Flexible photovoltaic support cable force intelligent real-time detection system and method

Abstract

The invention relates to the technical field of cable force detection, and discloses a flexible photovoltaic support cable force intelligent real-time detection system and method. Comprising the following steps: S01, simulating a bracket digital model; s02, judging whether a cable force real-time regulation and control mode is entered or not; s03, carrying out real-time cable force regulation and control; s04, drawing a time period cable force curve graph and a periodic cable force curve graph; s05, carrying out future cable force regulation and control; according to the invention, the dynamic change of the flexible photovoltaic support can be mapped in real time on a virtual level, the limitation of passive data acquisition under a single point location or a specific point location is avoided, and the cable force change trend of the flexible photovoltaic support in a historical period can be analyzed from a plurality of different dimensions. The cable force abnormity false alarm phenomenon caused by abnormal fluctuation of single-point data at different moments is avoided, and the accuracy of cable force abnormity diagnosis is further improved.

Description

A Smart Real-Time Detection System and Method for Flexible Photovoltaic Support Cable Stress Technical Field

[0001] This invention relates to the field of cable stress detection technology, and more specifically, to an intelligent real-time detection system and method for the cable stress of flexible photovoltaic supports. Background Technology

[0002] Flexible photovoltaic (PV) support structures use long-span steel cables as the main load-bearing units to support and fix PV panels. However, during long-term operation, flexible PV support structures are affected by wind loads, temperature changes, and structural relaxation, which can cause changes in cable force. This can lead to problems such as attitude deviation, uneven structural stress distribution, local fatigue damage, and instability risk in PV modules. Therefore, monitoring the cable force data on the steel cables of flexible PV support structures is extremely important.

[0003] Reference patent application CN116380155A discloses a flexible support detection system and control method, equipment, and medium, including a support module comprising two support piles and a steel cable connecting the two support piles; a first acquisition module, mounted on the steel cable, for acquiring the tension value of the steel cable; a second acquisition module, mounted on the steel cable, for acquiring wind direction and wind speed; and a central processing unit electrically connected to the first and second acquisition modules. Existing cable force detection systems typically employ a method of deploying sensors at a single point or specific location on the steel cable for detection. This method only achieves passive, single-point measurement of cable force and cannot meet the needs of multi-span, comprehensive, and real-time cable force monitoring for flexible supports. Furthermore, when analyzing and calculating the acquired data, it cannot effectively avoid the negative impact of multiple different dimensions of data on cable force, making abnormal fluctuations in single-point data prone to causing errors in cable force analysis results and leading to misjudgments, thereby reducing the accuracy of cable force anomaly diagnosis.

[0004] In view of this, the present invention proposes an intelligent real-time detection system and method for the cable tension of flexible photovoltaic supports to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the existing technology and achieve the above objectives, the present invention provides the following technical solution: a method for intelligent real-time detection of cable force in flexible photovoltaic support structures, applied to a cloud server, comprising: S01: constructing a three-dimensional model of the support structure using structural design parameters, and dynamically mapping the comprehensive cable force data to the three-dimensional model of the support structure to simulate a digital model of the support structure. The comprehensive cable force data includes cable force values, strain values, acceleration, and temperature values; S02: after constructing a smooth cable force sequence, performing cable force time-series analysis to estimate the cable force estimate at the current acquisition time, and determining whether to enter the real-time cable force control mode; if entering... In the real-time cable stress control mode, execute S03; if not entering the real-time cable stress control mode, execute S04; S03: In the real-time cable stress control mode, determine the abnormal level and perform real-time cable stress control on the flexible photovoltaic support; S04: Determine the smoothing period and prediction cycle, draw the period cable stress curve based on the first cable stress characteristic simulated by the support digital model, and draw the period cable stress curve based on the second cable stress characteristic simulated by the support digital model; S05: Perform fusion analysis of the period cable stress curve and the period cable stress curve to determine the cable stress health status and perform future cable stress control on the flexible photovoltaic support.

[0006] Furthermore, the structural design parameters include point cloud data, material data, and appearance data. The method for constructing the 3D model of the support structure is as follows: the point cloud data is preprocessed by removing duplicates and supplementing missing ones, and the preprocessed point cloud data is imported into the same spatial coordinate system for coordinate alignment. Using the point cloud data as the position reference, continuous mesh units are simulated through mesh reconstruction technology, and the mesh units at adjacent positions are spliced ​​together to generate the 3D outline of the support structure. The steel cable area in the 3D outline of the support structure is identified through visual inspection technology, and material skeletons corresponding to the material data are added to the steel cable area. Using the appearance data as the rendering reference, the 3D outline of the support structure is rendered into a 3D model of the support structure through rendering technology.

[0007] Furthermore, the method for constructing the smooth cable force sequence is as follows: a blank sequence with four sequence positions is constructed, and the first three sequence positions are designated as independent sequence positions and the last sequence position is designated as dependent sequence position, from left to right; strain, acceleration, and temperature are used as independent factors, and cable force is used as dependent factor, and the comprehensive cable force data at B acquisition times are arranged to generate B factor units; the independent factors of the factor units are imported into the three independent sequence positions one by one, the dependent factors are imported into the dependent sequence positions, and the acquisition time is marked, thereby converting the blank sequence into a smooth cable force sequence.

[0008] Furthermore, the estimation method for cable force is as follows: Any three smooth cable force sequences at consecutive acquisition times are bound into a sequence group, resulting in C sequence groups; a ternary linear function is constructed using strain coefficient, acceleration coefficient, and temperature coefficient as unknowns, and strain value, acceleration value, temperature value, and cable force value at the same acquisition time as knowns; the sequence groups and the ternary linear function are fused into a system of ternary linear equations to calculate the strain coefficient, acceleration coefficient, and temperature coefficient; the strain coefficient, acceleration coefficient, and temperature coefficient of the C sequence groups are summed and averaged to calculate the standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient; the standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient are combined with the ternary linear function to construct a cable force estimation function; the strain value, acceleration value, and temperature value at the current acquisition time are substituted into the cable force estimation function to calculate the cable force estimation value; the cable force difference is calculated by subtracting the cable force estimation value from the cable force value at the current acquisition time; when the cable force difference is greater than the calibrated upper limit of the cable force difference, the real-time cable force control mode is entered.

[0009] Furthermore, the anomaly level includes a first-level deviation level and a second-level deviation level. The method for determining the anomaly level is as follows: divide the cable force difference of the flexible photovoltaic support by the cable force estimate to calculate the anomaly ratio. When the anomaly ratio is greater than the calibrated anomaly ratio threshold, the anomaly level is a first-level deviation level. When the anomaly ratio is less than or equal to the calibrated anomaly ratio threshold, the anomaly level is a second-level deviation level.

[0010] Furthermore, when determining the smoothing period, the duration of change in strain, acceleration, temperature, and cable force values ​​during the historical period is statistically analyzed. The maximum duration of change is recorded as the average variation duration. The current acquisition time is taken as the start of the period, and the first acquisition time after one average variation duration is taken as the end of the period. The period between the start and end of the period is recorded as the smoothing period. When determining the prediction period, the current acquisition time is taken as the start of the period, and the first acquisition time after D smoothing periods is taken as the end of the period. The period between the start and end of the period is recorded as the prediction period.

[0011] Furthermore, the first cable force feature includes time points and time-period cable force values. The method for drawing the time-period cable force curve is as follows: arranging all smooth cable force sequences in the historical time period in chronological order to generate a sequence queue; merging the sequence queue with the scaffold digital model and inputting it into simulation software, using the duration of the smooth time period as the simulation duration, to simulate the first simulation scenario; in the first simulation scenario, using the smooth cable force sequence as the simulation unit, identifying the cable force change trend of the sequence queue through deep learning technology, and extracting F time points and F time-period cable force values ​​from the cable force change trend; using the time points as the horizontal axis and the time-period cable force values ​​as the vertical axis, connecting the points of the F time points and F time-period cable force values ​​sequentially to draw the time-period cable force curve.

[0012] Furthermore, the second cable force feature includes periodic time points and periodic cable force values. The method for drawing the periodic cable force curve is as follows: after merging the sequence queue with the scaffold digital model, the data is input into the simulation software, and the simulation duration is set as the predicted period to simulate the second simulation scenario. In the second simulation scenario, the smooth cable force sequence is used as the simulation unit, and the cable force change trend of the sequence queue is identified through deep learning technology. H periodic time points and H periodic cable force values ​​are extracted from the cable force change trend. The periodic cable force curve is drawn by connecting the H periodic time points and H periodic cable force values ​​sequentially with the periodic time points as the horizontal axis and the periodic cable force values ​​as the vertical axis.

[0013] Furthermore, the health status of the sol-force includes short-term sub-health, long-term sub-health, overall sub-health, and normal status. The method for determining the sol-force health status is as follows: The average of all sol-force values ​​within the previous smoothing period is calculated; the difference between each of the F period sol-force values ​​and the average period sol-force is calculated; and the time point where the difference exceeds the calibrated upper limit of the sol-force difference is recorded as the first abnormal time. The average of all period sol-force values ​​within the previous prediction period is calculated to obtain the periodic sol-force average. The cable force values ​​for each of the H cycles are subtracted from the average cable force value for each cycle to calculate the cycle difference. The cycle point where the cycle difference is greater than the upper limit of the calibrated cable force difference is recorded as the second abnormal moment. When only the first abnormal moment exists, the cable force health status is a short-term sub-healthy state. When only the second abnormal moment exists, the cable force health status is a long-term sub-healthy state. When both the first and second abnormal moments exist, the cable force health status is a fully sub-healthy state. When neither the first nor the second abnormal moment exists, the cable force health status is a normal state.

[0014] A flexible photovoltaic (PV) support cable force intelligent real-time detection system, applied to a cloud server, is used to implement a method for intelligent real-time detection of cable force in flexible PV supports. The system includes: a model building module for constructing a three-dimensional model of the support using structural design parameters and dynamically mapping comprehensive cable force data to the three-dimensional model to simulate a digital model of the support; an estimation and analysis module for constructing a smoothed cable force sequence and performing time-series analysis to estimate the cable force at the current acquisition time and determine whether to enter a real-time cable force control mode; a real-time control module for determining the anomaly level and performing real-time cable force control on the flexible PV support under the real-time cable force control mode; a predictive analysis module for determining the smoothed period and prediction period, plotting a periodic cable force curve based on the first cable force characteristic simulated by the support digital model, and plotting a periodic cable force curve based on the second cable force characteristic simulated by the support digital model; and a fusion analysis module for fusing and analyzing the periodic and periodic cable force curves to determine the cable force health status and perform future cable force control on the flexible PV support.

[0015] The technical effects of the intelligent real-time detection system and method for cable force of flexible photovoltaic support in this invention are as follows: (1): By dynamically mapping the three-dimensional model of the support with the comprehensive cable force data and simulating the digital model of the support, the dynamic changes of the flexible photovoltaic support can be mapped in real time at the virtual level. This can avoid the limitations of passively collecting data at a single point or a specific location, and can also visualize the dynamic changes of the flexible photovoltaic support, making it convenient to collect and detect the multi-dimensional data that causes cable force changes in a comprehensive and accurate manner.

[0016] (2): By constructing a smooth cable force sequence and establishing a cable force estimation function between cable force and strain, acceleration and temperature, this invention can analyze the cable force change trend of flexible photovoltaic support in historical periods from multiple different dimensions. By comparing the cable force value estimated by the cable force change trend with the actual cable force value, it can avoid the cable force anomaly false alarm caused by abnormal fluctuation of single point data at different times, thereby improving the accuracy of cable force anomaly diagnosis.

[0017] (3): By drawing the cable force change curves of smooth time period and prediction period, this invention can predict the cable force change of flexible photovoltaic support in the short and long term in advance, and can timely and accurately regulate and optimize the cable force of flexible photovoltaic support before the cable force changes abnormally, thereby transforming post-fault maintenance into pre-fault maintenance, effectively avoiding potential structural risks and safety accidents, and realizing the adaptive regulation effect of cable force along the time line. Attached Figure Description

[0018] Figure 1 is a schematic diagram of a flexible photovoltaic support cable force intelligent real-time detection system provided in Embodiment 1 of the present invention; Figure 2 is a flowchart of a flexible photovoltaic support cable force intelligent real-time detection method provided in Embodiment 2 of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: As shown in Figure 1, this example describes a flexible photovoltaic support cable force intelligent real-time detection system applied to a cloud server. It includes a model building module that collects structural design parameters of the flexible photovoltaic support, constructs a three-dimensional model of the support, and dynamically maps the comprehensive cable force data to the three-dimensional model of the support using digital twin technology to simulate a digital model of the support. During installation and use, the flexible photovoltaic support requires the coordinated operation of steel cables, anchor assemblies, cable force fine-tuning actuators, multi-source sensor mechanisms, signal acquisition and control units, intelligent analysis units, and communication units. The two ends of the steel cable are connected to the anchor assemblies, which fix the two ends of the steel cable to concrete columns. Force sensors are installed inside the anchor assemblies to measure the actual cable force. The middle of the steel cable is supported by concrete. The support column provides stability. The multi-source sensor mechanism includes a fiber optic grating sensor, an accelerometer, and a temperature sensor. The fiber optic grating sensor is positioned on the steel cable near the anchor assembly to measure strain data. The accelerometer is installed at the mid-span of the steel cable to monitor cable vibration and changes. The temperature sensor is installed at the end of the steel cable to measure the ambient temperature. The cable force fine-tuning actuator is installed between the anchor assembly and the steel cable and includes an electric screw, a geared drive motor, an elastic support assembly, and a mechanical locking structure. Under control commands, the cable force fine-tuning actuator can automatically adjust the cable force within a certain range. The signal acquisition and control unit, the intelligent analysis unit, and the communication unit are used to acquire data and output control signals, analyze cable force trends and detect anomalies, and remotely visualize data, respectively.

[0021] Because the steel cables on the flexible photovoltaic support have a large span and long length, it is impossible to deploy high-density, multi-point sensors on the cables. To avoid the limitations of single-point sensor detection data, a 3D model of the support, consistent with the structure and shape of the flexible steel cable support, needs to be constructed in the virtual layer using 3D modeling technology. When constructing the 3D model of the support, structural design parameters that can physically represent the position, structure, and dimensions of the flexible photovoltaic support are first obtained. The structural design parameters include point cloud data, material data, and appearance data. Point cloud data refers to the 3D spatial location information of all points on the flexible photovoltaic support, material data refers to the material type information of the steel cables on the flexible photovoltaic support, and appearance data refers to the color and texture information displayed on the exterior of the flexible photovoltaic support. In this embodiment, point cloud data is obtained by scanning with a laser scanner, material data is obtained by querying construction design drawings, and appearance data is obtained by capturing images with a camera.

[0022] The method for constructing the 3D model of the support structure is as follows: Point cloud data undergoes preprocessing, including duplicate removal and missing data filling. The preprocessed point cloud data is then imported into the same spatial coordinate system and aligned with the origin. Using the aligned point cloud data as a positional reference, continuously distributed grid cells are simulated using mesh reconstruction technology. Adjacent grid cells are then sequentially stitched together to generate the 3D outline of the support structure. Visual inspection technology is used to identify the cable regions within the 3D outline. Material skeletons corresponding to the material data are added to these cable regions. Using the appearance data as a rendering reference, the 3D outline of the support structure is rendered using rendering technology to construct the 3D model. The material skeleton is a virtual parameter used to realistically match the cable regions with the material data, ensuring that the material type of the cable regions matches that of the cables in the flexible photovoltaic support structure.

[0023] The 3D model of the support structure can only provide a static, hardware-level virtual representation of the flexible photovoltaic support structure. It cannot accurately map the dynamic changes of the flexible photovoltaic support structure in real time. Therefore, it is necessary to dynamically map the 3D model of the support structure and upgrade it into a digital model of the support structure. In this embodiment, the digital model of the support structure is a digital twin model that dynamically maps the 3D model of the support structure with the comprehensive cable force data in real time through digital twin technology, thereby achieving a real-time dynamic mapping effect of the flexible photovoltaic support structure at the virtual level.

[0024] When upgrading the 3D model of the support structure to a digital model, it is necessary to first obtain comprehensive cable force data that can represent the dynamic changes of the flexible photovoltaic module at different times. Specifically, the comprehensive cable force data includes cable force value, strain value, acceleration, and temperature value. Cable force value refers to the actual cable force on the steel cable, strain value refers to the relative deformation of the steel cable under the action of external force or non-uniform temperature field, acceleration refers to the swaying rate of the steel cable's up-and-down vibration response, and temperature value refers to the temperature of the environment in which the steel cable is currently located.

[0025] When collecting comprehensive cable force data, force sensors, fiber optic grating sensors, acceleration sensors, and temperature sensors are used to acquire the data, and a real-time acquisition method is adopted to achieve dynamic acquisition of comprehensive cable force data at each acquisition moment.

[0026] In this embodiment, when simulating the digital model of the scaffold, the comprehensive cable force data at the same moment are time-aligned to obtain A data sets. According to the order of acquisition time, the A data sets are dynamically mapped to the 3D model of the scaffold using digital twin technology, and a time axis positively correlated with the acquisition time is configured to upgrade the 3D model of the scaffold to the digital model of the scaffold.

[0027] It should be noted that by configuring a timeline, a basis for time traceability can be provided for data collected at different times on the digital model of the stent, ensuring that data from different times can be aggregated and collected. Through logical settings that are positively correlated with the time of collection, the timeline is ensured to progress sequentially from front to back, ensuring the reliability and accuracy of the dynamic mapping of the digital model of the stent.

[0028] The estimation and analysis module constructs a smoothed cable force sequence, performs time-series analysis on the smoothed cable force sequence, estimates the cable force value at the current acquisition moment, and determines whether to enter the real-time cable force control mode. Since the support digital model is a dynamic mapping representation of the flexible photovoltaic support, it contains comprehensive cable force data from all historical acquisition moments of the flexible photovoltaic support. The comprehensive cable force data at different acquisition moments have unique data characteristics and meanings. Therefore, it is necessary to summarize and analyze the data in the support digital model and use the smoothed cable force sequence to represent the data change trends and situations in the support digital model. In this embodiment, the smoothed cable force sequence is a data sequence used to arrange and summarize the comprehensive cable force data of the flexible photovoltaic support at each acquisition moment within a historical period. Therefore, the smoothed cable force sequence needs to include the cable force value, strain value, acceleration, and temperature value at each acquisition moment.

[0029] The method for constructing the smooth cable force sequence is as follows: A blank sequence with four sequence positions is constructed. Following a left-to-right approach, the first three sequence positions are designated as independent sequence positions, and the last sequence position is designated as dependent sequence position. Using strain, acceleration, and temperature as independent factors and cable force as the dependent factor, the comprehensive cable force data from B acquisition times in the digital model of the support structure are arranged to generate B factor units. The independent factors in each factor unit are imported one by one into the three independent sequence positions, and the dependent factors are imported into the dependent sequence positions. The acquisition time of each factor unit is marked, thus converting the blank sequence into a smooth cable force sequence. The smooth cable force sequence is as follows: , ;in, For the first Strain values ​​at each acquisition time, For the first Acceleration at each acquisition moment, For the first Temperature values ​​at each data collection time. For the first The cable force value at each acquisition moment.

[0030] After constructing the smooth cable force sequence, cable force time series analysis can be performed on the smooth cable force sequence to analyze the numerical change law between the independent and dependent factors at each acquisition time, thereby estimating the cable force estimate at the current acquisition time. In this embodiment, the cable force estimate is the cable force value at the current acquisition time estimated based on the specific numerical changes in the smooth cable force sequence, and is used as the object for analysis and comparison with the actual cable force value at the current acquisition time, thus providing a basis for whether to enter the real-time cable force control mode. Specifically, the method for estimating the cable force is as follows: any three smooth cable force sequences at consecutive acquisition times are bound into sequence groups to obtain C sequence groups; one sequence group contains three consecutive acquisition times. A smooth cable force sequence was obtained, providing a foundation for the subsequent construction of a ternary linear function. Using strain coefficient, acceleration coefficient, and temperature coefficient as unknowns, and strain, acceleration, temperature, and cable force values ​​at the same acquisition moment as knowns, a ternary linear function was constructed. The sequence group and the ternary linear function were fused into a system of ternary linear equations to calculate the strain coefficient, acceleration coefficient, and temperature coefficient. The strain coefficient, acceleration coefficient, and temperature coefficient of the C sequence groups were then summed and averaged to calculate the standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient. The standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient were combined with the ternary linear function to construct a cable force estimation function. The expression for the cable force estimation function is: In the formula, The strain value, The standard strain coefficient, For acceleration, The standard acceleration factor, This is the temperature value. Standard temperature coefficient, The cable force value is calculated by substituting the strain, acceleration, and temperature values ​​at the current acquisition moment into the cable force estimation function.

[0031] It should be noted that in the constructed cable force estimation function, the strain value, acceleration value, temperature value and cable force value are all dimensionless, so that the cable force estimation function only contains direct numerical calculations, so as to facilitate the rapid calculation of cable force estimation values.

[0032] After estimating the cable force, the estimated cable force is compared and analyzed with the cable force value collected at the current moment to determine whether there is an anomaly in the cable force of the flexible photovoltaic support at the current moment, and thus determine whether to enter the real-time cable force control mode. The method for determining whether to enter the real-time cable force control mode is as follows: calculate the cable force difference by subtracting the cable force estimated value at the current collection moment from the cable force value. When the cable force difference is greater than the calibrated upper limit of the cable force difference, it indicates that there is a large deviation between the actual cable force value and the estimated cable force value of the flexible photovoltaic support at the current collection moment, and real-time cable force control is required. In this case, it is determined to enter the real-time cable force control mode. When the cable force difference is less than or equal to the calibrated upper limit of the cable force difference, it indicates that there is a small deviation between the actual cable force value and the estimated cable force value of the flexible photovoltaic support at the current collection moment, and real-time cable force control is not required. In this case, it is determined not to enter the real-time cable force control mode.

[0033] In this embodiment, the calibrated upper limit of cable force difference refers to the maximum deviation between the actual cable force value and the estimated cable force value of the flexible photovoltaic support cable, thus serving as a direct basis for judging whether the cable force has deviated significantly.

[0034] The real-time control module, in the real-time cable force control mode, determines the abnormality level of the flexible photovoltaic support and performs real-time cable force control on the flexible photovoltaic support. If the real-time cable force control mode is entered, it is necessary to analyze the degree of cable force abnormality of the flexible photovoltaic support and determine the abnormality level of the cable force deviation phenomenon, so as to provide the necessary basis for subsequent cable force magnitude control. Specifically, the abnormality level includes a first-level deviation level and a second-level deviation level; and the cable force deviation amplitude of the first-level deviation level is greater than that of the second-level deviation level.

[0035] The method for determining the anomaly level is as follows: divide the cable force difference of the flexible photovoltaic support by the estimated cable force value to calculate the anomaly ratio; when the anomaly ratio is greater than the calibrated anomaly ratio threshold, it indicates that the cable force of the flexible photovoltaic support is seriously abnormal, and the anomaly level is determined to be a first-level deviation level; when the anomaly ratio is less than or equal to the calibrated anomaly ratio threshold, it indicates that the cable force of the flexible photovoltaic support is slightly abnormal, and the anomaly level is determined to be a second-level deviation level.

[0036] In this embodiment, the calibrated abnormality percentage threshold refers to a pre-set critical value used to distinguish between the first-level deviation level and the second-level deviation level. The calibrated abnormality percentage threshold is set according to actual needs. For example, the calibrated abnormality percentage threshold is 1.5%.

[0037] After determining the anomaly level, the cable tension of the flexible photovoltaic support needs to be adjusted to increase the cable tension value. This ensures that the collected actual cable tension value remains the same as or close to the estimated cable tension value, until the real-time cable tension adjustment mode is no longer entered. Specifically, during real-time cable tension adjustment of the flexible photovoltaic support, the communication unit uploads the cable tension estimate, cable tension value, cable tension difference, anomaly level, and other relevant information obtained by the intelligent analysis unit to the cloud server. The cloud server sends a cable tension adjustment command to the signal acquisition and control unit, which then applies the command to the cable tension fine-tuning actuator. At this time, the mechanical locking structure of the cable tension fine-tuning actuator opens, the reduction drive motor is energized to drive the electric lead screw to rotate, and the elastic support component is driven to extend outward elastically, increasing the pulling force on the cable and fine-tuning the cable tension value in real time. The adjustment operation stops when the cable tension difference is less than or equal to the calibrated upper limit of the cable tension difference.

[0038] The predictive analysis module, if it does not enter the real-time cable force control mode, determines the smoothing period and prediction cycle, simulates the first cable force characteristics of the support digital model during the smoothing period, and plots the cable force curve for the period. It also simulates the second cable force characteristics of the support digital model during the prediction cycle and plots the period cable force curve. If it does not enter the real-time cable force control mode, it means that the flexible photovoltaic support has not experienced cable force deviation or imbalance at the current acquisition time. Therefore, it is necessary to predict the cable force change trend of the flexible photovoltaic support in the short and long term in the future.

[0039] When predicting the cable force value of a flexible photovoltaic support in the future, it is necessary to first determine the time periods of the shorter and longer future periods, and record them as the smoothing period and the prediction period, respectively. The smoothing period is a time period consisting of multiple consecutive acquisition times starting from the current acquisition time and proceeding along the time line. In this embodiment, when determining the smoothing period, the duration of change in strain value, acceleration, temperature value and cable force value during the historical period is counted, the maximum value of the change duration is recorded as the average change duration, and the time period between the start and end of the period is recorded as the smoothing period, with the current acquisition time as the start time and the first acquisition time after one average change duration as the end time.

[0040] For example, the duration of the smoothing period is 2 hours or 3 hours.

[0041] The prediction period is a time period consisting of multiple consecutive smoothed time periods starting from the current acquisition time and proceeding along the timeline. In this embodiment, when determining the prediction period, the current acquisition time is taken as the start of the period, and the first acquisition time after D smoothed time periods is taken as the end of the period. The time period between the start of the period and the end of the period is recorded as the prediction period.

[0042] For example, the forecast period is 2 days or 3 days.

[0043] Once the smoothing period is determined, the digital model of the support can be simulated with a small lead time to predict the first cable force characteristics of the smoothing period. The cable force values ​​of the flexible photovoltaic support at different times within the smoothing period can then be simulated and predicted based on the first cable force characteristics. The first cable force characteristics include the time point of the time period and the cable force value of the time period. The time point of the time period refers to each acquisition time within the smoothing period, and the cable force value of the time period refers to the cable force value corresponding to each acquisition time within the smoothing period.

[0044] After obtaining the first cable force characteristic, a time-period cable force curve can be plotted based on this characteristic to represent the trend of cable force value changes in the digital model of the stent during the smooth period. The method for plotting the time-period cable force curve is as follows: All smooth cable force sequences from historical time periods are arranged sequentially to generate a sequence queue; the sequence queue is then integrated with the digital model of the stent and synchronously input into the simulation software, using the duration of the smooth period as the simulation duration to simulate the first simulation scenario; the first simulation scenario is used to analyze the cable force values ​​of the digital model of the stent at different acquisition times within the smooth period. The simulated virtual scenario ensures the advanced prediction simulation effect of the cable force change trend within a smooth time period. In the first simulation scenario, the smooth cable force sequence is used as the simulation unit. The cable force change trend of the sequence queue is identified by deep learning technology, and F time period time points and F time period cable force values ​​are extracted from the cable force change trend. The cable force change trend refers to the distribution of cable force values ​​at different acquisition times, which can provide numerical basis for the change of cable force values. With the time period time point as the horizontal axis and the time period cable force value as the vertical axis, the points of the F time period time points and F time period cable force values ​​are connected in sequence to draw the time period cable force curve.

[0045] Once the prediction period is determined, the digital model of the support structure can be simulated at significantly ahead moments to predict the second cable force characteristics during smooth periods. The cable force characteristics are then used to simulate and predict the cable force values ​​of the flexible photovoltaic support structure at different times within the prediction period. The second cable force characteristics include the periodic time points and the periodic cable force values. The periodic time points refer to each acquisition time within the prediction period, and the periodic cable force values ​​refer to the cable force values ​​corresponding to each acquisition time within the prediction period.

[0046] After obtaining the second cable force characteristic, a periodic cable force curve can be plotted based on this characteristic to represent the trend of cable force value changes in the digital stent model within the prediction period. The method for plotting the periodic cable force curve is as follows: the sequence queue and the digital stent model are merged and synchronously input into the simulation software, and the simulation duration is set as the prediction period to simulate the second simulation scenario. The second simulation scenario is a virtual scenario used to simulate the cable force values ​​of the digital stent model at different acquisition times within the prediction period, ensuring the advanced prediction simulation effect of the cable force change trend within the prediction period. In the second simulation scenario, a smooth cable force sequence is used as the simulation unit. The cable force change trend of the sequence queue is identified through deep learning technology, and H periodic time points and H periodic cable force values ​​are extracted from the cable force change trend. The periodic cable force curve is plotted by connecting the H periodic time points and H periodic cable force values ​​sequentially with the periodic time points as the horizontal axis and the periodic cable force values ​​as the vertical axis.

[0047] It should be noted that the constructed time-period cable force curve and periodic cable force curve can separately represent the cable force change trend of the flexible photovoltaic support over a shorter and longer time span in the future, respectively, thus providing a data foundation for the analysis of the overall cable force change of the flexible photovoltaic support in the future with two time dimensions.

[0048] The fusion analysis module performs a combined analysis of the time-period cable force curve and the periodic cable force curve to determine the cable force health status of the flexible photovoltaic support and to adjust the cable force of the flexible photovoltaic support in the future. After obtaining the time-period cable force curve and the periodic cable force curve, it is necessary to perform a combined analysis of the two curves to analyze the overall trend of cable force changes in the flexible photovoltaic support over a short and long time span in the future. This allows for advance knowledge of the cable force value of the flexible photovoltaic support at future moments, enabling adaptive adjustment of the cable force value. When adaptively adjusting the cable force value, it is necessary to first determine the cable force health status of the flexible photovoltaic support at future moments based on the results of the future cable force fusion analysis, which serves as the basis for adaptive adjustment of the cable force.

[0049] Specifically, the Soli health status includes short-term sub-health status, long-term sub-health status, full sub-health status, and normal status. Among them, short-term sub-health status refers to abnormal fluctuations in Soli value during the smoothing period; long-term sub-health status refers to abnormal fluctuations in Soli value during the prediction period; full sub-health status refers to abnormal fluctuations in Soli value during both the smoothing period and the prediction period; and normal status refers to no abnormal fluctuations in Soli value during either the smoothing period or the prediction period.

[0050] The method for determining the cable force health status is as follows: The average cable force value is calculated by summing all cable force values ​​within the previous smoothing period; the cable force difference is calculated by subtracting the average cable force value from each of the F time points, and the time point where the time difference exceeds the calibrated upper limit of the cable force difference is recorded as the first abnormal time point; the average periodic cable force value is calculated by summing all periodic cable force values ​​within the previous prediction period; the periodic cable force difference is calculated by subtracting the average periodic cable force value from each of the H periodic time points, and the periodic difference is recorded as the periodic time point where the periodic difference exceeds the calibrated upper limit of the cable force difference. The first abnormal moment is recorded as the second abnormal moment. If the first abnormal moment exists within the smoothing period and there is no second abnormal moment within the prediction period, the health status of Soli is determined to be a short-term sub-healthy state. If there is no first abnormal moment within the smoothing period and there is a second abnormal moment within the prediction period, the health status of Soli is determined to be a long-term sub-healthy state. If the first abnormal moment exists within the smoothing period and there is a second abnormal moment within the prediction period, the health status of Soli is determined to be a full sub-healthy state. If there is no first abnormal moment within the smoothing period and there is no second abnormal moment within the prediction period, the health status of Soli is determined to be a normal state.

[0051] Once the cable stress health status of the flexible photovoltaic (PV) support system during the smoothing period and the prediction cycle is determined, potential cable stress imbalances that may occur in the future can be controlled to prevent excessive cable stress fluctuations and loss of control, achieving a proactive and adaptive control effect. Specifically, when controlling the cable stress of the flexible PV support system in the future, the communication unit will upload relevant information such as the cable stress health status, the first abnormal moment, and the second abnormal moment obtained from the intelligent analysis unit to the cloud server, and will also upload the time point of the previous period before the first abnormal moment, or the time point of the second abnormal moment. Taking the previous cycle time point as the target time, the cloud server sends a cable force adjustment command to the signal acquisition and control unit, which then sends the cable force adjustment command to the cable force fine-tuning actuator. At the target time, the mechanical locking structure of the cable force fine-tuning actuator is opened, the reduction drive motor is energized to drive the electric lead screw to rotate, and synchronously drives the elastic support component to extend outward elastically, increasing the pulling force on the steel cable, thus fine-tuning the cable force value in advance, and stopping when the time difference or cycle difference is less than or equal to the calibrated upper limit of cable force difference. This achieves the effect of advance adaptive control of cable force on the flexible photovoltaic support.

[0052] Example 2: Please refer to Figure 2. For details not described in this example, please refer to the description in Example 1. This example provides a method for intelligent real-time detection of cable force in flexible photovoltaic (PV) brackets, applied to a cloud server. It is implemented based on the aforementioned intelligent real-time detection system for cable force in flexible PV brackets, including: S01: Constructing a three-dimensional model of the bracket using structural design parameters, and dynamically mapping the comprehensive cable force data to the three-dimensional model to simulate a digital model of the bracket. The comprehensive cable force data includes cable force values, strain values, acceleration, and temperature values; S02: After constructing a smooth cable force sequence, performing cable force time-series analysis to estimate the cable force estimate at the current acquisition time, and determining whether... Enter the real-time cable force control mode; if the real-time cable force control mode is entered, execute S03; if the real-time cable force control mode is not entered, execute S04; S03: In the real-time cable force control mode, determine the abnormal level and perform real-time cable force control on the flexible photovoltaic support; S04: Determine the smoothing period and prediction period, draw the period cable force curve based on the first cable force characteristic simulated by the support digital model, and draw the period cable force curve based on the second cable force characteristic simulated by the support digital model; S05: Perform fusion analysis of the period cable force curve and the period cable force curve to determine the cable force health status and perform future cable force control on the flexible photovoltaic support.

[0053] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent real-time detection of cable tension in flexible photovoltaic supports, applied to a cloud server, characterized in that, include: S01: Construct a three-dimensional model of the support structure using structural design parameters, and dynamically map the comprehensive cable force data to the three-dimensional model of the support structure to simulate a digital model of the support structure. The comprehensive cable force data includes cable force values, strain values, acceleration and temperature values. S 02: After constructing a smooth cable force sequence, perform cable force time series analysis to estimate the cable force at the current acquisition time and determine whether to enter the real-time cable force control mode; If you enter the real-time control mode for cable tension, execute S03; If you do not enter the real-time control mode for cable tension, execute S04; S 03: Under the real-time cable force control mode, the abnormal level is determined, and the cable force of the flexible photovoltaic support is controlled in real time; S04: The smoothing period and prediction period are determined. The cable force curve of the period is plotted by simulating the first cable force characteristic through the support digital model, and the cable force curve of the period is plotted by simulating the second cable force characteristic through the support digital model; S05: The cable force curve of the period and the cable force curve of the period are integrated and analyzed to determine the cable force health status, and the cable force of the flexible photovoltaic support is controlled in the future.

2. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 1, characterized in that, The structural design parameters include point cloud data, material data, and appearance data. The method for constructing the 3D model of the support structure is as follows: the point cloud data is preprocessed by removing duplicates and supplementing missing ones, and the preprocessed point cloud data is imported into the same spatial coordinate system for coordinate alignment. Using the point cloud data as the position reference, continuous mesh units are simulated through mesh reconstruction technology, and adjacent mesh units are spliced ​​together to generate the 3D outline of the support structure. The steel cable area in the 3D outline of the support structure is identified through visual inspection technology, and material skeletons corresponding to the material data are added to the steel cable area. Using the appearance data as the rendering reference, the 3D outline of the support structure is rendered into a 3D model of the support structure through rendering technology.

3. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 2, characterized in that, The method for constructing a smooth cable force sequence is as follows: a blank sequence with four sequence positions is constructed, and the first three sequence positions are designated as independent sequence positions and the last sequence position is designated as dependent sequence position, from left to right; strain, acceleration, and temperature are used as independent factors, and cable force is used as dependent factor; the comprehensive cable force data at B acquisition times are arranged to generate B factor units; the independent factors of the factor units are imported into the three independent sequence positions one by one, the dependent factors are imported into the dependent sequence positions, and the acquisition time is marked, thereby converting the blank sequence into a smooth cable force sequence.

4. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 3, characterized in that, The estimation method for cable force is as follows: Any three smooth cable force sequences at consecutive acquisition times are bound into a sequence group, resulting in C sequence groups. A ternary linear function is constructed using strain coefficient, acceleration coefficient, and temperature coefficient as unknowns, and strain value, acceleration value, temperature value, and cable force value at the same acquisition time as knowns. The sequence groups and the ternary linear function are fused into a system of ternary linear equations to calculate the strain coefficient, acceleration coefficient, and temperature coefficient. The strain coefficient, acceleration coefficient, and temperature coefficient of the C sequence groups are summed and averaged to calculate the standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient. The standard strain coefficient, standard acceleration coefficient, and standard temperature coefficient are combined with the ternary linear function to construct a cable force estimation function. The strain value, acceleration value, and temperature value at the current acquisition time are substituted into the cable force estimation function to calculate the cable force estimate. The cable force difference is calculated by subtracting the cable force estimate from the cable force value at the current acquisition time. When the cable force difference is greater than the calibrated upper limit of cable force difference, the cable force real-time control mode is entered.

5. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 4, characterized in that, The anomaly level includes Level 1 deviation level and Level 2 deviation level. The method for determining the anomaly level is as follows: divide the cable force difference of the flexible photovoltaic support by the cable force estimate to calculate the anomaly ratio. When the anomaly ratio is greater than the calibrated anomaly ratio threshold, the anomaly level is Level 1 deviation level. When the anomaly ratio is less than or equal to the calibrated anomaly ratio threshold, the anomaly level is Level 2 deviation level.

6. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 5, characterized in that, When the smoothing period is determined, the duration of change in strain, acceleration, temperature, and cable force values ​​during the historical period is statistically analyzed. The maximum duration of change is recorded as the average variation duration. The current acquisition time is taken as the start of the period, and the first acquisition time after one average variation duration is taken as the end of the period. The period between the start and end of the period is recorded as the smoothing period. When the prediction period is determined, the current acquisition time is taken as the start of the period, and the first acquisition time after D smoothing periods is taken as the end of the period. The period between the start and end of the period is recorded as the prediction period.

7. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 6, characterized in that, The first cable force feature includes the time point of the time period and the cable force value of the time period; the method for drawing the cable force curve of the time period is as follows: arrange all the smooth cable force sequences in the historical time period in chronological order to generate a sequence queue; after the sequence queue is integrated with the digital model of the support, it is input into the simulation simulation software, and the simulation duration is used as the duration of the smooth time period to simulate the first simulation scenario; In the first simulation scenario, a smooth cable force sequence is used as the simulation unit. The cable force change trend of the sequence queue is identified by deep learning technology, and F time points and F time point values ​​are extracted from the cable force change trend. Using the time points of each time period as the horizontal axis and the cable force values ​​of each time period as the vertical axis, draw a cable force curve by connecting the points of the F time points and the F cable force values ​​of each time period in sequence.

8. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 7, characterized in that, The second cable force feature includes the periodic time points and the periodic cable force value; the method for drawing the periodic cable force curve is as follows: after merging the sequence queue with the digital model of the support, the data is input into the simulation software, and the simulation duration is taken as the predicted period to simulate the second simulation scenario; in the second simulation scenario, the smooth cable force sequence is used as the simulation unit, and the cable force change trend of the sequence queue is identified by deep learning technology, and H periodic time points and H periodic cable force values ​​are extracted from the cable force change trend; Using the periodic time points as the horizontal axis and the periodic cable force values ​​as the vertical axis, draw a periodic cable force curve by connecting the H periodic time points and H periodic cable force values ​​in sequence.

9. The intelligent real-time detection method for the cable force of a flexible photovoltaic support according to claim 8, characterized in that, Soli's health status includes short-term sub-health status, long-term sub-health status, overall sub-health status, and normal status; The method for determining the cable force health status is as follows: sum up all the cable force values ​​in the previous smoothing period and calculate the average cable force value for each period. Then, calculate the time difference value by subtracting each of the F time period cable force values ​​from the average cable force value for each period. The time point when the time difference value is greater than the upper limit of the cable force difference is recorded as the first abnormal time. The average value of the periodic cable force is calculated by summing all the periodic cable force values ​​in the previous prediction period. The periodic difference is calculated by subtracting each of the H periodic cable force values ​​from the average value of the periodic cable force. The periodic time point when the periodic difference is greater than the upper limit of the cable force difference is recorded as the second abnormal time. When only the first abnormal moment exists, Soli's health status is a short-term sub-healthy state; When only the second abnormal moment exists, Soli's health status is a long-term sub-healthy state; when both the first and second abnormal moments exist, Soli's health status is a fully sub-healthy state; when neither the first nor the second abnormal moment exists, Soli's health status is a normal state.

10. A flexible photovoltaic support cable tension intelligent real-time detection system, applied to a cloud server, for implementing the flexible photovoltaic support cable tension intelligent real-time detection method according to any one of claims 1-9, characterized in that, include: The model building module is used to construct a three-dimensional model of the support structure based on structural design parameters, and to dynamically map the comprehensive cable force data to the three-dimensional model of the support structure to simulate the digital model of the support structure. The estimation and analysis module is used to construct a smooth cable force sequence and then perform cable force time series analysis to estimate the cable force at the current acquisition time and determine whether to enter the real-time cable force control mode. The real-time control module is used to determine the level of abnormality and to control the cable force of the flexible photovoltaic support in real time under the cable force real-time control mode. The predictive analysis module is used to determine the smoothing period and the prediction cycle. It draws the periodic cable force curve based on the first cable force characteristics simulated by the digital model of the support structure, and draws the periodic cable force curve based on the second cable force characteristics simulated by the digital model of the support structure. The fusion analysis module is used to perform fusion analysis on the time period cable force curve and the periodic cable force curve to determine the cable force health status and to adjust the cable force of the flexible photovoltaic support in the future.

Images