Cement pole structure health monitoring method and system

Abstract

The application provides a cement pole structure health monitoring method and system, comprising collecting inclination, displacement and vibration original monitoring data, calculating sampling period change rate, obtaining fusion characteristic values representing data change intensity and multi-dimensional state comprehensive change degree, and converting into data event importance indexes, and triggering data transmission when the indexes exceed thresholds; under a fat tree network topology, a candidate path set from a data source to a convergence center is generated, a path comprehensive weight is calculated in combination with an intermediate node transmission success rate and a cement pole health parameter, network channel state parameters are counted and path correlation degrees are solved; the number of redundant check subblocks is determined in combination with event importance and channel state, a block coding matrix is configured to generate redundant data subblocks containing system code subblocks and redundant check subblocks; a path correlation degree tolerance threshold is set according to event importance, and redundant data subblocks are transmitted through optimal path combination screening, until the convergence center can completely reconstruct the original data block.

Description

Technical Field

[0001] This application belongs to the field of cement poles, and in particular relates to a method and system for monitoring the structural health of cement poles. Background Technology

[0002] As crucial load-bearing components of power and communication infrastructure, concrete poles are susceptible to natural weathering, geological subsidence, and external damage during long-term operation, leading to safety hazards such as tilting, cracking, and even collapse. Currently, sensors for tilt angle, displacement, and vibration are typically installed on concrete poles, and a structural health monitoring system is constructed using a wireless sensor network to acquire operational status data. During the network transmission of monitoring data to the aggregation center, given the chain-like or branch-like wide-area distribution of concrete poles, some systems employ multi-path redundancy transmission mechanisms to ensure communication continuity.

[0003] However, existing multi-path routing strategies typically rely solely on link quality, hop count, or remaining energy as the basis for path selection and weight calculation, failing to adequately consider the structural state of the concrete poles to which relay nodes are attached. If the concrete pole containing a relay node tilts or collapses, it may lead to node failure and interruption of related communication links. Furthermore, existing strategies struggle to dynamically adjust redundant transmission intensity based on the current network channel state and the importance of specific data events, easily resulting in wasted communication resources or insufficient transmission reliability. If path correlation is not considered, multiple paths may pass through the same high-risk relay node, reducing the fault tolerance of redundant transmission to node failure risks. Therefore, how to rationally schedule transmission timing based on the multidimensional changes in monitoring data, and how to formulate redundant segmentation and multi-path routing strategies by combining the state of the node's supporting entity with the path topology, are pressing issues that need to be addressed in the field of concrete pole structural health monitoring. Summary of the Invention

[0004] According to a first aspect of the present invention, a method for monitoring the health of cement pole structures is proposed to address the problem that existing methods cannot be adjusted according to the current network channel status and the importance of specific data events, which easily leads to resource waste or substandard transmission reliability. The method includes the following steps: The system acquires raw monitoring data from tilt, displacement, and vibration sensors collected by monitoring nodes on cement poles, extracts the rate of change between adjacent sampling periods, and calculates a fusion feature value representing the degree of data change. The fusion feature value is then used as a data event importance index. When the data event importance index exceeds a preset threshold, data transmission is initiated. In the pre-defined fat-tree network topology, available paths from the data source node to the aggregation center are searched to form a candidate path set; based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement poles they are on, the comprehensive path weight of each candidate path is calculated; network channel state parameters are evaluated and generated based on the statistical characteristics of the comprehensive path weights; and the path correlation between candidate paths is calculated. By combining the data event importance index and the network channel state parameters, the target number of redundant check sub-blocks is determined; the original monitoring data is packaged into original data blocks, and a block coding matrix is ​​configured according to the target number and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundant check sub-blocks; a tolerance threshold for path correlation is set based on the data event importance index; multiple paths are selected from the candidate path set to form an optimal path combination, using the path comprehensive weight as the sorting basis and the tolerance threshold as the constraint; the redundant data sub-blocks are allocated to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

[0005] Optionally, the step of acquiring the raw monitoring data of tilt angle, displacement, and vibration sensors collected by the monitoring nodes on the cement pole, extracting the rate of change between adjacent sampling periods, and calculating the fusion feature value representing the degree of data change includes: The absolute values ​​of the output differences of the tilt sensor, displacement sensor, and vibration sensor within two adjacent sampling periods are collected and divided by the sampling time interval to obtain the tilt rate of change, displacement rate of change, and vibration rate of change. The tilt angle change rate, displacement change rate and vibration change rate are mapped to a set interval using normalization coefficients to generate normalized tilt angle change rate, normalized displacement change rate and normalized vibration change rate respectively. Calculate the product of the normalized tilt angle change rate and 0.4, the product of the normalized displacement change rate and 0.3, and the product of the normalized vibration change rate and 0.3. Then add the products of the normalized tilt angle change rate and 0.4, the normalized displacement change rate and 0.3, and the normalized vibration change rate and 0.3 to obtain the fused feature value.

[0006] Optionally, the step of calculating the fused feature value as a data event importance indicator includes: Extract the preset background base parameters and upper limit magnification constant; The intermediate value is obtained by adding 1 to the fusion feature value, taking the natural logarithm, and then adding the background basis parameter. The intermediate value is multiplied by the upper limit ratio constant and then rounded down. The minimum value between the rounded result and the preset scale upper limit value is used to obtain the data event importance index.

[0007] Optionally, the step of calculating the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement pole to which they are located includes: Within a preset historical time window, the data transmission success rate of each intermediate network node in the candidate path is statistically analyzed. The absolute value of the tilt angle offset of the cement pole where the intermediate network node is located is obtained as a structural health parameter; For each intermediate network node in the candidate path, calculate the ratio of the absolute value of the tilt angle offset of the cement pole where the intermediate network node is located to the set maximum allowable tilt angle offset constant; Calculate the difference between 1 and the ratio, and take the maximum value between the difference and 0 as the tilt angle health. Multiply the data transmission success rate of the intermediate network node by 0.6, and add the tilt health score multiplied by 0.4 to obtain the node weight of the intermediate network node. The combined weight of the path is obtained by multiplying the node weights of all intermediate network nodes included in the candidate path.

[0008] Optionally, the step of evaluating and generating network channel state parameters based on the statistical characteristics of the combined weights of all paths includes: Obtain the comprehensive weight of all paths in the candidate path set, calculate the expected value and standard deviation of the comprehensive weight set of paths, and set a minimum positive number; Calculate the numerator by multiplying the expected value by 2; Calculate the sum of the standard deviation multiplied by 1.5 and then added to 1, and use this as the denominator; Calculate the quotient of the numerator and the denominator; The quotient of the numerator and the denominator is compared with the smallest positive number, and the maximum value of the two is taken as the network channel state parameter.

[0009] Optionally, calculating the path correlation between candidate paths includes: Compare the list of transmission links with the list of node addresses for any two candidate paths; The number of intermediate network nodes and the number of communication links shared by the two candidate paths are counted. Calculate the product of the number of shared intermediate network nodes and 3, and the product of the number of shared communication links and 2. Add the two products together to obtain the initial value of the correlation degree. The minimum value between the initial correlation degree and the preset upper limit of path correlation degree is taken as the path correlation degree.

[0010] Optionally, determining the target number of redundancy check sub-blocks by combining the data event importance index and the network channel state parameters includes: Based on the data event importance index and the network channel state parameters, calculate the quotient of twice the data event importance index and the network channel state parameters, and determine the target number of redundant check sub-blocks by combining the preset lower limit of the number of redundant check sub-blocks and the maximum number of redundant check sub-blocks constant. The initial number of sub-blocks is obtained by adding 1 to the quotient of twice the importance index of the data event and the network channel state parameter, and then rounding down. The number of intermediate sub-blocks is obtained by taking the maximum value between the initial number of sub-blocks and 2. The minimum value between the number of intermediate sub-blocks and the constant value of the maximum redundancy check sub-blocks is taken as the target number of redundancy check sub-blocks.

[0011] Optionally, setting the tolerance threshold for path relevance based on the data event importance index includes: Set the upper limit of path correlation, the buffer constant, and the lower limit of the tolerance threshold constant; The result is calculated by subtracting the data event importance index from the upper limit value and adding the buffer constant; The minimum value between the calculation result and the upper limit value is taken to obtain the upper limit constraint result; The maximum value between the upper limit constraint result and the lower limit constant of the tolerance threshold is taken as the tolerance threshold of the path correlation.

[0012] According to a second aspect of the present invention, a health monitoring system for cement pole structures is provided, comprising the following modules: The calculation module is used to acquire the raw monitoring data of tilt angle, displacement and vibration sensors collected by the monitoring nodes on the cement pole, extract the rate of change of adjacent sampling periods, and calculate the fusion feature value representing the degree of data change; the fusion feature value is used to measure the importance index of the data event; when the importance index of the data event exceeds the preset threshold, data transmission is started. The generation module is used to search for available paths from data source nodes to the aggregation center in a preset fat-tree network topology, forming a candidate path set; calculate the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement poles they are on; evaluate the channel state parameters of the generated network based on the statistical characteristics of the comprehensive path weights; and calculate the path correlation between candidate paths. The determination module is used to jointly determine the target number of redundant check sub-blocks by combining the data event importance index and the network channel state parameters; package the original monitoring data into original data blocks, and configure a block coding matrix according to the target number and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundant check sub-blocks; set a tolerance threshold for path correlation based on the data event importance index; select multiple paths from the candidate path set to form an optimal path combination based on the path comprehensive weight and the tolerance threshold as a constraint; and allocate the redundant data sub-blocks to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

[0013] Preferably, the step of acquiring the original monitoring data of tilt angle, displacement, and vibration sensors collected by the monitoring nodes on the cement pole, extracting the rate of change between adjacent sampling periods, and calculating the fusion feature value representing the degree of data change includes: The absolute values ​​of the output differences of the tilt sensor, displacement sensor, and vibration sensor within two adjacent sampling periods are collected and divided by the sampling time interval to obtain the tilt rate of change, displacement rate of change, and vibration rate of change. The tilt angle change rate, displacement change rate and vibration change rate are mapped to a set interval using normalization coefficients to generate normalized tilt angle change rate, normalized displacement change rate and normalized vibration change rate respectively. Calculate the product of the normalized tilt angle change rate and 0.4, the product of the normalized displacement change rate and 0.3, and the product of the normalized vibration change rate and 0.3. Then add the products of the normalized tilt angle change rate and 0.4, the normalized displacement change rate and 0.3, and the normalized vibration change rate and 0.3 to obtain the fused feature value.

[0014] Preferably, the step of calculating the fused feature value as a data event importance indicator includes: Extract the preset background base parameters and upper limit magnification constant; The intermediate value is obtained by adding 1 to the fusion feature value, taking the natural logarithm, and then adding the background basis parameter. The intermediate value is multiplied by the upper limit ratio constant and then rounded down. The minimum value between the rounded result and the preset scale upper limit value is used to obtain the data event importance index.

[0015] Preferably, the step of calculating the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement pole to which they are located includes: Within a preset historical time window, the data transmission success rate of each intermediate network node in the candidate path is statistically analyzed. The absolute value of the tilt angle offset of the cement pole where the intermediate network node is located is obtained as a structural health parameter; For each intermediate network node in the candidate path, calculate the ratio of the absolute value of the tilt angle offset of the cement pole where the intermediate network node is located to the set maximum allowable tilt angle offset constant; Calculate the difference between 1 and the ratio, and take the maximum value between the difference and 0 as the tilt angle health. Multiply the data transmission success rate of the intermediate network node by 0.6, and add the tilt health score multiplied by 0.4 to obtain the node weight of the intermediate network node. The combined weight of the path is obtained by multiplying the node weights of all intermediate network nodes included in the candidate path.

[0016] Preferably, the step of evaluating and generating network channel state parameters based on the statistical characteristics of the combined weights of all paths includes: Obtain the comprehensive weight of all paths in the candidate path set, calculate the expected value and standard deviation of the comprehensive weight set of paths, and set a minimum positive number; Calculate the numerator by multiplying the expected value by 2; Calculate the sum of the standard deviation multiplied by 1.5 and then added to 1, and use this as the denominator; Calculate the quotient of the numerator and the denominator; The quotient of the numerator and the denominator is compared with the smallest positive number, and the maximum value of the two is taken as the network channel state parameter.

[0017] Preferably, the calculation of path correlation between candidate paths includes: Compare the list of transmission links with the list of node addresses for any two candidate paths; The number of intermediate network nodes and the number of communication links shared by the two candidate paths are counted. Calculate the product of the number of shared intermediate network nodes and 3, and the product of the number of shared communication links and 2. Add the two products together to obtain the initial value of the correlation degree. The minimum value between the initial correlation degree and the preset upper limit of path correlation degree is taken as the path correlation degree.

[0018] Preferably, determining the target number of redundancy check sub-blocks by combining the data event importance index and the network channel state parameters includes: Based on the data event importance index and the network channel state parameters, calculate the quotient of twice the data event importance index and the network channel state parameters, and determine the target number of redundant check sub-blocks by combining the preset lower limit of the number of redundant check sub-blocks and the maximum number of redundant check sub-blocks constant. The initial number of sub-blocks is obtained by adding 1 to the quotient of twice the importance index of the data event and the network channel state parameter, and then rounding down. The number of intermediate sub-blocks is obtained by taking the maximum value between the initial number of sub-blocks and 2. The minimum value between the number of intermediate sub-blocks and the constant value of the maximum redundancy check sub-blocks is taken as the target number of redundancy check sub-blocks.

[0019] Preferably, the step of setting a tolerance threshold for path relevance based on the data event importance index includes: Set the upper limit of path correlation, the buffer constant, and the lower limit of the tolerance threshold constant; The result is calculated by subtracting the data event importance index from the upper limit value and adding the buffer constant; The minimum value between the calculation result and the upper limit value is taken to obtain the upper limit constraint result; The maximum value between the upper limit constraint result and the lower limit constant of the tolerance threshold is taken as the tolerance threshold of the path correlation.

[0020] The present invention has at least the following beneficial effects: This invention extracts the rate of change of multidimensional monitoring data and calculates fused feature values ​​to assess the importance of data events, thereby triggering data transmission on demand and reducing redundant communication overhead and node energy consumption. During routing, the invention combines the historical transmission success rate of intermediate nodes with the structural health parameters of the concrete poles they reside on to calculate the comprehensive weight of the path, reducing the probability of data transmission via nodes with high structural risks and improving the stability of the transmission path. Simultaneously, the invention determines the number of redundant check sub-blocks by combining the importance of data events with the network channel status, thereby generating redundant data sub-blocks for transmission reconstruction; and sets a path correlation tolerance threshold based on the importance of data events, selecting multiple relatively independent paths for transmission. This achieves a match between data redundancy, path independence, and data importance, dispersing the risks of network congestion and single-point failures, and improving the system's fault tolerance and data reconstruction reliability in the event of node anomalies. Attached Figure Description

[0021] Figure 1 A flowchart of the first embodiment; Figure 2 A schematic diagram of the evaluation indicators for intermediate nodes of candidate paths; Figure 3 This diagram illustrates the relationship between the number of redundant check sub-blocks and the importance index of data events. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In the first embodiment, the present invention proposes a method for health monitoring of cement pole structures, such as... Figure 1 It includes the following steps: S1. Acquire the raw monitoring data of tilt angle, displacement and vibration sensors collected by the monitoring nodes on the cement pole, extract the rate of change of adjacent sampling periods, and calculate the fusion feature value representing the degree of data change; calculate the fusion feature value as the data event importance index; when the data event importance index exceeds the preset threshold, start data transmission.

[0024] The analog signals output by the tilt sensor, rope displacement sensor, and piezoelectric vibration sensor within the current sampling period are read and converted from analog to digital to obtain digital tilt angle, displacement, and vibration acceleration values, which are used as raw monitoring data. The differences between each raw monitoring data point in the current sampling period and the corresponding data from the previous sampling period are calculated. The absolute value of the difference is then divided by the sampling time interval to obtain the tilt angle change rate, displacement change rate, and vibration change rate. These rates are then multiplied by their respective preset normalization coefficients, and the smaller value between the product and 1 is taken to obtain the normalized change rate, which is limited to the interval [0,1]. The normalized rates of change of tilt angle, displacement, and vibration are weighted and summed according to preset weights to obtain a fused feature value. This fused feature value is the sum of the products of the normalized rates of change of tilt angle and 0.4, the normalized rates of change of displacement and 0.3, and the normalized rates of change of vibration and 0.3. The preset weights characterize the importance of different monitoring dimensions to structural state change events. The fused feature value is incremented by 1, its natural logarithm is taken, and then the background baseline parameter is added to obtain an intermediate value. This intermediate value is multiplied by an upper limit constant of 10 and rounded down, with the rounded result limited to a preset scale upper limit, to calculate a data event importance index. This data event importance index is compared with a preset threshold. If the index is greater than the preset threshold, the data transmission process is initiated.

[0025] In an optional embodiment, the step of acquiring the raw monitoring data of tilt angle, displacement, and vibration sensors collected by the monitoring nodes on the cement pole, extracting the rate of change between adjacent sampling periods, and calculating the fusion feature value representing the degree of data change includes: The absolute values ​​of the output differences of the tilt sensor, displacement sensor, and vibration sensor within two adjacent sampling periods are collected and divided by the sampling time interval to obtain the tilt rate of change, displacement rate of change, and vibration rate of change. The tilt angle change rate, displacement change rate and vibration change rate are mapped to a set interval using normalization coefficients to generate normalized tilt angle change rate, normalized displacement change rate and normalized vibration change rate respectively. Calculate the product of the normalized tilt angle change rate and 0.4, the product of the normalized displacement change rate and 0.3, and the product of the normalized vibration change rate and 0.3. Then add the products of the normalized tilt angle change rate and 0.4, the normalized displacement change rate and 0.3, and the normalized vibration change rate and 0.3 to obtain the fused feature value.

[0026] During the set sampling period, for example The system reads the multi-dimensional sensor data transmitted from the cement pole monitoring nodes at the current and previous moments, and calculates the absolute value of the difference between the corresponding quantities. For example, if the tilt angle values ​​of two adjacent cycles are 5.2 degrees and 5.0 degrees, the displacement values ​​are 12.5 mm and 12.0 mm, and the vibration accelerations are 0.08g and 0.05g, then by subtracting the absolute values, the absolute difference between the tilt angle sensor output is 0.2 degrees, the absolute difference between the displacement sensor output is 0.5 mm, and the absolute difference between the vibration sensor output is 0.03g. With a sampling period of 1 second, these values ​​are divided by 1 second to obtain the corresponding tilt angle change rate, displacement change rate, and vibration change rate. A preset normalization coefficient is extracted to unify the data dimensions, and the mapping interval is set to [0,1]. Assuming the normalization coefficients for tilt angle, displacement, and vibration are configured to be 0.5, 0.2, and 2.0, respectively, a linear mapping operation is performed by multiplying the normalization coefficients by their corresponding rates of change. The smaller value of the product and 1 is then taken to obtain the normalized rate of change within the interval [0,1]. Based on this operation, the normalized rate of change for tilt angle is calculated to be 0.1 (0.2 × 0.5), the normalized rate of change for displacement is 0.1 (0.5 × 0.2), and the normalized rate of change for vibration is 0.06 (0.03 × 2.0). The calculations are: tilt angle rate of change × 0.4 = 0.04, displacement rate of change × 0.3 = 0.03, and vibration rate of change × 0.3 = 0.018. The fusion eigenvalue is calculated as 0.04 + 0.03 + 0.018 = 0.088, representing the severity of the multidimensional state changes of the cement pole and the combined contribution of different monitoring dimensions to the same structural state change event.

[0027] In an optional embodiment, the step of calculating the fused feature value as a data event importance indicator includes: Extract the preset background base parameters and upper limit magnification constant; The intermediate value is obtained by adding 1 to the fusion feature value, taking the natural logarithm, and then adding the background basis parameter. The intermediate value is multiplied by the upper limit ratio constant and then rounded down. The minimum value between the rounded result and the preset scale upper limit value is used to obtain the data event importance index.

[0028] Extract the preset background baseline parameter of 0.15 and the upper limit multiplier constant of 10. Calculate the median value = ln(0.088+1)+0.15≈0.2343. This logarithmic transformation is used to compress the impact of outliers on the overall indicator evaluation. The background baseline parameter is introduced to provide an offset when the monitored data changes gradually. Calculate the data event importance index = =2, output the integer 2 as the data event importance indicator for this period.

[0029] S2. In the preset fat-tree network topology, search for available paths from the data source node to the aggregation center to form a candidate path set; calculate the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement poles they are on; evaluate and generate network channel state parameters based on the statistical characteristics of the comprehensive path weights; calculate the path correlation between candidate paths.

[0030] In a predefined fat-tree network topology comprising an edge layer, a aggregation layer, and a core layer, starting from the data source node on the concrete pole currently generating data and ending at the aggregation center, available end-to-end links connecting the starting point to the ending point are searched based on network routing information. These searched available end-to-end links form a candidate path set. For each candidate path in the set, the data transmission success rate of each intermediate node is calculated by statistically analyzing the proportion of successfully forwarded data packets to the total number of received data packets within a predefined historical time window (e.g., the past 24 hours). The absolute value of the tilt angle offset of each intermediate node is read, and its ratio to the maximum allowable tilt angle offset constant is calculated. The result of subtracting this ratio from 1, which is not less than 0, is used as the tilt angle health. The data transmission success rate and tilt angle health of each intermediate node are weighted and fused to obtain the node weight. The node weights of all intermediate nodes in the candidate path are multiplied together to obtain the path's comprehensive weight. The average and standard deviation of the comprehensive weights of all paths in the candidate path set are calculated. The network channel state parameters are calculated by multiplying the expected value by 2 as the numerator and the standard deviation by 1.5 plus 1 as the denominator. For any two paths in the candidate path set, count the number of intermediate nodes they share and the number of communication links they share. Calculate the product of the number of shared intermediate nodes and 3, and the product of the number of shared communication links and 2. Add these two products to obtain an initial correlation value. Then, take the smaller value between this initial correlation value and the preset upper limit of path correlation to obtain the path correlation value in the range of 0 to 10. The evaluation index for intermediate nodes in candidate paths is illustrated below. Figure 2 As shown.

[0031] In an optional embodiment, calculating the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement pole to which they are located includes: Within a preset historical time window, the data transmission success rate of each intermediate network node in the candidate path is statistically analyzed. The absolute value of the tilt angle offset of the cement pole where the intermediate network node is located is obtained as a structural health parameter; For each intermediate network node in the candidate path, calculate the ratio of the absolute value of the tilt angle offset of the cement pole where the intermediate network node is located to the set maximum allowable tilt angle offset constant; Calculate the difference between 1 and the ratio, and take the maximum value between the difference and 0 as the tilt angle health. Multiply the data transmission success rate of the intermediate network node by 0.6, and add the tilt health score multiplied by 0.4 to obtain the node weight of the intermediate network node. The combined weight of the path is obtained by multiplying the node weights of all intermediate network nodes included in the candidate path.

[0032] Within a preset historical time window, such as the past 24 hours, the historical transmission success rate of each intermediate node on the current candidate path is calculated. Assume the path contains three intermediate nodes: N1, N2, and N3. The historical transmission success rate of node N1 is 0.95. Obtain the structural health parameters of the concrete pole containing node N1, assuming its tilt angle offset is 3 degrees, and normalize it using a preset upper limit of 10 degrees, resulting in a ratio of 3 / 10 = 0.3. Calculate the tilt angle health of node N1 as max(1-0.3,0) = 0.7 to avoid a negative health score when the tilt angle offset exceeds the limit. Then calculate the node weight of node N1 as 0.95 × 0.6 + 0.7 × 0.4 = 0.85. Calculate the node weights of nodes N2 and N3 in the same way, assuming they are 0.90 and 0.80 respectively. The overall path weight of the current candidate path is the product of the weights of each intermediate node, i.e., 0.85 × 0.90 × 0.80 = 0.612.

[0033] In an optional embodiment, the step of evaluating and generating network channel state parameters based on the statistical characteristics of the combined weights of all paths includes: Obtain the comprehensive weight of all paths in the candidate path set, calculate the expected value and standard deviation of the comprehensive weight set of paths, and set a minimum positive number; Calculate the numerator by multiplying the expected value by 2; Calculate the sum of the standard deviation multiplied by 1.5 and then added to 1, and use this as the denominator; Calculate the quotient of the numerator and the denominator; The quotient of the numerator and the denominator is compared with the smallest positive number, and the maximum value of the two is taken as the network channel state parameter.

[0034] Obtain the combined weights of all paths in the candidate path set and construct a numerical set. Assume the current candidate path set contains 5 paths with corresponding combined weights of 0.85, 0.82, 0.88, 0.75, and 0.90, respectively. Calculate the expected value of this weight set to 0.84, the sample standard deviation to approximately 0.0525, and assign a very small positive number, such as 0.01.

[0035] Calculate the numerator = 0.84 × 2 = 1.68. Calculate the denominator = 0.0525 × 1.5 + 1 = 1.07875. Calculate the quotient = 1.68 / 1.07875 ≈ 1.557. Calculate the network channel state parameter = max(1.557, 0.01) = 1.557.

[0036] In an optional embodiment, calculating the path correlation between candidate paths includes: Compare the list of transmission links with the list of node addresses for any two candidate paths; The number of intermediate network nodes and the number of communication links shared by the two candidate paths are counted. Calculate the product of the number of shared intermediate network nodes and 3, and the product of the number of shared communication links and 2. Add the two products together to obtain the initial value of the correlation degree. The minimum value between the initial correlation degree and the preset upper limit of path correlation degree is taken as the path correlation degree.

[0037] Compare any two candidate paths in the candidate path set, let them be path P1 and path P2. Obtain the transmission link list and node address list for path P1 and path P2 respectively, and count the number of intermediate nodes and communication links they share. If path P1 and path P2 share 1 intermediate node and 0 communication links, then calculate the initial value of path correlation = 1×3 + 0×2 = 3; then compare the initial value of path correlation with the preset upper limit value of path correlation of 10, and take the smaller value as the path correlation, that is, the path correlation of 3, which is used to represent the degree of overlap between the two candidate paths.

[0038] S3. Combine the data event importance index and the network channel state parameters to determine the target number of redundant check sub-blocks; package the original monitoring data into original data blocks, and configure a block coding matrix according to the target number and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundant check sub-blocks; set a tolerance threshold for path correlation based on the data event importance index; select multiple paths from the candidate path set to form an optimal path combination based on the path comprehensive weight and the tolerance threshold; allocate the redundant data sub-blocks to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

[0039] Based on the data event importance index and network channel state parameters, the quotient of twice the data event importance index and the network channel state parameters is calculated. This quotient is then incremented by 1 and rounded down to obtain the initial number of sub-blocks. This initial number is then combined with the maximum redundancy check (RCC) sub-block number constant to determine the target number R of RCC sub-blocks. The relationship between the number of RCC sub-blocks and the data event importance index is as follows: Figure 3As shown. At the data source node, the original monitoring data of tilt angle, displacement, and vibration over a continuous time period are concatenated according to timestamp order and a cyclic redundancy check (CRC) code is added to package them into a fixed-byte original data block. The number of basic data blocks K is determined according to the length of the original data block or a preset block division rule, and the total number of encoded output sub-blocks N=K+R is determined. The original data block is divided into K basic data blocks. The Vandermonde matrix dimension of the Reed-Solomon coding is determined using the number of basic data blocks K and the target number of redundancy check sub-blocks R. Encoding operations are performed by multiplying the basic data blocks with the block coding matrix to generate N redundant data sub-blocks. The redundant data sub-blocks include K system code sub-blocks and R redundancy check sub-blocks.

[0040] A path relevance tolerance threshold and a data event importance index are inversely proportional to each other, so that the higher the data event importance index, the lower the path relevance tolerance threshold. Paths in the candidate path set are sorted from largest to smallest according to their comprehensive path weight, and the first-ranked path is added to the optimal path combination. Paths are then iterated sequentially. If the path relevance of the currently iterated path to all existing paths in the optimal path combination is less than the tolerance threshold, the path is added to the optimal path combination; otherwise, it is discarded. This process continues until the number of paths in the optimal path combination reaches the total number of encoded output sub-blocks or all candidate paths have been traversed. If the number of paths is insufficient after traversal, additional paths are selected from the remaining candidate paths according to their comprehensive path weight, or the path relevance constraint is reduced and the selection is re-selected. If the number of candidate paths is still insufficient, the selected paths are used as the optimal path combination, and multiple redundant data sub-blocks are allowed to be carried on the same path. Redundant data sub-blocks are assigned to the paths in the optimal path combination and sent to the aggregation center. The aggregation center performs decoding when the received sub-blocks meet the decoding quantity requirements and the corresponding encoding matrix sub-matrix is ​​full rank. After successfully reconstructing the original data block, the remaining redundant data sub-blocks are stopped from being sent.

[0041] In an optional embodiment, determining the target number of redundancy check sub-blocks by combining the data event importance index and the network channel state parameters includes: Based on the data event importance index and the network channel state parameters, calculate the quotient of twice the data event importance index and the network channel state parameters, and determine the target number of redundant check sub-blocks by combining the preset lower limit of the number of redundant check sub-blocks and the maximum number of redundant check sub-blocks constant. The initial number of sub-blocks is obtained by adding 1 to the quotient of twice the importance index of the data event and the network channel state parameter, and then rounding down. The number of intermediate sub-blocks is obtained by taking the maximum value between the initial number of sub-blocks and 2. The minimum value between the number of intermediate sub-blocks and the constant value of the maximum redundancy check sub-blocks is taken as the target number of redundancy check sub-blocks.

[0042] Obtain the data event importance index and network channel status parameters for the current period. For example, the data event importance index is 2 and the network channel status parameter is 1.557. Read the maximum redundancy check (RCC) subblock number constant, assuming it has a value of 5. First, calculate the quotient of twice the data event importance index and the network channel status parameter, i.e., quotient = (2 × 2) / 1.557 ≈ 2.569; then add 1 to the quotient and round down to obtain the initial subblock number of 3. Combining this with the lower limit of the RCC subblock number of 2, we obtain the intermediate subblock number max(3,2) = 3; then, combining this with the maximum RCC subblock number constant, we obtain the target RCC subblock number min(3,5) = 3, and output the integer 3 as the target RCC subblock number for this period.

[0043] In this example, the original data block is divided into two basic data blocks. When the target number of redundancy check sub-blocks is three, five redundant data sub-blocks are generated. These five redundant data sub-blocks include two system code sub-blocks and three redundancy check sub-blocks. Once the aggregation center receives sub-blocks that are no less than the number of basic data blocks and whose corresponding encoding matrix sub-matrices are at full rank, decoding is performed and the original data block is reconstructed.

[0044] In an optional embodiment, setting the tolerance threshold for path relevance based on the data event importance index includes: Set the upper limit of path correlation, the buffer constant, and the lower limit of the tolerance threshold constant; The result is calculated by subtracting the data event importance index from the upper limit value and adding the buffer constant; The minimum value between the calculation result and the upper limit value is taken to obtain the upper limit constraint result; The maximum value between the upper limit constraint result and the lower limit constant of the tolerance threshold is taken as the tolerance threshold of the path correlation.

[0045] Set the upper limit constant for path relevance to 10, the buffer constant to 2, and the lower limit constant for tolerance threshold, for example, 3. Obtain the importance index of data events in the current period. Assuming its value is 2, calculate the candidate threshold value as 10 - 2 + 2 = 10. If the importance index of data events changes to 8, calculate the candidate threshold value as 10 - 8 + 2 = 4. After completing the algebraic calculation, first compare the candidate threshold value with the upper limit constant for path relevance and take the smaller value. Then compare the result with the lower limit constant for tolerance threshold and take the larger value to obtain the tolerance threshold for path relevance.

[0046] In a second embodiment, the present invention also proposes a health monitoring system for cement pole structures, comprising the following modules: The calculation module is used to acquire the raw monitoring data of tilt angle, displacement and vibration sensors collected by the monitoring nodes on the cement pole, extract the rate of change of adjacent sampling periods, and calculate the fusion feature value representing the degree of data change; the fusion feature value is used to measure the importance index of the data event; when the importance index of the data event exceeds the preset threshold, data transmission is started. The generation module is used to search for available paths from data source nodes to the aggregation center in a preset fat-tree network topology, forming a candidate path set; calculate the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement poles they are on; evaluate the channel state parameters of the generated network based on the statistical characteristics of the comprehensive path weights; and calculate the path correlation between candidate paths. The determination module is used to jointly determine the target number of redundant check sub-blocks by combining the data event importance index and the network channel state parameters; package the original monitoring data into original data blocks, and configure a block coding matrix according to the target number and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundant check sub-blocks; set a tolerance threshold for path correlation based on the data event importance index; select multiple paths from the candidate path set to form an optimal path combination based on the path comprehensive weight and the tolerance threshold as a constraint; and allocate the redundant data sub-blocks to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

[0047] The above description represents the preferred embodiments of the present invention. It should be noted that, for those skilled in the art, various improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for monitoring the structural health of cement poles, characterized in that, include: The raw monitoring data of tilt, displacement and vibration sensors collected by the monitoring nodes on the cement pole are obtained, the rate of change of adjacent sampling periods is extracted, and the fusion feature value representing the degree of data change is calculated. The fused feature values ​​are calculated as a data event importance index; when the data event importance index exceeds a preset threshold, data transmission is initiated. In the preset fat tree network topology, search for available paths from the data source node to the aggregation center to form a candidate path set; calculate the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes through which each candidate path passes and the structural health parameters of the cement pole where it is located. Network channel state parameters are generated based on the statistical characteristics of the combined weights of all paths. Calculate the path correlation between candidate paths; By combining the data event importance index and the network channel state parameters, the target number of redundancy check sub-blocks is determined; The original monitoring data is packaged into original data blocks, and a block coding matrix is ​​configured according to the target quantity and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundancy check sub-blocks. A tolerance threshold for path relevance is set based on the aforementioned data event importance index; Based on the comprehensive weight of the paths and the tolerance threshold, multiple paths are selected from the candidate path set to form an optimal path combination. The redundant data sub-blocks are then distributed to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

2. The method according to claim 1, characterized in that, The process of acquiring raw monitoring data from tilt, displacement, and vibration sensors collected at monitoring nodes on the cement pole, extracting the rate of change between adjacent sampling periods, and calculating fusion feature values ​​representing the degree of data change includes: The absolute values ​​of the output differences of the tilt sensor, displacement sensor, and vibration sensor within two adjacent sampling periods are collected and divided by the sampling time interval to obtain the tilt rate of change, displacement rate of change, and vibration rate of change. The tilt angle change rate, displacement change rate and vibration change rate are mapped to a set interval using normalization coefficients to generate normalized tilt angle change rate, normalized displacement change rate and normalized vibration change rate respectively. Calculate the product of the normalized tilt angle change rate and 0.4, the product of the normalized displacement change rate and 0.3, and the product of the normalized vibration change rate and 0.

3. Then add the products of the normalized tilt angle change rate and 0.4, the normalized displacement change rate and 0.3, and the normalized vibration change rate and 0.3 to obtain the fused feature value.

3. The method according to claim 1 or 2, characterized in that, The step of calculating the fused feature value as a data event importance indicator includes: Extract the preset background base parameters and upper limit magnification constant; The intermediate value is obtained by adding 1 to the fusion feature value, taking the natural logarithm, and then adding the background basis parameter. The intermediate value is multiplied by the upper limit ratio constant and then rounded down. The minimum value between the rounded result and the preset scale upper limit value is used to obtain the data event importance index.

4. The method according to claim 1, characterized in that, The comprehensive path weight of each candidate path is calculated based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement pole to which they are located, including: Within a preset historical time window, the data transmission success rate of each intermediate network node in the candidate path is statistically analyzed. The absolute value of the tilt angle offset of the cement pole where the intermediate network node is located is obtained as a structural health parameter; For each intermediate network node in the candidate path, calculate the ratio of the absolute value of the tilt angle offset of the cement pole where the intermediate network node is located to the set maximum allowable tilt angle offset constant; Calculate the difference between 1 and the ratio, and take the maximum value between the difference and 0 as the tilt angle health. Multiply the data transmission success rate of the intermediate network node by 0.6, and add the tilt health score multiplied by 0.4 to obtain the node weight of the intermediate network node. The combined weight of the path is obtained by multiplying the node weights of all intermediate network nodes included in the candidate path.

5. The method according to claim 1, characterized in that, The process of generating network channel state parameters based on the statistical characteristics of the combined weights of all paths includes: Obtain the comprehensive weight of all paths in the candidate path set, calculate the expected value and standard deviation of the comprehensive weight set of paths, and set a minimum positive number; Calculate the numerator by multiplying the expected value by 2; Calculate the sum of the standard deviation multiplied by 1.5 and then added to 1, and use this as the denominator; Calculate the quotient of the numerator and the denominator; The quotient of the numerator and the denominator is compared with the smallest positive number, and the maximum value of the two is taken as the network channel state parameter.

6. The method according to claim 1, characterized in that, The calculation of path correlation between candidate paths includes: Compare the list of transmission links with the list of node addresses for any two candidate paths; The number of intermediate network nodes and the number of communication links shared by the two candidate paths are counted. Calculate the product of the number of shared intermediate network nodes and 3, and the product of the number of shared communication links and 2. Add the two products together to obtain the initial value of the correlation degree. The minimum value between the initial correlation degree and the preset upper limit of path correlation degree is taken as the path correlation degree.

7. The method according to claim 1 or 6, characterized in that, The determination of the target number of redundancy check sub-blocks by combining the data event importance index and the network channel state parameters includes: Based on the data event importance index and the network channel state parameters, calculate the quotient of twice the data event importance index and the network channel state parameters, and determine the target number of redundant check sub-blocks by combining the preset lower limit of the number of redundant check sub-blocks and the maximum number of redundant check sub-blocks constant. The initial number of sub-blocks is obtained by adding 1 to the quotient of twice the importance index of the data event and the network channel state parameter, and then rounding down. The number of intermediate sub-blocks is obtained by taking the maximum value between the initial number of sub-blocks and 2. The minimum value between the number of intermediate sub-blocks and the constant value of the maximum redundancy check sub-blocks is taken as the target number of redundancy check sub-blocks.

8. The method according to claim 1, characterized in that, The tolerance threshold for path relevance based on the data event importance index includes: Set the upper limit of path correlation, the buffer constant, and the lower limit of the tolerance threshold constant; The result is calculated by subtracting the data event importance index from the upper limit value and adding the buffer constant; The minimum value between the calculation result and the upper limit value is taken to obtain the upper limit constraint result; The maximum value between the upper limit constraint result and the lower limit constant of the tolerance threshold is taken as the tolerance threshold of the path correlation.

9. A structural health monitoring system for cement poles, characterized in that, include: The calculation module is used to acquire the raw monitoring data of tilt angle, displacement and vibration sensors collected by the monitoring nodes on the cement pole, extract the rate of change of adjacent sampling periods, and calculate the fusion feature value representing the degree of data change; The fused feature values ​​are calculated as a data event importance index; when the data event importance index exceeds a preset threshold, data transmission is initiated. The generation module is used to search for available paths from the data source node to the aggregation center in the preset fat tree network topology, forming a candidate path set; calculate the comprehensive path weight of each candidate path based on the historical transmission success rate of the intermediate nodes traversed by each candidate path and the structural health parameters of the cement poles they are on; and evaluate the channel state parameters of the generated network based on the statistical characteristics of the comprehensive path weights. Calculate the path correlation between candidate paths; The determination module is used to determine the target number of redundancy check sub-blocks by combining the data event importance index and the network channel state parameters. The original monitoring data is packaged into original data blocks, and a block coding matrix is ​​configured according to the target quantity and the number of basic data blocks to generate redundant data sub-blocks containing system code sub-blocks and redundancy check sub-blocks. A tolerance threshold for path relevance is set based on the aforementioned data event importance index; Based on the comprehensive weight of the paths and the tolerance threshold, multiple paths are selected from the candidate path set to form an optimal path combination. The redundant data sub-blocks are then distributed to the optimal path combination for transmission until the aggregation center can completely reconstruct the original data blocks.

10. The system according to claim 9, characterized in that, The process of acquiring raw monitoring data from tilt, displacement, and vibration sensors collected at monitoring nodes on the cement pole, extracting the rate of change between adjacent sampling periods, and calculating fusion feature values ​​representing the degree of data change includes: The absolute values ​​of the output differences of the tilt sensor, displacement sensor, and vibration sensor within two adjacent sampling periods are collected and divided by the sampling time interval to obtain the tilt rate of change, displacement rate of change, and vibration rate of change. The tilt angle change rate, displacement change rate and vibration change rate are mapped to a set interval using normalization coefficients to generate normalized tilt angle change rate, normalized displacement change rate and normalized vibration change rate respectively. Calculate the product of the normalized tilt angle change rate and 0.4, the product of the normalized displacement change rate and 0.3, and the product of the normalized vibration change rate and 0.

3. Then add the products of the normalized tilt angle change rate and 0.4, the normalized displacement change rate and 0.3, and the normalized vibration change rate and 0.3 to obtain the fused feature value.

Images