Stress monitoring method for flexible photovoltaic support system

Abstract

The invention relates to the technical field of stress monitoring, in particular to a stress monitoring method for a flexible photovoltaic supporting system, and the method comprises the steps: obtaining stress data, cable force data, vibration data, wind speed data and wind direction data of the photovoltaic supporting system; acquiring a wind load influence coefficient of each point location at each moment, acquiring a vibration response coefficient of each point location in each interval, and further calculating a consistency coefficient of each point location; acquiring a first fluctuation coefficient of the single-row support structure according to the fluctuation form of the consistency coefficient of the single-row support structure; acquiring a second fluctuation coefficient of the single-row supporting structure according to the cable force data of the single-row supporting structure; combining the stress data to construct a stress state abnormal value of the single-row support structure; and monitoring the stress state of the flexible photovoltaic support system according to the stress state abnormal value. According to the invention, the stress monitoring precision of the flexible photovoltaic support system can be improved, and the stress stability of the flexible photovoltaic support system can be accurately monitored.

Description

Technical Field

[0001] The present application relates to the technical field of stress monitoring, and in particular to a stress monitoring method for a flexible photovoltaic support system. Background Art

[0002] Affected by geological changes, climate, oxidation, corrosion and aging factors, flexible photovoltaic support structures are easily damaged under long-term static and dynamic loads. Stress monitoring can be used to grasp the stress state of the support structure in real time and promptly discover potential structural safety hazards.

[0003] The stress monitoring process of the support structure is easily disturbed by environmental factors. Temperature changes will directly affect the thermal expansion and contraction of the material, and changes in wind speed and direction have a more significant impact on the stress of the flexible photovoltaic support system. Strong winds will cause the structure to vibrate and respond dynamically, leading to stress fluctuations and even changes in the stress concentration area. In addition, during the power generation process, the photovoltaic panels will automatically change angles with the direction of sunlight, making it more susceptible to interference from wind loads during stress monitoring. Conventional methods fail to fully consider the impact of wind loads and changes in the angle of photovoltaic panels during the monitoring process, making it difficult to detect potential abnormal risks in the photovoltaic support system, and the reliability of stress monitoring methods is low. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a stress monitoring method for a flexible photovoltaic support system to solve the existing problems.

[0005] A stress monitoring method for a flexible photovoltaic support system of the present application adopts the following technical solution: An embodiment of the present application provides a stress monitoring method for a flexible photovoltaic support system, the method comprising the following steps: S1: Obtain stress data, cable force data, vibration data of each point of each row of support structures in the photovoltaic support system at each time, as well as wind speed and wind direction data in the area where the photovoltaic support system is located; S2: Calculate the wind load influence coefficient of each photovoltaic panel at each moment: obtain the unit normal vector of the plane where each photovoltaic panel is located along the back direction at each moment, use the modulus of each unit normal vector in the vertical direction as the inclination of each photovoltaic panel at each moment, calculate the absolute value of the cosine value corresponding to the angle formed by the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment, obtain the ratio of the angle formed by the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment to 180 degrees, use the sum of the absolute value and the ratio as the wind direction influence coefficient of each photovoltaic panel at each moment, and record the ratio of the product of the wind direction influence coefficient of each photovoltaic panel at each moment and the wind speed to the inclination as the wind load influence coefficient of each photovoltaic panel at each moment; S3: Divide the vibration data of each point into multiple intervals; obtain the vibration response coefficient of each point in each interval based on the difference of the vibration data of each point in each interval and the average level of the vibration data; combine the correlation between the wind load influence coefficient and the vibration response coefficient of each point in each interval to construct the consistency coefficient of each point; S4: According to the degree of dispersion and the degree of variation of the consistency coefficient of all points in the single-row support structure, obtain the first fluctuation coefficient of the single-row support structure; according to the degree of dispersion and the degree of variation of the cable force data of all points in the single-row support structure, obtain the second fluctuation coefficient of the single-row support structure; analyze the degree of fluctuation of the stress data of all points in the single-row support structure, and obtain the stress state abnormal value of the single-row support structure by combining the mean of the first fluctuation coefficient and the second fluctuation coefficient; S5: Monitor the stress state of each row of supporting structures in combination with the abnormal values ​​of the stress state of each row of supporting structures.

[0006] Preferably, the method for determining the division into multiple intervals is: The vibration data of each point is used as the input of the segmentation algorithm, and the output is the positions of all segmentation points in the vibration data; If the time interval between two adjacent segmentation points is less than or equal to a preset value, the result of rounding up the median of the corresponding moments of the two adjacent segmentation points is used as a new segmentation point to replace the original two adjacent segmentation points, and the segmentation points are merged; The vibration data of each point is divided into multiple intervals according to the segmentation points after the merging operation.

[0007] Preferably, the vibration response coefficient of each point in each interval is the product of the entropy and the mean of all vibration data of each point in each interval.

[0008] Preferably, the method for determining the consistency coefficient of each point is: arrange the mean of the wind load influence coefficient and the vibration response coefficient of each point in all intervals in ascending time order to obtain the wind load influence sequence and vibration response sequence of each point; and use the Pearson correlation coefficient between the wind load influence sequence and the vibration response sequence of each point as the consistency coefficient of each point.

[0009] Preferably, the method for determining the first fluctuation coefficient of the single-row support structure is: , where is the first wave coefficient of the j-th row support structure, is the cumulative sum of the shortest distances between each data in the consensus sequence of the j-th row support structure and the fitting curve of the consensus sequence of the j-th row support structure; It is the cumulative sum of the absolute values ​​of the slopes of the fitting curve of the j-th row of support structures at each point, wherein the consistency sequence of the j-th row of support structures is obtained by arranging the consistency coefficients of each point in the j-th row of support structures in the order of their positions.

[0010] Preferably, the method for determining the second fluctuation coefficient of the single-row support structure is: The average of all the cable force data at each point is calculated to obtain the average cable force data at each point, and the sequence composed of the average cable force data at all points of each row of supporting structures is recorded as the cable force average sequence; The expression of the second wave coefficient of the j-th row support structure is: , where is the second wave coefficient of the j-th row support structure, is the cumulative sum of the shortest distances between each data in the cable force mean sequence of the j-th row of support structures and the fitting curve of the cable force mean sequence, It is the cumulative sum of the absolute values ​​of the slopes of the fitting curve of the cable force mean sequence of the j-th row of support structures at each point.

[0011] Preferably, the calculation method of the stress state abnormal value of the single-row support structure is the sum of the standard deviation of the stress data of all points in the single-row support structure and the mean of the first fluctuation coefficient and the second fluctuation coefficient.

[0012] Preferably, the monitoring of the stress state of each row of support structures further comprises: if the stress state abnormal value is greater than or equal to an abnormal threshold, then the stress state of the corresponding support structure is abnormal.

[0013] This application has at least the following beneficial effects: This application calculates the wind load influence coefficient by deeply analyzing the inclination of photovoltaic panels and the relationship between photovoltaic panels and wind direction, so that the stress monitoring results are corrected by wind loads, improving the accuracy of stress monitoring; further considering the degree of correlation between wind load influence and bracket vibration data and the complexity of data fluctuation patterns of each row of support structures, the stress state abnormal values ​​of each row of support structures are calculated, and the stress state of each row of support structures is no longer evaluated at a single point, but the stress state of each row of support structures is evaluated, making the evaluation structure more comprehensive. Compared with the conventional method of directly comparing real-time stress data, its advantage is that it can accurately evaluate the potential impact of wind loads on the support system and obtain comprehensive stress monitoring results of each row of support structures in the photovoltaic support system, which helps to improve the reliability of stress monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0015] Figure 1 A flow chart of the steps of a stress monitoring method for a flexible photovoltaic support system provided in this application; Figure 2 A flowchart for constructing stress state anomalies provided for this application. DETAILED DESCRIPTION

[0016] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the stress monitoring method of a flexible photovoltaic support system proposed in the present application, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0017] Unless defined otherwise, 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 application belongs.

[0018] The specific scheme of a stress monitoring method for a flexible photovoltaic support system provided by the present application is described in detail below with reference to the accompanying drawings.

[0019] An embodiment of the present application provides a stress monitoring method for a flexible photovoltaic support system. Specifically, the following stress monitoring method for a flexible photovoltaic support system is provided. Figure 1 , the method comprises the following steps: Example: S1: Obtain stress data, cable force data, vibration data of each point of each row of support structures in the photovoltaic support system at each time, as well as wind speed and wind direction data in the area where the photovoltaic support system is located.

[0020] In the stress monitoring process of the flexible photovoltaic support system, it is very important to monitor the stress, cable force and support vibration data of the support. The support is the foundation of the support system and directly bears the weight load of the superstructure. Real-time monitoring of the support stress can timely discover potential structural problems. The cable maintains the balance of the structure through pre-tension, and the change of cable force may affect the overall stability. The vibration characteristics of the support directly affect the dynamic stability of the structure, especially under the action of wind load. And the entire support structure is composed of multiple supports, steel cables, and fixings. When subjected to stress changes, the stress states of different points affect each other. Therefore, the present application collects data at multiple points in the entire support structure, and the specific points can be selected according to the type and actual situation of the flexible photovoltaic support system. Among them, strain gauges are installed at each point to collect stress data of the support, cable force meters are installed to collect cable force data of the cable, and accelerometers are installed to collect vibration data of the support. In order to evaluate the impact of wind load on the stress of the photovoltaic support system, the wind speed and wind direction data of the photovoltaic support system area are collected by an anemometer. The collection frequency of each data is set to 10Hz.

[0021] At this point, the stress data, cable force data, vibration data, wind speed and wind direction data of each point at each time have been obtained. Through 5G wireless transmission, the data collected by each sensor is sent to the monitoring platform, and data analysis is performed on the monitoring platform.

[0022] S2: Analyze the inclination of the panel surface of each photovoltaic panel to obtain the inclination of each photovoltaic panel at each moment, combine the wind speed at each moment, and the relationship between the position of the photovoltaic panel and the wind direction at each moment, and obtain the wind load influence coefficient of each point at each moment.

[0023] After receiving the data collected by the sensor, the monitoring platform performs real-time analysis and processing on the data. By analyzing the deep features between different types of data, the stress state of the photovoltaic support system can be evaluated.

[0024] In flexible photovoltaic support systems, wind load is an important factor affecting its stress state. Since photovoltaic modules usually have a large area, light weight and low stiffness, they are highly sensitive to wind loads. Under strong winds, the support structure is prone to deformation and vibration. In addition, during the power generation process of photovoltaic modules, their placement will be dynamically adjusted according to the angle of sunlight, and the wind direction may also change. The wind speed and the angle between the photovoltaic module and the wind direction will have a significant impact on the wind load, resulting in different stress states.

[0025] Further, the following analysis is performed on the variation characteristics between the photovoltaic panel angle and the wind direction. It should be noted that each point corresponds to a photovoltaic panel. The wind direction data collected by the wind speed and direction meter in this application is in the horizontal direction. When the angle between the photovoltaic panel and the horizontal plane is larger, the area of ​​the photovoltaic panel subjected to wind force is larger. In addition, the wind may also blow to the photovoltaic panel at different angles. If the wind direction is perpendicular to the photovoltaic panel, the force of the wind load is stronger at this time, and when the wind direction is parallel to the photovoltaic panel, the force area is smaller at this time, and the force of the wind load is weaker. Moreover, in the windward and leeward states, the size of the wind load is also different. Affected by the complex structure and rough surface of the back of the photovoltaic panel, the wind load in the leeward is usually larger than that in the windward. Therefore, the influence characteristics of the wind load on stress are obtained by combining the inclination angle of the photovoltaic panel itself with the horizontal plane direction and the angle between the wind direction and the photovoltaic panel.

[0026] This application obtains the position state of the photovoltaic panel at each moment through the adjustment device of the photovoltaic support system, and takes any photovoltaic panel at the i-th moment as an example for detailed description. In order to consider the difference in wind loads of photovoltaic panels in the windward and leeward conditions, first obtain the unit normal vector of the plane where the photovoltaic panel is located along its back direction, where the back of the photovoltaic panel is the side facing away from the sun. In view of the fact that the influence of wind load on photovoltaic panels is mainly reflected in two key factors: its own inclination angle and the relationship between the position of the photovoltaic panel and the wind direction. The modulus of the obtained unit normal vector in the vertical direction is used as the inclination of the photovoltaic panel, is the inclination of the photovoltaic panel at the i-th moment, The larger the value is, the closer the normal vector of the photovoltaic panel plane is to the vertical direction at the i-th moment, the closer the tilt direction of the photovoltaic panel is to the horizontal direction, and the smaller the degree of influence of the wind load is. When the unit normal vector is nearly parallel to the wind direction, the photovoltaic panel is in a posture facing or facing away from the wind direction, and the closer the angle between the unit normal vector and the wind direction is to 0 degrees or 180 degrees, the greater the influence of the wind force on it, and the greater the influence of the wind load in the leeward state; and when the unit normal vector is perpendicular to the wind direction, the photovoltaic panel is in a posture facing the wind direction sideways, and the closer the angle between the unit normal vector and the wind direction is to 90 degrees, the smaller the influence of the wind force on it. Therefore, first calculate the absolute value of the cosine value corresponding to the angle between the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment, and then calculate the ratio of the angle between the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment to 180 degrees. The larger the ratio is, the closer the photovoltaic panel is to the leeward state, and the smaller the ratio is, the closer the photovoltaic panel is to the windward state. The sum of the absolute value and the ratio is recorded as the wind direction influence coefficient of each photovoltaic panel at each moment. It represents the wind direction influence coefficient of the photovoltaic panel at the i-th moment. The larger it is, the more the photovoltaic panels are affected by wind direction.

[0027] In addition, the greater the wind speed, the higher the wind load on the photovoltaic panel. The formula for calculating the wind load influence coefficient at the i-th moment is: , where is the wind load influence coefficient of the photovoltaic panel at the i-th moment, is the wind speed data at the i-th moment, is the wind direction influence coefficient of the photovoltaic panel at the i-th moment, is the inclination of the photovoltaic panel at the i-th moment. The larger the value is, the greater the impact of wind load on the photovoltaic panel is at that moment.

[0028] S3: Divide the vibration data of each point into multiple intervals; obtain the vibration response coefficient of each point in each interval based on the difference in vibration data of each point in each interval and the average level of vibration data; combine the correlation between the wind load influence coefficient and the vibration response coefficient of each point in each interval to construct the consistency coefficient of each point.

[0029] Furthermore, the cable force and the vibration characteristics of the bracket directly affect the dynamic stability of the photovoltaic support structure, especially under the action of wind load. Since the flexible photovoltaic support system adopts prestressed flexible rope support, its vibration amplitude is larger than that of the rigid support. The mutual influence between the various parts is also quite significant. The local deployment of stabilizing cables, anchors and other dragging structures makes its vibration more complicated than that of the rigid support. In addition, the action of wind load has certain pulsating characteristics, which means that the intensity of wind action on the support structure is not constant, but presents irregular periodic changes, which in turn causes the vibration data of the bracket to have corresponding irregular periodic changes. When the stability of the support structure is worse, the vibration degree of the bracket will change greatly with the increase of wind load; when the stability of the support structure is better, the vibration degree of the bracket will change slightly with the increase of wind load. Therefore, when the stability of the support structure is worse, the correlation characteristics between the corresponding wind load and the vibration characteristics of the bracket are stronger.

[0030] Thus, the irregular periodic changes of vibration data caused by the pulsating characteristics of wind load are further analyzed. Since the stress state of the photovoltaic support system responds to the wind relatively quickly, the duration of each stress monitoring period should not be too long. Preferably, the time length of each stress monitoring period is set to 20 seconds in this embodiment. The support vibration data has a high frequency of small-range up and down jitters. When subjected to the force of strong winds, the amplitude of the corresponding vibration data will increase significantly and rapidly. The force of the wind may be unstable for a short period of time. When the strong wind ends, the amplitude of the corresponding vibration data will decrease significantly and rapidly, and the process is intermittent. The increase and decrease amplitude of the vibration data is usually greater than its jitter amplitude. Taking a certain period as an example, this application uses the Bernaola Galvan (BG) segmentation algorithm to divide the period of the vibration data, and the output of the algorithm is the position of all segmentation points in the vibration data. Since the support vibration data still jitters during the rapid increase and decrease process, there may be a situation where there are multiple segmentation points in the local range, and these segmentation points that are relatively close will interfere with the period division. The multiple segmentation points within the local range are further merged. Specifically, if the difference between the corresponding time values ​​of two adjacent segmentation points is less than or equal to m, preferably, in this embodiment, m is 9, then the result of rounding up the median of the corresponding time of the two segmentation points is used as the new segmentation point, thereby dividing the vibration data into multiple intervals according to the segmentation points after the merge operation.

[0031] Under the influence of high-intensity wind force, the bracket vibrates more violently, and the more unstable the jitter of the vibration data in the corresponding interval is, the vibration characteristics in each interval are obtained. The product of the Shannon entropy and the mean of the corresponding vibration data in each interval is calculated as the vibration response coefficient corresponding to each interval. The larger the obtained vibration response coefficient, the greater the vibration degree of the bracket in the interval. The wind load influence coefficient is calculated at each moment in the interval, and the mean of all wind load influence coefficients in each interval is calculated respectively. The mean of the wind load influence coefficient and the vibration response coefficient are arranged in ascending order of time, and the wind load influence sequence and the vibration response sequence can be obtained. Based on the correlation characteristics between the wind load influence and the vibration degree of the bracket, the Pearson correlation coefficient between the above two sequences is calculated as the consistency coefficient of the wind load and the vibration characteristics of the bracket. The larger the consistency coefficient obtained, the worse the stability of the bracket at the point in the corresponding period. At this point, the consistency coefficient corresponding to each point in the photovoltaic support system is obtained.

[0032] S4: According to the degree of discreteness and the degree of variation of the consistency coefficient of all points in the single-row support structure, obtain the first fluctuation coefficient of the single-row support structure; according to the degree of discreteness and the degree of variation of the cable tension data of all points in the single-row support structure, obtain the second fluctuation coefficient of the single-row support structure; analyze the degree of fluctuation of the stress data of all points in the single-row support structure, and obtain the stress state abnormal value of the single-row support structure by combining the average of the first fluctuation coefficient and the second fluctuation coefficient.

[0033] The flexible photovoltaic support system adopts an array design, and each photovoltaic panel unit in each row is connected by a cable. Due to its large span, the single-row support structure is prone to bending vibrations with wave characteristics. The placement angle of the photovoltaic panel changes dynamically during the power generation process, and the wind direction may also change. The wind speed and the angle between the photovoltaic panel and the wind direction will affect the wind load, thereby causing different degrees of stress influence. The data sequence of the single-row support structure may show different fluctuation patterns. Among them, if the fluctuation pattern is manifested as an overall smooth fluctuation, it means that the more stable the support structure is, the more balanced the stress distribution is; if the fluctuation pattern is manifested as multiple complex waveforms, it means that the greater the degree of distortion of the support structure, the more uneven the stress distribution is, indicating that there is a greater possibility of corrosion aging and loose connectors. This application can comprehensively evaluate the stress of the support system by analyzing the data fluctuation pattern of the single-row support structure.

[0034] Taking the j-th row of support structures as an example, the greater the fluctuation of the point, the worse the stability of the support, and the larger the corresponding consistency coefficient. Therefore, the fluctuation shape of the single-row support structure can be reflected by the consistency coefficient at continuous points. The consistency coefficients of each point in the single-row support structure are arranged in the order of the positions of each point to obtain the consistency sequence of the corresponding structure. It should be noted that the position sequence in this embodiment is to sort the points in sequence from one end of the support structure to the other end, and the implementer does not make special requirements for the specific starting endpoint. In order to analyze the complexity of the fluctuation, the polynomial fitting technology is used to fit the consistency sequence to obtain the corresponding fitting curve, and then the absolute value of the slope of the fitting curve at each point is obtained, and the shortest distance between each data in the consistency sequence and the fitting curve is calculated respectively. Then the first fluctuation coefficient of the j-th row of support structures is obtained. , the specific formula is: , where is the first fluctuation coefficient of the j-th row support structure, which is used to characterize the unevenness of the stability of each point in the single-row support structure. is the cumulative sum of the shortest distances between each data and the fitting curve in the consistency sequence of the j-th row of support structures. This value reflects the discrete degree of the vibration waveform of the single-row support structure. It is the cumulative sum of the absolute values ​​of the slopes of the fitting curves at each point, which reflects the severity of the structural changes at each point. The higher the complexity coefficient of the obtained fluctuation pattern, the more unstable the single-row flexible photovoltaic support is.

[0035] The greater the fluctuation, the greater the cable force is usually. The cable force data at each point has a fluctuation characteristic similar to the consistency coefficient to a certain extent. The average cable force data of each point in a stress monitoring period is averaged to obtain the average cable force data of each point. The sequence consisting of the average cable force data from one end to the other of a single-row support structure is recorded as the cable force mean sequence. The cable force mean sequence is used as the input of the polynomial fitting technology to obtain the corresponding fitting curve. Based on the cable force mean sequence of the j-th row support structure, the corresponding second fluctuation coefficient can be obtained. , the specific formula is: , where is the second fluctuation coefficient of the j-th row support structure, which is used to characterize the imbalance of the cable force at each point in the single-row support structure. is the cumulative sum of the shortest distances between each data in the cable force mean sequence of the j-th row of support structures and the fitting curve corresponding to the cable force mean sequence, It is the cumulative sum of the absolute values ​​of the slopes of the fitting curve of the cable force mean sequence of the j-th row of support structures at each point.

[0036] The support is the force fulcrum of the bracket and the cable. The vibration of the bracket and the force of the cable will cause the stress change of the bracket. Therefore, the stress state is evaluated by combining the change of stress data and the size of the complexity coefficient of each fluctuation form. The sum of the standard deviation of the stress data of all points in the single-row support structure and the mean of the first fluctuation coefficient and the second fluctuation coefficient is taken as the stress state abnormal value of the single-row support structure. The construction flow chart of the stress state abnormal value is as follows: Figure 2 The obtained stress state abnormal value reflects the stress instability degree of the corresponding single-row support structure.

[0037] S5: Monitor the stress state of each row of supporting structures in combination with the abnormal values ​​of the stress state of each row of supporting structures.

[0038] Based on the stress state abnormal value, this embodiment uses the softmax function to normalize the obtained stress state abnormal value, and sets the abnormal threshold to 0.7, and then judges the size of the normalized result and the abnormal threshold. If the stress state abnormal value of the j-th row of support structures is greater than or equal to the abnormal threshold, it means that the stress state of the corresponding support structure is abnormal, and the stability of the row of support structures in the array is poor, and an early warning message needs to be issued to the staff; if the stress state abnormal value of the j-th row of support structures is less than the abnormal threshold, it means that the row of support structures in the array is stable.

[0039] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0040] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0041] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A stress monitoring method for a flexible photovoltaic support system, characterized in that: The method comprises the following steps: S1: Obtain stress data, cable force data, vibration data of each point of each row of support structures in the photovoltaic support system at each time, as well as wind speed and wind direction data in the area where the photovoltaic support system is located; S2: Calculate the wind load influence coefficient of each photovoltaic panel at each moment: obtain the unit normal vector of the plane where each photovoltaic panel is located along the back direction at each moment, take the modulus of each unit normal vector in the vertical direction as the inclination of each photovoltaic panel at each moment, calculate the absolute value of the cosine value corresponding to the angle between the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment, obtain the ratio of the angle between the unit normal vector of each photovoltaic panel at each moment and the wind direction at each moment to 180 degrees, take the sum of the absolute value and the ratio as the wind direction influence coefficient of each photovoltaic panel at each moment, and take the ratio of the product of the wind direction influence coefficient of each photovoltaic panel at each moment and the wind speed to the inclination as the wind load influence coefficient of each photovoltaic panel at each moment; S3: Divide the vibration data of each point into multiple intervals; obtain the vibration response coefficient of each point in each interval based on the difference of the vibration data of each point in each interval and the average level of the vibration data; combine the correlation between the wind load influence coefficient and the vibration response coefficient of each point in each interval to construct the consistency coefficient of each point; S4: According to the degree of dispersion and the degree of variation of the consistency coefficient of all points in the single-row support structure, obtain the first fluctuation coefficient of the single-row support structure; according to the degree of dispersion and the degree of variation of the cable force data of all points in the single-row support structure, obtain the second fluctuation coefficient of the single-row support structure; analyze the degree of fluctuation of the stress data of all points in the single-row support structure, and obtain the stress state abnormal value of the single-row support structure by combining the mean of the first fluctuation coefficient and the second fluctuation coefficient; S5: Monitor the stress state of each row of supporting structures in combination with the abnormal values ​​of the stress state of each row of supporting structures.

2. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The method for determining the division into multiple intervals is: The vibration data of each point is used as the input of the segmentation algorithm, and the output is the positions of all segmentation points in the vibration data; If the time interval between two adjacent segmentation points is less than or equal to a preset value, the result of rounding up the median of the corresponding moments of the two adjacent segmentation points is used as a new segmentation point to replace the original two adjacent segmentation points, and the segmentation points are merged; The vibration data of each point is divided into multiple intervals according to the segmentation points after the merging operation.

3. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The vibration response coefficient of each point in each interval is the product of the entropy and the mean of all vibration data of each point in each interval.

4. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The method for determining the consistency coefficient of each point is as follows: the mean of the wind load influence coefficient and the vibration response coefficient of each point in all intervals are arranged in ascending time order to obtain the wind load influence sequence and the vibration response sequence of each point; and the Pearson correlation coefficient between the wind load influence sequence and the vibration response sequence of each point is used as the consistency coefficient of each point.

5. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The method for determining the first fluctuation coefficient of the single-row support structure is: , where is the first wave coefficient of the j-th row support structure, is the cumulative sum of the shortest distances between each data in the consensus sequence of the j-th row support structure and the fitting curve of the consensus sequence of the j-th row support structure; It is the cumulative sum of the absolute values ​​of the slopes of the fitting curve of the j-th row of support structures at each point, wherein the consistency sequence of the j-th row of support structures is obtained by arranging the consistency coefficients of each point in the j-th row of support structures in the order of their positions.

6. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The method for determining the second fluctuation coefficient of the single-row support structure is: The average of all the cable force data at each point is calculated to obtain the average cable force data at each point, and the sequence composed of the average cable force data at all points of each row of supporting structures is recorded as the cable force average sequence; The expression of the second wave coefficient of the j-th row support structure is: , where is the second wave coefficient of the j-th row support structure, is the cumulative sum of the shortest distances between each data in the cable force mean sequence of the j-th row of support structures and the fitting curve of the cable force mean sequence, It is the cumulative sum of the absolute values ​​of the slopes of the fitting curve of the cable force mean sequence of the j-th row of support structures at each point.

7. A stress monitoring method for a flexible photovoltaic support system according to claim 1, characterized in that: The calculation method of the stress state abnormal value of the single-row support structure is the sum of the standard deviation of the stress data of all points in the single-row support structure and the mean of the first fluctuation coefficient and the second fluctuation coefficient.

8. A stress monitoring method for a flexible photovoltaic support system as claimed in claim 1, characterized in that: The monitoring of the stress state of each row of support structures further includes: if the stress state abnormal value is greater than or equal to the abnormal threshold, then the stress state of the corresponding support structure is abnormal.

Images