SF6 online monitoring-oriented wireless Mesh network dynamic routing optimization networking method

Abstract

The invention relates to the technical field of industrial wireless communication, and discloses an SF6 online monitoring-oriented wireless Mesh network dynamic routing optimization networking method, which comprises the following steps of: firstly, acquiring component connection position coordinates of gas insulated metal-enclosed switchgear, flange bolt spacing and other construction parameters, and electromagnetic interference spectrum data generated by switching operation; determining axial resonance distribution according to a component connection position, determining circumferential slot radiation capability according to a flange bolt distance, and generating structural leakage source field intensity in combination with an interference spectrum; then, node disturbed field intensity is calculated based on a spatial attenuation rule, port common-mode coupling response is calculated in combination with node cable size, and node transient electromagnetic vulnerability is generated; and finally, calculating a structural risk weight according to the vulnerability degrees of the nodes at the two ends of the link, and superposing the structural risk weight to standard link measurement to generate composite routing cost and complete networking. According to the invention, electromagnetic leakage hot spots caused by equipment construction can be effectively avoided, and the communication reliability under strong interference is ensured.

Description

Technical Field

[0001] This invention relates to the field of industrial wireless communication technology, and more specifically, to a dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring. Background Technology

[0002] In smart grid construction, gas-insulated metal-enclosed switchgear (GIS) is a core piece of equipment in ultra-high voltage substations. To monitor the insulation status inside the GIS in real time, a large number of sulfur hexafluoride (SF6) density, moisture content, and partial discharge monitoring sensors are typically deployed on the equipment itself and its associated pipe racks. Due to the complex high-voltage environment at substation sites, the high cost and difficulty of laying communication cables, wireless mesh network technologies such as IEEE 802.11s are widely used for transmitting the aforementioned monitoring data.

[0003] Existing wireless mesh networking protocols (such as HWMP) typically use air interface time link metric as the basis for routing selection. This metric mainly examines the average packet loss rate, transmission rate, and protocol overhead of the communication link over a period of time, and tends to select the path with the best statistical average quality for data forwarding.

[0004] However, in the actual operation of GIS substations, this routing mechanism based on historical average statistics has significant limitations. When GIS performs disconnecting switch operations, it generates extremely fast transient overvoltages and transient ground potential rises. This electromagnetic interference exhibits strong spatial non-uniformity and temporal suddenness. Because GIS uses a metal enclosed shell structure, its metal flange connections, grounding down conductors, and insulation basin positions constitute axial discontinuities, while the bolt spacing on the flanges forms a circumferential array of gaps.

[0005] When the switch is activated, the high-frequency electromagnetic energy inside the housing does not radiate outwards uniformly. Instead, due to the structural design of the device itself, it forms high-intensity electromagnetic leakage hotspots at specific locations. Simultaneously, secondary cables (such as power cords and network cables) connected to the wireless aggregation node, at certain lengths, also exhibit strong common-mode coupling effects at specific frequencies, making the device ports highly susceptible to damage or communication interruptions.

[0006] Existing routing methods have the following significant drawbacks: Unable to detect structural risks: Traditional air interface time metrics only focus on the signal-to-noise ratio or bit error rate of the wireless channel, failing to identify potential electromagnetic leakage risks arising from the structural design of GIS equipment (such as flange gaps and grounding point locations). Some wireless nodes may have high signal strength and low average packet loss rate under normal circumstances, but their installation locations may happen to be in high-risk areas where flange gap resonance and shell standing waves overlap.

[0007] Deterministic failure during critical windows: Within the short window of switching operations (typically milliseconds to seconds), these high-risk nodes are subject to sudden, strong electromagnetic intrusions. Due to the lag in updates to traditional routing protocols and the lack of analysis on the correlation between device architecture and interference spectrum, the network will still direct critical alarm data (such as position change signals) to these seemingly high-quality but actually high-risk paths, causing communication disruptions to occur at the moment when the monitoring system most needs to transmit data.

[0008] Lack of protection against port intrusion: Existing technologies typically treat interference as pure air interface noise, ignoring the destructive mechanism of high-frequency interference intruding into equipment ports through cable conduction. Relying solely on air interface quality for routing is insufficient to prevent functional paralysis of nodes caused by resonance between cable size and interference frequency. Summary of the Invention

[0009] This invention provides a dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring, which solves the technical problems mentioned in the background.

[0010] This invention provides a dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring, including: Acquire the equipment component parameters and electromagnetic interference spectrum data generated by the switching operation of the gas-insulated metal-enclosed switchgear. The equipment component parameters include at least the component connection position coordinates representing axial discontinuities and the flange bolt spacing representing circumferential radial gaps. The axial resonance distribution value of the high-frequency current in the shell is determined based on the coordinates of the component connection position, the electromagnetic radiation capability value of the flange gap is determined based on the flange bolt spacing, and the axial resonance distribution value and the electromagnetic radiation capability value are superimposed on the main frequency point of the electromagnetic interference spectrum data to generate the electromagnetic field strength of the structural leakage source. The electromagnetic field strength of the structural leakage source is mapped to the location of the wireless node using a preset electromagnetic wave spatial attenuation model to obtain the spatial disturbance field strength of the node. The common-mode coupling response value of the cable to the main frequency point is calculated by combining the cable size parameters of the wireless node. The product of the spatial disturbance field strength of the node and the common-mode coupling response value is calculated as the transient electromagnetic vulnerability of the node. The structural risk weight of the link is determined based on the transient electromagnetic vulnerability of the wireless nodes at both ends of the communication link. The structural risk weight of the link is then introduced into the standard communication link quality metric to generate a composite routing cost. The network transmission path is then calculated based on the composite routing cost to complete the network formation.

[0011] The beneficial effects of this invention include: by integrating structural information such as the axial connection distribution of components and the circumferential bolt spacing of flanges in gas-insulated metal-enclosed switchgear, and combining it with the spectral characteristics of electromagnetic interference during switch operation, a risk assessment mechanism capable of predicting the distribution of electromagnetic leakage hotspots is established. Furthermore, the vulnerability analysis of port common-mode coupling caused by cable size is introduced, and structural risk weights are superimposed on the quality metric of the wireless communication link. This invention effectively overcomes the shortcomings of traditional routing mechanisms that cannot identify the non-uniform electromagnetic leakage risk caused by the equipment's structural design, achieving deterministic avoidance of high-risk nodes within the transient window of switch operation, and significantly improving the transmission reliability and equipment port security of sulfur hexafluoride online monitoring data in environments with strong electromagnetic interference. Attached Figure Description

[0012] Figure 1 This is a flowchart of the dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring according to the present invention; Figure 2 This is a schematic diagram illustrating a specific implementation of the present invention. Detailed Implementation

[0013] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0014] like Figure 1 As shown, the dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring includes: Acquire the equipment component parameters and electromagnetic interference spectrum data generated by the switching operation of the gas-insulated metal-enclosed switchgear. The equipment component parameters include at least the component connection position coordinates representing axial discontinuities and the flange bolt spacing representing circumferential radial gaps. The axial resonance distribution value of the high-frequency current in the shell is determined based on the coordinates of the component connection position, the electromagnetic radiation capability value of the flange gap is determined based on the flange bolt spacing, and the axial resonance distribution value and the electromagnetic radiation capability value are superimposed on the main frequency point of the electromagnetic interference spectrum data to generate the electromagnetic field strength of the structural leakage source. The electromagnetic field strength of the structural leakage source is mapped to the location of the wireless node using a preset electromagnetic wave spatial attenuation model to obtain the spatial disturbance field strength of the node. The common-mode coupling response value of the cable to the main frequency point is calculated by combining the cable size parameters of the wireless node. The product of the spatial disturbance field strength of the node and the common-mode coupling response value is calculated as the transient electromagnetic vulnerability of the node. The structural risk weight of the link is determined based on the transient electromagnetic vulnerability of the wireless nodes at both ends of the communication link. The structural risk weight of the link is then introduced into the standard communication link quality metric to generate a composite routing cost. The network transmission path is then calculated based on the composite routing cost to complete the network formation.

[0015] In a preferred embodiment, the method for acquiring and discretizing the electromagnetic interference spectrum data includes: Preset includes The discrete main frequency set of characteristic frequencies And acquire the measurement sensor at each main frequency point. Amplitude-frequency response function at ... ; collection The second switch operation event is calculated. The original spectral amplitude of the event The electromagnetic interference spectrum data after statistical averaging is calculated using the following formula. : in, The first in the discrete main frequency set Each frequency point, This represents the total number of switch operation events.

[0016] In a preferred embodiment, the axial resonance distribution value includes: For each dominant frequency point in the discrete dominant frequency set The following cosine square function model is used to calculate the first... Axial resonance distribution at the connection points of each component : in, For the first The coordinates of the connection points of each component For reference point coordinates, Main frequency The corresponding surface wave wavelength of the shell, The calibrated surface wave propagation velocity of the shell.

[0017] In a preferred embodiment, the electromagnetic radiation capability value includes: For each dominant frequency point in the discrete dominant frequency set Calculate the first using the following Lorentz-type function. Electromagnetic radiation capability at the connection points of each component : in, For the first The resonant frequency of the flange gap corresponding to the connection position of each component. This is the resonant half-power bandwidth of the flange gap.

[0018] In a preferred embodiment, the flange gap resonant frequency includes: The following modified half-wavelength resonance model is used to calculate the first... The flange gap resonant frequency corresponding to the connection position of each component : in, The speed of light in a vacuum. For the first Flange bolt spacing at the connection points of each component This is the structural correction factor. The equivalent medium parameters at the gap; and These are offline calibration constants.

[0019] In a preferred embodiment, the electromagnetic field strength of the structural leakage source includes: Calculate the first frequency domain superposition formula using the following formula. Electromagnetic field strength of structural leakage sources at component connection points : in, For discrete dominant frequencies, For the set of the first One frequency point; This refers to the electromagnetic interference spectrum data at the corresponding frequency points. These are the axial resonance distribution values ​​at the corresponding frequency points and locations. This represents the electromagnetic radiation capability value at the corresponding frequency point and location.

[0020] In a preferred embodiment, the disturbance field strength in the node space includes: The following exponential decay superposition model is used to calculate the first... The disturbance field strength in the node space of each wireless node : in, This represents the total number of connection points for the component. For the first Electromagnetic field strength of structural leakage sources at the connection points of individual components; For the first Coordinates of each wireless node With the Coordinates of the connection points of each component The Euclidean spatial distance between them; This is the spatial attenuation scale constant calibrated offline.

[0021] In a preferred embodiment, the common-mode coupling response value includes: The following sinusoidal square resonance response model is used to calculate the first... Common-mode coupling response value of a wireless node : in, For discrete dominant frequencies, For the set of the first One frequency point; This refers to the electromagnetic interference spectrum data at the corresponding frequency points; For the first Equivalent cable length for each wireless node This refers to the calibrated cable propagation speed.

[0022] In a preferred embodiment, the link structural risk weight includes: The following formula for coupling mean and variance is used to calculate the connection of the first... The wireless node and the first Link structural risk weights of communication links for each wireless node : in, and The first The wireless node and the first The transient electromagnetic vulnerability of a wireless node; The difference penalty weights are obtained in advance through offline calibration.

[0023] In a preferred embodiment, the mechanism for generating and updating the composite routing cost includes: The standard communication link quality metric is calculated using the following formula. Combined routing costs : in, For physical layer overhead, To test the frame bit count, The current link transmission rate, This represents the current link frame error rate. The total weight of lattice risk as determined offline. Weights for structural risks in the link; Every fixed period ,renew and And recalculate Calculate the minimum cost path across the entire network using Dijkstra's algorithm: in For from the first From one wireless node to the gateway node The path.

[0024] In a preferred embodiment, the offline calibration process for the parameters includes: The surface wave propagation velocity of the shell is calibrated using the following formula. Spatial attenuation scale constant and cable propagation speed : in, The axial distance between the two sensors on the housing. The time delay difference for the same transient event; Radial distance from the leak source The average electric field strength at that location, To fit the intercept, a logarithmic linear fit was obtained from the multi-point measurement data. ; To determine the known physical length of the cable used for calibration, The pulse round-trip time is measured by a time-domain reflectometer.

[0025] This represents the number of characteristic frequencies contained in the discrete dominant frequency set, and is an integer parameter used to limit the range of electromagnetic interference spectrum data acquisition and calculation. The preferred value is usually determined by referring to the typical clustered frequency distribution of ultrafast transient overvoltages (VFTO) or transient ground potential rises (TGR), combined with the statistical regularities of engineering measured data. A common preferred value is 4 (corresponding to the frequency set...). The method of obtaining the data is as follows: First, the dominant frequency distribution characteristics of electromagnetic interference from GIS switch operations are analyzed through literature review. Then, combined with the voltage level and structural type of the GIS equipment in the target substation, multiple switch operations are performed for verification to screen out the characteristic frequencies with the most concentrated energy and the most significant impact on communication. Finally, the data is determined. The specific value can be set. For example, in a 1000kV GIS scenario, actual measurements showed that four main frequency points could cover more than 90% of the interference energy, therefore, the value can be set... .

[0026] It includes The discrete set of dominant frequencies with characteristic frequencies, i.e. This method focuses on the core frequency components of electromagnetic interference from GIS switch operations, avoiding redundancy caused by full-band calculations. The preferred values ​​are based on large-scale statistical data from existing research, focusing on typical clustering frequencies of ultra-fast transient overvoltages or transient ground potential rises. Common preferred sets are as follows: The acquisition method is as follows: First, collect measured electromagnetic interference spectrum data of GIS equipment with different voltage levels and structures during switching operations. Then, through statistical analysis, identify the frequency points with the highest frequency and largest amplitude. Finally, combine the sensor measurement range and computational efficiency to filter out... This set consists of several core frequency points. For example, for a 220kV GIS device, multiple measurements revealed that 5.5MHz and 18.7MHz are the frequency points with the most concentrated interference energy. By adding adjacent characteristic frequency points, a discrete set of main frequencies is finally formed.

[0027] Represents the set of discrete dominant frequencies The first in Frequency points ( The frequency reference is the core frequency standard for key indicators such as electromagnetic interference spectrum data calculation, axial resonance distribution value, and electromagnetic radiation capability value. Optimal values ​​are selected based on... The preferred set corresponds to each These are typical focal points of ultrafast transient overvoltages or transient ground potential rises, such as 4.5MHz and 5.5MHz. The acquisition method is: from a predetermined set of discrete dominant frequencies... Extract in order, the first The element is Its specific values ​​need to be verified through both literature review and on-site measurement to ensure accurate characterization of the electromagnetic interference frequency characteristics of the target GIS equipment. For example, when hour, , And so on.

[0028] It measures the sensor at the dominant frequency point. The amplitude-frequency response function at a given frequency is used to calibrate the raw spectral data acquired by the sensor, eliminating the influence of the sensor's own frequency response characteristics on the measurement results. This parameter needs to be determined based on the selected sensor model and performance. It is obtained through offline calibration: using a standard signal source with known spectral characteristics (such as a comb signal source), at a discrete set of dominant frequencies... Each frequency point The system outputs a standard amplitude signal. A sensor acquires this signal and records the output amplitude. The ratio of the standard signal amplitude to the sensor output amplitude is calculated, which is the amplitude-frequency response function at that frequency point. For example, in At that point, the standard signal source output amplitude is The output amplitude collected by the sensor is ,but .

[0029] This represents the total number of switch operation events collected. It's used to reduce random errors in electromagnetic interference spectrum data and improve data reliability through statistical averaging of multiple events. The optimal value needs to balance statistical accuracy and acquisition efficiency; a common optimal range is 10-30 times. The acquisition method is as follows: during the normal operation of the target GIS equipment, when disconnecting switch operations are performed, the data is continuously recorded through the data acquisition system. The original time-domain waveform of each complete switching operation is used to ensure that the operating conditions (such as operating phase, gas conditions, etc.) of each operation are representative and to avoid data distortion caused by special operating conditions.

[0030] It is the first The secondary switch operation event occurs at the main frequency point. The original spectral amplitude at point [number] is the fundamental raw data for electromagnetic interference (EMI) spectral data. The value of this parameter is determined by factors such as the EMI intensity of the switching operation and sensor sensitivity. It is obtained by: [doing something related to point [number]]. The original time-domain waveform acquired during the second switching operation was converted into frequency-domain data using a Fast Fourier Transform (FFT), and then the dominant frequency point of this frequency-domain data was extracted. The amplitude at that point is .

[0031] This is the statistically averaged electromagnetic interference spectrum data, used to characterize the electromagnetic interference of the target GIS equipment during switching operations at the dominant frequency point. The average intensity at a given point eliminates the influence of random fluctuations from a single operation. This parameter is determined by the statistical average of multiple original spectral amplitudes. It is obtained by first using the amplitude-frequency response function... The raw spectral amplitude for each event Perform deconvolution amplitude calibration to obtain the calibrated spectral amplitude. And then The calibrated amplitudes of the events are then averaged arithmetically. .

[0032] It is the first The component connection points are located at the main frequency point. The axial resonance distribution value at a given location is used to characterize the peak probability of the current standing wave formed by the reflection of the descending wave at a specific frequency at that location. The value range is [value missing]. A value closer to 1 indicates a higher probability of peak standing wave (VSWR). This parameter is determined by the coordinates of the component connection location, the coordinates of the reference point, and the wavelength of the surface wave on the shell. It is obtained by calculating the VSWR envelope model based on the square cosine function, using the following formula: ,in Main frequency The corresponding surface wave wavelength of the shell.

[0033] It is the first The axial coordinates of each component connection point are used to locate the component connection point along the axial direction of the GIS housing. This parameter is determined by the structural design and installation layout of the GIS equipment, and a measurement accuracy of 1-5 cm is recommended. The method of acquisition is offline calibration: using a total station or high-precision measuring ruler, with the geometric reference plane of the module containing the GIS circuit breaker operating mechanism as a reference, measure the axial coordinates of the first component connection point. The axial distance from the center section of each component connection point (such as a flange or grounding lead-down point) to the reference datum plane is... For example, if the axial distance from the center section of the second flange connection location to the reference datum plane is measured to be 5.2m, then... .

[0034] This is the reference point coordinate, used to provide a unified benchmark for axial coordinate measurements, ensuring consistency in axial coordinates across all component connection locations. This parameter requires a unique benchmark determined based on the structural characteristics of the GIS equipment. It is obtained by selecting a fixed and easily identifiable geometric reference plane on the GIS equipment (usually the geometric reference plane of the module containing the circuit breaker operating mechanism), and setting the axial coordinates of this reference plane to... Once set, this setting is used uniformly throughout the entire station and cannot be changed. For example, if the front face of the GIS circuit breaker operating mechanism is set as the reference plane with an axial coordinate of 0m, then... .

[0035] It is the main frequency point The corresponding surface wave wavelength is used to describe the propagation characteristics of high-frequency electromagnetic waves on the surface of a GIS shell. This parameter is determined by both the surface wave propagation velocity and the dominant frequency. It is obtained using the formula... Calculation, where The calibrated surface wave propagation velocity of the shell. The first in the discrete dominant frequency set A frequency point. For example, when the surface wave propagation speed of the shell. main frequency When the frequency point is reached, the wavelength of the surface wave of the shell can be calculated using a formula.

[0036] This is the calibrated surface wave propagation velocity of the GIS shell, used to describe how fast high-frequency electromagnetic waves propagate on the surface of the GIS shell. This parameter is determined by factors such as the material properties and structural design of the GIS shell. It is obtained through offline calibration: two TEV sensors are placed at different axial positions on the GIS shell, and the time delay difference between the two sensors is measured for the transient waveform generated by the same switching operation. Simultaneously measure the axial distance between the two sensors on the housing. Through formula Calculated.

[0037] It is the first The component connection points are located at the main frequency point. The electromagnetic radiation capability value at a given location is used to characterize the normalized efficiency of electromagnetic energy radiated outward from the flange gap when the interference frequency is close to the resonant frequency of the flange gap. The value range is [value missing]. A value closer to 1 indicates higher radiation efficiency. This parameter is determined by the dominant frequency, the flange gap resonant frequency, and the resonant half-power bandwidth. It is obtained by calculating using a Lorentz-type resonance curve function, with the following formula: For example, the resonant frequency of the flange gap at the fourth component connection location is 18.7MHz, and the resonant half-power bandwidth is 1.5MHz, at the dominant frequency point. At this location, it can be calculated This indicates that the radiation efficiency is highest at this frequency.

[0038] It is the first The resonant frequency of the flange gap corresponding to the connection position of each component is used to characterize the resonant frequency when the gap formed by the flange bolt spacing is used as a slot antenna. This parameter is determined by the flange bolt spacing, structural correction factor, equivalent medium parameters, and speed of light. It is obtained by calculating using a modified half-wavelength resonance model incorporating the structural correction factor, as shown in the formula: For example, the bolt spacing of the first flange is 0.15m, and the structural correction factor is obtained after calibration. Equivalent medium parameters , substitute the speed of light The gap resonant frequency of the flange can then be calculated. .

[0039] It is the first The resonant half-power bandwidth of the flange gap corresponding to the connection position of each component is used to control the decay rate of the Lorentz-type resonance curve, reflecting the range of radiation efficiency variation near the resonant frequency of the flange gap. A preferred value can be set as follows: (Based on the narrowband characteristics of the slot antenna, the engineering fixed value) can also be optimized through offline calibration. The method of obtaining this value is as follows: perform frequency sweep measurements on the sample flange to obtain the curve of its leakage response amplitude changing with frequency, and find the main peak frequency (i.e., The frequencies on either side of the half-power point (-3dB) corresponding to this. and Through formula The calculation yields the following result. For example, if the frequency to the left of the main peak half-power point of a certain flange's swept frequency response is 18.2MHz and the frequency to the right is 19.2MHz, then the resonant half-power bandwidth of this flange is... .

[0040] It is the speed of light in a vacuum, used to calculate the resonant frequency of the flange gap, and its value is fixed. .

[0041] It is the first The flange bolt spacing at each component connection point is the arc distance between the centers of two adjacent bolt holes on the flange circumference. This parameter is determined by the design specifications of the GIS flange. It is obtained through offline calibration: measuring the total circumference along the flange circumference using a flexible measuring tape. Count the number of bolts on the flange Through formula The calculation is possible; alternatively, the bolt pitch can be obtained by directly querying the manufacturing drawings of the GIS equipment. For example, if a flange has a total perimeter of 1.884m and 12 bolts, then the bolt spacing of this flange is... .

[0042] This is a structural correction factor used to correct the deviation between the actual resonant length of the flange gap and the ideal half-wavelength in the half-wavelength resonance model. It reflects the influence of factors such as the structural end effect and gap shape of the flange gap on the resonant frequency. This parameter is an offline calibration constant and needs to be derived from the measured data of the sample flanges. The method for obtaining it is offline calibration: select several representative flanges (with different numbers of bolts and different gaskets / sealing structures), and perform frequency sweep measurements on each sample flange to obtain the measured resonant frequency. Through formula Calculate multiple samples Values, after averaging, are fixed. (or other set values), derived by reverse calculation to determine The value of .

[0043] This is the equivalent medium parameter at the flange gap, comprehensively reflecting the common dielectric properties of various media such as air, gasket, and sealing material within the gap. This parameter is an offline calibration constant and needs to be derived by back-calculating from the measured data of a sample flange. The acquisition method and structural correction factor are also relevant. The calibration is performed simultaneously: after obtaining After averaging, fix The value of (e.g.) ), through formula The calculation yielded, where For multiple sample flanges The average value.

[0044] It is the first The electromagnetic field strength of the structural leakage source at the component connection location is used to characterize the total electromagnetic energy leakage intensity at that location, determined by the discontinuity of the GIS structure (axial) and the radiation from the flange gap (circumferential). This parameter is determined by electromagnetic interference spectrum data, axial resonance distribution value, and electromagnetic radiation capability value. It is obtained by calculation using a frequency domain superposition formula, the formula being: This involves iterating through all frequency points in the discrete main frequency set, calculating the product of the squared amplitude of the electromagnetic interference spectrum data, the axial resonance distribution value, and the electromagnetic radiation capability value for each frequency point, and then summing all the products. For example, the product of the connection positions of the second component at the four main frequency points is as follows: The electromagnetic field strength of the structural leakage source at that location .

[0045] It is the first The spatially disturbed field strength of each wireless node characterizes the combined interference intensity experienced at the node's location from electromagnetic leakage at all component connection points. This parameter is determined by the electromagnetic field strength of the structural leakage source, the spatial distance between the node and component connection points, and the spatial attenuation scale constant. It is obtained by calculating using an exponential attenuation superposition model based on spatial distance, as shown in the formula: ,in This represents the Euclidean distance between the node and the component connection point.

[0046] This refers to the total number of component connection points on the GIS equipment. These connection points include axial discontinuities such as flange connections, grounding leads, and abrupt changes in the shell structure. This parameter is used to limit the range of electromagnetic field strength superposition calculations for structural leakage sources. This parameter is determined by the structural design and installation layout of the target GIS equipment. It is obtained by extracting all component connection points from the GIS as-built drawings / CAD and on-site inspection records, numbering them, and then counting the total number of these numbers. For example, if a GIS device has 8 flange connection points and 2 grounding lead-down points, and no other structural abrupt changes, then... .

[0047] It is the first The wireless node and the first The Euclidean spatial distance between the connection points of each component is used to quantify the spatial relationship between the node and the electromagnetic leakage source. This parameter is determined by the three-dimensional coordinates of the node and component connection points. It is obtained by measuring the distance between the nodes and component connection points using a total station or laser scanner. Three-dimensional coordinates of a wireless node and the Three-dimensional coordinates of the connection points of each component Using the Euclidean distance formula Calculated.

[0048] It is the first The 3D coordinates of each wireless node are used to accurately locate their spatial position within the GIS equipment environment. This parameter is determined by the installation location of the wireless node. The data is obtained through offline calibration: using a total station or laser scanner, the 3D coordinates of all installed wireless mesh nodes are measured, and the 3D coordinate data of each node is recorded. During measurement, a fixed reference point of the GIS equipment should be used as the origin of the coordinates to ensure that the measurement reference of the coordinates of all nodes is consistent.

[0049] It is the first The three-dimensional coordinates of each component connection location are used to accurately locate the spatial position of the component connection location on the GIS equipment site. This parameter is determined by the actual installation layout of the component connection location. The acquisition method is offline calibration: using a total station or laser scanner, the three-dimensional coordinates of all component connection locations (such as flanges, grounding leads) on the GIS equipment are measured, and the three-dimensional coordinate data of each location is recorded. The measurement reference must be consistent with the measurement reference of the wireless node coordinates to ensure the accuracy of spatial distance calculation.

[0050] This is an offline-calibrated spatial attenuation scale constant, used to characterize the exponential attenuation rate of electromagnetic waves generated by GIS switch operations in space with distance. This parameter is determined by the on-site electromagnetic environment and the characteristics of the propagation medium. It is obtained through offline calibration: selecting a typical hot spot flange (structural leakage source electromagnetic field strength)... For a larger flange, multiple measuring points are arranged radially along the flange. Using a TEV / electric field probe, the electric field amplitude at each measuring point is recorded during several switching operations, and the average electric field strength corresponding to each distance is calculated. By fitting an exponential decay model (in (For the fitting intercept), the spatial attenuation scale constant is obtained. For example, by fitting data from five measurement points at different radial distances, the slope is obtained as follows: ,but .

[0051] It is the first The common-mode coupling response value of a wireless node characterizes the sensitivity of the wireless node's cable to electromagnetic interference at discrete dominant frequencies, reflecting the probability of a common-mode current peak caused by the cable's length resonating with the interference frequency. This parameter is determined by the electromagnetic interference spectrum data, the cable's equivalent length, and the cable's propagation speed. It is obtained by calculating using a sinusoidal squared resonance response model, with the formula: For example, the equivalent cable length of the fourth wireless node is 1.2m, and the cable propagation speed is... Iterate through all major frequency points to calculate the product term corresponding to each frequency point, and then sum them up to obtain the product term. .

[0052] It is the first The equivalent cable length of a wireless node is the single equivalent length of all exposed cables (power lines, communication lines, antenna feeders, etc.) connected to the node. This parameter is determined by the node's cable connection configuration. It is obtained through offline calibration: the physical lengths of all exposed cables (power lines, communication lines, shielded grounding wires, etc.) are measured on-site. The lengths are then summed according to the rule of adding the maximum exposed length to the length of the exposed section in the cable tray and the length of the unshielded section inside the cabinet, and converted to the equivalent single length. For example, the exposed length of a wireless node's power supply line is 0.5m, the exposed length of its communication line is 0.8m, and the exposed length of its antenna feeder is 0.3m. The equivalent lengths after conversion are... .

[0053] This is the calibrated cable propagation speed, used to describe how quickly electromagnetic interference signals propagate through a cable. This parameter is determined by the cable's material properties, structural type, and installation method. It is obtained through offline calibration: a cable sample of the same model and installation method as the wireless node cable is selected, and an open-circuit or short-circuit termination experiment is performed using a time-domain reflectometer (TDR) to measure the round-trip time of the pulse signal in the cable. The physical length of the cable sample is known. Through formula Calculated. For example, if the cable sample length is 10m, the pulse round-trip time measured by TDR is... Then the cable propagation speed .

[0054] It is the connection of the first The and the first The link structural risk weight of a wireless node's communication link is used to quantify the structural failure risk of the link within the transient window of switching operation, comprehensively considering the average vulnerability and vulnerability difference between the nodes at both ends of the link. This parameter is jointly determined by the transient electromagnetic vulnerability and the difference penalty weight of the nodes at both ends of the link. It is obtained by calculating using a mean-variance coupled formula, the formula is as follows: The first term is the basic risk term (the arithmetic mean of the vulnerability of the two nodes), and the second term is the imbalance penalty term (the product of the square of the difference in vulnerability between the two nodes and the difference penalty weight). For example, the transient electromagnetic vulnerability of the second node is... The transient electromagnetic vulnerability of the 5th node is Differential penalty weight The structural risk weights of links 2-5 can then be calculated using the formula. .

[0055] It is the first The transient electromagnetic vulnerability of a wireless node is used to comprehensively characterize the failure risk of a node within the transient window of switching operation. It is a key indicator for connecting environmental interference and route optimization; a higher value indicates that the node is more susceptible to failure due to electromagnetic interference. This parameter is jointly determined by the node's spatial disturbance field strength and common-mode coupling response value. It is obtained using the formula... Calculation, where The perturbation field strength in the nodal space. This represents the common-mode coupling response of the node. The product of these two values ​​directly reflects the node's vulnerability under the combined effects of external spatial interference and port intrusion interference. For example, the spatial disturbance field strength of a node... Common-mode coupling response value The transient electromagnetic vulnerability of this node .

[0056] This is a pre-obtained difference penalty weight through offline calibration, used to amplify the impact of differences in transient electromagnetic vulnerability between the two ends of the communication link on link risk, reflecting the link failure characteristics dominated by the vulnerable end. This parameter needs to be determined by fitting the packet loss rate data of the transient window in the field. The acquisition method is offline calibration: collecting transient window datasets of multiple switching operation events, and calculating the packet loss rate of the window for each link. Define the fitting objective function (in For the Sigmoid function, (where is the fitting constant), the optimal objective function is obtained by minimizing the objective function using a grid search or gradient descent algorithm. Values. For example, the difference penalty weights are obtained by fitting transient window data from 20 switching operations. .

[0057] It is a standard communication link quality metric, specifically the Airtime link metric in the 802.11s protocol. It reflects the average occupancy time of a communication link, comprehensively considering factors such as physical layer overhead, transmission rate, and frame error rate. It is a core indicator for evaluating the daily communication quality of a link. This parameter is determined by the real-time transmission status of the link. It is obtained based on the real-time transmission rate of the communication link. and frame error rate Statistical data, through formula The calculation yielded, where For physical layer overhead, This is for the number of test frame bits. For example, the physical layer overhead of a certain link. Test frame bit count Real-time transmission rate Frame error rate The standard communication link quality metric for that link can then be calculated using a formula. .

[0058] This refers to the physical layer overhead, including fixed overhead such as training sequences, frame headers, and access overhead for the wireless communication physical layer. This parameter varies depending on the PHY layer protocol used (e.g., 802.11b / g / n). It is obtained through offline calibration: on an isolated test channel, the node is set to operate at a fixed rate. Continuous transmission length is For the test frames, use a spectrum analyzer or packet capture device to record the average air interface occupancy time per unit frame. Through formula The calculations are then performed, and the calibration results at multiple rates are averaged to obtain the final result. Value. For example, in rate. Below, the average air interface occupancy time per unit frame was measured to be: Test frame bit count Then, the estimated physical layer overhead at that rate can be calculated and determined after multi-rate averaging. .

[0059] This is the number of test frame bits, used to calculate the transmission time component in standard communication link quality metrics. It is a fixed protocol-related parameter. Preferred values ​​follow the IEEE 802.11s protocol recommendations; common preferred values ​​are... This value is widely used in simulation tools such as ns-3 and in actual engineering implementations, exhibiting good compatibility and versatility. It is obtained by directly using the fixed value recommended by the protocol, eliminating the need for on-site calibration and ensuring that the calculation results are consistent with the standard protocol.

[0060] It is the current connection number The and the first The real-time transmission rate of the communication link for each node reflects the current data transmission speed of the link. This parameter changes dynamically with factors such as the wireless channel quality and signal strength. It is obtained by collecting real-time transmission rate statistics through the network card driver or protocol stack of the wireless mesh node. For example, the Adaptive Rate Adjustment (ARF) mechanism in the 802.11 protocol dynamically selects the transmission rate based on channel quality, and the node can directly obtain the currently used rate as the transmission rate. For example, at a certain moment, the wireless channel quality of link 2-3 is good, and the adaptive rate adjustment mechanism selects a transmission rate of... Then at this time .

[0061] It is the current connection number The and the first The frame error rate (FERR) of a communication link for each node, which is the ratio of the number of frames that err during transmission to the total number of transmitted frames, reflects the current communication reliability of the link. This parameter changes dynamically with factors such as wireless channel interference and signal attenuation. It is obtained by using the protocol stack of the wireless mesh node to statistically analyze the frame transmission status of the link in real time, calculating the ratio of the number of erroneous frames to the total number of transmitted frames over a certain period (e.g., 1 second), and using this ratio as the current link's FERR. For example, if a link transmits 1000 frames in 1 second, and 20 of those frames are corrupted, then the frame error rate at that moment is... .

[0062] It is the connection of the first The and the first The composite routing cost of a communication link for each node is used to comprehensively evaluate the daily communication quality and transient structural risks of the link; a smaller value indicates a better link. This parameter is determined by a standard communication link quality metric and a link structural risk weight. It is obtained using the formula... Calculation, where The total weight of lattice risk is used to penalize links with transient high risks by superimposing structural risk weights onto standard link metrics. For example, a standard communication link quality metric for a given link... Link structural risk weight Total weight of lattice risk Then the composite routing cost of this link .

[0063] This is the offline-calibrated total weight of lattice risk, used to adjust the influence of link structural risk weights on composite routing costs and control the penalty for transient structural risks on route selection. This parameter needs to be determined by fitting based on the intensity of transient interference and communication reliability requirements in the field. (Acquisition method and differential penalty weight) The calibration is performed simultaneously: offline calibration At that time, using the same set of transient window datasets and fitted objective function The optimal solution can be obtained simultaneously using either grid search or gradient descent algorithms. Values. For example, the total weight of lattice risk is obtained by fitting transient window data from 20 switching operations. .

[0064] This is the fixed update period for dynamic routing, used to control the frequency of recalculation of composite routing costs and reconstruction of the entire network routing topology, balancing the dynamic adaptability of routes and network overhead. This parameter is determined by the time scale of link metric changes. It is obtained through offline calibration: recording standard communication link quality metrics in the target deployment environment. Calculate the autocorrelation function of the time series. Find satisfaction The smallest ,Will Set as For example, the relevant time after calculating link metrics. The fixed update cycle of dynamic routing. .

[0065] From the first From one wireless node to the gateway node The transmission path, i.e., a sequence of nodes consisting of a series of adjacent communication links, is used to represent the data transmission path from the source node (the first node). From individual nodes to the aggregation node (gateway node) The transmission path of ( ). This parameter is determined by the network topology. It is obtained by: in the adjacency graph of the wireless mesh network, all paths from the ( )... Starting from each node, the final destination is the gateway node. All combinations of links represent possible transmission paths. For example, from the 4th node to the gateway node. Possible paths include , wait.

[0066] It is the first From one wireless node to the gateway node The optimal path, i.e., the path with the minimum composite routing cost among all possible transmission paths, is the final output of dynamic route optimization. This parameter is determined by the composite routing cost of each path. It is obtained by calculating the composite routing cost of all paths from the first path... From node to gateway node transmission path Path cost The path with the minimum cost is found using a shortest path algorithm (such as Dijkstra's algorithm). For example, from the 4th node to the gateway node. The costs of the two possible paths are respectively , Then the optimal path .

[0067] This is the gateway node (aggregation node) of the wireless mesh network. It receives SF6 online monitoring data uploaded by all wireless nodes and forwards the data to the backend monitoring system. It is the root node of the network routing. This parameter is determined by the network deployment plan. It is obtained by selecting one or more nodes with wired backhaul links (such as Ethernet) as gateway nodes during network deployment, specifying their node numbers and locations. Gateway nodes are typically deployed in the control room of GIS equipment or in areas with minimal interference to ensure the stability of data forwarding.

[0068] This refers to the axial distance between two TEV sensors used to calibrate the surface wave propagation velocity on the GIS housing, specifically the axial distance between the two sensors on the GIS housing. This parameter needs to be determined based on the length of the GIS housing and the calibration accuracy requirements, typically a distance of 1-3 meters. It is obtained through offline calibration: two TEV sensors are installed at two different locations along the GIS housing's axis, and the axial distance between the sensor installation locations is measured using a total station or a high-precision measuring ruler. During measurement, ensure that both sensors are installed at the same height and are firmly attached to the housing surface. For example, if the two sensors are installed at positions 0m and 2m along the axial direction of the GIS housing, then... .

[0069] This is the time delay difference between two TEV sensors for a transient event generated during the same switching operation; that is, the difference between the time it takes for the transient waveform to reach the first sensor and the time it takes to reach the second sensor. This parameter is determined by the axial distance between the two sensors and the propagation velocity of the surface waves on the housing. It is acquired through offline calibration: the transient waveforms of both sensors are simultaneously acquired during the same switching operation, cross-correlation analysis is performed on the two waveforms, and the time difference corresponding to the peak value of the cross-correlation function is found. ,in and These are transient waveforms acquired by two sensors. For example, performing cross-correlation analysis on the two waveforms of the same transient event yields the time difference corresponding to the peak value. ,but .

[0070] It is the radial distance from the electromagnetic leakage source. The average field strength at a given location is used to characterize the average interference intensity of electromagnetic waves at different distances in space, and is a constant for calibrating spatial attenuation scales. The basic data. This parameter is determined by both the intensity and distance of the leakage source. It is obtained by: arranging multiple measuring points radially along the typical hot spot flange, with each measuring point's radial distance being... Using a TEV / electric field probe, the electric field amplitude at each measuring point was recorded during several switching operations. The arithmetic mean of the electric field amplitudes from multiple operations was then taken to obtain the distance. Corresponding average field strength For example, at a measuring point 3m away from the hot flange, the measured field strength amplitudes after 10 switching operations were as follows: Then the average field strength at that distance .

[0071] It is the intercept parameter when fitting the exponential decay model, used to characterize the theoretical field strength amplitude at a radial distance of 0 from the electromagnetic leakage source, and is the calibration spatial decay scale constant. The auxiliary parameter is determined by the fitted data. It is obtained by taking data from multiple different radial distances. Corresponding average field strength Taking the natural logarithm, we get ,by x-axis Perform a linear fit on the ordinate to obtain the fitted line. The intercept of the fitted line is . After exponentiate, we get .

[0072] It is the radial distance from the measuring point to the electromagnetic leakage source (typically a hot spot flange), used to calibrate the attenuation law of electromagnetic waves in space and to obtain the average field strength. The basic distance parameter is determined by the measurement point layout scheme. It is obtained by setting multiple measurement point positions radially, using the center of a typical hot spot flange as the origin, and measuring the straight-line distance from each measurement point to the origin using a total station or laser distance meter. The spacing between measuring points is typically 1-2 meters to ensure accurate reflection of the field strength attenuation with distance. For example, three measuring points arranged radially on a hot flange at distances of 1m, 2m, and 3m would correspond to... They are 1m, 2m, and 3m respectively.

[0073] This is the known physical length of the calibration cable used to calibrate the cable propagation speed. This parameter needs to be selected based on the calibration accuracy requirements, typically using a known cable length of 5-20m. The method for obtaining this length is as follows: Select a cable sample of the same model and laying method as the wireless node cable, accurately measure its physical length using a tape measure or laser rangefinder, and record this known length as... When taking measurements, ensure the cable is in a naturally straight position to avoid measurement errors caused by bending.

[0074] This is the round-trip time of a pulse signal in the calibration cable, as measured by a time-domain reflectometer (TDR). It represents the total time it takes for the pulse signal to travel from one end of the cable, be reflected back to the transmitting end, and return to the transmitting end. This parameter is determined by the physical length of the cable and its propagation speed. It is obtained through offline calibration: connect one end of the calibration cable to the transmitting port of the TDR, and open or short-circuit the other end. Start the TDR to transmit a pulse signal, and record the transmission time and the reception time of the reflected signal. The difference between these two times is the time to travel. For example, the time domain reflectometer emits a pulse at a time when... The time to receive the reflected pulse is Then the pulse round-trip time .

[0075] like Figure 2 As shown, Figure 2This image showcases a GIS (Gas Insulated Switchgear) metal-enclosed switchgear with an online monitoring system. The GIS device in the image is a long cylinder composed of multiple flange segments. Flange gaps and bolt spacing are marked at the flange connections. Curved ripples are drawn from these gaps, representing transient electromagnetic leakage. Two TEV sensors are installed on the device to monitor these leaks. These sensors are connected via cables to multiple wireless mesh nodes distributed across the GIS body and an adjacent control cabinet. All data is ultimately aggregated at a convergence node located on the control cabinet. The entire system forms a network through these wireless nodes for monitoring and collecting electromagnetic signal data from the GIS device.

[0076] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring, characterized in that, include: Acquire the equipment component parameters and electromagnetic interference spectrum data generated by the switching operation of the gas-insulated metal-enclosed switchgear. The equipment component parameters include at least the component connection position coordinates representing axial discontinuities and the flange bolt spacing representing circumferential radial gaps. The axial resonance distribution value of the high-frequency current in the shell is determined based on the coordinates of the component connection position, the electromagnetic radiation capability value of the flange gap is determined based on the flange bolt spacing, and the axial resonance distribution value and the electromagnetic radiation capability value are superimposed on the main frequency point of the electromagnetic interference spectrum data to generate the electromagnetic field strength of the structural leakage source. The electromagnetic field strength of the structural leakage source is mapped to the location of the wireless node using a preset electromagnetic wave spatial attenuation model to obtain the spatial disturbance field strength of the node. The common-mode coupling response value of the cable to the main frequency point is calculated by combining the cable size parameters of the wireless node. The product of the spatial disturbance field strength of the node and the common-mode coupling response value is calculated as the transient electromagnetic vulnerability of the node. The structural risk weight of the link is determined based on the transient electromagnetic vulnerability of the wireless nodes at both ends of the communication link. The structural risk weight of the link is then introduced into the standard communication link quality metric to generate a composite routing cost. The network transmission path is then calculated based on the composite routing cost to complete the network formation.

2. The dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The acquisition and discretization processing methods for the electromagnetic interference spectrum data include: A discrete set of dominant frequencies is preset, which includes typical clustering frequencies of ultra-fast transient overvoltages or transient ground potential rises, and the amplitude-frequency response function of the measurement sensor for the discrete set of dominant frequencies is obtained; The original time-domain waveforms generated by multiple switching operations are acquired and converted into frequency-domain amplitudes. The frequency-domain amplitudes are deconvolved and calibrated using the amplitude-frequency response function. The frequency-domain amplitudes of the calibrated events are then arithmetically averaged, and the resulting set of average amplitudes is used as the electromagnetic interference spectrum data.

3. The dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The axial resonance distribution value includes: The axial resonance distribution value is calculated using a standing wave envelope model based on the square cosine function; The standing wave envelope model uses the axial distance difference between the component connection position coordinates and the reference point coordinates as the independent variable, and the shell surface wave wavelength at the corresponding frequency as the periodic modulation parameter; the axial resonance distribution value characterizes the peak probability of the current standing wave formed by the downwave reflection on the shell surface at a specific frequency.

4. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The electromagnetic radiation capability value includes: The electromagnetic radiation capability value was calculated using the Lorentz-type resonance curve function. The Lorentz-type resonance curve function uses the frequency difference between the current dominant frequency point in the electromagnetic interference spectrum data and the resonant frequency of the flange gap as the independent variable, and uses the resonant half-power bandwidth of the flange gap as the attenuation control parameter; the electromagnetic radiation capability value characterizes the normalized efficiency of electromagnetic energy radiated outward by the flange gap when the interference frequency is close to the resonant frequency of the flange gap.

5. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The resonant frequency of the flange gap includes: The resonant frequency of the flange gap was calculated using a modified half-wavelength resonance model that incorporates structural correction coefficients. The modified half-wavelength resonance model defines the flange gap resonant frequency as being proportional to the speed of light and inversely proportional to the product of the flange bolt spacing, the structural correction coefficient, and the square root of the equivalent medium parameter; wherein, the structural correction coefficient and the equivalent medium parameter are fixed calibration constants obtained in advance by performing frequency sweep measurements on sample flanges and back-calculating based on the measured resonant peak values.

6. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The electromagnetic field strength of the structural leakage source includes: For each component connection location, iterate through all discrete dominant frequency points contained in the electromagnetic interference spectrum data; Calculate the product of the amplitude square, axial resonance distribution value, and electromagnetic radiation capability value of the electromagnetic interference spectrum data corresponding to each discrete main frequency point, and sum the products calculated for all discrete main frequency points. The summation result is used as the electromagnetic field strength of the structural leakage source at the connection location of the component. The electromagnetic field strength of the structural leakage source characterizes the total intensity of electromagnetic energy leakage at this location, which is jointly determined by structural discontinuities and gap radiation.

7. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The disturbed field strength in the node space includes: The spatial disturbance field strength of the node is calculated using an exponential decay superposition model based on spatial distance; Determine the Euclidean space distance between the location coordinates of the wireless node and the connection location coordinates of each component; The exponential decay factor is calculated by using the negative of the ratio of the Euclidean spatial distance to the preset spatial attenuation scale constant as the exponent; the product of the electromagnetic field strength of the structural leakage source at each component connection location and the corresponding exponential decay factor is calculated, and the products of all component connection locations are summed to obtain the node spatial disturbance field strength of the wireless node.

8. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The common-mode coupling response value includes: A sinusoidal square resonant response model is adopted to construct a sinusoidal square function with the product of pi, the current discrete dominant frequency point, and the equivalent length of the cable as the independent variable and the cable propagation speed as the independent variable. Traverse all discrete dominant frequency points in the electromagnetic interference spectrum data, and for each discrete dominant frequency point, calculate the product of the square of the amplitude of the electromagnetic interference spectrum data corresponding to that dominant frequency point and the value of the sine square function; The products calculated from all discrete main frequency points are summed to obtain the common-mode coupling response value.

9. The method for dynamic routing optimization networking of wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The link structural risk weights include: Calculate the arithmetic mean of the transient electromagnetic vulnerability of wireless nodes at both ends of the communication link, and use it as the basic risk item; Calculate the difference in transient electromagnetic vulnerability between wireless nodes at both ends of the communication link, calculate the square of the difference and multiply it by a preset difference penalty weight as an imbalance penalty term; The basic risk term is added to the imbalance penalty term to obtain the link structural risk weight; the link structural risk weight is used to quantify the structural failure risk of the communication link within the transient window of the switching operation.

10. The dynamic routing optimization networking method for wireless mesh networks for SF6 online monitoring according to claim 1, characterized in that, The mechanism for generating and updating the composite routing cost includes: Based on the real-time transmission rate and frame error rate statistics of the communication link, a standard communication link quality metric reflecting the average occupancy time of the wireless channel is calculated. The product of the link structural risk weight and the total lattice risk weight is used as a gain term to weight and amplify the standard communication link quality metric, thus obtaining the composite routing cost. The standard communication link quality metric is updated at a fixed time period, while keeping the link structural risk weight unchanged. The composite routing cost is recalculated, and the entire network routing topology is reconstructed based on the updated composite routing cost using the shortest path algorithm.

Images