A communication method and system for dynamic networking of SF6 remote density meters

Abstract

The application relates to the technical field of power detection equipment and discloses a communication method and system for dynamic networking of an SF6 remote density meter, wherein the method is used for establishing a wireless connection with a started field sensor, acquiring signal strength and position information, calculating a signal intermediate value as a screening value, dynamically adjusting a screening condition in combination with a position distribution uniform value until the distribution meets the requirements, dividing a position interval according to the requirements and selecting a networking sensor, and controlling the field sensor to network with the networking sensor; the networking sensor responds to a collection instruction, generates data by itself, wakes up the field sensor to collect data, and sleeps together after being packaged and uploaded. The scheme is suitable for remote density monitoring of the power system and can adapt to different scenes through dynamic networking, improve data transmission reliability, and reduce energy consumption.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power detection equipment, in particular to a communication method and system for dynamic networking of an SF6 remote density meter. BACKGROUND

[0002] In the operation process of a power system, a density sensor is a key monitoring device for ensuring safe and stable operation of equipment and is widely used in density monitoring of key mediums such as transformer oil and SF6 gas. The monitoring data can reflect the internal medium state of the power equipment in real time and provide an important basis for equipment fault early warning and maintenance decision-making. Traditional density sensors in power systems are mostly local monitoring types, and data needs to be read by staff on site, which is not only low in efficiency but also difficult to meet the real-time monitoring needs in special scenarios such as remote areas and high voltage levels.

[0003] With the penetration of Internet of Things technology in the power field, remote density sensors gradually replace traditional local sensors and become mainstream. At present, most remote density sensors on the market directly upload data to the power monitoring background in a direct communication mode. Although this direct connection mode can realize basic data transmission function in short-distance and low-complexity monitoring scenarios and promote the automation development of density monitoring in power systems.

[0004] However, there are problems such as electromagnetic interference of high-voltage equipment and complex terrain shielding in the monitoring area of the power system, which leads to unstable signal strength between the sensor and the background, easy occurrence of data packet loss and transmission delay, and affects the real-time and accuracy of the monitoring data. The networking structure is fixed and cannot adapt to the differences in communication environment in different power scenarios. SUMMARY

[0005] In order to improve the remote stability of the remote density sensor, the application provides a communication method and system for dynamic networking of an SF6 remote density meter.

[0006] In the first aspect, the application provides a communication method for dynamic networking of an SF6 remote density meter, which adopts the following technical scheme:

[0007] A communication method for dynamic networking of an SF6 remote density meter, comprising the following steps:

[0008] Wireless connection is established with a plurality of activated field sensors to obtain signal strength of all field sensors and position information of the field sensors;

[0009] A signal intermediate value is calculated using a preset screening algorithm according to a plurality of signal strengths;

[0010] The intermediate value of the signal is taken as a signal screening value, the position information of the on-site sensor with a signal strength greater than the signal screening value is screened according to the signal screening value, and the position distribution uniform value is calculated using a preset distribution degree algorithm according to the screened position information;

[0011] If the position distribution uniform value is greater than a preset distribution requirement value, a difference value between the distribution requirement value and the position distribution uniform value is calculated as a distribution difference value, the signal screening value is adjusted according to the distribution difference value, the smaller the distribution difference value is, the greater the down-regulation amplitude of the signal screening value is, and the signal screening value is adjusted in a loop until the position distribution uniform value is less than or equal to the distribution requirement value;

[0012] A plurality of position intervals are divided according to the screened position information after the adjustment is completed, the on-site sensor closest to the center position in the position interval is obtained as a networking sensor, and the other on-site sensors in the position interval are controlled to network with the networking sensor;

[0013] The networking sensor obtains a collection instruction, generates sensing data and generates a start work instruction in response to the collection instruction, and sends the start work instruction to the on-site sensors in the network, the on-site sensors generate sensing data and send the sensing data to the networking sensor in response to the start work instruction after starting the collection work, and the networking sensor packages and uploads the received sensing data and the sensing data of itself, and then the networking sensor and the on-site sensors enter a sleep state.

[0014] By adopting the above technical solution, through dynamic networking, data flows between grids, compared with the traditional direct connection mode, it can effectively cope with the problems of electromagnetic interference of high-voltage equipment in power system monitoring area, complex terrain shielding and other problems, increase the probability of successful data transmission, reduce data packet loss and transmission delay phenomenon, and ensure the real-time and accuracy of monitoring data. According to the signal strength and position distribution, the screening condition is dynamically adjusted and the position interval is divided for networking, which is no longer limited by the fixed networking structure, and can well adapt to the communication environment difference in different power scenarios. The in-group sensors are close to each other, and start collection and transmission work synchronously, which increases the probability of correct data transmission and further ensures the reliable transmission of sensing data. The networking sensor and the on-site sensor enter a sleep state after completing data collection and uploading, reducing unnecessary energy consumption.

[0015] Optionally, the method further comprises the following steps:

[0016] Obtaining the sensor numbers of all networking sensors;

[0017] According to the signal strength, the obtained sensor numbers are sorted to obtain a networking number sequence;

[0018] After sorting the networking numbers, all networking sensors are woken up for the first time, and all other on-site sensors are woken up by the networking sensors;

[0019] all other field sensors are hibernated by the networking sensor, and all networking sensors are hibernated again;

[0020] According to a preset setting period, a sensor number is randomly selected from the networking number sequence, and a plurality of sensor numbers are selected in the two directions of the networking number sequence with the selected sensor number as an interval, and all the selected sensor numbers are not adjacent to each other;

[0021] The selected sensor numbers are marked and sorted to obtain a first number sequence, and the unselected sensor numbers are marked and sorted to obtain a second number sequence;

[0022] The corresponding networking sensors are sequentially woken up according to the first number sequence, and the corresponding field sensors are woken up through the networking sensors; and the corresponding networking sensors are sequentially woken up according to the second number sequence, and the corresponding field sensors are woken up through the networking sensors.

[0023] By adopting the above technical solution, the inter-group time-sharing wake-up is adopted to avoid the mutual interference of signals between sensors and to ensure the signal transmission quality; during installation, the sensors automatically acquire the fragmentation information and the signal strength, automatically divide the communication groups and negotiate different wake-up times, and realize the automation and intelligentization of wake-up; the reasonable wake-up time and number selection strategy of different groups and within the groups balance the sensor energy consumption and prolong the overall service life of the network; the random and interval selection of sensor numbers enhances the system fault tolerance.

[0024] Optionally, the method further comprises the following steps:

[0025] The networking sensor corresponding to the first selected sensor number is defined as a temporary center node, and the temporary center node acquires the packaged data of other selected networking sensors and forwards the acquired packaged data for uploading;

[0026] After the packaged data transmission of the current networking sensor is completed, the next networking sensor is woken up; and the wake-up time interval of the next networking sensor is adjusted according to the positive correlation of the data amount of the currently acquired packaged data;

[0027] After the packaged data transmission of all the networking sensors is completed, the networking sensor with the largest data amount of the packaged data is selected as the next temporary center node.

[0028] By adopting the technical scheme, the temporary center node collects data and updates the time, eliminates the wake-up error, and realizes the time synchronization within and between groups; the next group of network sensors is dynamically adjusted according to the current data volume, the transmission rhythm is optimized, and the load is balanced; the temporary center node is preferentially synchronized with the platform clock, the accuracy of the clock interval between groups is guaranteed, and the interference is reduced; the network sensor with the largest data volume is dynamically selected as the next temporary center node, the node with stronger processing capacity is made to undertake the center task, and the adaptability of the system to different scenes is enhanced.

[0029] Optionally, the method further comprises the following steps:

[0030] The first numbering sequence and the second numbering sequence operate simultaneously;

[0031] The network sensor corresponding to the first selected sensor number in the first numbering sequence is defined as a temporary center node;

[0032] The network sensor corresponding to the first selected sensor number in the second numbering sequence is defined as a verification center node;

[0033] The temporary center node and the verification center node both obtain the packaged data of the other selected network sensors, the packaged data does not contain the sensor data of the temporary center node and the verification center node, and the temporary center node sends the obtained packaged data to the verification center node;

[0034] The verification center node verifies the consistency of the packaged data obtained by itself and the packaged data from the temporary center node;

[0035] If the consistency verification is passed, the verification center node packages and sends the sensor data of itself to the temporary center node, and the temporary center node uploads all the obtained packaged data.

[0036] By adopting the technical scheme, the data consistency is cross-verified by the two nodes, forming a double-check barrier, greatly improving the data verification accuracy and reducing the error; the two numbering sequences operate synchronously, and the two nodes obtain data respectively, constructing a double-backup transmission path to avoid the loss of key data due to single-point failure; the verification center node only transmits its own data to the temporary center node, and the temporary center node uploads all the obtained packaged data, which simplifies the verification process, reduces the redundant link, and improves the system operation efficiency; the two nodes cooperate to make the system have the fault self-checking and data self-repairing capability, and even if a single node is abnormal or offline, the other node can still guarantee the operation.

[0037] Optionally, the method further comprises the following steps:

[0038] The signal strength of the last plurality of temporary center nodes is recorded as a temporary strength, and the signal strength of the last plurality of verification center nodes is recorded as a verification strength;

[0039] The temporary comprehensive strength is calculated according to the plurality of temporary strengths, and the verification comprehensive strength is calculated according to the plurality of verification strengths;

[0040] The node strength ratio is calculated as a ratio of the temporary comprehensive strength and the verification comprehensive strength;

[0041] If the node strength ratio is less than a preset required strength ratio, the operation instructions of the temporary central node and the verification central node are interchanged.

[0042] By using the above technical solution, the signal comprehensive strengths of the recent plurality of temporary central nodes and verification central nodes are recorded and calculated to obtain the node strength ratio, and when the ratio does not meet the preset requirement, the operation instructions of the two types of nodes are interchanged, so that the node with stronger signal undertakes the more core data processing and transmission task, effectively avoids the data receiving deviation caused by the imbalance of node signal strength, and further improves the data verification accuracy.

[0043] Optionally, the method further comprises the following steps:

[0044] The first numbered sequence and the second numbered sequence operate simultaneously; the communication strength or the power of the networking sensor in the sequence is detected;

[0045] The networking sensor corresponding to the sensor number with the maximum communication strength or the most power in the first numbered sequence is defined as the temporary central node;

[0046] The networking sensor corresponding to the sensor number with the maximum communication strength or the most power in the second numbered sequence is defined as the verification central node;

[0047] The temporary central node and the verification central node both obtain the packaged data of the other selected networking sensors, and the packaged data does not contain the sensing data of the temporary central node and the verification central node; the temporary central node sends the obtained packaged data to the verification central node;

[0048] The verification central node verifies the consistency of the packaged data obtained by itself and the packaged data from the temporary central node;

[0049] If the consistency verification is passed, the verification central node packages and sends the sensing data of itself to the temporary central node; and the temporary central node uploads all the obtained packaged data.

[0050] By adopting the technical scheme, the networking sensor with the maximum communication strength or the most power in the first and second numbering sequences is selected as the temporary center node and the verification center node respectively, so that the double center nodes have better signal transmission capability or longer continuous working capability, and a reliable hardware foundation is provided for data receiving and verification; the double center nodes synchronously acquire other node data and cross-verify consistency, which not only retains the guarantee of data accuracy by double verification, but also reduces data transmission errors caused by weak signals or insufficient power due to the performance advantages of the nodes, and further improves the reliability of data verification and the stability of system operation.

[0051] Optionally, the method further comprises the following steps:

[0052] An interval range area of the location interval is acquired based on a GIS map;

[0053] If the interval range area is greater than a preset required area, a new location interval is divided by averaging, and a new networking sensor is selected in the new location interval;

[0054] The communication strength of the new networking sensor is acquired, and the interval range area is adjusted in positive correlation with the communication strength.

[0055] By adopting the technical scheme, the interval area is accurately acquired, the interval is averagely divided and a new networking sensor is added if the interval area is greater than a preset area, the networking layout is optimized, and the transmission blind area in a large interval is avoided; the interval range is adjusted in positive correlation with the communication strength of the new networking sensor, the adaptation of strong and weak signals to the interval size is realized, data transmission is stabilized, and the deployment cost is reduced; the layout is flexibly adjusted by combining the communication strength with the terrain features reflected by the GIS map, and the adaptability to complex terrains is improved.

[0056] Optionally, the method further comprises the following steps:

[0057] If the communication with the platform end cannot be completed, the disabled temporary center node starts a networking sensor closest to the disabled temporary center node as a new temporary center node;

[0058] The packaged data in the disabled temporary center node is sent to the new temporary center node, and the mark of the disabled temporary center node is restored to a field sensor.

[0059] By adopting the technical scheme, the disabled fault node is started, and a networking sensor closest to the disabled fault node is started as a new temporary center node, so that seamless switching of nodes is realized, communication link continuity is ensured, and monitoring data disconnection is avoided; the packaged data of the disabled node is transmitted to the new node, so that seamless migration of data is realized, and data safety is ensured without loss; the disabled node is restored to a field sensor to reuse resources, and the response time is shortened by switching the node nearby, so that the system fault tolerance efficiency and resource utilization rate are improved.

[0060] Optionally, the method further comprises the following steps:

[0061] Obtaining historical communication intensity data and historical residual power data of all selected networking sensors in the first numbering sequence;

[0062] Performing normalized and weighted calculation on the historical communication intensity data to obtain a communication performance value, and performing normalized and weighted calculation on the historical residual power data to obtain a endurance performance value;

[0063] Summing the communication performance value and the endurance performance value to obtain a comprehensive performance score;

[0064] Selecting a networking sensor with the highest comprehensive performance score as a temporary center node of the first numbering sequence; wherein the weight coefficient of the communication performance value and the weight coefficient of the endurance performance value are dynamically adjusted according to the electromagnetic interference level of the current monitoring area; when the electromagnetic interference level is higher than a preset threshold, the weight coefficient of the communication performance value is higher than the weight coefficient of the endurance performance value; when the electromagnetic interference level is lower than or equal to the preset threshold, the weight coefficient of the endurance performance value is higher than the weight coefficient of the communication performance value.

[0065] By adopting the above technical solution, the comprehensive performance score is obtained by performing normalized and weighted calculation on the historical communication intensity and residual power data, ensuring the comparability of different dimension performance indicators, and making the selection of the temporary center node more scientific and objective. At the same time, the weight coefficients of the communication performance value and the endurance performance value are dynamically adjusted according to the electromagnetic interference level of the monitoring area, the signal transmission capability of the center node is prioritized to reduce data packet loss in a strong interference environment, and the endurance of the node is focused on improving to extend the network working time in a low interference environment, realizing the precise adaptation of the center node selection strategy to the actual communication environment, ensuring the stability of data transmission, and optimizing the system energy consumption.

[0066] In a second aspect, the application provides a communication system for SF6 remote density meter dynamic networking, which adopts the following technical solution:

[0067] A communication system for SF6 remote density meter dynamic networking, comprising a processor, wherein the processor executes the steps of the communication method for SF6 remote density meter dynamic networking according to any one of the above.

[0068] In summary, the present application includes at least one of the following beneficial technical effects: by dynamic networking, double verification and transmission load dynamic adjustment, the data transmission stability and accuracy are greatly improved, and the power monitoring data is reliable; relying on dynamic layout, flexible networking and intelligent awakening, highly adapt to complex power scenarios, enhance the adaptability of networking environment; with hibernation mechanism, balanced awakening and resource reuse, optimize system energy consumption and resource utilization, prolong the life of sensor network; through fault rapid switching, data migration and anti-fault design, strengthen the fault tolerance of the system, and guarantee the continuity of the monitoring link. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 is a step diagram of a communication method of SF6 remote density meter dynamic networking.

[0070] Figure 2 is a step diagram of an awakening mechanism further optimizing the basis of dynamic networking.

[0071] Figure 3 is a step diagram of realizing networking optimization on the basis of time-sharing awakening. DETAILED DESCRIPTION

[0072] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0073] In the description of the present specification, the description referring to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the described embodiments or examples are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily mean the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0074] The communication method for dynamic networking of SF6 remote density meters disclosed by the embodiments of the present application is suitable for the dense installation scene of remote density meters of SF6 gas insulated equipment and transformer oil pillow in a substation. In such a scene, 30-150 remote density meters need to be deployed in a single area, the installation density is high, the single working energy consumption of the equipment needs to be controlled within 80 mWh, the power consumption limit is low, and it is affected by high-voltage equipment electromagnetic radiation and metal cabinet shielding. The platform signal is unstable due to the shielding of the metal cabinet, and the signals between sensors are easily interfered with each other due to close-range wireless communication, so the anti-interference demand is strong, and improper networking can easily lead to data transmission failure. At the same time, all field sensors and networking sensors are configured with lithium batteries, small photovoltaic panels and super capacitors to form a composite power system. The photovoltaic panel stores power in the super capacitor during daily work, and the capacitor can meet the continuous power supply for 2-3 days on sunny days. When the capacitor power is lower than 15%, the lithium battery is automatically switched to power supply, taking into account the low power consumption and endurance requirements. For specific methods, refer to Figure 1 , the steps are as follows:

[0075] Taking the SF6 gas circuit breaker monitoring area of a certain 110 kV substation as an example, 82 SF6 remote density sensors, i.e. SF6 remote density meters, are deployed in the area, distributed in 12 circuit breaker intervals, and there are problems of high-voltage bus electromagnetic interference and circuit breaker cabinet shielding signals. After the system is powered on and started, it first establishes a connection with all 82 field sensors started through the LoRa wireless communication module, and synchronously acquires signal strength data and location information data.

[0076] The signal strength data is acquired by collecting the real-time communication signal strength of each field sensor and the LoRa gateway in the area, unit: dBm, excluding 2 sensors with initial connection failure, and subsequent automatic retry connection, and the effective signal strength data is 80 groups,

[0077] The location information data is acquired by the GPS positioning module built-in the sensor, with a positioning accuracy of ≤2 meters, and the latitude and longitude coordinates of the 80 field sensors are acquired, providing accurate spatial basis for subsequent position distribution calculation.

[0078] The signal intermediate value is calculated according to the signal strength data of the 80 field sensors, using the algorithm of screening out one-tenth; that is, the signal strength of the highest 8 and the lowest 8 is removed; the signal strength of the highest 8 may be interfered by the gateway in close range, and the data is not representative, and the signal strength of the lowest 8 may be in a signal blind area, with poor transmission stability, and the average value of the signal strength of the remaining 64 sensors is taken to obtain the signal intermediate value, assuming that the calculation result is -78 dBm, which is taken as the initial signal screening value.

[0079] Position distribution uniform value calculation and judgment: with -78 dBm as the signal screening value, 52 field sensors with signal strength greater than the value are screened out; the regional variance method, i.e. the preset distribution degree algorithm, is used to calculate the position distribution uniform value; the monitoring area is divided into 12 basic areas according to the circuit breaker interval, the number of screened sensors in each area is counted, the variance of the number of sensors in each area and the average number is calculated, the greater the variance, the greater the position distribution uniform value, representing the more uneven distribution. If the calculated position distribution uniform value is 7.3, and the preset distribution requirement value is 4, i.e. the variance ≤ 4, the sensor distribution can meet the networking coverage requirement, then the distribution difference value is 4-7.3=-3.3; according to the distribution difference value adjustment rule, the smaller the distribution difference value, the greater the signal screening value down adjustment, to include more sensors with slightly weak signals but can optimize the distribution, and the signal screening value is adjusted to -83 dBm.

[0080] Screening value cycle adjustment to meet the standard: with -83 dBm as the new signal screening value, 60 field sensors are re-screened; the position distribution uniform value is calculated again to be 3.8, ≤ 4, meeting the distribution requirement, stopping the signal screening value adjustment, and determining the final screened sensors to be 60.

[0081] Combined with the physical layout of the circuit breaker interval of the substation, the signal attenuation caused by cross-interval networking is avoided, the position information of the 60 field sensors is matched with the interval boundary, 8 position intervals are divided, each interval corresponds to 1-2 circuit breaker intervals, the maximum diameter of the interval ≤ 25 meters, ensuring that the sensors in the group are close and reducing the transmission loss; for each position interval, the center point coordinates of the interval are calculated by the geometric center method, 1 field sensor closest to the center point is selected as the networking sensor, a total of 8, and the remaining 52 field sensors in the interval are controlled by wireless instructions to establish a sub-network connection with the corresponding networking sensor, the group communication adopts a short-distance LoRa protocol, the transmission distance ≤ 25 meters, the signal strength is stable, and the mutual interference between sensors can be reduced.

[0082] When the SF6 gas density collection instruction is issued by the power monitoring platform, the 8 networking sensors preferentially start working through capacitor power supply, and perform two-step operations after responding to the instruction:

[0083] Self-collection: the density detection module built-in the networking sensor collects SF6 gas density and temperature data to generate initial sensing data.

[0084] In-group wake-up: the networking sensor sends a start working instruction to all field sensors in the group, the field sensors start working through capacitor power supply to avoid frequent use of lithium batteries, and after the collection is completed, the sensing data is sent to the networking sensor, the in-group transmission time ≤ 0.3 seconds, the signal exposure time is reduced, and the mutual interference is reduced.

[0085] After receiving all the sensor data in the group, the group networking sensor packs its own data with the group data, with each packet data volume ≤1.2KB, which meets the low-power transmission requirements. The data is uploaded to the power monitoring platform through the LoRa gateway, and the success rate of uploading is 99.2%. The data packet that fails to upload is automatically retried once.

[0086] After all the data is uploaded, the total time for single collection and uploading is ≤2.5 seconds. The group networking sensor and the on-site sensor immediately enter the sleep state, with a sleep current ≤8μA, which is much lower than the working current 12mA. Only the power detection module and the timing wake-up module are working. If the capacitor power detection shows less than 15%, it will automatically switch to lithium battery power before sleeping, ensuring that the device starts normally next time.

[0087] Through dynamic networking, data flows between grids, which can effectively deal with electromagnetic interference and terrain shielding problems in power monitoring areas, improve data transmission success rate, reduce packet loss and delay, and ensure real-time and accurate monitoring data. According to signal strength and location distribution, dynamic adjustment and screening conditions can be used to divide intervals and network, breaking the fixed structure limit and adapting to the communication differences of different power scenes. The in-group sensors work synchronously at close range, further improving the correctness of data transmission. After completing the collection and uploading, the sensor sleeps, reducing unnecessary energy consumption.

[0088] Referring to Figure 2 , in order to further optimize the wake-up mechanism of dynamic networking, it is suitable for dense deployment of remote density meters in transformer substations and converter stations. In such scenarios, there can be 10-30 single-area networking sensors. If they are woken up at the same time, signal collision may occur, and centralized power supply may cause energy consumption to increase sharply. Therefore, time-sharing wake-up and orderly networking are needed to solve the interference and power consumption problems. Combined with the power configuration of the sensor photoelectric capacitor priority power supply, the specific method further includes the following steps:

[0089] Based on the 8 location intervals divided in the foregoing, the system reads the unique hardware numbers of the 8 networking sensors through the wireless communication protocol (such as LoRa), such as SN1001-SN1008, to form an initial number list, ensuring that subsequent operations can accurately locate a single device.

[0090] The real-time signal strength of each networking sensor and platform gateway is collected, such as -73 dBm for SN1001, -72 dBm for SN1002, -70 dBm for SN1003, and the like. The networking numbers are sorted in order from strong to weak, and the networking number sequence is obtained: SN1003 (-70 dBm) → SN1002 (-72 dBm) → SN1001 (-73 dBm) → SN1006 (-76 dBm) → SN1005 (-78 dBm) → SN1008 (-80 dBm) → SN1007 (-82 dBm) → SN1004 (-85 dBm). This sorting can preferentially guarantee that sensors with strong signals transmit more stably when waking up, reducing initial communication failures.

[0091] After the system is started or restarted for the first time, all 8 networking sensors are synchronized to wake up in the networking number sequence, and the capacitor is preferentially used for power supply to avoid lithium battery power consumption; after the networking sensors wake up, the field sensors in each interval are sequentially woken up through short-distance communication (such as Bluetooth low power consumption BLE) in the group, such as 8 field sensors in the interval corresponding to SN1003, to realize the full network wake-up of the networking sensor-field sensor, complete the device state detection and communication link test.

[0092] After all sensors complete the state self-checking, such as capacitor power and collection module fault diagnosis, the system first issues a sleep instruction through the networking sensor to let all field sensors enter a sleep state, and the sleep current is ≤8 μA; after confirming that the field sensors are in a sleep state, the 8 networking sensors are controlled to sleep, only the timing wake-up module and the signal detection module are kept working, and the energy consumption is ≤5 μA. This step can avoid device long standby power consumption and adapt to low-power consumption limit requirements.

[0093] The preset wake-up period is 15 minutes, which can be adjusted according to power monitoring requirements, such as the SF6 gas density monitoring period is usually 10-30 minutes. When the period is started, one sensor number is randomly selected as a starting point from the networking number sequence SN1003 → SN1002 → SN1001 → SN1006 → SN1005 → SN1008 → SN1007 → SN1004, and it is assumed that SN1005 is randomly selected.

[0094] Take SN1005 as the starting point, select the numbers in the two directions of the network numbering sequence, forward: SN1005→SN1008→SN1007→SN1004; reverse: SN1005→SN1006→SN1001→SN1002→SN1003; select according to the rule of interval 1 number; forward select SN1005, SN1007, reverse select SN1005, SN1001, SN1003, ensure that all selected numbers are not adjacent to each other, avoid signal interference caused by adjacent sensors waking up at the same time; finally mark the selected numbers: SN1005, SN1007, SN1001, SN1003, and sort according to the original network numbering sequence, get the first number sequence: SN1001→SN1003→SN1005→SN1007; The unselected numbers SN1002, SN1006, SN1008, SN1004, after sorting, get the second number sequence: SN1002→SN1004→SN1006→SN1008.

[0095] The step of "ensuring that the selected numbers in the same wake-up sequence are not adjacent to each other" cooperates with "time-sharing wake-up double sequence" to have essential differences in anti-signal interference, balanced energy consumption, and enhanced anti-fault capability compared with direct sequence selection networking, which is a key link to realize the optimization of dynamic networking wake-up mechanism.

[0096] The applicable scope of "not adjacent to each other" is defined: this constraint is only for "in the same wake-up sequence" (such as the first number sequence or the second number sequence), not for the global number sorting after merging the two sequences. That is, to ensure that the sensor numbers in the same batch wake-up have no adjacent relationship, not the global numbers without adjacent.

[0097] The core difference between direct sequence selection networking and the present application:

[0098] Direct sequence selection networking: it is easy to have sensors with consecutive numbers in the same batch wake-up (physical positions are often adjacent), leading to wireless signal collision interference;

[0099] The present application: first, interval selection to realize "numbers in the same sequence are not adjacent to each other", then, two batches of time-sharing wake-up first and second sequences, from "space (numbers not adjacent) + time (time-sharing)" to avoid adjacent sensors working at the same time, completely solve the interference problem.

[0100] The value of this step lies in that it can avoid the frequency band overlap of adjacent sensors in the same batch, adapt to the dense deployment demand of power scene sensors; dispersed selection of numbers prevents the sudden increase of energy consumption caused by local sensor concentrated wake-up, prolongs the network life; random starting point + interval selection, fault nodes only affect locally, avoid the risk of chain failure of direct sequence selection.

[0101] First, according to the first numbering sequence, the corresponding group network sensor is woken up in turn, each wake-up interval is 2 seconds, to avoid signal collision caused by simultaneous start, after each group network sensor is woken up, it is powered by capacitor preferentially, and the field sensor in its interval is woken up immediately to complete data collection and uploading; after all sensors in the first numbering sequence complete work and hibernate, interval 5 seconds is confirmed to hibernate, then the corresponding group network sensor and field sensor are woken up in turn according to the second numbering sequence, and the same operation is executed. The total time of the whole time-sharing wake-up process is ≤30 seconds, which is much lower than the cycle interval, and no sensor works at the same time, which completely avoids signal interference.

[0102] The technical scheme avoids signal interference between sensors through group time-sharing wake-up, ensures transmission quality; when installed, the sensor automatically acquires fragmentation and signal strength information, automatically divides the communication group and negotiates the wake-up time, realizes the automation and intelligence of wake-up; with the help of reasonable wake-up time and numbering selection strategy, the sensor energy consumption is balanced, and the overall service life of the network is prolonged; by randomly and interval selecting sensor number, the system fault resistance is enhanced.

[0103] Reference Figure 3 In order to realize group network optimization on the basis of time-sharing wake-up, it is suitable for substation multi-interval equipment monitoring, new energy power station dispersed density monitoring and other scenes; in such scenes, the data amount collected by sensors in different time periods is quite different, for example, the data amount increases by 30% during equipment maintenance period, and the transmission conflict caused by cross-interval clock deviation needs to be solved through temporary central node and dynamic adjustment. Combined with the power configuration of the sensor photoelectric capacitor preferential power supply, the specific method includes the following steps:

[0104] Taking the remote density monitoring network of three main transformer intervals of a certain 220kV substation as an example, five location intervals are divided, corresponding to five group network sensors, numbered S1-S5, in a certain monitoring period, the sensor number selected for the first time according to the interval selection rule is S2, at this time S2 is defined as a temporary central node. This node preferentially enables capacitor power supply to ensure power supply stability and avoid data interruption caused by lithium battery switching, and simultaneously starts the data receiving module and clock synchronization module.

[0105] The temporary central node S2 first sends a data request instruction to other selected group network sensors (assuming S1, S4) in the same group through the LoRa protocol; S1 and S4 send the packaged data (S1 data packet size is 1.2KB, S4 data packet size is 1.5KB) of SF6 density value + temperature and pressure collected by themselves to S2 after receiving the instruction; after receiving and verifying the data integrity (excluding wrong packets through CRC verification), S2 merges the packaged data of S1 and S4 with 1.3KB data collected by itself to form a total size of 4KB summary data packet, which is forwarded to the power monitoring platform through the 4G module before uploading, and the clock is synchronized with the platform before uploading to ensure the accuracy of data timestamp.

[0106] After the current networking sensor (such as S1) completes the 1.2 KB packaged data transmission, the temporary center node S2 immediately counts the data volume of the data packet, and calculates the wake-up time interval of the next networking sensor (S4) according to the positive correlation adjustment rule of the data volume; the preset basic interval is 5 seconds, and the interval increases by 1 second for each 0.5 KB increase in data volume; the data volume of S1 is 1.2 KB, which is 0.7 KB more than the basic data volume (0.5 KB), so the wake-up interval of S4 is adjusted to 5+1=6 seconds, rounded up.

[0107] If the data volume collected by a networking sensor is large, such as the data volume increases to 2.5 KB when the equipment is abnormal, extending the wake-up interval can avoid the congestion of subsequent node data transmission and the current node; if the data volume is small, such as 0.8 KB under normal working conditions, then shortening the interval ensures the transmission efficiency. This dynamic adjustment mechanism controls the total transmission time of 5 networking sensors within 30 seconds in one monitoring period, avoids the overall transmission delay caused by the excessive data volume of a node, balances the network load, and reduces unnecessary continuous power consumption of the capacitor.

[0108] When the packaged data of the 5 networking sensors (S1-S5) are all transmitted, the temporary center node S2 counts the data packet size of each node: S1 is 1.2 KB, S2 is 1.3 KB, S3 is 1.1 KB, S4 is 1.5 KB, and S5 is 1.0 KB. Select S4 with the largest data volume as the temporary center node of the next monitoring period, and cancel the temporary center node mark of S2, restoring it to a normal networking sensor.

[0109] Networking sensors with large data packet volumes (such as S4) usually mean that they cover more field sensors or collect more comprehensive parameters, indirectly reflecting their stronger data processing and storage capabilities, such as larger built-in storage chip capacity and faster CPU processing speed. Selecting such nodes as temporary center nodes can more efficiently complete data aggregation, verification, and forwarding, avoiding data backlog caused by insufficient processing capacity of the center node. At the same time, the rotation mechanism avoids a single node bearing the center task for a long time; if S2 is the center node for a long time, the daily energy consumption will be higher than that of a normal node, and after rotation, the consumption of the capacitor and lithium battery of each node can be balanced, prolonging the overall network life.

[0110] Collecting data and updating the timing through the temporary center node eliminates the wake-up error to achieve time synchronization within and between groups, and it prioritizes synchronization with the platform clock to ensure accurate clock interval between groups to reduce interference; dynamically adjusting the wake-up interval of the next networking sensor according to the current data volume optimizes the transmission rhythm and balances the load; at the same time, dynamically selecting the networking sensor with the largest data volume as the next temporary center node allows the node with stronger processing capacity to bear the center task, enhancing the adaptability of the system to different scenarios.

[0111] In order to achieve optimization on the basis of time-sharing wake-up and temporary central node, it is suitable for substation SF6 gas leakage monitoring, transformer oil density early warning and other scenes with high requirements for data accuracy; In such scenarios, data errors may cause equipment failure misjudgment, and single node failure may cause key data loss, which requires the cooperation of temporary central node and verification central node to build a data security and transmission protection system. Combined with the power configuration of sensor photoelectric capacitor priority power supply, the specific method includes the following steps:

[0112] Taking the SF6 circuit breaker monitoring network of a 500kV substation as an example, the network is divided into 6 location intervals, corresponding to 6 networking sensors, numbered Z1-Z6, according to the rules in the previous section: the first number sequence: Z2→Z4→Z6, the second number sequence: Z1→Z3→Z5. When the monitoring cycle starts, both sequences start operating at the same time to avoid data collection gaps caused by single sequence delay, and both sequences preferentially enable capacitor power supply to ensure initial operation stability.

[0113] The first selected sensor number in the first number sequence is Z2, which is defined as the temporary central node, mainly responsible for data aggregation and upload core tasks; The first selected sensor number in the second number sequence is Z1, which is defined as the verification central node, focusing on data consistency verification. After the start of the double nodes, the initial handshake is completed through short distance communication, and the current cycle collection parameters such as data sampling frequency and verification rules are synchronized.

[0114] The temporary central node Z2 needs to obtain the packaged data of other selected sensors (Z4, Z6) in the first number sequence, and the verification central node Z1 needs to obtain the packaged data of other selected sensors (Z3, Z5) in the second number sequence, and both types of packaged data do not contain the sensor data of the double central nodes themselves, to avoid interference with the verification results. For example, the packaged data of Z4 is the SF6 density data of 8 field sensors in its coverage interval (size 1.8KB), and the packaged data of Z6 is the data of 6 field sensors (size 1.5KB). Z2 receives these two types of data through LoRa protocol; Similarly, Z1 receives the packaged data of Z3 (1.6KB) and Z5 (1.4KB).

[0115] After receiving the packaged data of Z4 and Z6, the temporary central node Z2 immediately sends the two pieces of data to the verification central node Z1 through the encrypted communication link to avoid interference and tampering during data transmission. After receiving the data from Z2, the verification central node Z1 starts double comparison verification. First, it compares whether the Z3 and Z5 data obtained by itself are consistent with the preset format, which is the basic format verification. Then, it focuses on comparing whether the Z4 and Z6 data from Z2 are logically consistent with itself, such as whether the density data deviation between the same device intervals exceeds 0.001 MPa. If all data meet the consistency requirements, i.e., the verification pass rate is 100%, it is determined that the consistency verification is passed.

[0116] After the consistency verification is passed, the verification central node Z1 packages the sensor data it collects (7 pieces of on-site sensor data in the coverage interval, 1.7 KB in size) and sends it to the temporary central node Z2. At this time, Z2 has gathered the data of Z4 and Z6 (3.3 KB in total) and Z1 (1.7 KB), plus its own 1.6 KB of collected data, forming a complete data package with a total size of 6.6 KB.

[0117] The temporary central node Z2 performs the final verification on the complete data package, such as CRC (Cyclic Redundancy Check), and uploads it to the power monitoring platform through the 4G / 5G module after confirming that there is no error. After the upload is completed, Z2 sends an upload success instruction to Z1, and the two nodes synchronize the control of the sensors in their respective sequences to enter a sleep state with a sleep current ≤8 μA. If the capacitor power is less than 15%, it switches to lithium battery power supply to ensure normal startup in the next cycle. The entire process avoids the link redundancy caused by separate uploads of the two nodes, reducing invalid energy consumption.

[0118] This technical solution cross-verifies data through temporary and verification dual central nodes, forming a double verification barrier to significantly improve data verification accuracy. Dual numbering sequences are operated synchronously, and data is obtained by two nodes respectively, constructing a double backup transmission path to avoid loss of critical data due to single point failure. The verification central node only transmits its own data to the temporary central node for unified upload, simplifying the process, reducing redundant links, and improving system operation efficiency. The cooperation of the two nodes enables the system to have fault self-checking and data self-repairing capabilities, even if a single node is abnormal or offline, the other node can still ensure operation.

[0119] To further optimize the dual central node verification mechanism, it is suitable for scenarios such as the periphery of high-voltage equipment in substations and along power transmission and transformation lines where signal fluctuations are frequent. In such scenarios, the signal strength of the temporary central node and the verification central node may become unbalanced due to equipment start-stop and weather changes (such as signal attenuation caused by thunderstorms). If the strong signal node only bears the verification task and the weak signal node bears the core transmission, it is easy to cause data packet loss, which requires "signal strength monitoring + instruction exchange" to ensure the adaptation of core node capabilities. Combined with the power configuration of the sensor "photoelectric capacitor priority power supply", the specific method includes the following steps:

[0120] Take the SF6 density monitoring network of a certain 220 kV substation as an example, the system continuously monitors the real-time signal strength (unit: dBm) of the temporary central node (such as Z2 in the foregoing text) and the verification central node (such as Z1 in the foregoing text), sets the “last 5 monitoring periods” as the recording range, each period is 15 minutes, a total of 75 minutes, to ensure that the data can reflect the recent signal stability of the node, and avoid the influence of single accidental fluctuation on the judgment. For example, the signal strength of the temporary central node Z2 in the last 5 periods is -72 dBm, -74 dBm, -73 dBm, -76 dBm, and -75 dBm, respectively; and the signal strength of the verification central node Z1 is -68 dBm, -69 dBm, -70 dBm, -67 dBm, and -68 dBm, respectively.

[0121] During the recording process, abnormal data is automatically excluded, such as -90 dBm extreme value caused by instantaneous electromagnetic interference. If the signal strength data is missing in a certain period, such as temporary hibernation of the node, the average value of the previous 3 periods is used to complete it, to ensure the integrity of the recent multiple data. At the same time, the signal strength collection is preferentially carried out when the node is awake and the capacitor power supply is stable, to avoid the influence of lithium battery power supply fluctuation on the signal detection module, and to ensure the accuracy of the data.

[0122] The weighted average method is used to calculate the temporary comprehensive strength and the verification comprehensive strength. Since the recent data can better reflect the current signal state of the node, different weights are given to the signal strength of the last 5 periods, the first period weight is 0.1, the second period weight is 0.2, the third period weight is 0.2, the fourth period weight is 0.25, and the fifth period weight is 0.25. Taking the temporary central node Z2 as an example, the temporary comprehensive strength = (-72 x 0.1) + (-74 x 0.2) + (-73 x 0.2) + (-76 x 0.25) + (-75 x 0.25) = -74.35 dBm; and the verification comprehensive strength of the verification central node Z1 = (-68 x 0.1) + (-69 x 0.2) + (-70 x 0.2) + (-67 x 0.25) + (-68 x 0.25) = -68.35 dBm, note: the closer the signal strength value is to 0, the stronger the signal is.

[0123] The ratio of the temporary comprehensive strength to the verification comprehensive strength is calculated, since the signal strength is negative, the absolute value is taken during calculation, that is, |-74.35| / |-68.35|≈1.088. The preset required strength ratio is 1.0, that is, the absolute value of the temporary comprehensive strength ≤ the absolute value of the verification comprehensive strength, to ensure that the signal of the temporary central node is not weaker than that of the verification central node. If the current ratio 1.088>1.0, it means that the signal of the temporary central node is weaker than that of the verification central node, which does not meet the preset requirement, and the instruction interchanging mechanism is triggered; if the ratio ≤1.0, the original operation instruction of the two nodes remains unchanged.

[0124] When the instruction exchange is triggered, the system informs the two nodes through encrypted wireless instruction synchronization; the operation instruction of the original temporary center node Z2 is switched to the verification center node instruction, and the subsequent task is only to bear data verification and its own data transmission; the operation instruction of the original verification center node Z1 is switched to the temporary center node instruction, and the subsequent task is to bear data aggregation and unified upload. At the same time, the system updates the working parameters of the two nodes, such as the temporary center node needs to start the data aggregation module, and the verification center node needs to close the upload module, to ensure that the function is adapted after the instruction switching.

[0125] After the instruction exchange, the new temporary center node Z1 (the original verification node) starts the core module by priority through the capacitor power supply, synchronizes the clock with the platform first, and then reacquires the data of other nodes in the group to avoid data interruption in the switching process; the new verification center node Z2 (the original temporary node) focuses on verifying data, and if it finds that the data transmitted by the new temporary node has deviation, it will immediately start the secondary verification. The whole switching process takes ≤3 seconds, and the data transmission success rate after switching is improved to ensure the stable operation of the network.

[0126] By recording and calculating the signal comprehensive strength of the recent multiple temporary and verification center nodes, the node strength ratio is obtained. When the ratio does not meet the preset requirement, the operation instructions of the two types of nodes are exchanged, and the node with stronger signal is assigned to bear the core data processing and transmission task, effectively avoiding the data receiving deviation caused by the imbalance of node signal strength, and further improving the data verification accuracy.

[0127] In order to further optimize the dual-center node cooperation mechanism, it is suitable for the communication intensity fluctuation and energy consumption sensitive scenarios such as dense deployment of multiple intervals in substations and outdoor dispersed monitoring of new energy power stations. In such scenarios, part of the networking sensors may be affected by high-voltage electromagnetic interference, resulting in communication attenuation, or due to long-term rainy days, the capacitor power is insufficient. If the center node is randomly selected, it may cause data transmission failure, and the node needs to be selected by communication strength and power to ensure that the core task is borne by the strong performance equipment. Combined with the power configuration of the sensor photoelectric capacitor priority power supply, the specific method includes the following steps:

[0128] Taking the transformer oil density monitoring network of a certain 220 kV substation as an example, the network is divided into a first number sequence (containing 5 networking sensors, numbered A1-A5) and a second number sequence (containing 4 networking sensors, numbered B1-B4) according to the region. When the monitoring cycle starts, both sequences enter the working state at the same time to avoid long data acquisition interval caused by single sequence delay; all sensors preferentially use super capacitor power supply, and only switch to lithium battery when the capacitor power is lower than 15% to ensure the stability of the initial power supply.

[0129] The system performs real-time performance detection on all networking sensors in the two sequences:

[0130] Measure the real-time signal strength of each node and the area gateway through the LoRa protocol, unit: dBm, for example, A1 is -70 dBm, A2 is -75 dBm, A3 is -68 dBm, the closer the value is to 0, the stronger the signal; B1 is -72 dBm, B2 is -69 dBm, B3 is -76 dBm.

[0131] Collect the remaining capacity of each node (percentage) through the built-in power meter, for example, A1 is 85%, A2 is 40%, A3 is 90%; B1 is 30%, B2 is 88%, B3 is 50%. The detection process takes ≤1 second to avoid excessive power consumption.

[0132] In the first numbered sequence, compare the communication strength and power of each node; A3 has the strongest communication strength (-68 dBm) and sufficient power (90%), which is better than A1 (slightly weaker signal) and A2 (insufficient power), so it is defined as the temporary center node, which undertakes the task of data aggregation and final upload; The nodes covered by A1, A2, etc. will be ordinary networking sensors, only need to send packaged data to A3.

[0133] In the second numbered sequence, B2 has the strongest communication strength (-69 dBm) and the highest power (88%), which is better than B1 (low power) and B3 (weak signal), so it is defined as the verification center node, which focuses on data consistency verification; B1, B3, etc. send packaged data to B2, and B2 completes cross-checking with the temporary center node.

[0134] Note: If there is a conflict between the node with the strongest communication strength and the node with the highest power in the sequence, the node with stronger communication strength is preferred; weak signal will directly lead to data transmission failure, while insufficient power can be temporarily compensated by switching lithium batteries. This rule is suitable for the priority demand of real-time data in the power scene.

[0135] The temporary center node A3 obtains the packaged data of A1 and A2 in the first sequence, A1 data contains 6 on-site sensor oil density values, size 1.2KB; A2 data contains 5 on-site sensor data, size 1.0KB; The verification center node B2 obtains the packaged data of B1 and B3 in the second sequence, B1 data size 0.9KB, B3 data size 1.1KB, both types of data do not contain the sensor data of the double center node itself, to avoid verification interference.

[0136] A3 encrypts and sends the packaged data of A1 and A2 to B2, and B2 initiates double verification; first, the format integrity of the two data is compared, such as whether it contains a timestamp and a check code, and then the data logic is verified by threshold comparison method, such as whether the oil density value is within the normal range of 0.89-0.92 g / cm³, and after confirming the consistency, B2 sends the 1.3 KB data collected by itself to A3; A3 collects all the data, the total size is 1.2+1.0+0.9+1.1+1.3=5.5 KB, which is uploaded to the platform through the 4G module, and then the double-node controls the respective sequence sensor to sleep, and the sleep current is ≤8 μA.

[0137] In the first and second numbered sequences, the networking sensor with the maximum communication strength or the most power is selected as the temporary and verification center node, ensuring that the double center node has better signal transmission capability or longer endurance, laying a reliable hardware foundation for data reception and verification; the double center nodes synchronously acquire data and cross-verify, which not only retains the double verification to ensure the accuracy of the data, but also reduces the transmission errors caused by weak signals or insufficient power by taking advantage of the node performance, further improving the reliability of data verification and the stability of system operation.

[0138] In order to further optimize the spatial scheme of dynamic networking layout, it is suitable for complex terrain and wide monitoring range scenarios such as mountainous transformer substations and hilly power transmission lines; in such scenarios, the actual coverage area may far exceed the preset value due to the terrain undulations, and if the interval is too large, signal transmission blind areas such as valley shadow areas may occur, which requires precise mapping through GIS maps, dynamic interval segmentation, and communication strength adaptation to optimize the networking layout. Combined with the power configuration of the sensor photoelectric capacitor priority power supply, the specific method includes the following steps:

[0139] Taking the SF6 switch monitoring network of a certain mountainous 35 kV transformer substation as an example, the system accesses the GIS map of the transformer substation with an accuracy of 1 meter, which contains information such as terrain elevation, equipment distribution, and vegetation shielding, and analyzes the actual range of the five location intervals divided in the previous section through map coordinates; for example, interval C covers three switch intervals on the east side of the transformer substation, and GIS calculates its actual area as 1200 square meters, while the preset required area is 800 square meters. According to the maximum effective transmission distance of 50 meters within the group, 800 square meters is the critical value for stable signal coverage.

[0140] By comparing the actual area and the preset required area of the interval, interval C (1200㎡) > 800㎡, it is determined that it is an over-limit interval, and the segmentation process needs to be started; the areas of the remaining four intervals are within the range of 600-750㎡, and no adjustment is needed. GIS map synchronously marks the terrain features within interval C: including two slopes with a height difference of more than 8 meters (which are easy to cause signal shielding) and one dense shrub (which causes serious signal attenuation), providing terrain reference for subsequent segmentation.

[0141] For the 1200㎡ area of interval C, the average segmentation along the contour line is adopted to avoid the terrain crossing caused by traditional straight-line segmentation, and it is divided into two new intervals C1 (uphill area) and C2 (downhill area) with an area of 600㎡ each. The segmentation avoids the dense area of high-voltage equipment such as switch cabinet cluster, ensures that there is no metal obstruction between the field sensors and networking sensors in the new interval, and reduces signal transmission loss.

[0142] In the new interval C1, the field sensor with the closest distance to the geometric center and the highest communication strength (number C1-05, signal strength -70dBm, capacitor capacity 92%) is selected as the new networking sensor; in C2, the field sensor with number C2-03 (signal strength -72dBm, capacitor capacity 88%) is selected as the new networking sensor, both of which meet the basic requirement of communication strength > -75dBm to ensure the stability of the transmission within the group.

[0143] The real-time communication strength of the new networking sensors C1-05 and C2-03 is collected and converted to a dimensionless value of 0-1 through a normalization formula; the unit value = (measured strength - minimum strength threshold) / (maximum strength threshold - minimum strength threshold). For example, set the effective range of communication strength as -85dBm (minimum threshold) to -65dBm (maximum threshold), then the unit value of C1-05 = (-70 - (-85)) / (-65 - (-85)) = 15 / 20 = 0.75; the unit value of C2-03 = (-72 + 85) / 20 = 13 / 20 = 0.65.

[0144] According to the unit value, the interval range is dynamically adjusted; the preset basic area is 600㎡, and the adjustment coefficient is 200㎡ / unit value, i.e. the actual area = basic area + (unit value x adjustment coefficient). Therefore, the interval area of C1-05 = 600 + 0.75 x 200 = 750㎡, the signal is strong and the coverage is appropriately expanded; the interval area of C2-03 = 600 + 0.65 x 200 = 730㎡, the signal is slightly weak and the coverage is slightly smaller. After adjustment, both intervals avoid the terrain obstruction area, and the maximum transmission distance of the sensors within the group is controlled within 40 meters, the signal attenuation is ≤5dB, ensuring stable data transmission.

[0145] By accurately obtaining the position interval area, the interval is evenly divided and a new networking sensor is added for the area exceeding the preset area, and the layout is optimized to avoid large interval transmission blind area; according to the positive correlation adjustment of the communication strength of the new networking sensor, the interval range is adjusted to realize the adaptation of strong and weak signals to the interval size, ensuring stable data transmission and reducing deployment cost; at the same time, with the help of the terrain features of GIS map, the layout is adjusted flexibly combined with communication strength, improving the adaptability to complex terrain.

[0146] This step is the fault-tolerant optimization of the dynamic networking system, suitable for substation core equipment monitoring, long-distance power transmission line monitoring, and other scenarios with extremely high requirements for communication continuity. In such scenarios, if the temporary central node is interrupted in communication with the platform due to high-voltage electromagnetic shock, hardware failure, or power depletion, it may cause the entire data transmission to stall, and the monitoring link needs to be ensured uninterrupted through the rapid disablement of the faulty node, the switching of nearby nodes, and the reuse of data migration. Combined with the power configuration of the sensor photoelectric capacitor priority power supply, the specific method includes the following steps:

[0147] Taking the GIS equipment density monitoring network of a certain 500kV substation as an example, the temporary central node T5 covers 3 intervals and 12 field sensors. During the data upload phase, the system determines that T5 cannot complete communication with the platform, triggers the fault handling process, and the timeout time is set to 5 seconds per attempt. The heartbeat signal with the verification central node V3 is interrupted, with a heartbeat interval of 10 seconds. At this time, the capacitor power of T5 is displayed as 12%, which has been switched to lithium battery power supply, excluding temporary communication interruption caused by insufficient power, and the lithium battery can support at least 2 uploads.

[0148] The system sends an encrypted disablement instruction to T5, closes its data aggregation, upload control, and other core modules, retains basic communication and storage functions, and avoids the interference of the faulty node sending invalid signals to the network. At the same time, T5 is marked as a node to be restored in the network topology diagram, and its participation in subsequent networking decisions is suspended to ensure that the new node is not disturbed.

[0149] The system retrieves the location information (latitude and longitude coordinates) of T5 and the location data of the surrounding networking sensors, calculates the nearest networking sensor to T5; node T6 is only 18 meters away from T5, located in the adjacent interval, with a communication strength of -71 dBm, a capacitor power of 85%, and normal communication with the platform, and the latest upload time is only 2 minutes away from the current time. It is selected as the new temporary central node. The selection process excludes nodes with signal strength < -80 dBm or power < 30% to ensure that the new node has stable working capacity.

[0150] The system sends a start temporary central node instruction to T6, which completes module switching within 2 seconds, opens the data aggregation interface, and synchronizes the platform clock. At the same time, the disabled node T5 transmits the 15 pieces of unuploaded packaged data (total size 2.8KB) stored to T6 through short-distance LoRa communication with the maximum transmission power to ensure stable transmission within 18 meters. After receiving the data, T6 confirms the integrity through CRC verification (verification pass rate 100%) and supplements 3 pieces of data collected by itself to form a complete data packet.

[0151] After the data migration is completed, the system sends a role restoration instruction to T5, restores it from a temporary central node to a normal on-site sensor, only retains the data collection and data sending function to the networking sensor, and is included in the coverage interval of T6. After restoration, the power consumption of T5 is reduced from 15 mA in central node mode to 8 mA in on-site sensor mode, and the capacitor is preferentially used for power supply. If the photoelectric charging is more than 20% later, the lithium battery consumption is reduced.

[0152] The system synchronizes the information of the new temporary central node T6 to all networking sensors, updates the interval coverage relationship, and changes the 12 on-site sensors originally covered by T5 to be managed by T6. The new topology structure is marked in the GIS map for real-time viewing by the operation and maintenance platform. The entire switching process takes ≤15 seconds, which is much lower than the monitoring period of 15 minutes, ensuring continuous data transmission link.

[0153] When the temporary central node cannot communicate with the platform, the disabled fault node is bypassed and the nearest networking sensor is started as a new temporary central node, realizing seamless node switching to ensure continuous communication link and avoid monitoring data disconnection; the packaged data of the disabled node is transmitted to the new node to realize seamless data migration, ensuring data safety without loss; at the same time, the disabled node is restored to an on-site sensor to reuse resources, and the nearby switching shortens the response time, effectively improving the fault tolerance efficiency and resource utilization of the system.

[0154] In order to optimize the selection strategy of temporary central nodes, it is suitable for different scenarios such as urban substations (with large electromagnetic interference fluctuations) and remote areas (with prominent endurance requirements). In such scenarios, the electromagnetic interference level may change dramatically with the running state of the equipment (such as transformer start-stop), and a single communication priority or endurance priority selection logic cannot adapt to complex environments. It is necessary to realize accurate matching of central nodes and scenes through historical data modeling and dynamic weight adjustment. Combined with the power configuration of sensor photoelectric capacitor priority power supply, the specific method includes the following steps:

[0155] Taking the remote density monitoring network of a certain 110kV urban substation as an example, the first numbered sequence contains 6 networking sensors, numbered D1-D6. The system extracts two types of core data of each node in the last 30 monitoring periods (every period is 15 minutes, a total of 7.5 hours):

[0156] The communication strength (dBm) is collected 3 times per period, and the average value is taken as the period data. For example, the 30 period data of D1 ranges from -68 to -75 dBm, and the data of D2 ranges from -70 to -78 dBm.

[0157] The remaining capacity of the capacitor is collected every period (%), for example, the capacity of D1 ranges from 72% to 95% (sunny photoelectric charging is sufficient), and the capacity of D3 ranges from 45% to 60% (long-term rainy days).

[0158] Remove the extreme value of -90dBm caused by transient electromagnetic shock, accounting for less than 2%, and smooth the jump value caused by switching lithium batteries in the power data (such as from 18% to 90%), to ensure that the historical data reflects the true performance of the node.

[0159] Convert the two types of data to 0-1 dimensionless values using min-max standardization to eliminate dimensional differences:

[0160] Communication intensity normalized value = (measured value - minimum intensity) / (maximum intensity - minimum intensity), for example, the average communication intensity of D1 is -71dBm, and the calculation is 0.8, the higher the value, the stronger the signal;

[0161] Remaining power normalized value = (measured value - minimum power) / (maximum power - minimum power), D1's average power is 82%, and the calculation is 0.75, the higher the value, the better the endurance.

[0162] Communication performance value = Σ (each period normalized communication intensity x time decay coefficient), recent data has higher weight, and far-term data has decreasing weight; such as the last 5 periods weight 0.8, and the first 5 periods weight 0.2; D1's final communication performance value is 0.78;

[0163] Endurance performance value = Σ (each period normalized power x power supply mode coefficient), the coefficient is 1.0 when powered by a capacitor, and the coefficient is 0.8 when powered by a lithium battery, because the priority of lithium battery endurance is low, D1's final endurance performance value is 0.72.

[0164] Comprehensive performance score: score = communication performance value x weight coefficient + endurance performance value x (1 - weight coefficient), the weight coefficient is dynamically adjusted according to the electromagnetic interference level.

[0165] Electromagnetic interference level monitoring: real-time monitoring of regional electromagnetic interference level through the built-in interference detector in the gateway, unit: V / m, set the preset threshold to 5V / m, corresponding to the interference level when the substation equipment is running normally:

[0166] When the breaker closing causes the interference level to rise to 7V / m (> 5V / m), the communication performance value weight coefficient is adjusted to 0.7, and the endurance performance value weight coefficient is 0.3 (priority is given to signal transmission);

[0167] When the equipment is running stably, the interference level decreases to 3V / m (≤ 5V / m), the weight coefficient is reversed, communication 0.3, endurance 0.7, focusing on energy saving.

[0168] Score results and node selection:

[0169] Strong interference scenario (7V / m): D1 comprehensive score = 0.78*0.7+0.72*0.3=0.762, D4 due to communication performance value 0.82 (better than D1), score 0.794, selected as temporary center node;

[0170] Low interference scenario (3V / m): D1 comprehensive score = 0.78*0.3+0.72*0.7=0.738, D5 due to endurance performance value 0.85 (sufficient capacitor power), score 0.801, selected as temporary center node.

[0171] The comprehensive performance score is obtained by normalizing and weighting the historical communication intensity and residual power data, the dimension difference of the index is eliminated, the selection of the temporary center node is more scientific and objective; meanwhile, the weight coefficients of the communication and endurance performance values are dynamically adjusted according to the electromagnetic interference level of the monitoring area; when the interference is strong, the signal transmission capability of the center node is preferentially guaranteed to reduce packet loss, when the interference is low, the endurance of the node is focused on to prolong the network time, the precise adaptation of the selection strategy to the actual environment is realized, the stable data transmission is guaranteed, and the system energy consumption is optimized.

[0172] The embodiment of the application further discloses a communication system of SF6 remote density gauge dynamic networking, comprising a processor, wherein the processor executes the steps of the communication method of the SF6 remote density gauge dynamic networking according to any one of the above.

[0173] Although the embodiments of the application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the application.

Claims

1. A communication method for SF6 remote density meter dynamic networking, characterized in that, The method comprises the following steps: wireless connection with multiple activated field sensors, obtaining signal strength of all field sensors and position information of field sensors; signal intermediate value is calculated from multiple signal strengths using a preset screening algorithm; signal screening value is taken as the signal intermediate value, position information of field sensors with signal strength greater than the signal screening value is screened according to the signal screening value, and position distribution uniform value is calculated from the screened position information using a preset distribution degree algorithm; if the position distribution uniform value is greater than a preset distribution requirement value, a distribution difference value is calculated as a difference between the distribution requirement value and the position distribution uniform value, the signal screening value is adjusted according to the distribution difference value, the smaller the distribution difference value is, the greater the signal screening value is adjusted down, and the signal screening value is adjusted in a loop until the position distribution uniform value is less than or equal to the distribution requirement value; multiple position intervals are divided according to the screened position information after the adjustment, field sensors closest to the center position in the position intervals are taken as networking sensors, and other field sensors in the position intervals are controlled to network with the networking sensors; the networking sensors obtain collection instructions, generate sensing data and start work instructions in response to the collection instructions, and send the start work instructions to the field sensors, the field sensors generate sensing data and send the sensing data to the networking sensors in response to the start work instructions, the networking sensors package the received sensing data and the sensing data of the networking sensors and then upload the packaged data, and then the networking sensors and the field sensors enter a sleep state; sensor numbers of all networking sensors are obtained; the obtained sensor numbers are sorted according to signal strength to obtain a networking number sequence; all networking sensors are woken up for the first time after the networking number sorting, and all other field sensors are woken up through the networking sensors; all other field sensors are put to sleep through the networking sensors, and all networking sensors are also put to sleep; a sensor number is randomly selected from the networking number sequence according to a preset setting period, and multiple sensor numbers are selected in both directions of the networking number sequence at intervals with the selected sensor number as the center; the selected sensor numbers are marked and sorted to obtain a first number sequence, and the unselected sensor numbers are marked and sorted to obtain a second number sequence; the corresponding networking sensors are woken up in sequence according to the first number sequence, and the corresponding field sensors are woken up through the networking sensors; then the corresponding networking sensors are woken up in sequence according to the second number sequence, and the corresponding field sensors are woken up through the networking sensors; the first number sequence and the second number sequence are operated at the same time; the networking sensor corresponding to the first selected sensor number in the first number sequence is defined as a temporary center node; the networking sensor corresponding to the first selected sensor number in the second number sequence is defined as a verification center node; the temporary center node and the verification center node both obtain packaged data of other selected networking sensors, the packaged data does not contain sensing data of the temporary center node and the verification center node, and the temporary center node sends the obtained packaged data to the verification center node; The verification center node verifies the consistency of the packaged data obtained by itself and the packaged data from the temporary center node; If the consistency verification is passed, the verification center node sends its sensing data to the temporary center node in a packaged manner; and the temporary center node uploads all the obtained packaged data.

2. The communication method for SF6 remote transmission density gauge dynamic networking according to claim 1, characterized in that, The method further comprises the following steps: The first selected sensor number corresponds to a networking sensor defined as a temporary center node, which obtains the packaged data of other selected networking sensors and forwards the obtained packaged data for uploading; After the transmission of the packaged data of the current networking sensor is completed, the next networking sensor is woken up; and the wake-up time interval of the next networking sensor is adjusted according to the positive correlation of the data amount of the currently obtained packaged data; After the transmission of the packaged data of all networking sensors is completed, the networking sensor with the largest data amount of packaged data is selected as the next temporary center node.

3. The communication method of claim 1, wherein the SF6 remote density meter dynamic networking is characterized by, The method further comprises the following steps: The signal strength of the last plurality of temporary center nodes is recorded as a temporary strength, and the signal strength of the last plurality of verification center nodes is recorded as a verification strength; The temporary comprehensive strength is calculated according to the plurality of temporary strengths, and the verification comprehensive strength is calculated according to the plurality of verification strengths; The node strength ratio is calculated as the ratio of the temporary comprehensive strength to the verification comprehensive strength; If the node strength ratio is less than a preset required strength ratio, the operation instructions of the temporary center node and the verification center node are interchanged.

4. The communication method for SF6 remote transmission density meter dynamic networking according to claim 1, characterized in that, The method further comprises the following steps: The interval range area of the location interval is obtained based on a GIS map; If the interval range area is greater than a preset required area, the interval range area exceeding the required area is evenly divided to obtain a plurality of new location intervals, and new networking sensors are selected in the new location intervals; The communication strength of the new networking sensor is obtained, and the interval range area of the location interval is adjusted according to the positive correlation of the communication strength.

5. The communication method of claim 1 or 2, wherein, The method further comprises the following steps: If the temporary center node cannot complete communication with the platform end, the networking sensor closest to the disabled temporary center node is started as a new temporary center node; The packaged data in the disabled temporary center node is sent to the new temporary center node, and the label of the disabled temporary center node is restored to an on-site sensor.

6. The communication method for SF6 remote transmission density meter dynamic networking according to claim 1, characterized in that, The method further comprises the following steps: The historical communication strength data and the historical remaining power data of all selected networking sensors in the first number sequence are obtained; The historical communication strength data is normalized and weighted to obtain a communication performance value, and the historical remaining power data is normalized and weighted to obtain a endurance performance value; The communication performance value and the endurance performance value are summed to obtain a comprehensive performance score; The networking sensor with the highest comprehensive performance score is selected as the temporary center node of the first number sequence; wherein the weight coefficient of the communication performance value and the weight coefficient of the endurance performance value are dynamically adjusted according to the electromagnetic interference level of the current monitoring area; when the electromagnetic interference level is higher than a preset threshold, the weight coefficient of the communication performance value is higher than the weight coefficient of the endurance performance value; when the electromagnetic interference level is lower than or equal to the preset threshold, the weight coefficient of the endurance performance value is higher than the weight coefficient of the communication performance value.

7. A communication system for SF6 remote density meter dynamic networking, characterized in that, The SF6 remote density meter dynamic networking communication method comprises a processor, and the processor executes the steps of the SF6 remote density meter dynamic networking communication method as claimed in any one of claims 1-6.

Images