A method and system for multi-source data integration and visualization monitoring of cable hoisting in mountainous areas
By establishing wireless relay nodes and a mesh transmission network in the cable hoisting system in mountainous areas, spatiotemporal alignment and reliability assessment of multi-source data were achieved, solving the problems of inconsistent data formats and signal obstruction, ensuring the integrity and reliability of the data, and providing visual support for construction safety monitoring.
Patent Information
- Application Number
- CN202611620877.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-20
- Publication Date
- 2026-11-17
AI Technical Summary
In mountainous cable hoisting systems, the inconsistent formats, time bases, and spatial coordinate systems of multi-source data prevent direct data fusion and use. The terrain in mountainous areas causes frequent data packet loss due to wireless signal obstruction. Furthermore, the lack of data reliability assessment methods affects the data support capabilities of the visualization monitoring system.
Wireless relay nodes are established on the tops of the towers on both sides of the hoisting area to build a wireless mesh acquisition and transmission network covering the entire hoisting area. Multi-source monitoring data is collected in real time and additionally equipped with device identification codes and timestamps. Spatiotemporal alignment and packet loss detection are performed. After correcting the data to a unified benchmark through spatial-temporal joint reconstruction compensation and comprehensive reliability assessment, the data is visualized in a 3D scene model.
It enables correlation analysis of multi-source data under a unified spatiotemporal framework, repairs data loss caused by signal obstruction, ensures data integrity and reliability, and provides a reliable data foundation to support construction safety monitoring.
Smart Images

Figure CN123123408A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a method and system for integrating and visually monitoring multi-source data of cable hoisting in mountainous areas. Background Technology
[0002] Mountain cable-stayed hoisting systems involve various monitoring data, including main load-bearing cable force, traction cable force, hoisting point load, trolley positioning, component attitude, motor status, and ambient wind speed, all originating from different types of sensors and positioning terminals. Existing technologies face several challenges in processing this multi-source data: First, the inconsistent formats, sampling frequencies, time bases, and spatial coordinate systems of the data sources prevent direct data fusion. Second, wireless signal obstruction in mountainous canyon terrain leads to frequent packet loss, compromising data integrity. Third, differences in installation location, spatial attitude, and ambient temperature among different data sources result in a lack of physical comparability. Fourth, the lack of quantitative assessment methods for the reliability of various data sources allows low-quality data to potentially contaminate the overall dataset. These issues hinder the effective integration of multi-source monitoring data into a unified and reliable comprehensive dataset, limiting the data support capabilities of visualization monitoring systems. Therefore, a data integration and visualization monitoring method specifically tailored to the characteristics of multi-source data from mountain cable-stayed hoisting systems is urgently needed. Summary of the Invention
[0003] This invention provides a method and system for integrating and visually monitoring multi-source data in cable hoisting in mountainous areas, which can effectively solve the problems in the background art.
[0004] This invention provides a method for multi-source data integration and visualization monitoring of cable hoisting in mountainous areas, comprising the following steps: S1: Establish wireless relay nodes on the tops of the towers on both sides of the hoisting area and construct a wireless mesh data acquisition and transmission network covering the entire hoisting area; S2: Real-time acquisition of multi-source monitoring data during cable hoisting process; S3: Add a device identification code, a serial number that increments independently with the device identification code, and a timestamp to the monitoring data collected from each data source to form a raw data packet, and perform spatiotemporal alignment of all raw data packets under a unified time reference and spatial coordinate system to form an aligned data packet; S4: Detect packet loss in each data source based on the continuity of sequence numbers in the aligned data packets and perform spatial-temporal joint reconstruction compensation on the monitoring data; S5: Calculate the overall credibility of each data source from three dimensions: continuous arrival rate, statistical consistency rate, and spatial consistency rate, and label the credibility of the monitoring data of each data source accordingly; at the same time, uniformly correct the cable force value of the main load-bearing cable to the same spatial and temperature reference. S6: Combine the overall credibility and timeliness of each data source to obtain the fused representative value of each data type, and synchronously map and display the fused representative value in the 3D scene model.
[0005] Furthermore, in step S2, the multi-source monitoring data includes at least the following data types: the cable force value of each main load-bearing cable, the temperature of each main load-bearing cable, and the weight of each suspension point; and each data type has at least two data sources.
[0006] Furthermore, in step S3, the spatiotemporal alignment of all raw data packets specifically involves: The original data packet includes at least the device identification code field, timestamp field, data type field, serial number field, and monitoring data field; After receiving the raw data packets from each data source, the relay node extracts the timestamps of each raw data packet. Using the relay node's time as a unified time reference, the timestamps of each raw data packet are converted into the relay node's time value. A hoisting area coordinate system is established with the pre-established reference point position as the origin, the longitudinal direction of the bridge as the X-axis, the transverse direction of the bridge as the Y-axis, and the vertical direction as the Z-axis. The position coordinates and positioning data of each data source are uniformly converted into the hoisting area coordinate system to form hoisting area coordinates, which are then appended to the raw data packets to form aligned data packets.
[0007] Furthermore, based on the continuity of sequence numbers in the aligned data packets, the packet loss situation of each data source is detected, and spatial-temporal joint reconstruction compensation is performed on the monitoring data, specifically as follows: After grouping the alignment data packets by device identifier, check the continuity of the sequence number of each group of alignment data packets. When a sequence number jump of the alignment data packet is detected, record the corresponding data source and the time of loss. Suppose that the r-th data source loses data at time t, and there are M data sources corresponding to the data type of the r-th data source, with the most recent data x before and after it. r (t prev ) and x r (t next ); Calculate time linear interpolation x time,r (t)=x r (t prev )+[x r (t next )-x r (t prev )]×(tt prev ) / (t next -t prev ); Calculate the spatial correlation coefficient ρ between the j-th data source and the r-th data source in this data type. rj =1 / (1+d rj / dref ); In the formula, d rj Let d be the spatial distance between the j-th data source and the r-th data source. ref This is the arithmetic mean of the pairwise distances between all data sources in this data type; Calculate the spatial estimate x space,r (t): ; In the formula, x j (t) represents the monitoring data of the j-th data source in this data type at time t; Calculate the confidence level τ of time interpolation time =exp(-λ·T loss ); In the formula, λ is the attenuation coefficient, and T loss Duration of loss; Computational spatial estimation confidence ; Calculate the joint compensation value x comp,r (t): x comp,r (t)=[τ time / (τ time +τ space )]·x time,r (t)+[τ space / (τ time +τ space )]·x space,r (t); x comp,r (t) is used as monitoring data and is combined with the missing sequence number to form the missing aligned data packet.
[0008] Further, in step S5, the overall credibility of each data source is calculated, specifically as follows: Set the monitoring period T; Calculate the continuous arrival rate R of the i-th data source. arr,i =N rec,i / N exp,i ; In the formula, N rec,i N represents the total number of aligned data packets successfully received from the i-th data source within time period T. exp,i This represents the total number of raw data packets that the i-th data source should send within time period T; Calculate the statistical similarity rate R of the same source. sta,i =M cnt,i / N rec,i ; In the formula, M cnt,i For the monitoring data of the i-th data source within time period T, satisfy M avg,i -2Sstd,i ≤x i,k ≤M avg,i +2S std,i The number of packages, M avg,i S is the average value of the monitoring data from the i-th data source within time period T. std,i Let be the standard deviation of the monitoring data from the i-th data source within time period T; Calculate the spatial consistency rate R spa,i =1-|log2(x i / G avg,i )|; In the formula, G avg,i This is the geometric mean of the monitored data from all data sources in this data type at the same time; when R... spa,i When R is less than 0, take R. spa,i =0, when R spa,i When the value is greater than 1, take R. spa,i =1; Calculate the overall credibility of the i-th data source: C i =(R arr,i ) 0.60 ·(R sta,i ) 0.25 ·(R spa,i ) 0.15 .
[0009] Furthermore, in step S5, the cable force values of the main load-bearing cables are uniformly corrected to the same spatial and temperature references, specifically as follows: Calculate the spatial position correction coefficient K for the k-th main load-bearing cable. geo,k =1+Kfr k ·Ang k ; In the formula, Kfr k Let be the coefficient of friction between the k-th main load-bearing cable and the saddle roller; Ang k Let be the contact wrap angle of the k-th main load-bearing cable on the saddle roller; Calculate the lever arm balance correction coefficient K for the k-th main load-bearing cable. arm,k =l k / l ref ; In the formula, l k Let l be the lever arm length of the k-th main load-bearing cable on the balance wheel. ref Let be the reference lever arm length of the k-th main load-bearing cable; Calculate the temperature correction factor K for the k-th main load-bearing cable. temp,k =1+α steel ·(T avg,k -T 0,k ); In the formula, α steelThe coefficient of thermal expansion of the main load-bearing cable, T avg,k Let T be the current temperature of the k-th main load-bearing cable. 0,k Initial installation temperature; Comprehensive correction factor and physical reference correction: W cal,k =K geo,k ·K arm,k ·K temp,k ; F cal,k =W cal,k ·F k ; In the formula, W cal,k F is the comprehensive correction factor for the k-th main load-bearing cable. cal,k F is the cable force value after physical benchmark correction for the k-th main load-bearing cable. k This represents the original monitored cable force value of the k-th main load-bearing cable; After the calculation is completed, F cal,k The force value of the k-th main load-bearing cable will be used for subsequent steps.
[0010] Furthermore, in step S6, the fusion representative value of each data type is obtained by combining the overall credibility and timeliness of each data source as follows: Calculate the timeliness factor ΔT for each data source factor,m =1 / (1+η·Δt m ); In the formula, ΔT factor,m Let η be the timeliness factor of the m-th data source, η be the time delay attenuation coefficient, and Δt be the time factor. m This represents the time delay from the data source acquisition time to the current processing time. All data sources are divided into multiple data types. For data type d, let N be the total number of data types. d There are 1 data source; the overall credibility of the m-th data source is C. m ; Calculate the fusion weight W of the m-th data source fus,m : ; Calculate the representative value of data type d at the current moment: ; In the formula, Y cal,m This refers to the monitoring data from the m-th data source; The combined representative value of all data types and the overall credibility C of each data source. i The coordinates of the hoisting areas corresponding to each data source are encapsulated into a comprehensive monitoring dataset.
[0011] Furthermore, in step S6, the synchronous mapping and display of the fused representative values in the 3D scene model specifically involves: Construct a 3D scene model of the cable hoisting area; The fusion representative value of each data type in the comprehensive monitoring dataset is mapped to the corresponding position in the 3D scene model according to the average value of the hoisting area coordinates of each data source in that data type; and the mapping color of the fusion representative value is determined according to the range of the average value of the comprehensive credibility of each data source in that data type.
[0012] Furthermore, in step S6, the comprehensive monitoring dataset is transmitted to the cloud platform for storage. The remote terminal accesses the cloud platform through a webpage to obtain real-time data, which is then synchronously displayed in the 3D scene model on the webpage.
[0013] The present invention also provides a multi-source data integration and visualization system for cable hoisting in mountainous areas, including a storage unit and a processor. The storage unit is used to store one or more program instructions; the processor is used to run one or more program instructions to execute the steps of the above-mentioned multi-source data integration and visualization monitoring method for cable hoisting in mountainous areas.
[0014] The technical solution of this invention can achieve the following technical effects: This invention addresses the issue of inconsistent data formats, time bases, and spatial coordinate systems among multiple sources used in cable hoisting in mountainous areas. Through unified data format conversion, BeiDou time synchronization, and 3D spatial coordinate mapping, it enables correlation analysis of various data sources within a unified spatiotemporal framework. Furthermore, to address packet loss caused by mountainous terrain, this invention designs a spatial-temporal joint compensation mechanism based on exponential decay weights. This mechanism dynamically adjusts the weight ratio of time interpolation and spatial estimation based on the duration of data loss, effectively repairing data gaps caused by signal obstruction. It also comprehensively evaluates the reliability of each data source, automatically identifying and downweighting low-quality data to prevent contamination of the fusion results. This achieves seamless integration of multi-source monitoring data from acquisition and integration to visual monitoring in mountainous cable hoisting, providing a reliable data foundation for construction safety monitoring. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] Figure 1 A flowchart illustrating a method for integrating and visually monitoring multi-source data in cable hoisting in mountainous areas; Figure 2A logic diagram for a method of integrating and visually monitoring multi-source data for cable hoisting in mountainous areas. Detailed Implementation
[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0019] This invention relates to a method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas, such as... Figures 1-2 As shown, the method comprises six steps, S1 to S6. Through these six steps, the scattered raw data is progressively transformed into a comprehensive monitoring dataset with unified format, spatiotemporal alignment, missing data compensation, confidence labeling, and physical correction. Finally, it achieves synchronous visualization in a 3D scene model, forming a complete processing chain from raw data to usable information. The specific details of each step are as follows: S1: Establish wireless relay nodes on the tower tops on both sides of the hoisting area and construct a wireless mesh data acquisition and transmission network covering the entire hoisting area. The cable hoisting area spans both sides of a canyon, with a large spatial span. Sensors are distributed across multiple locations, including the tower tops, main cables, trolleys, hoists, and winches. Wired transmission would be difficult and costly; single-point wireless coverage would suffer from severe signal obstruction due to the mountainous terrain. This step addresses this by establishing relay nodes on the tower tops on both sides, allowing each sensor to connect to the nearest relay node, avoiding signal attenuation over long distances. A backbone link is formed between the relay nodes on the two tower tops, and all relay nodes are interconnected to form a mesh topology. The advantage of a mesh topology is that when a link is blocked or a node fails, data can be transmitted via alternative paths, and the overall reliability of the system does not depend on any single link.
[0020] S2: Real-time acquisition of multi-source monitoring data during cable hoisting. During cable hoisting operations, it is necessary to monitor changes in several physical quantities: the tension in the main load-bearing cables reflects the distribution of the hoisting load among the main load-bearing cables; the temperature of each main load-bearing cable reflects the influence of ambient temperature on the linear expansion of the cable material, serving as a necessary input for subsequent temperature correction of the tension value; the lifting weight at the four hoisting points reflects the attitude and balance state of the hoisted component. These data come from different types of sensors, collected independently, and together constitute a multi-source monitoring dataset. In addition, data such as wind speed and winch temperature can be additionally monitored as needed.
[0021] S3: Add a device identification code, a serial number that increments independently with the device identification code, and a timestamp to the monitoring data collected from each data source to form a raw data packet. Then, perform spatiotemporal alignment on all raw data packets under a unified time reference and spatial coordinate system to form an aligned data packet.
[0022] The raw data output by each sensor has different formats, sampling frequencies, time bases, and spatial coordinate systems. Without standardization, these data cannot be correlated and analyzed temporally and spatially. This step achieves data traceability by attaching a device identification code to each set of data; it achieves packet loss detection by attaching a serial number that increments independently with the device identification code; and it achieves spatiotemporal alignment of multi-source data by unifying the time base and spatial coordinate system, enabling any two sets of data to be correlated and analyzed at the same time and spatial location.
[0023] S4: Detect packet loss in each data source based on the continuity of sequence numbers in the aligned data packets and perform spatial-temporal joint reconstruction compensation on the monitored data. In mountainous environments, wireless signals are easily lost due to object obstruction and excessive transmission distance. This step determines which data packets are lost by detecting the continuity of sequence numbers, and then reconstructs the data by combining information from both the temporal dimension (the historical trend of the data source itself) and the spatial dimension (synchronous measurements from other data sources within the same data type), effectively repairing data loss caused by signal obstruction.
[0024] S5: Each data source has three evaluation dimensions: continuous arrival rate, statistical consistency rate, and spatial consistency rate. The continuous arrival rate reflects the reliability of the communication link; the statistical consistency rate reflects the stability of the sensor itself; and the spatial consistency rate reflects the consistency between the data source and other data sources of the same type. Each data source can obtain a comprehensive credibility score through these three dimensions, which reflects the reliability of the communication link, and the credibility of the monitoring data from each data source is labeled accordingly. Simultaneously, the cable force values of the main load-bearing cables are uniformly calibrated to the same spatial and temperature references, eliminating monitoring deviations caused by factors such as monitoring equipment and installation environment, thereby restoring the original readings to the true stress on the main load-bearing cables.
[0025] S6: After the aforementioned steps, multiple data sources of the same data type each provide monitoring data that has been compensated, credited, and physically corrected. This step combines the overall creditworthiness and timeliness of each data source to obtain a fused representative value for each data type. This value, along with auxiliary information such as the overall creditworthiness of each data source and the coordinates of the hoisting area, is encapsulated into a comprehensive monitoring dataset and mapped onto a 3D scene model for visualization.
[0026] Preferably, in step S2, multi-source monitoring data is collected in real time during the cable hoisting process, specifically as follows: Multi-source monitoring data is categorized into multiple data types based on physical quantities. A data type refers to a collection of multiple data sources measuring the same physical quantity. For example, multiple force sensors are installed on the main load-bearing cable, each sensor acting as a data source and outputting the force value of one main load-bearing cable. Together, they constitute the data type "force value of the main load-bearing cable." When there are multiple main load-bearing cables, each cable is treated as a separate data type, such as "force value of main load-bearing cable #1," "force value of main load-bearing cable #2," etc. Multi-source monitoring data must include at least the following data types: force value of each main load-bearing cable, temperature of each main load-bearing cable, and weight at each suspension point; and each data type must have at least two data sources to ensure multiple measurement samples for the same physical quantity.
[0027] Preferably, in step S3, spatiotemporal alignment of all raw data packets is performed as follows: Because the data sources are collected from different sensors, the raw data output by each sensor has a different time base (due to deviations in the internal clocks of each sensor) and a different spatial coordinate system (each sensor records data using its own installation position as a reference point). These differences prevent direct correlation analysis of data from different sensors. For example, the "1150kN at 10:00:00" recorded by the cable sensor and the "vehicle position at 10:00:00" recorded by the positioning sensor may not represent the actual state at the same time, as their internal clocks may differ by hundreds of milliseconds. This step addresses this issue through four layers of processing: attaching a device identifier to each raw data packet for data traceability; attaching a serial number that increments independently with the device identifier to each raw data packet for packet loss detection; unifying the timestamps of each raw data packet to the same time base; and unifying the spatial coordinates of each data source to the same spatial coordinate system. The aligned data packets formed after these four layers of processing allow any two to be aligned on the time axis by timestamp and correlated in spatial location by spatial coordinates, providing a unified data foundation for subsequent packet loss detection, spatial correlation analysis, and 3D visualization mapping.
[0028] The raw data packet includes at least the device identification code field, timestamp field, data type field, serial number field, and monitoring data field. For ease of understanding, take a complete set of data as an example: the device identification code "01-No4-NET03" indicates the cable force sensor (01), the 4th main load-bearing cable (No4), and the device with the network address NET03; the timestamp "08-07-14:32:15.328" indicates the time of data collection; the data type "cable force" indicates that the data belongs to the main load-bearing cable force value data type; the serial number "117" indicates that the device has sent the 7th raw data packet; and the monitoring data field value is "1150".
[0029] The device identifier includes the device type number (e.g., "01" for a tension sensor, "02" for a temperature sensor, and "03" for a weight sensor), the installation location number, and the device address within the network. These three elements combine to form a globally unique device identifier. Relay nodes assign a uniquely incrementing sequence number to each raw data packet, based on the device identifier. Each device identifier has its own independent sequence number space, starting from 111 and incrementing. The purpose of the sequence number is to allow for the identification of packet loss by checking the continuity of sequence numbers for the same device identifier. For example, if raw data packets with sequence numbers 111, 112, and 114 are received but sequence number 113 is missing, it can be determined that packet 113 is lost. Furthermore, if a break (or partial loss) occurs during transmission, the next node will receive 0 data, and the sequence number will become something like 003, 103, or 013. This allows for the timely identification of data packets with breakpoints or low reliability.
[0030] Each relay node synchronizes its time using the BeiDou satellite system time as a reference, ensuring time consistency across all relay nodes. After receiving raw data packets from various data sources, each relay node extracts the timestamp of each raw data packet and converts it into its own time value, using its own time as a unified time reference.
[0031] Each sensor records data with its own installation location as the origin. The cable tension sensor records the force at its location, while the positioning sensor records the coordinates of the positioning antenna. These locations are all different. Without a unified spatial coordinate system, the spatial relationship between the data from different sensors cannot be determined. For example, the position data of the upstream and downstream vehicles come from two independent positioning terminals. Only after converting to the same coordinate system can the actual distance between them be calculated. The relay node uses the pre-established reference point as the origin and establishes a hoisting area coordinate system with the longitudinal direction of the bridge as the X-axis, the transverse direction as the Y-axis, and the vertical direction as the Z-axis. It then converts the position coordinates and positioning data from each data source into the hoisting area coordinate system to form hoisting area coordinates, which are then appended to the original data packet to form an aligned data packet.
[0032] Each aligned data packet, after spatiotemporal alignment, contains five standardized basic fields plus a hoisting area coordinate field. Any two aligned data packets can be aligned along the timeline by timestamp and spatially correlated by hoisting area coordinates, providing a unified data foundation for subsequent steps such as packet loss detection, spatial correlation coefficient calculation, and 3D visualization mapping.
[0033] Preferably, in step S4, the detection of packet loss in each data source based on the continuity of sequence numbers in the aligned data packets and the spatial-temporal joint reconstruction compensation of the monitoring data specifically involves: After grouping the alignment data packets by device identifier, check the continuity of the sequence number in each group of alignment data packets. For the same device identifier, the sequence number should be a continuously increasing integer sequence. When a sequence number jump is detected (e.g., receiving sequence numbers 111, 112, and 114, but missing 113), it can be determined that the corresponding alignment data packet (sequence number 113) was lost during transmission, and the corresponding data source and the time of loss are recorded.
[0034] Suppose that the r-th data source loses data at time t, and there are M data sources corresponding to the r-th data source. Let the times of the most recent valid data before and after the time of data loss be t and t', respectively. prev and t next The corresponding data value is x r (t prev ) and x r (t next ).
[0035] First, time-linear interpolation is performed using the most recent valid data before and after the data source loss. When data from a data source is lost at time t, the physical quantities measured by that data source typically change slowly over continuous time, without abrupt changes. prev to t next Within a given time interval, the physical quantity can be approximated as changing linearly with time. Time-linear interpolation uses the two most recent valid data points before and after the data loss to determine the slope of this linear change, and then calculates the slope from t... prev Starting from the value at time t, calculate the value x lost at time t along this slope. time,r (t): x time,r (t)=x r (t prev )+[x r (t next )-x r (t prev )]×(tt prev ) / (t next -t prev ).
[0036] Within the same data type, different data sources monitor samples of the same physical quantity at different spatial locations. This physical quantity is typically continuously distributed in space, and values at adjacent locations do not show drastic differences. Spatial estimation of missing data can be performed using synchronized data from other data sources within the same data type. First, it is necessary to calculate the spatial correlation coefficient ρ between the j-th and r-th data sources within this data type. rj =1 / (1+d rj / d ref ); In the formula, d rj Let d be the spatial distance between the j-th data source and the r-th data source. ref This is the arithmetic mean of the pairwise distances between all data sources in this data type; ρ rj The reference value of other data sources of the same type to the lost data source was quantified. The closer other data sources are to the lost data source, the higher their spatial correlation coefficient, and the greater their weight in spatial inference.
[0037] The spatially estimated value x is obtained by weighting the values of other data sources in this data type at the same time using spatial correlation coefficients. space,r (t): ; In the formula, x j (t) represents the monitoring data of the j-th data source in this data type at time t; x space,r (t) is the spatial interpolation result obtained by weighting measurements from other data sources of the same type at the same time according to spatial distance. When the data sources of the same data type are densely distributed, the accuracy of spatial inference is high.
[0038] The reliability of time interpolation decreases as the duration of data loss increases. When data loss first occurs, the physical quantity has not yet changed significantly, and the time interpolation is close to the true value. However, as the loss time increases, the physical quantity may have actually changed, but the time interpolation is still based on the trend before the loss, leading to a gradual increase in accumulated error. Therefore, it is necessary to calculate the confidence level τ of the time interpolation. time =exp(-λ·T loss Simulate this smooth decrease in confidence over time; In the formula, λ is the attenuation coefficient, taken as 2 / s; T loss The duration of the loss is expressed in seconds (s). τ time This reflects the reliability of the time interpolation at the current moment. The shorter the duration of the data loss, the more reliable the time interpolation; the longer the duration of the data loss, the greater the cumulative error of the time interpolation, and the lower the reliability.
[0039] The reliability of spatial inference depends on the reference value of other data sources of the same type to the lost data source. If the spatial correlation coefficients between other data sources of the same type and the lost data source are generally high (i.e., the data sources are densely distributed), the reliability of the spatial inference is high; if the correlation coefficients are generally low (i.e., the data sources are sparsely distributed), the reliability of the spatial inference is low. Therefore, it is necessary to calculate the confidence level of the spatial inference. This means taking the average of the spatial correlation coefficients of all other data sources, reflecting the overall value of the entire data type as a spatial reference.
[0040] Joint compensation value x comp,r (t) according to τ time and τ space The weighted average of the two is calculated using the following formula: x comp,r (t)=[τ time / (τ time +τ space )]·x time,r (t)+[τ space / (τ time +τ space )]·x space,r (t).
[0041] After the calculation is complete, x comp,r (t) is used as monitoring data and is combined with the missing sequence number to form the missing aligned data packet.
[0042] Traditional timeout retransmission mechanisms require waiting for confirmation of a timeout before retransmission after each packet loss. When the signal is consistently poor, this leads to a continuous accumulation of data delays, meaning operators see data from seconds or even tens of seconds ago, completely failing to meet real-time monitoring requirements. This new step does not rely on retransmissions. Instead, it detects packet loss instantly by checking the continuity of sequence numbers, then uses existing data for compensation and reconstruction. It does not wait for any confirmation messages; packet loss is detected and compensated for the instant the data packet arrives, without introducing additional delays. Furthermore, it utilizes spatial dimension information to avoid the cumulative errors of temporal interpolation over long periods of data loss.
[0043] Preferably, in step S5, the overall credibility of each data source is calculated, specifically as follows: The monitoring period T is set, usually the most recent 60 seconds as the evaluation window. 60 seconds is short enough for the hoisting process to reflect the data quality status of the current period without introducing outdated historical information.
[0044] Calculate the continuous arrival rate R of the i-th data source. arr,i =N rec,i / N exp,i ; In the formula, N rec,iN represents the total number of aligned data packets successfully received from the i-th data source within time period T. exp,i This represents the total number of raw data packets that the i-th data source should send within time period T; R arr,i The closer the value is to 1, the more stable the transmission link of the data source; R arr,i A value significantly lower than 1 indicates that the data source is located in a place where it is blocked by mountains or too far from the relay node, resulting in poor communication quality and a corresponding decrease in the reliability of subsequent data.
[0045] Calculate the statistical similarity rate R of the same source. sta,i =M cnt,i / N rec,i ; In the formula, M cnt,i For the monitoring data of the i-th data source within time period T, satisfy M avg,i -2S std,i ≤x i,k ≤M avg,i +2S std,i The number of packages, M avg,i S is the average value of the monitoring data from the i-th data source within time period T. std,i Let be the standard deviation of the monitoring data from the i-th data source within time period T; R sta,i A value closer to 1 indicates that the time-series fluctuations of the monitored data source are stable, with no jumps exceeding the normal statistical range, and that the sensor itself is functioning normally; R sta,i A value significantly lower than 1 indicates that the data source may be experiencing sensor malfunction or severe environmental interference.
[0046] Calculate the spatial consistency rate R spa,i =1-|log2(x i / G avg,i )|; In the formula, G avg,i This is the geometric mean of the monitored data from all data sources in this data type at the same time; when R... spa,i When R is less than 0, take R. spa,i =0, when R spa,i When the value is greater than 1, take R. spa,i =1; Within the same data type, different data sources monitor different spatial samples of the same physical quantity (e.g., multiple main load-bearing cables share the same hoisting load; their cable force values should have consistent spatial distribution, although there may be differences, but the deviation should not be too large). Therefore, the monitoring data from different data sources should have spatial consistency. spa,i The closer the value is to 1, the more consistent the monitoring data of this data source is with the numerical level of other data sources in this data type at the same time; R spa,iA value significantly lower than 1 indicates a large discrepancy between this data source and other data sources, possibly due to loose installation of the data acquisition equipment or abnormal local stress. Statistical consistency and spatial consistency rates are calculated based on data corrected for physical reference.
[0047] Calculate the overall credibility of the i-th data source: C i =(R arr,i ) 0.60 ·(R sta,i ) 0.25 ·(R spa,i ) 0.15 ; C i Using a product approach, the overall reliability of the data source should be limited if any one of the three dimensions has a serious problem; for example, intermittent communication links (R... arr,i (Low) Even if the sensor readings are very stable and the spatial consistency is excellent, if the data cannot arrive in time, it is still unusable. C i A score below 0.60 is marked as a low-trust data source; C i Data with a value below 0.35 is marked as a discarded data source and will not participate in subsequent fusion calculations.
[0048] Preferably, in step S5, the cable force value of the main load-bearing cable is uniformly corrected to the same spatial and temperature reference. Specifically, this involves: The main load-bearing cable is composed of multiple main load-bearing cables connected in parallel. During actual installation, due to differences in the contact wrap angle with the saddle rollers, the lever arm length on the balance wheel, and the ambient temperature, the monitoring data of each main load-bearing cable exhibits systematic deviations from the physical benchmark. If used directly without calibration, even if the actual force on each main load-bearing cable is the same, the monitoring data will differ due to these physical factors, leading to misjudgments. This step uses triple calibration to eliminate the influence of these physical factors one by one, unifying the force values of each main load-bearing cable to the same physical benchmark.
[0049] Calculate the spatial position correction coefficient K for the k-th main load-bearing cable. geo,k =1+Kfr k ·Ang k ; In the formula, Kfr k Let be the coefficient of friction between the k-th main load-bearing cable and the saddle roller (determined based on the material and lubrication condition of the saddle roller, with a value ranging from 0.01 to 0.05); Ang k The contact wrap angle of the k-th main load-bearing cable on the saddle roller is expressed in radians (rad) and ranges from 0 to π.
[0050] K geo,kThe calculation principle is as follows: When the main load-bearing cable passes over the saddle roller, the tension on the rope-out side due to friction between the main load-bearing cable and the roller surface is less than the force transmission loss on the rope-in side. The friction loss is directly proportional to the product of the friction coefficient and the contact wrap angle. The larger the contact wrap angle, the longer the contact arc length between the main load-bearing cable and the roller, and the greater the friction loss. If the data acquisition equipment is located on the rope-out side, the measured main load-bearing cable force value has already deducted the friction loss and is lower than the true main load-bearing cable force value, requiring amplification correction. The correction factor is 1 plus the friction loss Kfr. k ·Ang k The ideal state with no frictional loss is used as a unified physical benchmark. The original monitoring data is multiplied by this coefficient to compensate for the force transmission loss caused by cable saddle friction, restoring it to the actual main load-bearing cable force level.
[0051] Calculate the lever arm balance correction coefficient K for the k-th main load-bearing cable. arm,k =l k / l ref ; In the formula, l k l is the lever arm length of the k-th main load-bearing cable on the balance wheel, in meters (m); ref The reference lever arm length is expressed in meters (m), and is usually taken as the average of the lever arm lengths of all main load-bearing cables in this data type.
[0052] K arm,k The calculation principle is as follows: The balance wheel is a circular component that collects the forces of each main load-bearing cable onto the lifting device. Each main load-bearing cable is fixed to a different radial position on the balance wheel through different connection points. When the balance wheel is in equilibrium, the sum of the torques of each main load-bearing cable on the balance wheel is zero. For the k-th main load-bearing cable, the torque it generates is F. k ·l k If the actual forces on the two main load-bearing cables are the same (both are F), but the lever arm lengths are different (l... A ≠l B If they do not contribute the same amount of torque to the balance wheel (F·l), then their contributions to the torque of the balance wheel will be different. A ≠F·l B To maintain the balance of the balance wheel, the sensor readings are adjusted accordingly, resulting in variations in the correspondence between the readings and the actual force depending on the lever arm. Without correction, two main load-bearing cables with the same actual force will display different readings. Using the reference lever arm length as a unified physical standard, a correction factor K is established. arm,k Take l k / l ref K arm,k >1 indicates that the lever arm of the main load-bearing cable is less than the reference value, and its monitoring data needs to be corrected upwards; K arm,k <1 indicates that the lever arm of the main load-bearing cable is greater than the reference value, and its monitoring data needs to be corrected downward.
[0053] Calculate the temperature correction factor K for the k-th main load-bearing cable. temp,k =1+α steel ·(T avg,k -T 0,k ); In the formula, α steel The coefficient of thermal expansion of the main load-bearing cable (taken as 1.2 × 10⁻⁶). -5 / ℃); T avg,k T represents the current temperature of the k-th main load-bearing cable, in °C. 0,k The initial installation temperature is expressed in °C.
[0054] K temp,k The calculation principle is as follows: Due to the long length of the main load-bearing cable, the length change caused by thermal expansion and contraction is not negligible. The material of the main load-bearing cable expands, and under the same geometric constraints, its internal tension decreases, leading to a lower force value measured by the strain gauge sensor. When the temperature decreases, the main load-bearing cable contracts, increasing its internal tension, resulting in a higher measured value. Different main load-bearing cables may be in different temperature environments (some on the sunny side, some on the shady side), and without correction, it is impossible to distinguish whether the reading difference comes from the actual force difference or the temperature difference. The correction factor is taken as 1+α. steel ·(T avg -T0), using the initial installation temperature as a unified temperature reference, the main load-bearing cable force values measured at different temperatures are uniformly corrected to the equivalent main load-bearing cable force values at the reference temperature.
[0055] Comprehensive correction factor and physical reference correction: W cal,k =K geo,k ·K arm,k ·K temp,k ; F cal,k =W cal,k ·F k ; In the formula, W cal,k F is the comprehensive correction factor for the k-th main load-bearing cable. cal,k F is the cable force value after physical benchmark correction for the k-th main load-bearing cable. k This represents the original monitoring cable force value of the k-th main load-bearing cable.
[0056] The three correction coefficients correspond to three independent sources of physical reference deviation: spatial location, lever arm length, and temperature. All three factors have a proportional scaling effect on the monitored values. When all three factors act simultaneously, the total correction effect is the product of the individual correction coefficients, and the product can be naturally superimposed. The corrected main load-bearing cable force value F cal,kThis eliminates system deviations caused by differences in the contact state between the main load-bearing cables and the saddle, differences in the lever arm length of the balance wheel, and differences in ambient temperature, allowing the force values of the dispersed main load-bearing cables to be directly compared on the same physical reference.
[0057] After the calculation is completed, F cal,k The force value of the k-th main load-bearing cable will be used for subsequent steps.
[0058] Preferably, in step S6, the fused representative value of each data type is obtained by combining the overall reliability and timeliness of each data source as follows: Calculate the timeliness factor ΔT for each data source factor,m =1 / (1+η·Δt m ); In the formula, ΔT factor,m η is the timeliness factor of the m-th data source, dimensionless, and ranging from 0 to 1; η is the time delay attenuation coefficient, taken as 5 / s; Δt m This represents the time delay from the data source acquisition time to the current processing time, expressed in seconds (s).
[0059] ΔT factor,m The calculation principle is as follows: there is a transmission and queuing delay between data collection and processing. The longer the delay, the lower the reliability of the data in reflecting the current real state. The role of the timeliness factor is to assign lower weights to data with larger delays in the fusion calculation. The smaller the delay, the closer the timeliness factor is to 1, and the higher its weight in the fusion.
[0060] Divide all data sources into multiple data types (the data types are consistent with those set in step S2). For data type d, let there be N data types. d There are 1 data source; the overall credibility of the m-th data source is C. m ; Calculate the fusion weight W of the m-th data source fus,m : ; The overall credibility reflects the reliability of the data source itself (higher values indicate more reliable data), while the timeliness factor reflects the freshness of the data (higher values indicate newer data). The product of these two factors automatically reduces the weight of low-credibility or high-latency data in the fusion process, effectively filtering data from both quality and timeliness dimensions. Data sources marked as discarded (C) m <0.35) does not participate in the fusion calculation.
[0061] Calculate the representative value of data type d at the current moment: ; In the formula, Y cal,m This refers to the monitoring data from the m-th data source; Y fusion,d It is a comprehensive representative value that is weighted by overall credibility and data timeliness. This value integrates information from multiple data sources and eliminates interference from low-credibility and high-latency data, so as to truly reflect the overall status of the data type at the current moment.
[0062] The combined representative value of all data types and the overall credibility C of each data source. i The coordinates of the hoisting areas corresponding to each data source are encapsulated into a comprehensive monitoring dataset.
[0063] Preferably, in step S6, the synchronous mapping and display of the fused representative value in the 3D scene model is specifically as follows: A 3D scene model of the cable hoisting area is constructed. The fused representative values of each data type within the comprehensive monitoring dataset are mapped to their corresponding positions in the 3D scene model based on the average coordinates of the hoisting area from each data source within that data type. The mapping color of the fused representative value is determined according to the range of the average overall reliability of each data source within that data type: green for an average value ≥ 0.60 (indicating high overall reliability of the data type, operators can confidently refer to the data); yellow for 0.35 ≤ average value < 0.60 (indicating low reliability of some data sources within that data type, requiring attention, and potential data quality issues); and red for an average value < 0.35 (indicating low overall reliability of the data type, the data may be unreliable, operators should refer to other information sources). This color mapping method allows operators to immediately determine the reliability of the data without needing to check the status of each data source individually.
[0064] The advantage of a 3D scene model is that operators no longer need to deal with a large amount of tabular data. Instead, they can intuitively see the distribution of each monitoring value in the hoisting area through the 3D model. For example, the main load-bearing cable force value is displayed in the form of a digital label at the location of the corresponding main load-bearing cable. The color is clear at a glance, making it easy to quickly identify the data quality status.
[0065] Managers of cable-stayed hoisting operations are typically not on-site and need to monitor the construction status from the project office or remotely via computers, tablets, or other terminals. Preferably, in step S6, the comprehensive monitoring dataset is transmitted to a cloud platform for storage. Remote terminals access the cloud platform via a webpage to obtain real-time data, which is then simultaneously displayed in the 3D scene model on the webpage.
[0066] The present invention also relates to a multi-source data integration and visualization system for cable hoisting in mountainous areas, comprising a storage unit and a processor. The storage unit is used to store one or more program instructions; the processor is used to run one or more program instructions to execute the steps of the above-described multi-source data integration and visualization monitoring method for cable hoisting in mountainous areas.
[0067] This embodiment uses the arch rib hoisting construction of a mountain canyon bridge as an application scenario. The main span of the bridge is 520m, and the maximum lifting weight of the arch rib segment is 5800kN (approximately 580t). Two sets of main cables are used in parallel operation, each set consisting of 6 main load-bearing cables with a nominal diameter of 52mm. The following uses the main load-bearing cable force value of one set of main cables as a representative example to illustrate the entire process of this method in more detail: 1. Data acquisition and spatiotemporal alignment: The force values and corresponding temperature values of the six main load-bearing cables were collected at a sampling frequency of 10Hz. During a continuous 60-second monitoring period (600 sampling times in total), each data source should theoretically send 600 data packets. Due to the obstruction caused by the mountain, data source No. 4 (corresponding to the 4th main load-bearing cable) actually successfully received 270 data packets, with a data packet loss rate of 55%; the other five data sources actually successfully received approximately 570-590 data packets, with a loss rate of approximately 2%-5%.
[0068] 2. Packet loss compensation verification: Taking the loss of data packet No. 4 at t=5.2s as an example, compensation verification is performed: the most recent valid data before and after the loss time are x4(5.1s)=938kN and x4(5.4s)=942kN. The duration of the data loss is T. loss =0.3s, time interpolation confidence level τ time =exp(-2×0.3)=0.549.
[0069] The values of the other five data sources within this data type (main load-bearing cable force value) at t=5.2s are as follows: No1 is 1152kN, No2 is 1082kN, No3 is 1021kN, No5 is 881kN, and No6 is 821kN. The spatial distance d between each data source and No4 in the hoisting area coordinate system is... 4j The distances are as follows: No. 1 is 2.1m, No. 2 is 2.8m, No. 3 is 3.5m, No. 5 is 4.2m, and No. 6 is 4.9m. The arithmetic mean d of the pairwise distances between all data sources within this data type is... ref =6.5m.
[0070] Calculate the spatial correlation coefficients: ρ 41 =1 / (1+2.1 / 6.5)=0.756, ρ 42 =1 / (1+2.8 / 6.5)=0.699, ρ 43 =1 / (1+3.5 / 6.5)=0.650, ρ 45 =1 / (1+4.2 / 6.5)=0.607, ρ 46 =1 / (1+4.9 / 6.5)=0.570. Σρ=3.282, spatial estimation confidence level τ space=3.282 / 5=0.656.
[0071] Time linear interpolation x time =938+(942-938)×(5.2-5.1) / (5.4-5.1)=939.3kN; Spatial estimate x space =(0.756×1152+0.699×1082+0.650×1021+0.607×881+0.570×821) / 3.282=1003.7kN.
[0072] Joint compensation value x comp =[0.549 / (0.549+0.656)]×939.3+[0.656 / (0.549+0.656)]×1003.7=974.3kN. The actual value at that moment (confirmed by subsequent data) was 941kN, and the deviation of the joint compensation value was (974.3-941) / 941=3.5%. After compensating for all lost data packets, the average deviation between the joint compensation value and the actual value was 2.3%, which is within the normal fluctuation range of the main load-bearing cable force monitoring, verifying the accuracy of the space-time joint compensation method in the event of data loss.
[0073] 3. Overall credibility calculation: During the 60-second monitoring period, the continuous arrival rate R of each data source arr,i =N rec,i / N exp,i The values are as follows: No. 1 is 0.97, No. 2 is 0.95, No. 3 is 0.98, No. 4 is 0.45, No. 5 is 0.96, and No. 6 is 0.95. No. 4 has a significantly lower consecutive arrival rate than other data sources due to its 55% loss rate.
[0074] Statistical consistency rate R among different data sources sta,i The values are approximately 0.92 to 0.95 for the five normal data sources and 0.85 for No.4 (the statistical consistency is slightly lower due to incomplete data).
[0075] At a typical moment, the monitored values from the six data sources were 1150, 1080, 1020, 940, 880, and 820 kN. The geometric mean G of all data sources in this data type is... avg =(1150×1080×1020×940×880×820) (1 / 6) =976kN. The spatial uniformity R of No.4 at that moment is... spa,4 =1-|log2(940 / 976)|=1-0.054=0.946.
[0076] The overall credibility of No.4 is C4 = (0.45). 0.60×(0.85) 0.25 ×(0.946) 0.15 =0.619 × 0.961 × 0.992 = 0.591. This value is below the credibility threshold of 0.60, and is therefore marked as a low-credibility data source. For comparison, the overall credibility of No.1 is C1 = (0.97). 0.60 ×(0.94) 0.25 ×(0.98) 0.15 =0.964, higher than 0.60, marked as a high-confidence data source. The only problem with No.4 is poor communication quality (R arr =0.45), while the other two dimensions were normal. Overall, it was identified as having low confidence but was not abandoned (still higher than 0.35), and was subsequently included in the fusion process with reduced weight.
[0077] 4. Physical reference calibration: Through on-site measurements, the contact wrap angles Ang1 to Ang6 between the six main load-bearing cables and the saddle rollers were 0.8, 1.0, 1.2, 1.4, 0.9, and 1.1 rad, respectively. The coefficient of friction Kfr between the saddle rollers and the main load-bearing cables was determined to be 0.03 through on-site measurement. The spatial position correction coefficient K for each main load-bearing cable... geo,k =1+0.03×Ang k The values are: 1.024, 1.030, 1.036, 1.042, 1.027, and 1.033.
[0078] By measuring the distance from the connection point of each main load-bearing cable on the balance wheel to the center of rotation of the balance wheel, the lever arm lengths l1~l6 were obtained as 1.00, 0.95, 0.90, 1.10, 1.05, and 0.85 m, respectively. The average value l of the lever arm length of all main load-bearing cables in this data type is... ref =0.975m. The lever arm balance correction coefficient K for each main load-bearing cable. arm,k =l k / l ref The values are: 1.026, 0.974, 0.923, 1.128, 1.077, and 0.872, respectively.
[0079] The initial installation temperature of each main load-bearing cable was 22°C, and the current temperature is 37°C (temperature difference 15°C). Temperature correction factor K. temp,k =1+1.2×10 -5 ×(37-22)=1.00018.
[0080] Comprehensive correction factor W for each main load-bearing cable cal,k =K geo ×K arm ×K tempThe values are: 1.051, 1.004, 0.957, 1.175, 1.106, and 0.901, respectively. ΣW cal =6.194. The main load-bearing cable force value F after physical benchmark correction. cal,k =W cal,k ×F k The values are as follows: No. 1 is 1209 kN, No. 2 is 1084 kN, No. 3 is 976 kN, No. 4 is 1105 kN, No. 5 is 973 kN, and No. 6 is 739 kN. All corrected values are in the range of 800–1150 kN, within the same order of magnitude as the original main load-bearing cable forces, and can be directly compared on the same physical reference. After physical reference correction, the cable force of No. 4 is 1105 kN, and the cable force of No. 6 is 739 kN, with the difference between the two increasing from the original 120 kN to 366 kN. This indicates that without lever arm correction, the actual force of No. 4 (with a larger lever arm) will be underestimated, while the actual force of No. 6 (with a smaller lever arm) will be overestimated, making it impossible for operators to accurately determine the actual force distribution of each main load-bearing cable. Physical reference correction is necessary.
[0081] 5. Weighted fusion and 3D visualization: The overall reliability of each data source, C1 to C6, are 0.964, 0.958, 0.951, 0.591, 0.965, and 0.940, respectively. The time delay Δt of each data packet is 0.02, 0.03, 0.02, 0.15, 0.02, and 0.03 seconds, respectively, and the timeliness factor ΔT is... factor =1 / (1+5Δt) are 0.909, 0.870, 0.909, 0.571, 0.909, and 0.870 respectively.
[0082] Fusion weighted molecule M m =C m ×ΔT factor,m The values are 0.876, 0.833, 0.864, 0.337, 0.877, and 0.818 respectively, with a total of 4.606. The fusion weight W fus,m The values are 0.190, 0.181, 0.187, 0.073, 0.190, and 0.177, respectively. Due to its low overall credibility (0.591) and large time delay (0.15s), the fusion weight of No.4 was reduced to 0.073, while the fusion weight of No.1, which has high credibility and low delay, was 0.190. This proves that the fusion weight formula can effectively identify and suppress the interference of low-quality data on the fusion results.
[0083] The representative value of the main load-bearing cable force F fusion=0.190×1209+0.181×1084+0.188×976+0.073×1105+0.190×973+0.178×739=1006.5kN. The total main load-bearing cable force of the 6 main load-bearing cables is 1036.0×6=1006.5×6=6039kN, which is 1.041 compared with the maximum lifting weight of 5800kN of the arch rib segment. Considering the weight of the lifting equipment and the dynamic impact coefficient, it is within a reasonable range.
[0084] The main load-bearing cable force value (1036.0 kN), the overall reliability of each data source, and the coordinates of the hoisting area of each data source are packaged into a comprehensive monitoring dataset and uploaded to the cloud platform. In the 3D scene model, the fused representative value is mapped to the corresponding position of the main cable. The average overall reliability (0.964+0.958+0.951+0.591+0.965+0.940) / 6=0.895≥0.60, displayed in green, indicating that the overall reliability of this data type is high. Operators can use this to judge that the main load-bearing cable force value data is reliable and the hoisting status is normal.
Claims
1. A method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas, characterized by the following steps: include: S1: Establish wireless relay nodes on the tops of the towers on both sides of the hoisting area and construct a wireless mesh data acquisition and transmission network covering the entire hoisting area; S2: Real-time acquisition of multi-source monitoring data during cable hoisting process; S3: Add a device identification code, a serial number that increments independently with the device identification code, and a timestamp to the monitoring data collected from each data source to form a raw data packet. Then, perform spatiotemporal alignment on all raw data packets under a unified time reference and spatial coordinate system to form an aligned data packet. S4: Detect packet loss in each data source based on the continuity of sequence numbers in the aligned data packets and perform spatial-temporal joint reconstruction compensation on the monitored data; S5: Calculate the overall credibility of each data source from three dimensions: continuous arrival rate, statistical consistency rate, and spatial consistency rate, and mark the credibility of the monitoring data of each data source accordingly; at the same time, uniformly correct the cable force value of the main load-bearing cable to the same spatial and temperature reference. S6: Combine the overall credibility and timeliness of each data source to obtain the fused representative value of each data type, and synchronously map and display the fused representative value in the 3D scene model.
2. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, In step S2, the multi-source monitoring data shall include at least the following data types: the cable force value of each main load-bearing cable, the temperature of each main load-bearing cable, and the weight of each suspension point; and each data type shall have at least two data sources.
3. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, In step S3, the spatiotemporal alignment of all raw data packets is specifically performed as follows: The original data packet includes at least the device identification code field, timestamp field, data type field, serial number field, and monitoring data field; After receiving the raw data packets from each data source, the relay node extracts the timestamps of each raw data packet. Using the relay node's time as a unified time reference, the timestamps of each raw data packet are converted into the relay node's time value. A hoisting area coordinate system is established with the pre-established reference point position as the origin, the longitudinal direction of the bridge as the X-axis, the transverse direction of the bridge as the Y-axis, and the vertical direction as the Z-axis. The position coordinates and positioning data of each data source are uniformly converted into the hoisting area coordinate system to form hoisting area coordinates, which are then appended to the raw data packets to form aligned data packets.
4. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, Based on the continuity of sequence numbers in the aligned data packets, packet loss in each data source is detected, and spatial-temporal joint reconstruction compensation is performed on the monitored data. Specifically: After grouping the alignment data packets by device identifier, check the continuity of the sequence number of each group of alignment data packets. When a sequence number jump of the alignment data packet is detected, record the corresponding data source and the time of loss. Suppose that the r-th data source loses data at time t, and there are M data sources corresponding to the data type of the r-th data source, with the most recent data x before and after it. r (t prev ) and x r (t next ); Calculate time linear interpolation x time,r (t)=x r (t prev )+[x r (t next )-x r (t prev )]×(tt prev ) / (t next -t prev ); Calculate the spatial correlation coefficient ρ between the j-th data source and the r-th data source in this data type. rj =1 / (1+d rj / d ref ); In the formula, d rj Let d be the spatial distance between the j-th data source and the r-th data source. ref This is the arithmetic mean of the pairwise distances between all data sources in this data type; Calculate the spatial estimate x space,r (t): ; In the formula, x j (t) represents the monitoring data of the j-th data source in this data type at time t; Calculate the confidence level τ of time interpolation time =exp(-λ·T loss ); In the formula, λ is the attenuation coefficient, and T loss Duration of loss; Computational spatial estimation confidence ; Calculate the joint compensation value x comp,r (t): x comp,r (t)=[τ time / (t time +t space )]·x time,r (t)+[τ space / (t time +t space )]·x space,r (t); x comp,r (t) is used as monitoring data and is combined with the missing sequence number to form the missing aligned data packet.
5. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 4, characterized in that, In step S5, the overall credibility of each data source is calculated, specifically as follows: Set the monitoring period T; Calculate the continuous arrival rate R of the i-th data source. arr,i =N rec,i / N exp,i ; In the formula, N rec,i N represents the total number of aligned data packets successfully received from the i-th data source within time period T. exp,i This represents the total number of raw data packets that the i-th data source should send within time period T; Calculate the statistical similarity rate R of the same source. sta,i =M cnt,i / N rec,i ; In the formula, M cnt,i For the monitoring data of the i-th data source within time period T, satisfy M avg,i -2S std,i ≤x i,k ≤M avg,i +2S std,i The number of packages, M avg,i S is the average value of the monitoring data from the i-th data source within time period T. std,i Let be the standard deviation of the monitoring data from the i-th data source within time period T; Calculate the spatial consistency rate R spa,i =1-|log2(x i / G avg,i )|; In the formula, G avg,i This is the geometric mean of the monitored data from all data sources in this data type at the same time; when R... spa,i When R is less than 0, take R. spa,i =0, when R spa,i When the value is greater than 1, take R. spa,i =1; Calculate the overall credibility of the i-th data source: C i =(R arr,i ) 0.60 ·(R sta,i ) 0.25 ·(R spa,i ) 0.15 。 6. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, In step S5, the cable force values of the main load-bearing cables are uniformly corrected to the same spatial and temperature references, specifically as follows: Calculate the spatial position correction coefficient K for the k-th main load-bearing cable. geo,k =1+Kfr k ·Ang k ; In the formula, Kfr k Let be the coefficient of friction between the k-th main load-bearing cable and the saddle roller; Ang k Let be the contact wrap angle of the k-th main load-bearing cable on the saddle roller; Calculate the lever arm balance correction coefficient K for the k-th main load-bearing cable. arm,k =l k / l ref ; In the formula, l k Let l be the lever arm length of the k-th main load-bearing cable on the balance wheel. ref Let be the reference lever arm length of the k-th main load-bearing cable; Calculate the temperature correction factor K for the k-th main load-bearing cable. temp,k =1+α steel ·(T avg,k -T 0,k ); In the formula, α steel The coefficient of thermal expansion of the main load-bearing cable, T avg,k Let T be the current temperature of the k-th main load-bearing cable. 0,k Initial installation temperature; Comprehensive correction factor and physical reference correction: W cal,k =K geo,k ·K arm,k ·K temp,k ; F cal,k =W cal,k ·F k ; In the formula, W cal,k F is the comprehensive correction factor for the k-th main load-bearing cable. cal,k F is the cable force value after physical benchmark correction for the k-th main load-bearing cable. k This represents the original monitored cable force value of the k-th main load-bearing cable; After the calculation is completed, F cal,k The force value of the k-th main load-bearing cable will be used for subsequent steps.
7. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, In step S6, the fused representative value for each data type is obtained by combining the overall credibility and timeliness of each data source as follows: Calculate the timeliness factor ΔT for each data source factor,m =1 / (1+η·Δt m ); In the formula, ΔT factor,m Let η be the timeliness factor of the m-th data source, η be the time delay attenuation coefficient, and Δt be the time factor. m This represents the time delay from the data source acquisition time to the current processing time. All data sources are divided into multiple data types. For data type d, let N be the total number of data types. d There are 1 data source; the overall credibility of the m-th data source is C. m ; Calculate the fusion weight W of the m-th data source fus,m : ; Calculate the representative value of data type d at the current moment: ; In the formula, Y cal,m This refers to the monitoring data from the m-th data source; The combined representative value of all data types and the overall credibility C of each data source. i The coordinates of the hoisting areas corresponding to each data source are encapsulated into a comprehensive monitoring dataset.
8. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 7, characterized in that, In step S6, the fusion representative value is synchronously mapped and displayed in the 3D scene model as follows: Construct a 3D scene model of the cable hoisting area; The fused representative value of each data type in the comprehensive monitoring dataset is mapped to the corresponding position in the 3D scene model according to the average value of the hoisting area coordinates of each data source in that data type. The mapping color of the fused representative value is determined according to the range of the average comprehensive credibility of each data source within this data type.
9. The method for multi-source data integration and visual monitoring of cable hoisting in mountainous areas according to claim 1, characterized in that, In step S6, the comprehensive monitoring dataset is transmitted to the cloud platform for storage. The remote terminal accesses the cloud platform through a webpage to obtain real-time data, which is then synchronously displayed in the 3D scene model on the webpage.
10. A multi-source data integration and visualization system for cable hoisting in mountainous areas, characterized in that, It includes a storage unit and a processor, wherein the storage unit is used to store one or more program instructions; and the processor is used to run one or more program instructions to perform the steps of the method for multi-source data integration and visualization monitoring of cable hoisting in mountainous areas as described in any one of claims 1 to 9.
