A method and system for compressing tower monitoring data based on short message communication

Abstract

The present application relates to a kind of tower monitoring data compression method and system based on short message communication, belong to transmission line online monitoring technical field.It includes: collecting microclimate data, tower posture data and tower structure response data;Through the pre-established mechanical coupling model, the coupling characteristic quantity between wind load, temperature stress and tower structure deformation is extracted;Based on coupling characteristic quantity, hierarchical compression coding is carried out, the predicted value of posture and structure response is calculated by mechanical model using current meteorological data in physical correlation compression layer, and the prediction residual is transmitted;Channel adaptive compression layer dynamically adjusts residual quantization accuracy and entropy coding according to channel bandwidth and signal quality;Monitoring center updates mechanical model after decoding, and visual display and safety evaluation are carried out in combination with three-dimensional model.The transmission efficiency and physical authenticity of tower monitoring data are significantly improved without relying on ground communication network.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of on-line monitoring of power transmission lines, and particularly relates to a tower monitoring data compression method and system based on short message communication. BACKGROUND

[0002] The tower of a power transmission line is a key infrastructure of a power system, and its structural safety is directly related to the stable operation of the power grid. The environment in which the tower is located is complex and changeable, and local strong winds, heavy rain, icing and other extreme weather conditions, as well as the resulting safety hazards such as tower tilting, vibration and structural fatigue, pose a continuous threat to the power transmission line. Therefore, synchronous and real-time monitoring of micro-meteorological parameters (temperature, humidity, wind speed, wind direction, air pressure) and structural state parameters (tilt angle, vibration frequency, strain at key positions) of the tower is an important means to timely detect hidden dangers and ensure the safety of the power grid.

[0003] However, a large number of power transmission towers are located in remote mountainous areas, Gobi deserts and other areas without ground communication network coverage, and traditional data transmission methods based on optical fiber or cellular networks are difficult to deploy. The short message communication service provided by the Beidou satellite navigation system has become a key technical means for power monitoring data transmission in such scenarios due to its wide coverage, lack of geographical environment restrictions and independence from ground infrastructure. There have been attempts in the prior art to apply Beidou short messages to tower deformation monitoring, micro-meteorological monitoring, image monitoring and other aspects, such as using a Beidou short message module to return tower tilt angle data and weather sensor data to a monitoring center.

[0004] However, the communication frequency and single communication capacity of the Beidou-3 short message communication are strictly limited, while tower monitoring involves multi-source heterogeneous data such as weather, attitude and structural response. If the data is not compressed and directly transmitted, the data volume far exceeds the single short message carrying capacity. If multiple packages are used for transmission, the communication delay and cost will increase dramatically, making it difficult to meet the real-time monitoring requirements. Therefore, how to achieve efficient transmission of tower multi-source monitoring data under the capacity constraints of short messages is a technical problem that needs to be solved urgently.

[0005] In view of the above problems, the prior art has proposed several data compression schemes. One type of scheme compresses specific types of monitoring data, such as compression transmission of Wide Area Measurement System (WAMS) measurement data, compression transmission of meteorological data, and progressive compression processing of power transmission line images, etc., but such schemes usually only face a single data type and do not involve the collaborative processing of multi-source heterogeneous data. Another type of scheme uses general compression algorithms (such as Lempel-Ziv-Welch (LZW) compression algorithm, Huffman coding, etc.) to compress monitoring data, but existing methods are mostly based on data format or statistical characteristics for classification, without fully considering the inherent physical correlation between micro-meteorological data and tower structure response data, such as the explicit mechanical mapping relationship between wind speed changes and tower inclination changes, and the physical coupling of thermal expansion and contraction between temperature changes and structural strain.

[0006] Therefore, there is a need for a tower monitoring data compression method and system based on short message communication, which can construct a unified compression framework for multi-source heterogeneous tower state data, use a tower structure mechanics coupling model to mine the physical correlation between micro-meteorological data and structure response data to eliminate physical redundancy, and adaptively adjust the compression strategy according to the time-varying characteristics of the short message channel, so as to realize efficient and reliable transmission of tower monitoring data under the constraint of satellite short message capacity. SUMMARY

[0007] To solve the above problems in the prior art, the present application provides a tower monitoring data compression method and system based on short message communication, and the purpose of the present application can be achieved by the following technical solutions: A tower monitoring data compression method based on short message communication, comprising: Collecting multi-source heterogeneous tower state data, including micro-meteorological data, tower attitude data, and tower structure response data; Based on a pre-established tower structure mechanics coupling model, performing physical correlation analysis on the multi-source heterogeneous tower state data, and extracting coupling characteristic quantities between the micro-meteorological data and the tower attitude data and the tower structure response data; the coupling characteristic quantities represent the mapping relationship between wind load, temperature stress and tower structure deformation; According to the coupling characteristic quantities, the multi-source heterogeneous tower state data is compressed and encoded in layers, including: Physical correlation compression: based on the current time micro-meteorological data, the predicted values of tower attitude and structure response are calculated through the tower structure mechanics coupling model, the predicted values are compared with the actual collected values to calculate the prediction residual, and the prediction residual value is transmitted; Channel adaptive compression: dynamically adjust the quantization precision of the predicted residual value according to the available bandwidth and signal quality of the current satellite short message channel, and use variable length coding to entropy encode the quantized residual value; The layered compressed and encoded data is sent to the monitoring center through the satellite short message channel. The monitoring center decodes the received layered compressed and encoded data step by step, updates the tower structure mechanics coupling model based on the decoded data, and combines the tower three-dimensional model to perform visual display and structure safety state evaluation.

[0008] Specifically, when collecting multi-source heterogeneous tower state data, different data types use different sampling frequencies: the sampling frequency of microclimate data is lower than that of tower attitude data, and the sampling frequency of tower attitude data is lower than that of tower structure response data; the data of different sampling frequencies are aligned on the time axis by interpolation and then input into the tower structure mechanics coupling model for physical correlation analysis.

[0009] Specifically, the channel adaptive compression further comprises: according to the remaining power of the one-time lithium battery built-in the tower monitoring device, cooperatively adjusting the quantization precision of the predicted residual value and the transmission frequency of the satellite short message, when the remaining power is lower than a preset threshold, quantizing the predicted residual value corresponding to the tower structure response data with a first quantization step, quantizing the predicted residual value corresponding to the microclimate data with a second quantization step, the first quantization step is smaller than the second quantization step, and transmitting the compressed package corresponding to the tower structure response data with a first transmission frequency and the compressed package corresponding to the microclimate data with a second transmission frequency, the first transmission frequency is higher than the second transmission frequency.

[0010] Specifically, the tower structure mechanics coupling model is pre-established by the following way: Based on the tower design parameters and material properties, a tower finite element model is constructed; Through historical monitoring data training, a mapping relationship network of wind load-temperature field-structure response is established, which is a physical information neural network or a radial basis function response surface model.

[0011] Specifically, the physical correlation compression further comprises: When the wind speed change rate is less than a first threshold and the temperature change rate is less than a second threshold, the microclimate data is compressed by using first-order difference coding based on time series prediction; When the wind speed change rate is greater than or equal to the first threshold, or the temperature change rate is greater than or equal to the second threshold, the tower attitude data and the tower structure response data are compressed by using key frame coding based on event triggering, the key frame coding takes the full amount of data at the starting time of the dramatic change event as the key frame, and encodes and transmits the incremental data during the event based on the key frame.

[0012] Specifically, before sending the layered compressed encoded data to the monitoring center through the satellite short message channel, the method further comprises marking the compressed data packets with priorities, wherein the priority marking comprises: an emergency priority, marking the compressed data packets corresponding to the abnormal values of the tower structure response data exceeding the safety threshold; and a regular priority, marking the compressed data packets corresponding to the normal change values of the tower posture data and the micro-meteorological data; and the satellite short message channel transmits the data packets in the order of the emergency priority data packets taking priority over the regular priority data packets.

[0013] Specifically, the calculation of the predicted residual values comprises: for the tower posture data, calculating a posture angle deviation vector in a three-dimensional space between the predicted values and the actually collected values; and for the tower structure response data, calculating stress-strain deviation amounts at key positions between the predicted values and the actually collected values. respectively taking the respective preset maximum allowable change amounts as the basis, performing extreme value normalization processing on the posture angle deviation vector and the stress-strain deviation amounts to generate dimensionless predicted residual values.

[0014] Specifically, the dynamic adjustment of the quantization precision of the predicted residual values comprises: when the available bandwidth of the satellite short message channel is greater than a first threshold value and the signal carrier-to-noise ratio is greater than a third threshold value, quantizing the predicted residual values with a first quantization step, retaining all significant digits; when the available bandwidth of the satellite short message channel is less than a second threshold value or the signal carrier-to-noise ratio is less than a fourth threshold value, quantizing the predicted residual values with a second quantization step, the second quantization step being greater than the first quantization step, and performing truncation or rounding processing on the quantized data before transmitting the data after enabling forward error correction coding; the first threshold value is greater than the second threshold value. when the satellite short message channel is unavailable or interrupted, buffering the compressed data in a local storage medium, and transmitting the data in batches after the channel is restored.

[0015] Specifically, the variable-length coding is Huffman coding or arithmetic coding; the quantized predicted residual values are subjected to probability statistics, different length code words are allocated according to the probabilities of the residual values, short code words are allocated to residual values with high probabilities, and long code words are allocated to residual values with low probabilities.

[0016] Specifically, the updating of the tower structure mechanical coupling model based on the decoded data comprises: comparing the decoded tower posture data and the decoded tower structure response data with the predicted values to calculate the prediction error of the tower structure mechanical coupling model. When the tower structure mechanics coupling model prediction error is greater than the preset tolerance at a preset number of continuous sampling time points, an online model correction process is triggered, and the mapping relationship network is retrained using the monitoring data in the latest time window; The corrected coupling characteristic quantity is sent to the tower monitoring device to update the physically associated compressed prediction reference.

[0017] Specifically, the structure safety state evaluation includes: Based on the decoded tower posture data and tower structure response data, the deviation degree of the current state of the tower from the design safety threshold is calculated; In combination with the updated tower structure mechanics coupling model, the tower structure response trend is calculated by taking the meteorological condition as input; When the deviation degree exceeds the warning threshold or the peak value in the structure response trend exceeds the design safety threshold, a hierarchical warning information is generated and returned to the tower monitoring device through the satellite short message channel; the tower monitoring device responds to the hierarchical warning information and switches the sampling frequency to a third sampling frequency higher than the regular sampling frequency, the compression rate to a second compression rate lower than the regular compression rate, and the sending frequency to a third sending frequency higher than the regular sending frequency.

[0018] Specifically, a tower monitoring data compression system based on short message communication includes: A monitoring device is arranged on the tower and includes a multi-source sensor array, a processor, and a satellite short message communication module. The processor has a tower structure mechanics coupling model calculation unit, a physically associated compression encoding unit, and a channel adaptive compression encoding unit, which are used to perform the data acquisition, coupling characteristic quantity extraction, and hierarchical compression encoding steps in the method, respectively. The satellite short message communication module is used to send the compression encoded data through the satellite short message channel; A satellite communication link connects the monitoring device and the monitoring center; The monitoring center includes a data receiving and decoding module, a model updating module, and a visual evaluation module, which are used to perform the data receiving and decoding, mechanics coupling model updating, and visual display and structure safety evaluation steps in the method, respectively.

[0019] The present application has the following advantages: Significantly improve the transmission efficiency of monitoring data under short message communication conditions. The present application extracts the coupling characteristic quantity between wind load, temperature stress and tower structure deformation based on the tower structure mechanics coupling model, uses the coupling characteristic quantity as the prediction reference of physical correlation compression, and only transmits the prediction residual value, which fundamentally eliminates the physical redundancy between multi-source heterogeneous data and greatly reduces the amount of data to be transmitted. Compared with directly transmitting raw data or using general lossless compression algorithm, the present application fully utilizes the mechanical coupling prior information between microclimate data and structural response data, significantly improves the compression ratio under the premise of ensuring the recovery accuracy at the decoding end, greatly increases the effective monitoring information carried by the unit short message frame, and effectively solves the technical problem that multi-source monitoring data is difficult to be completely transmitted under the condition of limited short message capacity.

[0020] Adapt the compression strategy to the short message channel. The present application further introduces channel adaptive compression on the basis of physical correlation compression, dynamically adjusts the quantization precision of prediction residual value according to the available bandwidth and signal quality of short message channel, and combines variable length entropy coding, so that the compression strategy can adapt to the time-varying characteristics of the channel. When the channel condition is good, high-precision quantization is adopted to retain complete monitoring information, and when the channel condition deteriorates, the precision of non-critical data is appropriately reduced at the cost of reliable delivery of data packets, avoiding the problem of sharp decline of communication efficiency caused by data packet loss or retransmission in traditional fixed compression scheme when the channel quality fluctuates.

[0021] Build a perception-compression-transmission-update closed loop to realize the continuous evolution of the mechanical model. The monitoring center updates the tower structure mechanics coupling model online based on the deviation between the measured data and the model prediction value after decoding, and feeds back the corrected coupling characteristic quantity to the monitoring device side, so that the prediction reference of physical correlation compression can be adaptively adjusted following the long-term service state of the tower (such as material aging, foundation settlement, etc.), ensuring that the compression accuracy remains stable throughout the life cycle, overcoming the defect that the prediction deviation of fixed model gradually increases after the tower structure characteristics change over time.

[0022] Ensure the continuous working ability of the monitoring device under severe working conditions. The present application manages the battery power and compression parameters cooperatively, prioritizes the transmission accuracy and frequency of tower structure response data when the power is limited, and combines a low-power management unit to effectively prolong the continuous working time of the device under the condition of no solar power supplement. At the same time, through the two-way closed loop feedback of early warning information, the monitoring center can actively trigger the monitoring device to switch to the emergency sampling mode, realizing seamless switching from regular monitoring to emergency response, and enhancing the rapid response ability of the system to sudden events. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to facilitate the understanding of those skilled in the art, the present application will be further described below with reference to the drawings.

[0024] Figure 1 This is a flowchart illustrating the overall method of the present invention; Figure 2 This is a diagram of the overall system architecture of the present invention; Figure 3 This is a flowchart of the layered compression encoding process of the present invention; Figure 4 This is a timing diagram of data acquisition and preprocessing in this invention. Detailed Implementation

[0025] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0026] Please see Figures 1-4 A method for compressing tower monitoring data based on short message communication, comprising: Collect multi-source heterogeneous tower status data, which includes micro-meteorological data, tower attitude data, and tower structural response data; Based on a pre-established coupled mechanical model of the tower structure, physical correlation analysis is performed on the state data of the multi-source heterogeneous towers to extract the coupling characteristic quantities between micro-meteorological data, tower attitude data, and tower structural response data; the coupling characteristic quantities characterize the mapping relationship between wind load, temperature stress, and tower structural deformation. Based on the coupling characteristic quantity, the multi-source heterogeneous tower status data is subjected to hierarchical compression coding, the hierarchical compression coding including: Physical correlation compression: Based on the micro-meteorological data at the current moment, the predicted values ​​of the tower attitude and structural response are calculated through the tower structure mechanical coupling model. The predicted values ​​are compared with the actual collected values ​​to calculate the prediction residuals, and the prediction residual values ​​are transmitted. Channel adaptive compression: Based on the available bandwidth and signal quality of the current satellite short message channel, the quantization accuracy of the predicted residual value is dynamically adjusted, and variable-length coding is used to entropy code the quantized residual value; The layered compressed and encoded data is sent to the monitoring center via a satellite short message channel; The monitoring center decodes the received layered compressed and encoded data step by step, updates the structural mechanical coupling model of the tower based on the decoded data, and combines the tower's three-dimensional model for visualization and structural safety status assessment.

[0027] Specifically, when collecting multi-source heterogeneous tower status data, different sampling frequencies are used for different data types: the sampling frequency of micro-meteorological data is lower than that of tower attitude data, and the sampling frequency of tower attitude data is lower than that of tower structural response data; data with different sampling frequencies are interpolated and aligned on the time axis before being sent to the tower structural mechanical coupling model for physical correlation analysis.

[0028] Specifically, the channel adaptive compression further includes: coordinating the quantization accuracy of the prediction residual value and the transmission frequency of satellite short messages based on the remaining power of the disposable lithium battery built into the tower monitoring device; when the remaining power is lower than a preset threshold, quantizing the prediction residual value corresponding to the tower structure response data with a first quantization step size and quantizing the prediction residual value corresponding to the micro-meteorological data with a second quantization step size, wherein the first quantization step size is smaller than the second quantization step size; and transmitting the compressed packet corresponding to the tower structure response data with a first transmission frequency and transmitting the compressed packet corresponding to the micro-meteorological data with a second transmission frequency, wherein the first transmission frequency is higher than the second transmission frequency.

[0029] Specifically, the mechanical coupling model of the tower structure is pre-established in the following manner: Based on the tower design parameters and material properties, a finite element model of the tower is constructed; By training with historical monitoring data, a mapping network of wind load, temperature field, and structural response is established. The mapping network is a physical information neural network or a radial basis function response surface model.

[0030] Specifically, the physical association compression also includes: When the rate of change of wind speed is less than the first threshold and the rate of change of temperature is less than the second threshold, the micrometeorological data is compressed using first-order differential coding based on time series prediction. When the rate of change of wind speed is greater than or equal to the first threshold, or the rate of change of temperature is greater than or equal to the second threshold, keyframe encoding based on event triggering is used to compress the tower attitude data and tower structure response data. The keyframe encoding uses the full data at the start of the drastic event as the keyframe, and uses the keyframe as the reference to encode and transmit the incremental data during the event.

[0031] Specifically, before sending the layered compressed and encoded data to the monitoring center via the satellite short message channel, the process further includes prioritizing the compressed data packets. The priority marking includes: Emergency priority is used to mark compressed data packets corresponding to abnormal values ​​in the tower structure response data that exceed the safety threshold; normal priority is used to mark compressed data packets corresponding to normal changes in the tower attitude data and micrometeorological data; the satellite short message channel sends emergency priority data packets in order of priority over normal priority data packets.

[0032] Specifically, the calculation of the predicted residual value includes: For the tower attitude data, calculate the attitude angle deviation vector between the predicted value and the actual acquired value in three-dimensional space; for the tower structural response data, calculate the stress-strain deviation between the predicted value and the actual acquired value at key parts. Using their respective preset maximum allowable variation as a benchmark, the attitude angle deviation vector and stress-strain deviation are subjected to extreme value normalization to generate dimensionless prediction residual values.

[0033] Specifically, the dynamic adjustment of the quantization precision of the predicted residual value includes: When the available bandwidth of the satellite short message channel is greater than the first threshold and the signal-to-noise ratio is greater than the third threshold, the predicted residual value is quantized with the first quantization step size, and all valid numbers are retained. When the available bandwidth of the satellite short message channel is less than the second threshold or the signal-to-noise ratio is less than the fourth threshold, the predicted residual value is quantized with a second quantization step size, the second quantization step size is greater than the first quantization step size, and the quantized data is truncated or rounded, and then sent after enabling forward error correction coding; the first threshold is greater than the second threshold. When the satellite short message channel is unavailable or interrupted, compressed data is cached on a local storage medium and transmitted in batches after the channel is restored.

[0034] Specifically, the variable-length encoding is Huffman coding or arithmetic coding; the quantized prediction residual values ​​are statistically analyzed, and codewords of different lengths are assigned according to the probability of each residual value. Residual values ​​with a high probability of occurrence are assigned short codewords, and residual values ​​with a low probability of occurrence are assigned long codewords.

[0035] Specifically, updating the tower structure mechanical coupling model based on the decoded data includes: The decoded tower attitude data and tower structural response data are compared with the predicted values ​​to calculate the prediction error of the tower structural mechanical coupling model. When the prediction error of the mechanical coupling model of the tower structure exceeds the preset tolerance for a series of consecutive preset sampling times, the online model correction process is triggered, and the mapping relationship network is retrained using the monitoring data in the most recent time window. The corrected coupling characteristic is sent to the tower monitoring device to update the prediction benchmark of the physical correlation compression.

[0036] Specifically, the structural safety status assessment includes: Based on the decoded tower attitude data and tower structural response data, the deviation between the current state of the tower and the design safety threshold is calculated; By combining the updated coupled mechanical model of the tower structure with predicted meteorological conditions as input, the response trend of the tower structure is calculated. When the deviation exceeds the warning threshold or the peak value in the structural response trend exceeds the design safety threshold, a graded warning message is generated and transmitted back to the tower monitoring device via a satellite short message channel. In response to the graded warning message, the tower monitoring device switches the sampling frequency to a third sampling frequency higher than the normal sampling frequency, switches the compression rate to a second compression rate lower than the normal compression rate, and switches the transmission frequency to a third transmission frequency higher than the normal transmission frequency.

[0037] Specifically, a tower monitoring data compression system based on short message communication includes: The monitoring device, deployed on a pole, includes a multi-source sensor array, a processor, and a satellite short message communication module. The processor has a built-in pole structure mechanical coupling model calculation unit, a physical correlation compression coding unit, and a channel adaptive compression coding unit, which are used to execute the data acquisition, coupling feature extraction, and hierarchical compression coding steps in the method, respectively. The satellite short message communication module is used to send the compressed and coded data through the satellite short message channel. A satellite communication link connects the monitoring device to the monitoring center; The monitoring center includes a data receiving and decoding module, a model updating module, and a visualization and evaluation module, which are used to execute the data receiving and decoding, mechanical coupling model updating, visualization and display, and structural safety evaluation steps in the method, respectively.

[0038] The claims and embodiments of this invention involve multiple sets of threshold, quantization step size, sampling frequency, and transmission frequency parameters. To facilitate those skilled in the art in determining the specific values ​​of these parameters, the following focuses on explaining the basis for determining each set of parameters, application scenarios, and typical value ranges.

[0039] I. Methods for determining the wind speed change rate threshold (first threshold) and temperature change rate threshold (second threshold) The wind speed change rate threshold (first threshold) and temperature change rate threshold (second threshold) are used to determine whether the current meteorological conditions are in a stable or rapidly changing state, and their determination is based on the following criteria: The method for determining the wind speed change rate threshold is as follows: Based on the structural dynamic characteristics of the monitored tower, the critical wind speed change rate at which the tower's structural response transitions from the linear to the nonlinear range is determined through wind tunnel testing or finite element dynamic analysis. When the wind speed change rate exceeds this critical value, the tower's structural response exhibits significant nonlinear characteristics, and the prediction accuracy of the tower's structural mechanical coupling model decreases. In this case, it is necessary to switch to keyframe encoding to ensure data integrity. The wind speed change rate threshold can also be adjusted according to the tower type (straight-line tower or tension tower), tower height, and the wind zone level: for tall towers located in strong wind zones, the threshold can be appropriately lowered; for short towers located in low wind zones, the threshold can be appropriately increased. The typical value range is 0.10–0.20 m / s², and in this embodiment, 0.15 m / s² is used.

[0040] The method for determining the temperature change rate threshold is as follows: based on the thermal expansion coefficient of steel and the thermal-structural coupling analysis of the tower structure, the critical temperature change rate at which temperature stress has a significant impact on the structural response is determined. When the temperature change rate exceeds this critical value, the change in structural response caused by temperature stress exceeds the tolerance range of the prediction model, and a more refined coding mode needs to be switched. The temperature change rate threshold can also be adjusted according to the season: in winter, when the diurnal temperature range is large, the threshold can be appropriately lowered; in summer, when the temperature range is small, the threshold can be appropriately increased. The typical value range is 0.5–2.0℃ / min, and in this embodiment, 1.0℃ / min is used.

[0041] II. Method for determining channel bandwidth thresholds (first threshold, second threshold) The first and second thresholds for channel bandwidth are used to determine whether the available bandwidth of the satellite short message channel is sufficient, moderate, or strained. The determination criteria are as follows: The first threshold (sufficient bandwidth threshold) is determined by calculating the maximum capacity of a single short message packet in the BeiDou-3 area (approximately 1400 bytes) and the system's expected maximum transmission frequency. When the available bandwidth is greater than or equal to the first threshold, the number of complete data packets that can be transmitted per unit time meets the maximum monitoring frequency requirement, and a high-precision quantization strategy can be adopted. Specifically, the first threshold = number of bytes per packet × expected number of transmissions per unit time. In this embodiment, a single packet is approximately 142 bytes, and the expected maximum transmission frequency is once every 10 minutes, i.e., 6 times / hour. Therefore, the first threshold ≈ 142 × 6 ≈ 850 bytes / hour, rounded down to 800 bytes / hour. The typical value range is 500–1000 bytes / hour.

[0042] The second threshold (bandwidth strain threshold) is determined by calculating it based on the system's minimum monitoring frequency requirement. When the available bandwidth is lower than the second threshold, only the minimum frequency of data transmission can be maintained, requiring a low-precision quantization strategy to compress the data volume per packet. The second threshold = minimum number of bytes per packet × minimum number of transmissions per unit time. In this embodiment, the minimum number of bytes per packet is approximately 50 bytes, and the minimum transmission frequency is once every 30 minutes, i.e., twice per hour. Therefore, the second threshold ≈ 50 × 2 = 100 bytes / hour. Considering channel fluctuations, data retransmissions, and the margin for occasional large packet transmissions, a safety factor of 3 is taken, resulting in a second threshold of 300 bytes / hour. A typical value range is 200–400 bytes / hour. The first threshold is greater than the second threshold.

[0043] III. Methods for determining the carrier-to-noise ratio thresholds (third threshold, fourth threshold) The third and fourth thresholds for carrier-to-noise ratio are used to determine whether the signal quality of the satellite short message channel is in a good or poor state, and their determination is based on the following criteria: The third threshold (good carrier-to-noise ratio threshold) is determined based on the receiving sensitivity and bit error rate requirements of the BeiDou short message communication module. When the carrier-to-noise ratio is greater than or equal to the third threshold, the bit error rate of the communication module is less than 10%. ﹣5 This ensures reliable transmission of high-precision quantized data without requiring forward error correction coding. This threshold can be obtained by consulting the communication module's technical manual or through actual channel testing. In this embodiment, 40 dBHz is used, with a typical range of 38–45 dBHz.

[0044] The fourth threshold (carrier-to-noise ratio (CNR) severity threshold) is determined based on the lowest available CNR of the communication module. When the CNR is less than the fourth threshold, the bit error rate increases significantly, requiring forward error correction coding to maintain reliable communication. This threshold can also be obtained by consulting the communication module's technical manual or through actual channel testing. In this embodiment, it is set to 30 dBHz, with a typical range of 25–35 dBHz. The third threshold is greater than the fourth threshold.

[0045] IV. Method for determining quantization step size (first quantization step size, second quantization step size, intermediate quantization step size) The first quantization step size, the second quantization step size, and the intermediate quantization step size are used to quantize the normalized prediction residual value, and their determination is based on the following criteria: The method for determining the first quantization step size (high-precision quantization step size) is as follows: it is determined by reverse calculation based on the engineering accuracy requirements of the monitoring data. Taking the tower tilt angle as an example, engineering typically requires the tilt angle monitoring accuracy to reach ±0.01°, while the normalized residual is normalized based on the maximum allowable change in tilt angle of ±5°. Therefore, the corresponding normalized quantization step size should not be greater than 0.01 / 5 = 0.002. Considering both coding efficiency and accuracy margin, 0.005 is taken in this embodiment. The typical value range is 0.001 to 0.01.

[0046] The second quantization step size (low-precision quantization step size) is determined based on channel capacity constraints and minimum monitoring accuracy requirements. When channel bandwidth is limited or battery power is insufficient, a slight reduction in data precision is permissible in exchange for improved transmission reliability or extended battery life. The second quantization step size is typically 5 to 20 times the first quantization step size. In this embodiment, it is set to 0.05, with a typical range of 0.02 to 0.10.

[0047] The intermediate quantization step size (medium precision quantization step size) is determined as a compromise between the first and second quantization step sizes, used when the channel is in a medium state. In this embodiment, it is set to 0.01, and the typical value range is 0.005 to 0.03. The relationship between the three quantization step sizes is: first quantization step size < intermediate quantization step size < second quantization step size.

[0048] V. Method for determining the transmission frequency (first transmission frequency, second transmission frequency) The first and second transmission frequencies are used to differentiate the transmission of structural response data and micro-meteorological data when the battery power is low, and their determination is based on the following criteria: The method for determining the first transmission frequency (high frequency of structural response data) is based on the real-time requirements of tower structure safety monitoring. Tower structural response (strain, displacement) is a key indicator for judging structural safety, and a high transmission frequency is required to ensure timely detection of anomalies. In this embodiment, the transmission frequency is set to once every 5 minutes, with a typical range of once every 3 to 10 minutes.

[0049] The second transmission frequency (low frequency of micro-meteorological data) is determined based on the slowness of change in meteorological parameters. Meteorological parameters such as temperature, humidity, and air pressure change slowly over a short period, so the transmission frequency can be appropriately reduced to save power. In this embodiment, it is set to once every 30 minutes, with a typical range of once every 15 to 60 minutes. The first transmission frequency is higher than the second transmission frequency.

[0050] VI. Method for determining emergency sampling parameters (third sampling frequency, second compression rate, third transmission frequency) The third sampling frequency, the second compression rate, and the third transmission frequency are used as parameter settings when the monitoring device switches to emergency sampling mode after the monitoring center triggers a graded early warning. Their determination is based on the following: The third sampling frequency is determined based on the data density required for hazard assessment. In emergency sampling mode, higher frequency data is needed to capture rapid changes in structural response. In this embodiment, the third sampling frequency for micrometeorological data is 1Hz (10 times the conventional 0.1Hz), the third sampling frequency for tower attitude data is 100Hz (5 times the conventional 20Hz), and the third sampling frequency for tower structural response data is 1000Hz (5 times the conventional 200Hz). Typical values ​​range from 5 to 20 times the conventional sampling frequency.

[0051] The second compression ratio is determined based on the principle of prioritizing data accuracy in emergency mode. During emergency sampling, data integrity is prioritized, allowing for a lower compression ratio. In this embodiment, a ratio of 3:1 is used (normally 9:1). A typical range is 2:1 to 5:1.

[0052] The third transmission frequency is determined based on the real-time requirements of the emergency response. In this embodiment, it is once every 1 minute (normally once every 10 minutes). The typical range is once every 30 seconds to 2 minutes. The third transmission frequency is higher than the normal transmission frequency.

[0053] It should be noted that the bandwidth threshold here refers to the effective data throughput of satellite short message communication, not the physical bandwidth of the channel. Because BeiDou short messages use an intermittent transmission mechanism, both the single packet capacity and transmission frequency are limited. Therefore, an effective throughput of several hundred bytes per hour is within the normal range and can meet the transmission requirements of pole monitoring data.

[0054] Example This embodiment uses a ±1100 kV ultra-high voltage direct current transmission line (hereinafter referred to as "the line") as an application scenario to illustrate the specific implementation of the method and system described in this invention.

[0055] The Gansu section of this route traverses a large amount of uninhabited areas such as the Gobi Desert and other deserts, where there is no ground communication network coverage. Some sections experience an average of 118 days of gale-force winds (level 7 or above) per year, making it the most typical application scenario for the technical solution of this invention.

[0056] I. Deployment of Monitoring Devices The monitoring devices of this invention were deployed on straight-line towers and tension towers in an uninhabited area of ​​the Gansu section of the railway line. Each tower's monitoring device consists of a base and a movable end. The base is permanently fixed to the tower's main structural member, angle iron, by welding or bolting. The movable end has a prefabricated magnetic attraction mechanism at its bottom. During on-site installation, the movable end is placed close to the base, and the magnetic attraction initially positions it. Then, two M8 bolts are used to lock the movable end to the base. The movable end has a built-in disposable high-capacity lithium battery power supply module with a rated capacity of 200Wh and is equipped with a low-power management unit. In normal monitoring mode, the total power consumption is approximately 0.5W, with a theoretical battery life of approximately 400 hours (approximately 16 days). The power monitoring unit detects the remaining power in real time. When the remaining power falls below a preset threshold (e.g., 15Wh, approximately 7.5% of the total capacity), a battery collaborative compression strategy is triggered (see Part 5 for details). The satellite short message communication module adopts the BeiDou-3 regional short message transceiver module, which supports uplink data transmission and downlink command reception, providing communication support for early warning information feedback and emergency sampling mode switching.

[0057] The mobile device can integrate the following sensor group: Micro-meteorological sensor group: ultrasonic wind speed and direction sensor (measurement range: wind speed 0~60m / s, accuracy ±0.2m / s; wind direction 0~360°, accuracy ±3°), temperature and humidity sensor (temperature -40~+80℃, accuracy ±0.3℃; humidity 0~100%RH, accuracy ±3%RH), barometric pressure sensor (300~1100hPa, accuracy ±0.5hPa); Micro-Electro-Mechanical System (MEMS) attitude sensor group: a three-axis MEMS accelerometer (range ±16g, resolution 0.06mg) and a three-axis gyroscope (range ±2000° / s, resolution 0.01° / s), used to collect the tower's three-axis tilt angle (accuracy ±0.01°) and vibration frequency (0.1~100Hz); Structural response sensor group: Fiber optic strain sensors (range ±5000με, accuracy ±1με) are installed on key stress-bearing parts of the main members of the tower legs and the main members of the crossarm under the crossarm to collect strain values; laser displacement sensors (range 0~200mm, accuracy ±0.5mm) are installed in the middle of the tower body to collect the lateral displacement of the tower under wind load.

[0058] II. Data Acquisition and Preprocessing The monitoring device acquires multi-source heterogeneous tower status data synchronously. The sampling frequency for micrometeorological data is set to 0.1Hz (once every 10 seconds), the sampling frequency for tower attitude data is set to 20Hz, and the sampling frequency for tower structural response data is set to 200Hz. These three frequencies exhibit an increasing relationship: the sampling frequency for micrometeorological data is lower than that for tower attitude data, and the sampling frequency for tower attitude data is lower than that for tower structural response data.

[0059] Data from different sampling frequencies are aligned on the time axis using linear interpolation: using the timestamp of the highest sampling frequency (200Hz) as a reference, cubic spline interpolation is performed on the 20Hz attitude data and the 0.1Hz micro-meteorological data to ensure strict temporal alignment with the 200Hz structural response data. The aligned data are then uniformly sent to the processor for further processing.

[0060] III. Establishment of the Coupled Mechanical Model of the Tower Structure Before the monitoring device is put into use, a structural mechanics coupling model of the specific monitoring tower is established. The modeling process is as follows: Based on the design drawings of the monitored tower, the tower's geometric dimensions (tower height, root opening, crossarm length, etc.), material properties (elastic modulus E=206GPa, yield strength fy=420MPa, density ρ=7850kg / m³ of Q420 high-strength steel), and cross-sectional parameters (angle steel type and corresponding cross-sectional area, moment of inertia) are obtained. Based on these parameters, a beam-rod finite element model of the tower is established in finite element software, discretizing the tower into several beam and rod elements, connected by nodes. Fixed-end constraints are applied to the tower foot nodes (constraining translational and rotational degrees of freedom in the X, Y, and Z directions), and equivalent concentrated loads (calculated based on conductor tension under design conditions) are applied to the conductor and ground wire suspension points. The first three natural frequencies of the tower (typically 0.5–3Hz) are obtained through modal analysis.

[0061] Using historical monitoring data (micrometeorological data and structural response data) from the first three months after the tower's commissioning, with wind speed (m / s), wind direction angle (°), and ambient temperature (°C) as input variables, and tower top tilt angle (°), strain at key locations (με), and lateral displacement of the tower body (mm) as output variables, a radial basis function response surface model was trained. Gaussian kernel functions were used for the radial basis functions, and the width parameter was determined through cross-validation. The number of hidden layer nodes in the model was determined by K-means clustering to locate the centers of each basis function. After training, this response surface model constitutes a mapping network of wind load, temperature field, and structural response.

[0062] In this embodiment, the coupling characteristic is defined as a partial derivative matrix between environmental load variables (including wind speed, wind direction, and temperature) and structural response variables (including tower tilt angle, strain at key components, and tower displacement). This matrix is ​​3 rows and 3 columns, where each element represents the degree of influence of a change in one environmental load variable on a specific structural response variable. For example, the element in the first row and first column represents the change in tower tilt angle for every unit change in wind speed; the element in the first row and second column represents the change in tower tilt angle for every unit change in wind direction; and the element in the first row and third column represents the change in tower tilt angle for every unit change in temperature. The elements in the second row represent the changes in strain at key components for every unit change in wind speed, wind direction, and temperature; and the elements in the third row represent the changes in tower displacement for every unit change in wind speed, wind direction, and temperature. Each element in this matrix is ​​obtained by taking the partial derivatives of the input variables using a response surface model, characterizing the sensitivity of small changes in wind speed, wind direction, and temperature to tower tilt angle, strain, and displacement. During model operation, this matrix is ​​continuously updated as monitoring data accumulates.

[0063] IV. Physical Association Compression Collect the current time (time t) micro-meteorological data—wind speed 12.5 m / s, wind direction 45°, temperature -5.3℃—and input it into the response surface sub-model of the tower structure mechanical coupling model to calculate the predicted values ​​of the tower attitude and structural response at time t+1: the predicted values ​​of the tower top tilt angle are 1.82° in the X direction and 1.15° in the Y direction, the predicted value of the strain in the key parts is +186με, and the predicted value of the lateral displacement of the tower body is 23.6 mm.

[0064] The actual tower attitude and structural response data collected at time t+1 are as follows: the actual values ​​of the tower top tilt angle are 1.79° in the X direction and 1.21° in the Y direction, the actual value of the strain in the key parts is +192με, and the actual value of the lateral displacement of the tower body is 24.1mm.

[0065] Calculate the predicted residuals: the attitude angle deviation vectors are -0.03° in the X direction and +0.06° in the Y direction, the stress-strain deviation is +6με, and the displacement deviation is +0.5mm. Extreme value normalization is performed based on their respective preset maximum allowable variations: the maximum allowable variation of tilt angle is set to ±5° (based on the tilt limit in the tower design code), the maximum allowable variation of strain is set to ±2000με (based on the yield strain of steel), and the maximum allowable variation of displacement is set to ±100mm (based on the deflection limit of the tower). After normalization, a dimensionless predicted residual vector is obtained, with four components of -0.006, 0.012, 0.003, and 0.005, corresponding to the tilt angle deviation in the X direction, the tilt angle deviation in the Y direction, the strain deviation, and the displacement deviation, respectively.

[0066] Since the coupled mechanical model of the tower structure has accurately captured the physical mapping relationship between wind speed and structural response, the predicted residual values ​​are mostly concentrated near zero (under typical operating conditions, the absolute value of the residual is less than 5% of the full scale). Transmitting only these residual values ​​is sufficient to accurately recover the original data at the decoding end, significantly reducing the amount of data transmitted. Taking a single acquisition in this embodiment as an example, the total amount of original multi-type sensor data is approximately 2.8KB (including timestamps, original values ​​of each sensor, and verification information). After physical correlation compression, the residual data size is approximately 0.3KB, with a compression ratio exceeding 9:1.

[0067] The monitoring device processor monitors the wind speed change rate and temperature change rate in real time. In this embodiment, the wind speed change in the past 10 seconds is 0.8 m / s (equivalent to a wind speed change rate of 0.08 m / s²), and the temperature change rate is 0.2℃ / min. The preset threshold for wind speed change rate is 0.15 m / s², and the preset threshold for temperature change rate is 1.0℃ / min. Since the current values ​​are both less than the corresponding thresholds, the meteorological conditions are determined to be in a "stable state." In the stable state, the micrometeorological data is compressed using first-order differential encoding: the first set of micrometeorological data at 00:00 of the day is used as the differential reference value, and only the difference between the current value and the previous value is transmitted in subsequent moments.

[0068] Suppose that the actual temperature at 03:00 on a certain day is -5.3℃, and the temperature at the previous time (02:59:50) is -5.1℃, then the difference is -0.2℃. After encoding, only this difference value (2 bytes) needs to be transmitted instead of the complete temperature value (4 bytes), which further reduces the transmission overhead of micro-meteorological data.

[0069] V. Channel Adaptive Compression and Battery Co-management The monitoring device monitors the status parameters of the BeiDou short message channel in real time: the current carrier-to-noise ratio is 42dBHz, and the historical transmission success rate is 98.5%. The preset thresholds for sufficient bandwidth are 800 bytes / hour, tight bandwidth is 300 bytes / hour, good carrier-to-noise ratio is 40dBHz, and poor carrier-to-noise ratio is 30dBHz.

[0070] The current available bandwidth is assessed at 650 bytes / hour, which is greater than the bandwidth shortage threshold (300 bytes / hour) but less than the bandwidth sufficiency threshold (800 bytes / hour). The carrier-to-noise ratio (CNR) of 42 dBHz is greater than the good CNR threshold (40 dBHz). Based on the assessment results, the channel is in a moderate state, and a medium-precision quantization strategy is adopted: the normalized residual value is quantized with an intermediate quantization step size (0.01), retaining two significant decimal places.

[0071] The four components of the normalized residual vector in the previous step were -0.006, 0.012, 0.003, and 0.005, which became -0.01, 0.01, 0.00, and 0.01 after quantization. The quantization error is within the allowable range (the maximum error does not exceed 0.005, which is much smaller than the engineering accuracy requirement of 0.01° tilt angle).

[0072] When channel conditions deteriorate—for example, during strong winds (up to 28.5 m / s, approximately force 11)—the strong winds cause a decrease in satellite communication link quality, reducing the current carrier-to-noise ratio (CNR) to 28 dBHz (below the severe CNR threshold of 30 dBHz), and the available bandwidth is assessed at 200 bytes / hour (below the bandwidth shortage threshold of 300 bytes / hour)—a low-precision quantization strategy is triggered: the quantization step size is increased from the intermediate quantization step size (0.01) to the low-precision quantization step size (0.05), and the normalized residual value is coarsely quantized. Taking a residual value of 0.006 as an example, after coarse quantization, it becomes 0.00. Although some precision is lost, the number of encoded bits is reduced from 4 bits to 1 bit. At the same time, forward error correction coding is enabled, adding a (7,4) Hamming code check bit to the data packet, enabling the data packet to be correctly decoded under low CNR conditions. If the channel is completely interrupted (carrier-to-noise ratio below 20dBHz and communication link cannot be established for more than 30 seconds), the monitoring device will cache the compressed data in the local storage medium and transmit it in batches after the channel is restored.

[0073] Meanwhile, the power monitoring unit detects the remaining power in real time. When the remaining power falls below a preset threshold (e.g., 12Wh, approximately 6% of the total capacity, resulting in less than 24 hours of battery life), a battery-coordinated compression strategy is triggered: the processor automatically adjusts the quantization precision and transmission frequency of various data types. The predicted residual values ​​corresponding to the tower structure response data (strain and displacement) are quantized with a high-precision quantization step size (0.005) to ensure that structural safety-related data does not lose accuracy due to insufficient power; the predicted residual values ​​corresponding to micro-meteorological data (temperature, humidity, wind speed, etc.) are quantized with a low-precision quantization step size (0.05), appropriately sacrificing the precision of meteorological data to save encoding bits. Simultaneously, tower structure response data compressed packets are sent at a high frequency (once every 5 minutes), while micro-meteorological data compressed packets are sent at a low frequency (once every 30 minutes), prioritizing the real-time performance of structural response data. Through these coordinated adjustments, the monitoring device can maintain the monitoring and transmission of critical data even when the power is severely insufficient.

[0074] Subsequently, probabilistic statistics were performed on the quantized residual values. Due to physical correlation compression, the residual values ​​are concentrated near zero. Statistical analysis shows that the probability of a residual value of 0.00 is approximately 65%, 0.01 is approximately 22%, -0.01 is approximately 10%, and other values ​​are approximately 3%. A variable-length code table was constructed using Huffman coding: 0.00 was assigned the codeword "0" (1 bit), 0.01 was assigned the codeword "10" (2 bits), -0.01 was assigned the codeword "110" (3 bits), and other values ​​were assigned "1110" plus the original value (5 bits). The total number of bits after encoding the four residual values ​​is 1+2+1+2=6 bits. If fixed-length coding (4 bits per value, 16 bits in total) is used, the compression rate reaches 62.5%.

[0075] VI. Priority Marking and Transmission The processor prioritizes the compressed data packets: outliers in the structural response data (strain and displacement) that exceed the safety threshold are marked as emergency priority; normal changes in tower attitude data and micrometeorological data are marked as regular priority.

[0076] In this embodiment, the strain value of 192 με is much smaller than the material yield strain of 2000 με, and the safety threshold is not triggered. The strain data is marked as normal priority. The tilt angle data is also within the normal range and is also marked as normal priority. Therefore, the current data packet is marked as normal priority. If the strain value suddenly jumps to 1800 με in a certain sampling (exceeding 80% of the preset safety threshold), the data packet is marked as emergency priority and is sent through the Beidou short message channel before other normal priority data packets in the queue.

[0077] The compressed data is encapsulated according to a preset unified frame format. The frame format is defined as follows: Field Length (bytes) Content example Frame header 2 0xAA55 Tower number 4 0x0001A2B3 Acquisition time (BCD encoding) 7 2024-01-15 14:32:05 Data length 2 0x008E (142 bytes) Data type identifier 1 0x07 (containing meteorological, attitude, and response data at the same time) Compressed data body Variable Residual vector + differential meteorological data + priority flag CRC16 check code 2 0x3F2A The total length of the encapsulated data is 142 bytes, which is less than the maximum capacity of a single packet of BeiDou-3 regional short message (approximately 1400 bytes). It is sent to the monitoring center in single packet form through the BeiDou short message channel.

[0078] VII. Monitoring Center Decoding and Model Update After receiving the data packet, the satellite short message receiving module at the monitoring center first performs a CRC16 check. If the check passes, the frame format is parsed. The step-by-step decoding module executes the following steps in reverse order of the encoding: Huffman decoding restores the quantization residual value → inverse normalization obtains the actual physical quantity residual → the residual is added to the model prediction value to restore the original acquired value.

[0079] The decoding and reconstruction results are as follows: the tower top tilt angle is 1.79° in the X direction and 1.21° in the Y direction, the strain is 192με, the displacement is 24.1mm, and the micro-meteorological data are as follows (temperature -5.3℃, humidity 62%RH, wind speed 12.5m / s, wind direction 45°, air pressure 89.7kPa).

[0080] The monitoring center's model update module compares the decoded measured values ​​with the predicted values ​​calculated by the mechanical model during the physical correlation compression stage. In this embodiment, the deviations between the predicted and measured attitude angle values ​​are -0.03° in the X direction and +0.06° in the Y direction, the strain prediction deviation is +6με, and the displacement prediction deviation is +0.5mm. All of these deviations are within the preset tolerance range (tilt angle tolerance ±0.5°, strain tolerance ±200με, displacement tolerance ±5mm), and model correction is not required.

[0081] If, during subsequent monitoring, the model prediction error exceeds the preset tolerance for 10 consecutive sampling times (i.e., one sampling point per second for 10 consecutive seconds)—for example, slow settlement of the tower foundation leading to changes in stiffness parameters and a systematically overestimation of the model's predicted values—then the online model correction process is triggered. The monitoring center uses monitoring data from the most recent 24-hour time window (including micrometeorological input and structural response output) to perform online correction of the stiffness parameters in the tower structure's mechanical coupling model using the recursive least squares method. The corrected coupling characteristic quantities (updated partial derivative matrix) are then sent to the monitoring device via the BeiDou short message channel to update the prediction benchmark for physical correlation compression, ensuring subsequent compression accuracy.

[0082] 8. 3D visualization, security assessment, and early warning linkage The monitoring center's 3D visualization module, based on decoded data, overlays and displays real-time micro-meteorological data fields (represented by wind speed vector arrows and temperature color patches), tower attitude deformation cloud maps (representing tilt angle distribution by color depth), and structural response heat maps (representing strain distribution by red gradient) onto the tower's 3D model. Based on coupled characteristic quantities, it dynamically renders structural deformation animations of the tower under the current combined wind load (12.5 m / s, 45° direction) and temperature (-5.3℃).

[0083] The structural safety assessment module performs three types of assessments simultaneously: Tilt Assessment: After decoding, the tower top tilt angle is 1.79° in the X direction and 1.21° in the Y direction, with a combined tilt angle of approximately 2.16°. The tilt limit specified in the tower design code is 3°. The current value of 2.16° is less than the warning threshold of 2.5°, and therefore does not trigger a tilt alarm.

[0084] Overload assessment: The strain at the critical part is 192με, which, when multiplied by the elastic modulus of 206GPa, yields a stress of approximately 39.6MPa. This is far less than the yield strength of Q420 steel (420MPa) (the safety margin exceeds 90%), and therefore does not trigger an overload alarm.

[0085] Resonance assessment: After decoding, the main vibration frequency is 1.2Hz, and the first natural frequency of the tower is 1.15Hz. The difference between the two is 0.05Hz, which is greater than the preset frequency threshold of 0.02Hz. Therefore, the resonance risk alarm is not triggered.

[0086] None of the three assessments triggered alarms, and the current safety status of the tower structure is assessed as "normal".

[0087] If, during subsequent monitoring, the deviation exceeds the warning threshold or the structural response trend indicates potential risk—for example, the tower top tilt angle increases continuously from 1.5° to 3.2° within 30 minutes, exceeding the tilt warning threshold of 2.5°, and combined with predicted meteorological conditions (a gale of level 8 is expected to pass through in the next 24 hours) predicts the tilt angle will reach 4.5° within the next 12 hours, exceeding the design safety threshold of 3.0°—the monitoring center generates a graded warning message, including the warning level, warning type, and recommended measures. This warning message is transmitted back to the monitoring device via the BeiDou short message channel. The satellite communication module of the monitoring device is a transceiver module. After successfully receiving the warning message, it automatically triggers the emergency sampling mode: the sampling frequency switches from 0.1Hz / 20Hz / 200Hz to 1Hz / 100Hz / 1000Hz, the compression ratio decreases from 9:1 to 3:1, and the transmission frequency increases from once every 10 minutes to once every 1 minute. In emergency sampling mode, the monitoring device collects and transmits data at a higher frequency, providing more intensive data support for the monitoring center to assess the development of the hazard and formulate a response plan. Once the tower's condition stabilizes, the monitoring center issues a release command via the BeiDou short message channel, and the monitoring device resumes its normal monitoring mode.

[0088] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for compressing tower monitoring data based on short message communication, characterized in that, include: Collect multi-source heterogeneous tower status data, which includes micro-meteorological data, tower attitude data, and tower structural response data; Based on a pre-established coupled mechanical model of the tower structure, physical correlation analysis is performed on the state data of the multi-source heterogeneous towers to extract the coupling characteristic quantities between micro-meteorological data, tower attitude data, and tower structural response data; the coupling characteristic quantities characterize the mapping relationship between wind load, temperature stress, and tower structural deformation. Based on the coupling characteristic quantity, the multi-source heterogeneous tower status data is subjected to hierarchical compression coding, the hierarchical compression coding including: Physical correlation compression: Based on the micro-meteorological data at the current moment, the predicted values ​​of the tower attitude and structural response are calculated through the tower structure mechanical coupling model. The predicted values ​​are compared with the actual collected values ​​to calculate the prediction residuals, and the prediction residual values ​​are transmitted. Channel adaptive compression: Based on the available bandwidth and signal quality of the current satellite short message channel, the quantization accuracy of the predicted residual value is dynamically adjusted, and variable-length coding is used to entropy code the quantized residual value; The layered compressed and encoded data is sent to the monitoring center via a satellite short message channel; The monitoring center decodes the received layered compressed and encoded data step by step, updates the structural mechanical coupling model of the tower based on the decoded data, and combines the tower's three-dimensional model for visualization and structural safety status assessment.

2. The method according to claim 1, characterized in that, When collecting multi-source heterogeneous tower status data, different sampling frequencies are used for different data types: the sampling frequency of micro-meteorological data is lower than that of tower attitude data, and the sampling frequency of tower attitude data is lower than that of tower structural response data; data with different sampling frequencies are interpolated and aligned on the time axis before being sent to the tower structural mechanical coupling model for physical correlation analysis.

3. The method according to claim 1, characterized in that, The channel adaptive compression further includes: adjusting the quantization accuracy of the prediction residual value and the transmission frequency of satellite short messages in coordination with the remaining power of the disposable lithium battery built into the tower monitoring device; when the remaining power is lower than a preset threshold, quantizing the prediction residual value corresponding to the tower structure response data with a first quantization step size and quantizing the prediction residual value corresponding to the micro-meteorological data with a second quantization step size, wherein the first quantization step size is smaller than the second quantization step size; and transmitting the compressed packet corresponding to the tower structure response data with a first transmission frequency and transmitting the compressed packet corresponding to the micro-meteorological data with a second transmission frequency, wherein the first transmission frequency is higher than the second transmission frequency.

4. The method according to claim 1, characterized in that, The mechanical coupling model of the tower structure is pre-established in the following way: Based on the tower design parameters and material properties, a finite element model of the tower is constructed; By training with historical monitoring data, a mapping network of wind load, temperature field, and structural response is established. The mapping network is a physical information neural network or a radial basis function response surface model.

5. The method according to claim 1, characterized in that, The physical association compression also includes: When the rate of change of wind speed is less than the first threshold and the rate of change of temperature is less than the second threshold, the micrometeorological data is compressed using first-order differential coding based on time series prediction. When the rate of change of wind speed is greater than or equal to the first threshold, or the rate of change of temperature is greater than or equal to the second threshold, keyframe encoding based on event triggering is used to compress the tower attitude data and tower structure response data. The keyframe encoding uses the full data at the start of the drastic event as the keyframe, and uses the keyframe as the reference to encode and transmit the incremental data during the event.

6. The method according to claim 1, characterized in that, Before transmitting the layered compressed and encoded data to the monitoring center via the satellite short message channel, the process further includes prioritizing the compressed data packets. The priority marking includes: Emergency priority is used to mark compressed data packets corresponding to abnormal values ​​in the tower structure response data that exceed the safety threshold; normal priority is used to mark compressed data packets corresponding to normal changes in the tower attitude data and micrometeorological data; the satellite short message channel sends emergency priority data packets in order of priority over normal priority data packets.

7. The method according to claim 1, characterized in that, The calculation of the predicted residual value includes: For the tower attitude data, calculate the attitude angle deviation vector between the predicted value and the actual acquired value in three-dimensional space; for the tower structural response data, calculate the stress-strain deviation between the predicted value and the actual acquired value at key parts. Using their respective preset maximum allowable variation as a benchmark, the attitude angle deviation vector and stress-strain deviation are subjected to extreme value normalization to generate dimensionless prediction residual values.

8. The method according to claim 1, characterized in that, The dynamic adjustment of the quantization precision of the predicted residual value includes: When the available bandwidth of the satellite short message channel is greater than the first threshold and the signal-to-noise ratio is greater than the third threshold, the predicted residual value is quantized with the first quantization step size, and all valid numbers are retained. When the available bandwidth of the satellite short message channel is less than the second threshold or the signal-to-noise ratio is less than the fourth threshold, the predicted residual value is quantized with a second quantization step size, the second quantization step size is greater than the first quantization step size, and the quantized data is truncated or rounded, and then sent after enabling forward error correction coding; the first threshold is greater than the second threshold. When the satellite short message channel is unavailable or interrupted, compressed data is cached on local storage media and transmitted in batches after the channel is restored.

9. The method according to claim 1, characterized in that, The variable-length encoding is either Huffman coding or arithmetic coding; the quantized prediction residual values ​​are statistically analyzed, and codewords of different lengths are assigned according to the probability of each residual value. Residual values ​​with a high probability of occurrence are assigned short codewords, and residual values ​​with a low probability of occurrence are assigned long codewords.

10. The method according to claim 1, characterized in that, The process of updating the tower structure mechanical coupling model based on decoded data includes: The decoded tower attitude data and tower structural response data are compared with the predicted values ​​to calculate the prediction error of the tower structural mechanical coupling model. When the prediction error of the mechanical coupling model of the tower structure exceeds the preset tolerance for a series of consecutive preset sampling times, the online model correction process is triggered, and the mapping relationship network is retrained using the monitoring data in the most recent time window. The corrected coupling characteristic is sent to the tower monitoring device to update the prediction benchmark of the physical correlation compression.

11. The method according to claim 1, characterized in that, The structural safety status assessment includes: Based on the decoded tower attitude data and tower structural response data, the deviation between the current state of the tower and the design safety threshold is calculated; combined with the updated tower structural mechanical coupling model, and with predicted meteorological conditions as input, the tower structural response trend is calculated. When the deviation exceeds the warning threshold or the peak value in the structural response trend exceeds the design safety threshold, a graded warning message is generated and transmitted back to the tower monitoring device via a satellite short message channel. In response to the graded warning message, the tower monitoring device switches the sampling frequency to a third sampling frequency higher than the normal sampling frequency, switches the compression rate to a second compression rate lower than the normal compression rate, and switches the transmission frequency to a third transmission frequency higher than the normal transmission frequency.

12. A tower monitoring data compression system based on short message communication, characterized in that, The system for performing the method of any one of claims 1 to 11 includes: The monitoring device, deployed on a pole, includes a multi-source sensor array, a processor, and a satellite short message communication module. The processor has a built-in pole structure mechanical coupling model calculation unit, a physical correlation compression coding unit, and a channel adaptive compression coding unit, which are used to execute the data acquisition, coupling feature extraction, and hierarchical compression coding steps in the method, respectively. The satellite short message communication module is used to send the compressed and coded data through the satellite short message channel. A satellite communication link connects the monitoring device to the monitoring center; The monitoring center includes a data receiving and decoding module, a model updating module, and a visualization and evaluation module, which are used to execute the data receiving and decoding, mechanical coupling model updating, visualization and display, and structural safety evaluation steps in the method, respectively.

Images