on the public Internet of Things. The system performance will be automatically updated as the public Internet of Things system is updated. Automatically upgraded, only maintenance and upgrade of the sensing node unit is required. The input required for the system is relatively small, and the system's expandability and adaptability are significantly enhanced. will be done. [Brief description of the drawings]

[0026] [Figure 1] FIG. 1 is a schematic diagram of a resistivity sensing system architecture for use in the present invention. [Diagram 2] FIG. 2 is a schematic diagram of a collection station structure and cabling scheme used in the present invention. [Diagram 3] FIG. 2 is a schematic diagram of a resistivity sensing system deployment scheme for use in the present invention. [Figure 4] FIG. 2 is a schematic diagram of the ground and well hole electrode installation on both sides of the road in accordance with the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0027] The invention will be further explained in connection with the following drawings.

[0028] 1. System Structure Well-ground collaboration's resistivity sensing system is based on a "cloud-edge-end" architecture The design is based on the sensing node (end) and the data collection process, which is located close to the sensing node. The edge cloud (edge), which is responsible for the coordination of processes, and the edge cloud (edge), which is responsible for centralized data processing and data analysis. We will tackle the central cloud computing platform (Cloud) and Cloud The system consists of wireless and wired transmission networks used for edge-to-end connections. The three layers are the sensing layer, the edge computing layer, and the central cloud computing layer. It is divided into four components (Figure 1).

[0029] a. Sensing layer The sensing layer consists of a number of resistivity sensing nodes, which are horizontally arranged along both sides of the urban road. or in conjunction with vertical well bores for street crossing and well resistivity imaging. This creates a four-dimensional stereoscopic sensing of the resistivity of the area below the street.

[0030] The resistivity sensing node is an independent resistivity sensor unit and is connected to the collection station (Figure 2 a) and a multichannel electrode conversion switch connected to a collection station; and a ground electrode connected to the multi-core high density electrical cable. A high density electric cable is a cable that has many cores and has equally spaced taps and conductive joints. Point to the cable.

[0031] There are three types of resistivity sensing node placement methods:

[0032] (1) Horizontal arrangement along urban roads At this time, the cable in the resistivity sensing node is a multi-core split cascade connection type high density electrical The split cascaded high density electric cable is a cascaded high density electric cable. The cables are connected in series as a whole via a pole-changing switch. The collection station is , connected to the end of one entire cable. Split cascaded high density cables include The electrodes are 8-10 in number. The connection method between the cable and the collection station is shown in Figure 2b. As shown.

[0033] (2) Arrangement along a vertical well hole At this time, the cable at the resistivity sensing node is single, and this cable is integral. The cable is designed with a centralized structure to ensure watertightness. The electrodes are installed at equal intervals, and each electrode junction is a ground electrode. The top of the cable is a concentrated electrode. The cable is connected to the collection station via a pole switch. The connection method is shown in Figure 2c.

[0034] (3) Horizontal and vertical well bore arrangements along urban roads At this time, the cable placed in the wellbore is still a single concentrated high density electric method wellbore cable. The cable has a plurality of electrode junctions arranged at equal intervals, and each electrode junction is A single earth electrode is used. The horizontally arranged cable is a multi-core split cascaded high density cable. However, at this time, a single concentrated high-density cable was placed in the wellbore. Then, through the centralized electrode conversion switch, the multi-core split cascade connection high density cable connect the cascaded cable to one end of the split cascade cable and connect the collection station to the other end of the split cascade cable. The cable and collection station must be connected together to form a single resistivity sensing node. The connection method with the jack is as shown in Figure 2d.

[0035] Here, the split cascade-connected high-density electrical cable is equipped with 8-10 electrodes. The concentrated high-density electric well cable is a watertight one-piece cable containing 30-60 electrode joints. It is a shaped cable.

[0036] The collection station includes a control module, a power supply module, a potential measurement module, and The collection station is composed of a communication module and a GPS module. The communication module performs the power supply or potential measurement tasks according to the command request of the communication module. is responsible for communication between collection stations and edge nodes that belong to them, relying on mobile communication technology. The control module receives the edge node command and controls other nodes of the collection station. Controls several modules to manage the operation and self-test of the collection station's system ,Communication with edge nodes and exchange of measurement roles (power supply / potential measurement) under collection command control , channel selection, collection process execution, data storage and measurement data upload, etc. Achieve control over each module of the system in a series of processes.

[0037] After receiving a power supply command from the edge node, the power supply module Select the appropriate electrode channel and connect the high-density electrical cable and electrode to it. The power supply is completed, and the power supply current is measured. Also upload the power supply channel number, measurement start time and power supply current value.

[0038] The electric potential measurement module receives an electric potential measurement command from the edge node and then Select the appropriate electrode channel via the After the measurement is completed, the node and its Upload the potential channel number, measurement start time and potential difference value. If the measurements belong to different sensing nodes, the edge node can detect the measurement start time of different nodes. Coordinate time (synchronization with GPS time signal) When power supply and potential measurement belong to the same node The collection station's own program coordinates the start time of power supply and potential measurement.

[0039] The communication module adopts 5G and above mobile communication module, and MEC edge access Supports edge computing and collection processes. It directly controls the process and coordinates the power supply / potential measurement channel selection between each sensing node. After the completion of the survey, the collected data is uploaded directly via the mobile communication network. The information is stored in the edge server.

[0040] The GPS module is used to provide accurate time information for each node. The power supply / potential measurement process is Regarding the coordination between different nodes, the accurate time signal using GPS satellites is It is an efficient and simple method to synchronize the GPS antenna and and an interface connection line.

[0041] b. Edge computing layer Resistivity sensing nodes are independent data collection units and have equal status with each other. Although they run independently, they also need to cooperate with each other to achieve combined measurements between different nodes. It is necessary to complete the setup (power supply / potential measurement) and realize imaging across the street. Coordination among multiple sensing nodes requires a higher-level control unit to plan the coordination. The traditional solution is to design one central console and communicate with all The aim is to remotely control sensing nodes. When there are many sensing nodes, network congestion and delays occur. There are many problems with the real-time nature, reliability, and collection efficiency of such a remote centralized control system. There are a number of problems with the centralized control system in the center console, which seems underpowered and It is difficult to continue. Therefore, the collection control "sinks" and moves forward closer to the collection node. This is implemented by moving the data to multiple mobile edge servers, forming a distributed edge control node, and It is necessary to achieve positional placement and proximal control.

[0042] Edge nodes are installed and distributed by a central cloud computing platform. A number of edge servers are distributed throughout the sensing network, and each edge server The system controls multiple sensing nodes in a domain by partition so that they work together. The tasks of the edge node mainly include: (1) Power supply and potential measurement within the domain. (2) Data collection to coordinate and control the selection and collection process of pole pairs. The data in the domain is selected, organized, and recorded in the edge cloud in a designed format. The data is stored and uploaded to the central cloud backup for subsequent Data storage and data transmission for processing during data set. (3) After data collection is completed, Real-time data, historical data, and models based on historical data (center crossover) The results of this area model calculation, which is fed back to the edge node by the loudspeaker, are compared and analyzed. This is a rudimentary process of data to compare differences. If there is an abnormal change, it will report the abnormal information and send it to the central Cloud computing platforms will enable more comprehensive analysis and processing. Make it easier.

[0043] The edge cloud layer simultaneously aggregates distributed wireless networks into wired public networks, It is also a converged network transmission layer that connects wireless mobile communication networks and wired Internet. The mobile communication network is a 5G network. Adopting G and above communication platforms, and using edge servers to collect and control data in the vicinity To achieve distributed collection nodes, efficient control of near-ends, and distributed storage of collected data. We demand that:

[0044] Mobile communication networks have the unparalleled mobility and flexibility of wired networks, especially Adaptable to dynamically increase or decrease the sensing nodes and adjust the positions of the sensing nodes; and Data transmission is distributed via nearby base stations to avoid channel congestion in wired transmission. The area division of mobile cellular base stations and the distributed network structure are It is highly compatible with domain division and distributed configuration, and is favorable for smooth transmission of commands and data. The advantage of using a mobile communication network to transmit data remotely is that it is possible to By making full use of the communication network, it is possible to avoid duplication of the construction of wired sensing networks and reduce costs. This will greatly reduce costs and investments, while providing efficient and stable public network resources. By making full use of the Internet, we can realize the seamless integration of wireless and wired networks, and This avoids the later system operation and maintenance costs of the network transmission layer. be.

[0045] c. Central cloud computing tier Central cloud computing is a common public cloud computing platform. This is realized by relying on network resources, and the scale of the center cloud is larger than that of the edge cloud. The parallel computing capabilities of the system are particularly suited to the demands of high-performance computing, such as large-scale sensor data processing. The combined data is used for storage and intelligent processing and analysis of the entire resistivity sensing data.

[0046] The main tasks of the center cloud include: (1) Operation and management of the entire sensing network. A method for setting up and configuring a distributed edge server, and distributing all resistances through the edge server (2) Global data processing and model inversion to Compare and mine time data against historical data, and send model results to edge servers (3) Anomaly alerts for data exceeding a threshold. Then, report to the urban brain to carry out comprehensive analysis and processing of multi-source data.

[0047] Therefore, the construction of a complete and viable resistivity sensing system requires a central cloud. ,The ,process ,is ,completed ,by ,the ,coordination ,and ,coordination ,of ,the ,edge ,cloud ,and ,the ,sensing ,nodes. The collaboration between the edge cloud and the network is driven by the federated computing paradigm constraints and task allocation. The central cloud and the edge cloud, and between the edge clouds, are based on the federated computing paradigm. The framework dynamically allocates task targets through cloud-edge collaboration and gaming. Realize collaboration and division of labor to jointly ensure the normal operation of the entire system and forward and reverse directions of data streams Maintain and ensure transmission (information feedback).

[0048] The urban brain is also built on the basis of the center cloud, and relatively speaking, the received multi Source data is aggregated, resulting in higher data integration and smart decision support capabilities. It is the final outlet for the results of light perception.

[0049] II. Data collection method 1. Resistivity Sensing Node Placement Resistivity sensing node placement mainly revolves around electrode placement, with surface horizontal placement and vertical well placement. There are two types of hole arrangement. Electrode pitch, total number of electrode channels, cable installation method type, collection station location (with respect to external power supply) and mobile communication antenna and GPS It is necessary to consider multiple factors such as antenna placement, etc. The cables are buried shallowly along green belts or sidewalks on both sides of the street, and the cables in the wells are The wells are laid in holes at the entrance or roadside of the road. The well holes are arranged in pairs on both sides of the road. It is recommended that the cables be placed at the appropriate distance (care should be taken to avoid underground duct cables). The cable in the well is placed to a length of 30 to 60 m, with an electrode pitch of 0.5 to 1 m. It is recommended that the location of the horizontal electrode points should satisfy the randomly distributed electrode placement principle, i.e. There are no special requirements for pole pitch and position, and the poles should be laid as uniformly as possible if conditions permit. After the installation of the electrodes is completed, the electrodes are measured using surveying equipment such as GPS and total stations. The three-dimensional geographic coordinates of the points are collected in a timely manner and entered into the system for subsequent data collection and data Used for data processing.

[0050] The collection station needs to be powered by a commercial power source, so the voltage must be boosted. Plan and design the location of the collection station according to the situation, build a fixed equipment box, and connect to the commercial power supply. It must be used for power access, power boosting and collection station installation. The collection station is a single equipment box (B03, S03, S04, and B04 in Fig. 3). S06, S07 and S08) may be considered for sharing.

[0051] The horizontal cable may be configured as a single wire, a U-shaped two wires, or an S-shaped wire. S02, S03, S06, etc. are single-wire, and S01 and S09 are U-shaped two-wire. Here, S01 is a U-shaped arrangement that connects both sides of the street via a street crossing cable, and S 14 is a U-shaped two-line street that is located on one side of the street. In this case, an S-shaped multiple measurement line (S05 line in Figure 3) may be arranged. The position, pitch and length of the wires may be randomly laid as required, with head-to-tail tandem connections. The tail end of the horizontal cable is also connected to the well cable in tandem to form a single collection station. In FIG. 3, S07 and B06 are independent. A separate collection station may be used to measure the B06 well cable. It may be connected to the tail end of B07 to save the collection station B06. Tandem connection of wells The advantage of this is that the well electrode measurements are completed directly inside the S07 collection station, making it seamless. The disadvantage is that it has more combination measurement methods. The drawback is that there are more points and the collection time is relatively long.

[0052] 2. Collection Parameter Settings The present invention adopts the random dipole device as a unified device type, and The filters are based on regular device types such as Wenner, Schlumberger, and dipole-dipole filters. This includes a wide variety of asymmetric and non-collinear device types, as well as a variety of other asymmetric and non-collinear device types. The random dipole device is a normalized representation of all device types and is The electromotive moment and interelectrode distance parameters are dynamically adjustable, providing wide applicability and flexibility. It is possible to flexibly implement observation settings for complex and special demands (crossing road junctions and setting electrodes at uneven intervals). and dipole-dipole devices have a relatively high detection resolution. In addition to the instrument type setting, the acquisition parameter setting determines the resolution and exploration of the actual measurement. Depth has a decisive influence on the detection efficiency, so it is important to select the appropriate acquisition parameters for optimal detection efficiency. It also needs to be designed.

[0053] (1) Optimal electrode distance The interelectrode distance is the distance between the electrodes placed in front and behind each other in the electrode array, and is randomly distributed. In a cloth system, the actual electrode positions may vary depending on the ground conditions. The distance between the electrodes determines the detection depth, imaging resolution, and system construction cost. Therefore, the distance between the electrodes can be changed by floating, but the relationship between the exploration depth and the resolution is Considering the balance point between the two, there is still an optimal range of the inter-electrode distance. When laying the electrodes, it is recommended to refer to the optimum inter-electrode distance electrode.

[0054] (2) Maximum isolation factor If the electrode pitch is p, the pitch of the power supply and potential measurement points is arranged as p, 2p, 3p, and 4p. A maximum row spacing of N*p (where N is the number of electrode channels in the instrument system) is possible. In actual measurements, the maximum isolation factor (m<=N) is often estimated from the maximum exploration depth h. Determine, JPEG2024534779000003.jpg1141(1) Here, λ = 2 to 3. When collecting data, the peaks of the power supply electrode pair AB and the potential measurement electrode pair MN are The pitch (dipole moment) between the AB and MN electrode pairs and the pitch (interelectrode distance) between the AB and MN electrode pairs are The separation factor is increased sequentially from 1 to m, and all possible ABMN position combination types are traced. To bathe.

[0055] (3) Effective measuring radius: For well site detection and monitoring, the location of power supply and potential measurement points in three-dimensional space The distribution and effective measurement radius issues must be considered. Due to the heterogeneity of the underground medium, the measurement points may vary. Since there are differences in the effective measurement range when the well is located in a different medium, The range is not a perfect, symmetric spherical domain of space. However, the effective measurement range itself is It has a certain elastic change space, which is affected by various factors such as the measurement accuracy and the magnitude of the power supply current. In order to consider the above, the measurement points are still simplified to one "effective measurement sphere domain". It is possible to select the effective measurement sphere domain and improve the measurement efficiency and effectiveness. The radius of the effective measurement sphere domain is The effective measurement radius is R. Outside the effective measurement radius R, the dipole-dipole device As the distance between the electrodes increases, the potential difference drops rapidly and quickly falls below the effective measurement accuracy of the meter. R<=n*a, where n is the effective radius coefficient and a is the distance between the power electrodes. The dipole moment of the sine wave is n = 6 to 8. By adopting the hipole moment measurement design, the accuracy of the reading of the meter can be effectively improved. Therefore, the actual sensing radius factor n can be chosen between 6 and 14. The purpose of setting the effective radius is to measure according to the effective measurement radius during actual data collection. By setting a constant threshold, most of the measurement processes that exceed the effective measurement radius are eliminated, improving data collection efficiency. The aim is to improve

[0056] At the same time, during the measurement, the pitch of the power supply electrode pair AB and the measurement electrode pair MN are adjusted in real time according to the following formula. Calculate in IM, The coordinates of points A and B are (x A , y A , z A ) and (x B , y B , z B ), then A The coordinates of point O at B are as follows: JPEG2024534779000004.jpg1350JPEG2024534779000005.jpg1349JPEG2024534779000006.jpg1348(2) The coordinates of points M and N are (x M , y M ) and (x N , y N ), then the point in MN The coordinates of O1 are as follows: JPEG2024534779000007.jpg1253JPEG2024534779000008.jpg1254JPEG2024534779000009.jpg1252(3) Then, OO' pitch L is as follows: JPEG2024534779000010.jpg15121(4) Then, it is compared with the effective measurement radius set, and the measurement of MN points that exceed it is canceled and data collection is performed. Accelerate the collection process.

[0057] 3. Selection of the power supply electrode pair and the measurement electrode pair In the data collection process of the present invention, the system supports a non-uniform, random distribution of measurement points. In order to support the AB and MN pitches, the separation factor (multiples of the electrode pitch) increases. The position information of all measurement points is obtained by positioning measurement before collection, and ABM The effective measurement radius changes dynamically according to the measurement process based on the position between N in real time. It is necessary to calculate the data in a time-series manner and control the collection and selection process.

[0058] The entire measurement process of the present invention revolves around the power supply process, i.e., All possible combinations of feed electrodes (feed electrode pair A and B) are traversed. For the combination, all possible combinations of potential measuring electrodes corresponding to this power supply electrode pair AB (a pair of potential measurement electrodes MN within the effective measurement sphere domain in the power supply node or in the surrounding nodes) The measurement process is performed by traversing and searching the edge node number. Then, all the power supply points are traversed in the order of the sensing node number until the last power supply point measurement is completed. Then, one data collection process for the entire measurement area is completed. Then, at the set time interval, The measurement process is repeated to realize four-dimensional dynamic sensing. If an abnormality is found in a certain area, By adjusting the observation frequency and performing density measurements on the entire area or on abnormal sections (edge ​​node settings), In addition, anomaly verification can be performed in conjunction with other detection means and on-site inspections.

[0059] The specific execution process is as follows:

[0060] (1) The central cloud computing platform is sequentially distributed to different edge nodes. Select the edge node and perform block-by-block measurement (the selected edge node is the active node). The other edge nodes that are not activated are in a sleeping state), The edge node selects one sensing node as a power supply node in the order of the sensing node number. Then, one electrode combination in the sensing node is called the power supply electrode pair AB, The electrode combination in the edge node domain is selected as a measurement electrode pair MN, and this measurement The fixed electrode pair MN belongs to the same sensing node, and the pitch between the measurement electrode pair MN and AB is the effective pitch of AB. If so, power is supplied and the potential is measured. If not, If not, move to the next ABMN combination and make a new measurement condition judgment. The traversal of all combinations of the power supply electrode pairs and the potential measurement electrode pairs in the detection node is completed. Then, the power supply and potential measurement process is completed when the sensing node is the power supply node. (2) Move to the next resistivity sensing node in sequence and perform the power supply and potential measurement process, and finally When all the combination electrode pairs of the sensing node are powered, the current edge node is powered and The entire measurement process is completed. (3) Then, move on to the next edge node until all edge nodes have been traversed. Proceed and perform the same powering and measuring process.

[0061] In addition, to ensure the integrity of collected data and improve the clarity of underground imaging, When selecting the power supply electrode pair, the distance between the electrodes AB is constantly changed in the following manner.

[0062] (1) When the edge node sequentially selects the sensing node as the power supply node, the power supply electrode pair AB is , traverse and select only this sensing node, and the selection of AB point is made according to the small electrode point number. Start at the end closest to the collection station (start point) and then The electrode point with the closest distance is electrode A, and the electrode point with the AB item number interval equal to 1 is electrode B. Then, while maintaining the interval between the item numbers of A and B, A and B are set to the next electrode point. When point B arrives at the last electrode point of the current sensing node, all of the AB items The powering process with interval equal to 1 is completed, (2) Then, from the starting point, select a measurement point that maintains an interval of two items between A and B and supply power. Then, by moving from A to B in order, when point B arrives at the last electrode point, there are two gaps between A and B. The power supply process is completed at the interval of item number. (3) Repeat changing the interval of A and B items until the maximum isolation factor is reached. The node power supply process is completed. Then, the next sensing node is selected and the power supply point selection process is repeated. The process is repeated to supply power.

[0063] Each time power is supplied, the edge node sets the power supply start time, power supply parameters, etc. After completion, the power supply electrode number, start time and power supply current value are saved. The entire data is uploaded to the edge node for processing and sorting.

[0064] At the same time, in order to improve the data collection rate, the processing of the measurement electrode pairs according to the invention is There are two collection methods: intra-node and inter-node.

[0065] (1) Within a node, sequential serial measurements are adopted: Within a node, measurements are taken every time power is supplied. According to the MN sequential table, one of the measurement electrode pairs MN is selected to measure the potential. Then, select another pair of MNs and perform the next power supply and potential difference measurement. The various combinations of MNs in the mode are determined by the combination of one of the potential measurement electrode pairs. It is necessary to select the first one first and then perform the measurement process.

[0066] (2) Parallel measurements are adopted between nodes simultaneously: MN is connected to other nodes other than the node where AB is located. When MN is located at a node, if AB is powered at each measurement, MN located at a different node The potential is measured simultaneously. The GPS time signal is used to measure the potential of multiple electrode pairs MN and the supply voltage between different nodes. The electrode pairs A and B are coordinated to operate simultaneously in parallel, achieving "one power supply, multiple measurements."

[0067] Furthermore, the data collection process of the present invention employs a method of searching for potential measurement points while supplying power. This method, although feasible, is too inefficient and involves a lot of duplicate calculations and reckless sizing. This process involves a survey and seriously affects the efficiency of data collection. Therefore, the collection table should be prepared in advance. The collection efficiency can be improved by using a pre-populated method, i.e., the sensing nodes are deployed at once. After the electrode point is fixed and has accurate position coordinates, A table of power supply and potential measurement collection is created in advance by calculating the node number, electrode number and The node number and electrode number of the corresponding potential measurement point MN can be arranged. Since there is a one-to-many relationship between the power supply electrodes, the collection table contains multiple There are potential measurement electrode pairs MN. MNs belonging to the same sensing node are placed in front and behind the column, The potential measurement points belonging to different sensing nodes are arranged in rows according to the node numbers. When collecting data, for each power supply point AB, the measurement electrode pair MN numbers from different nodes are sequentially After the measurement is completed, the node number, electrode number, and collection number are The collection time and the potential difference value are stored. Then, the next column is moved to the next column, and the different Extract the measurement electrode pair MN number of the node, notify the power supply point AB of power supply, and Notify the electrode of the corresponding number at the node of the number to measure the potential, and record and save it. The pointer then points to the MN electrode pair in the next row, extracts it, and uses it for potential measurement. When the MN in the column is empty, the AB electrode pair powering process is complete and moves to the next power point in the collection table. The above process is continued until the measurement of the last potential measurement point of the last power supply point in the collection table is completed. Repeating this process completes the entire data collection process for this edge node.

[0068] If there is an update to the sensing node (an increase or decrease in the number of sensing nodes), the updated information is submitted and the measurement after the update is performed Recalculate the collection table for the process. This table allows you to calculate the collection station's This can significantly reduce the amount of computational work and search time, improving collection efficiency.

[0069] 4. Upload and store measurement data After the collection operation is completed, the edge node uploads all the data of this operation. The edge node notifies each sensing node to upload the data. The data uploaded by each node is organized in chronological order. The data includes the measurement of the potential, and is collated by time and collection table, and the edge noise node number, sensing node number, measurement time, power supply point A number, power supply point B number, measurement point M number, measurement Create an electronic table of fixed point N number, power supply current I, potential difference V, device coefficient K, and apparent resistivity Ps where the device coefficient K and apparent resistivity Ps are determined by the ABMN position coordinates and the power supply voltage. The calculation is based on the current I and the potential difference V, and then added to the table to form a complete measurement data information table. do.

[0070] 5. Data Processing and Data Mining (Artificial Intelligence Cloud Computing) Artificial intelligence for big data processes data flow throughout the resistivity sensing system Based on the competitive collaboration and optimized deployment of the federated computing-based smart edge cloud, From forming an automatic optimization management of the sensing node data collection process to the central cloud Based on data mining and smart analysis of big data and machine learning, the sensing model The goal is to build a system that can automatically and quickly identify anomalies.

[0071] The most important feature of the present invention is the data storage between each component unit in the constructed system. The realm of smart bidirectional feedback is the existence of a sensing node. The edge nodes control the nodes and send their status information and collected data to the edge nodes. The edge node automatically uploads the collected parameters to the edge node in a timely manner. This makes it easy to adjust meter settings and update data collection frequency. Edge nodes are The results of the analysis are reported to the central cloud data center. Feedback based on smart analysis results based on historical data and other multi-source data Distribute it to each edge node and instruct each edge node to perform rapid anomaly analysis and risk identification. The urban brain receives model prediction results and early warning information sent from the data center. It is trusted and combined with other multi-source data for scientific analysis and decision-making. Multi-source data and its history are sent back to the data center to support model calibration and refinement. Support.

[0072] 6. Multi-source data analysis and smart decisions The sensing system is being built up day by day, obtaining large-scale apparent resistivity data, and the apparent resistivity The resistivity is merely a comprehensive reflection of the resistivity of the underground and spatial structure, and only by inverting the resistivity can the resistivity be determined. Three-dimensional and four-dimensional resistivity imaging are needed to obtain the imaging results. It requires computational resources and machine time. Inversion of data over a large range is neither economical nor practical. Therefore, the present invention adopts artificial intelligence algorithms at the cloud end to analyze the resistivity of big data. We perform smart analysis and data mining on the data, and identify abnormal points and differences with large changes. Identify and find normal regions, and perform precise 4D inversion on anomalous regions to determine the course of anomalous regions. Understand the time-varying characteristics, eliminate the causes of factors such as weather, and focus on the abnormal section. If abnormal changes tend to accelerate or expand in scope, Enter risk assessment mode: 1. Further increase the measurement frequency to perform dynamic real-time observation. 2. On-site drilling survey verification and other geophysical methods (radar, electromagnetic or seismic surveys) ) Verification of the site, including confirmation and verification. If an abnormality is ruled out through on-site verification, the cause is analyzed and If an abnormality is found during on-site inspection, a report is sent to the urban central office. Report the cause of the anomaly and activate multi-source data analysis and expert systems to determine the source and formation of the anomaly The system will then be used to make a decision and provide a temporary solution. Train a set optimization model to improve prediction performance.

Claims

1. A cloud-edge-end cooperation-based urban underground space resistivity sensing system, comprising: Adopting a cloud edge-end architecture design, central cloud computing platform, the central cloud computing platform distributed network Multiple edge servers connected to a distributed network, and each edge server connected to a distributed network a plurality of resistivity sensing nodes connected to the The central cloud computing platform manages the entire resistivity sensing system. Manage, perform global data processing and model inversion, and detect data anomalies that exceed the threshold. and to manage the entire resistivity sensing system. This involves setting up and configuring distributed edge servers, and distributing all resistance sensing through the edge servers. The above-mentioned global data processing and model inversion includes managing the knowledge nodes. Compare real-time data with historical data, mine it, and send the model results to the edge server. and provide guidance on basic data analysis. The edge server may be configured to allow multiple resistivity sensing nodes in a domain to work in a coordinated manner. An edge node for partition-specific control, which controls power supply and power supply within a domain. The selection of the position measurement electrode pair and the collection process are controlled in a coordinated manner, and the data obtained by the data collection are The system selects, organizes, and stores the collected data in a designed format, and Upload and back up data to a central cloud computing platform Once data collection is complete, real-time and historical data can be shared with a central cloud. Cloud computing platform feeds edge nodes based on historical data The results of the area model calculations are compared and analyzed to determine whether there are any abnormalities and to When there is a constant change, abnormal information is sent to the central cloud computing platform. and informing the user of the The resistivity sensing node is an end node, and the resistivity sensing nodes are and / or disposed vertically along the well bore, and each resistivity sensing node is independently A resistivity sensor unit is provided in a collection station and connected to the collection station. A multi-channel electrode conversion switch, a multi-core high-density electrical cable, and a multi-core high-density electrical and a ground electrode connected to the ground cable, and the resistivity sensing node is In response to a command request from a node, the power supply or potential measurement task is executed, and the measurement data is collected. The cloud-edge-end cooperation is characterized by uploading the data to a corresponding edge node. -based urban underground space resistivity sensing system.

2. When resistivity sensing nodes are arranged horizontally along city roads, The cable in this case is a multi-core split cascade-connected high-density electric cable, The corded cable is connected to the entire cable via a cascaded electrode conversion switch. and a collection station is connected to the end of one overall cable. When resistivity sensing nodes are placed along a vertical wellbore, The cable is a single-concentration type high-density electric well cable, and this cable has multiple The electrode junctions are spaced equally apart, each electrode junction is a ground electrode, and the top of the cable is connected to a collection station via a centralized electrode switch; Resistivity sensing nodes are placed horizontally along urban roads and in conjunction with well bores. In this case, the single concentrated high-density electric cable placed in the well hole is first connected to the concentrated electrode conversion switch. It is connected to one end of the multi-core split cascade-connected high-density electric cable on the ground via a switch, and The collection station is connected to the other end of the split cascaded high density electrical cable; and In a concentrated high-density electric cable, multiple electrode junctions are installed at equal intervals, and each electrode The cloud edge device according to claim 1, characterized in that the connection is made to one ground electrode. End-coordination based urban underground space resistivity sensing system.

3. The collection station includes a control module, a power supply module, and a potential measurement module. a communication module and a GPS module, The control module, under the direction of the edge node to which it belongs, controls other Controls several modules to manage the operation and self-test of the collection station's system , communication with edge nodes and mutual exchange of power supply / potential measurement functions under collection command control, It allows you to select channels, execute collection processes, save data, and upload measurement data. After receiving the power supply command, the power supply module controls the corresponding Select an electrode channel and connect it to a cable channel and the electrode to supply power underground. After the power supply is completed, the node itself and the power supply are Upload the channel number, measurement start time and power supply current value. After receiving the potential measurement command, the potential measurement module controls the corresponding Select the electrode channel to which you want to apply the potential via the cable channel and electrode connected to it. The measurement is performed and the magnitude of the potential difference is measured. After the measurement is completed, the node and its potential are Upload the measurement channel number, measurement start time and potential difference value. The GPS module is used for accurate time signaling and coordination of each node. The present invention relates to a cloud-edge-end cooperation-based well-ground cooperation resistivity sensing method. system.

4. The edge node and the end node transmit data remotely through a mobile communication network. The edge node and the central cloud computing platform 2. The method according to claim 1, wherein the remote data transmission is performed via a wired network. A well-ground coordinated resistivity sensing system based on cloud-edge-end cooperation.

5. A cloud-edge-end cooperation based urban underground space resistivity data collection method, comprising: The method is realized according to the system according to claim 1, and specifically includes the following steps: It includes steps such as: (1) Based on the actual condition of the target street, the maximum search depth, and the resolution of the underground detection target. Based on the above, a layout method and collection parameters of the resistivity sensing nodes are determined; (2) Deploy resistivity sensing nodes on target streets and use central cloud computing The platform assigns a unique system number to each edge node, and the edge node Each resistivity sensing node in the domain is assigned a unique system number. A unique system number is assigned to each electrode point in the system, and the three-dimensional location of each electrode point is calculated. Collect the physical coordinates, (3) The central cloud computing platform sequentially deploys different edge nodes Select the edge node to perform block-by-block measurement, and the selected edge node is the resistivity sensing node number. Select one sensing node as a power supply node in the order of The pole combination is the power supply electrode pair A and B, and the edge node domain to which the sensing node belongs is The electrode combination is selected as a potential measurement electrode pair MN, and the potential measurement electrode pair MN is The pitch of the measurement electrode pair MN and AB is within the effective measurement radius r of AB. If so, power is supplied and the potential is measured. If not, the next A / B Move to the position of the MN combination and perform a new measurement condition judgment, and set the effective measurement radius of the AB where n is the effective radius coefficient, n=6 to 14, and a is the AB pitch. All power supply electrode pairs and a plurality of potential measurement electrode pairs paired with the power supply electrode pairs in the sensing node When the traversal of the combinations is completed, the power supply and The potential measurement process is complete. (4) Move to the next resistivity sensing node in sequence and perform the power supply and potential measurement process, and finally When all the power supply electrode combinations of the sensing nodes are completed, the power supply and power supply of the current edge node are The entire position measurement process is completed. (5) Then, move on to the next edge node until all edge nodes have been traversed. Go ahead and perform the same power supply and potential measurement process. (6) After the collection operation is completed, the edge node sends the collected data and its own status information to The edge node then forwards the area data to each sensing node. The mat is organized and downloaded from a central cloud computing platform. The edge node performs a rudimentary processing and provides a processing analysis result by quickly comparing the results of the domain model. The analytical results are reported to the central cloud computing platform. The computing platform is capable of storing historical and other multi-source data. Based on the model results of the mart analysis, the model is fed back to each edge node and distributed for subsequent The system is characterized by instructing each edge node to perform rapid anomaly analysis and risk identification. A cloud-edge-end coordination-based urban underground space resistivity data collection method.

6. When A and B are a pair of power supply electrodes, the potential measurement electrodes located between different sensing nodes satisfy the conditions. The pole pair MN is coordinated by GPS module time signal, i.e., multiple The potential measurement electrode pair MN operates in parallel with one power supply electrode pair AB to perform one power supply multiple measurements. The cloud-edge-end collaborative based city according to claim 5, Subsurface resistivity data collection methods.

7. When selecting a power supply electrode pair, select the electrode pair in ascending order of item number and the collection station. Starting from one end, the electrode point closest to the collection station is called electrode A, and AB The electrode point with the item number interval equal to 1 is selected as electrode B to supply power, and the item number of A and B is The distance between A and B is kept constant, and A and B are moved to the next electrode points in order. Point B is the last electrode of the current sensing node. When the pole is reached, the power supply process is completed with all A and B item numbers having intervals equal to 1. Then, from the starting point, select a measurement point that maintains the interval of two item numbers between A and B and perform power supply. Then, moving from A to B in order, when point B arrives at the last electrode point, the gap between A and B becomes two terms. The power supply process is completed for the 1st interval. The AB interval is repeatedly changed, and when it reaches the maximum isolation factor, the supply of this sensing node is stopped.

6. The cloud edge-end cooperation method according to claim 5, wherein the power supply process is completed. -based urban underground resistivity data collection method.

8. The placement of the resistivity sensing nodes is completed at one time, and the position of each electrode point is fixed and accurate. After obtaining the position coordinates, the edge node corresponding to the resistivity sensing node is used to supply power and measure potential. A fixed collection table is calculated in advance and the sensing node number, electrode number and The sensing node numbers and electrode numbers of the corresponding potential measurement points MN are arranged in order to obtain the actual collection The data collection process is completed by following this table. The cloud edge-end cooperation based urban underground space resistance according to claim 5, 6 or 7. Resistance data collection method.

Images