Photovoltaic ground subsidence observation and early warning system based on multi-modal monitoring data

Abstract

This invention discloses a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data, belonging to the field of photovoltaic foundation settlement monitoring technology. Under standard environmental conditions, this invention calculates the theoretical reference power of each string using an irradiance power model, generating a power reference interval. Strings that continuously deviate from this power reference interval are marked as target strings. The angular offset of the target strings is iteratively solved, and foundation settlement data is collected by fixed-point sensors or manually based on the angular offset to complete verification and early warning. From a macroscopic positioning perspective, this invention significantly reduces the number of hardware deployments, lowers the data transmission and storage power consumption of the entire station, enables rapid location of large-scale foundation settlement hazards, and improves the efficiency of site-wide inspection. From a microscopic acquisition and verification perspective, this invention controls the collection of foundation settlement data based on the angular offset, and uses the foundation settlement data for settlement early warning, reducing the long-term monitoring and maintenance investment of the site while ensuring the reliability of photovoltaic support foundation safety early warning.

Description

Technical Field

[0001] This invention belongs to the field of photovoltaic foundation settlement monitoring technology, specifically a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data. Background Technology

[0002] Ground-mounted photovoltaic (PV) power stations occupy a large area and have a large number of support columns. During long-term operation, uneven foundation settlement is prone to occur in soft soil, backfill soil, and slope areas. Tilting of the support columns will directly cause the angle of the PV module panels to shift, significantly reducing the power generation efficiency of the strings, and in severe cases, causing the support columns to deform and collapse. Currently, PV foundation settlement monitoring mainly falls into two categories of existing technical solutions: one is the settlement sensor deployment solution, which installs a static level and tilt sensor at each support column of the power station, relying on a large number of measuring points to collect settlement data in real time to achieve full-site monitoring; the other is the periodic manual inspection and re-measurement solution, in which maintenance personnel regularly carry a level to manually measure the elevation of each column throughout the station to determine the settlement deformation.

[0003] The existing solution has significant drawbacks: First, the hardware investment for densely distributed sensors across the entire area is extremely high. The scale of the support structure for the site often involves thousands of columns. The procurement and construction costs of sensors, supporting cables, and data acquisition gateways are high. The continuous transmission and storage of massive amounts of measurement data significantly increases the computing power and maintenance costs of the site server. The workload for subsequent equipment calibration and fault repair is enormous. Second, the efficiency of manual re-measurement of the entire site is extremely low. A single complete measurement of a large-area photovoltaic site takes a long time, and the inspection cycle is long. It is impossible to capture the slow-developing uneven settlement risks in time. Moreover, there is no basis for targeted monitoring, so only indiscriminate measurement of the entire area is possible, resulting in high labor costs.

[0004] This invention provides a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data to solve the above-mentioned technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data.

[0006] To achieve the above objectives, a first aspect of the present invention provides a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data, comprising: The observation and early warning module is used to identify strings with abnormal output power fluctuations as target strings under standard environmental conditions; it calculates the angular offset of the target strings based on the output power deviation; wherein, the impact of scattered irradiance fluctuations on total irradiance under standard environmental conditions is negligible; and, This is used to control the acquisition of foundation settlement data of target groups based on angular offset, and to provide settlement early warning based on the foundation settlement data; wherein, the foundation settlement data is acquired through sensors.

[0007] In one possible implementation, identifying strings with abnormal output power fluctuations as target strings under standard environmental conditions includes: The theoretical reference power of each string under standard environmental conditions is calculated using an irradiation power model; the power reference range of each string is generated based on the theoretical reference power and power fluctuation error. When the measured output power of a string remains outside the power reference range for several consecutive sampling periods, the corresponding string is marked as the target string.

[0008] In one possible implementation, the irradiation power model is as follows: ;in, These are the string numbers in a photovoltaic power station. For the first The theoretical reference power of the string under current operating conditions. For the first The standard test conditions for the total rated power of the string are as follows: The total irradiance of the photovoltaic panel. The standard test total irradiance; The power temperature coefficient of the component; Real-time temperature of the component backplane. Standard test temperature; For the first The string correction factor.

[0009] In one possible implementation, the angular offset of the target string is calculated based on the output power deviation, including: An irradiance-panel tilt angle mapping model is constructed based on the geometric principles of solar radiation; the total irradiance deviation of the target string is estimated based on the output power deviation. The angular offset of the target string is obtained by iteratively solving the combined total irradiance deviation and irradiance-panel tilt mapping model; the angular offset includes pitch angle offset and roll angle offset.

[0010] In one possible implementation, the irradiance-plate tilt angle mapping model is as follows: ;in, The angle of incidence of the sun. The solar altitude angle, The azimuth of the sun. The tilt angle of the photovoltaic panel. The azimuth angle of the photovoltaic panel. Normal direct irradiance, This represents the irradiance scattered by the inclined plane.

[0011] In one possible implementation, estimating the total irradiance deviation of the target string based on the output power deviation includes: The average output power and theoretical reference power of the target string under standard environmental conditions are extracted; where the effect of scattered irradiance fluctuation on total irradiance under standard environmental conditions is negligible. The output power deviation is calculated based on the theoretical reference power and the average output power, and the output power deviation is used as the total irradiance deviation.

[0012] In one possible implementation, the angular offset of the target string is obtained by iteratively solving the combined total irradiance deviation and irradiance-plate tilt mapping model, including: Retrieve the model iteration parameters for two characteristic time points; where the difference in solar azimuth angle between the two characteristic time points is greater than 30°. Furthermore, one characteristic time is noon, and the model iteration parameters include solar altitude angle, solar azimuth angle, and total irradiance deviation; Based on the target string design angle value, the irradiance-panel tilt angle mapping model is iteratively solved through model iteration parameters to obtain the angle offset; among which, the design angle value includes the pitch angle and azimuth angle of the photovoltaic panel.

[0013] In one possible implementation, the acquisition of foundation settlement data in the target string is controlled based on the angular offset, including: Calculate the settlement value set of the target string based on the angle offset; determine the settlement risk area in the target string based on the settlement value set and mark it as the target area; The system collects ground settlement data within the target area using sensors and provides early warnings of ground settlement based on this data. The ground settlement data can be collected either through deployed sensors or manually using handheld sensors.

[0014] In one possible implementation, the settlement values ​​of the target string are calculated based on the angular offset, including: Extract the structural design parameters of the target string; among which, the structural design parameters include the pitch direction span and the tilt direction column spacing or the number of cover spans; The settlement numerical set is calculated based on the structural design parameters and angular offset; the settlement numerical set includes the average settlement gradient, the average settlement value of a single span, the total settlement value of the string, the settlement value of the columns in the same row, and the lateral settlement gradient.

[0015] In one possible implementation, the settlement risk area in the target string is determined based on the settlement data set, including: Extract settlement extreme points from the settlement data set; where the settlement extreme points are the points with the largest foundation settlement. Extend several support spans centered on the settlement extreme point, and use the area covered by these support spans as the target area; the number of support spans extended shall not be less than 2 spans.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention first calculates the theoretical reference power of each string under standard environmental conditions using an irradiation power model, generating a power reference interval. Strings that continuously deviate from this power reference interval are marked as target strings. Then, based on a solar radiation geometric model, the angular offset of the target strings is iteratively solved using a dual-feature time step. The monitoring target area is delineated based on the angular offset. Finally, verification and early warning are completed by collecting ground settlement data using fixed-point sensors or manually. From a macroscopic positioning perspective, this invention replaces the dense deployment of settlement sensors throughout the entire photovoltaic power station with a full-domain scanning of the output power of each string in the photovoltaic power station, only requiring monitoring in standard clear weather conditions. Under operating conditions, anomaly identification is carried out by screening suspected settlement areas based on power reference ranges. This eliminates the need to configure monitoring equipment for each support, significantly reducing the number of hardware deployments and the consumption of data transmission and storage computing power for the entire station. At the same time, it avoids invalid monitoring caused by random light fluctuations, enabling rapid location of potential foundation settlement hazards over a large area and improving the efficiency of site-wide inspections. From a microscopic acquisition and verification perspective, this invention controls the acquisition of foundation settlement data based on angular offset, and uses the foundation settlement data for settlement early warning. While ensuring the reliability of the safety early warning for photovoltaic support foundations, it reduces the long-term monitoring and maintenance investment of the site. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the workflow of the observation and early warning module in this invention. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. 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] Please see Figure 1 The first aspect of the present invention provides a photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data, comprising: The observation and early warning module is used to identify strings with abnormal output power fluctuations as target strings under standard environmental conditions; calculate the angular offset of the target strings based on the output power deviation; wherein, the impact of scattered irradiance fluctuations on total irradiance under standard environmental conditions is negligible; and to control the acquisition of foundation settlement data of the target strings based on the angular offset, and to provide settlement early warning based on the foundation settlement data; wherein, the foundation settlement data is acquired through sensors.

[0021] The workflow of the observation and early warning module in the photovoltaic foundation settlement observation and early warning system of this invention is as follows: A01: Identify the target string as the string with abnormal output power fluctuations under standard environmental conditions; calculate the angular offset of the target string based on the output power deviation; wherein, the angular offset includes pitch angle offset and roll angle offset; A string in a photovoltaic (PV) power plant is an electrical unit formed by connecting several PV modules in series. The output power of a single string is the sum of the power generated by all the PV modules within it. A PV module is the smallest physical unit of PV power generation; it receives sunlight on its front side and outputs direct current (DC), serving as the core carrier for photoelectric conversion.

[0022] A02: The foundation settlement data of the target string is controlled and collected based on the angular offset, and settlement early warning is given based on the foundation settlement data; the foundation settlement data is collected by sensors.

[0023] If the angular offset of the target string deviates, it may be due to foundation settlement of the photovoltaic power station. Therefore, foundation settlement data for the target string is collected, and settlement warnings are issued based on this data. If the foundation settlement data indicates that the target string has not experienced foundation settlement, then the output power fluctuation is not due to foundation settlement.

[0024] When the output power of a photovoltaic (PV) string fluctuates abnormally, it may be affected by various factors, such as hardware damage, weather conditions, and foundation settlement. These strings are marked as target strings for further analysis. The angular offset is calculated based on the output power deviation of the target strings, and a settlement value set is estimated based on the angular offset. This settlement value set is used to determine whether foundation settlement data needs to be collected. If so, the foundation settlement data of the target strings is collected, and finally, settlement early warning is provided using the foundation settlement data. This invention is based on macroscopic full-domain initial screening and microscopic directional verification. It eliminates the need to deploy sensors at every string and every span of the support structure. Sensors only need to be sparsely deployed at preset benchmark points and high-risk settlement areas to achieve full-field settlement control, significantly reducing the number of sensors and daily data processing workload. While ensuring the reliability of PV foundation settlement monitoring, it also reduces the hardware and maintenance costs of PV foundation settlement monitoring.

[0025] The implementation process of identifying strings with abnormal output power fluctuations as target strings under standard environmental conditions in this invention is as follows: B01: Calculate the theoretical reference power of each string under standard environmental conditions using the irradiation power model; generate the power reference range for each string based on the theoretical reference power and power fluctuation error; Based on the output characteristics of photovoltaic modules, an irradiance power model is constructed for each string in a photovoltaic power plant. The irradiance power model is as follows: ;in, These are the string numbers in a photovoltaic power station. For the first The theoretical reference power of the string under current operating conditions. For the first The standard test conditions for the string are the total rated power (e.g., 10.8kW). The total irradiance of the photovoltaic panel. The standard test total irradiance (e.g., 1000W / ㎡); The component power temperature coefficient (e.g., -0.41% / ℃); Real-time temperature of the component backplane. Standard test temperature (e.g., 25℃); For the first The string correction coefficient can be determined by fitting the sample dataset of the strings. It is used to compensate for the performance dispersion of the components and the difference in line loss. The value range is [0.96, 1.02]. If the fit exceeds this range, the upper or lower limit of the range is used. The total rated power, total irradiance, component power temperature coefficient, and standard test temperature of the irradiance power model standard test conditions are all set according to industry standards. The total irradiance of the photovoltaic panel and the real-time temperature of the component backsheet are measured values.

[0026] For example, during the initial operation of a photovoltaic power station (before foundation settlement and when string performance is in its initial state), steady-state operating data of the strings is collected. The total irradiance of the photovoltaic panel and the real-time temperature of the module backsheet required by the above model are extracted from the steady-state operating data and input into the irradiance power model to obtain the uncorrected theoretical reference power. The ratio of the uncorrected theoretical reference power to the measured output power of the strings under the same operating conditions is used as the string correction coefficient. The mean of the string correction coefficients corresponding to multiple samples can be taken as the final string correction coefficient, and the string correction coefficient should be updated regularly to avoid drift.

[0027] Power fluctuation error is used to define the normal fluctuation range of the string's output power. Essentially, it adds an upper and lower fluctuation threshold to the theoretical reference power, conforming to engineering realities, to prevent power fluctuations during normal operation from being misjudged as abnormal, thereby ensuring the reliability of the determined target string in foundation settlement inversion. The power fluctuation error is set based on engineering experience, with a default value of 2.5%. It can also be dynamically set according to the fluctuation characteristics under different lighting conditions. For example, during the midday direct sunlight period (e.g., 10:30-14:30), the solar altitude angle is high, the proportion of direct irradiance is large, and the light stability is strong, resulting in relatively small overall output power fluctuations; therefore, the power fluctuation error is set to 2%. At other times, the solar altitude angle is low, the proportion of diffuse irradiance is high, and the light itself fluctuates greatly; therefore, the power fluctuation error is set to 3.5% to prevent normal output power fluctuations under low light conditions from being judged as abnormal.

[0028] In engineering practice, real-time operating data and operating environment data of all strings in a photovoltaic power station can be collected at preset time intervals (such as 5 minutes). Parameters are extracted from the real-time operating data and operating environment data and input into the irradiance power model of the corresponding string to obtain the theoretical reference power under standard environmental conditions. The power fluctuation error is matched with the operating condition characteristics, and the power reference range of the string is generated by combining the theoretical reference power and the power fluctuation error.

[0029] B02: When the measured output power of a string is outside the power reference range for several consecutive sampling periods, the corresponding string will be marked as the target string.

[0030] If the measured output power of a string is outside its power reference range for several consecutive sampling periods, it is marked as a target string. Single or intermittent deviations from the power reference range are considered random irradiance fluctuations and are not marked. Marking a string as a target string does not necessarily mean that the abnormal output power deviation is caused by photovoltaic foundation settlement; it could also be due to module failure, loose connections, or partial shading, requiring further analysis of the target module's angular offset. For example, with a sampling period of 5 minutes per sampling, if the measured output power is outside the power reference range for 6 consecutive sampling periods, the corresponding string is marked as a target string.

[0031] The implementation process for calculating the angular offset of the target string based on the output power deviation in this invention is as follows: C01: Construct an irradiance-panel tilt angle mapping model based on the geometric principles of solar radiation; estimate the total irradiance deviation of the target string based on the output power deviation; An irradiance-panel tilt angle mapping model is constructed based on the geometric principles of solar radiation: the angle between the incident solar ray and the normal to the photovoltaic panel is the solar incidence angle. It is determined by the real-time solar angle parameters and the plate attitude, and the specific formula is as follows: ;in, The solar altitude angle, The azimuth of the sun. The tilt angle of the photovoltaic panel. The azimuth angle of the photovoltaic panel is the angle between the horizontal projection of the normal to the photovoltaic panel and the due south direction.

[0032] The direct irradiance received by the photovoltaic panel follows Lambert's cosine law, and the total irradiance of the photovoltaic panel... Angle of incidence of the sun The relationship is: ;in, Normal direct irradiance, The irradiance is the irradiance scattered by the inclined plane. Under clear and stable conditions, the proportion of irradiance scattered by the inclined plane is stable and is not affected by the slight tilt angle of the photovoltaic panel. Therefore, the irradiance deviation caused by the change in the tilt angle of the photovoltaic panel is determined by the normal direct irradiance.

[0033] The two equations above can be combined into an irradiance-plate tilt angle mapping model, namely: .

[0034] C02: The angular offset of the target string is obtained by iteratively solving the combined total irradiance deviation and irradiance-panel tilt mapping model; where the angular offset includes pitch angle offset and roll angle offset.

[0035] The aforementioned solar incidence angle equation is a transcendental equation and cannot be solved analytically directly. This invention utilizes the Newton-Raphson iterative method to solve two characteristic time data with significant differences in solar azimuth simultaneously, decoupling and solving the pitch angle offset and roll angle offset, thereby obtaining the actual attitude of the target string, i.e., the angle offset, which includes the pitch angle offset and the roll angle offset.

[0036] The implementation process for estimating the total irradiance deviation of the target string based on the output power deviation in this invention is as follows: D01: Extract the average output power and theoretical reference power of the target string under standard environmental conditions; where the effect of scattered irradiance fluctuation on total irradiance under standard environmental conditions is negligible. Under standard environmental conditions, the output power of the photovoltaic string has an approximately linear relationship with the total irradiance of the photovoltaic panel. Therefore, the deviation in the output power of the string is approximately equal to the deviation in the total irradiance of the photovoltaic panel. ;in, The theoretical reference power of the target string, This represents the average output power.

[0037] It should be noted that the total irradiance deviation and the output power deviation are approximately equal under standard environmental conditions. As mentioned above, this is because the fluctuation of the slope-scattered irradiance is small under standard environmental conditions, and its impact on the total irradiance is negligible. Standard environmental conditions refer to a condition where the scattered irradiance fluctuation is ≤3% within a short period of time (e.g., 10 minutes), that is, under conditions where the sky is clear with no obvious cloud cover and the proportion of scattered irradiance is stable.

[0038] It is worth noting that this invention uses the output power deviation instead of the total irradiance deviation when calculating the angular offset of the target string based on the actual total irradiance. This requires a stable proportion of scattered irradiance, which cannot be guaranteed under extreme weather conditions. Therefore, when calculating the output power deviation, it is necessary to ensure that the weather conditions meet the requirements of standard environmental conditions. Since the development of photovoltaic foundation settlement is a gradual process, photovoltaic foundation settlement monitoring will not be carried out in the event of extreme weather. Monitoring will only be conducted after the weather clears up, the proportion of scattered irradiance stabilizes, and the standard environmental conditions are met.

[0039] D02: Calculate the output power deviation based on the theoretical reference power and the average output power, and use the output power deviation as the total irradiance deviation.

[0040] The implementation process of obtaining the angular offset of the target string by iteratively solving the combined total irradiance deviation and irradiance-plate tilt angle mapping model in this invention is as follows: E01: Retrieve the model iteration parameters for two characteristic time points; where the difference in solar azimuth angle between the two characteristic time points is greater than... Furthermore, one characteristic time is noon, and the model iteration parameters include solar altitude angle, solar azimuth angle, and total irradiance deviation; E02: Based on the target string design angle value, the irradiance-panel tilt angle mapping model is iteratively solved through model iteration parameters to obtain the angle offset; among which, the design angle value includes the pitch angle and azimuth angle of the photovoltaic panel.

[0041] When the difference in solar azimuth angle between two moments is greater than At this time, the sensitivities of the two equations are significantly complementary: usually one time is close to noon (high sensitivity to pitch and low sensitivity to roll), and the other time is in the morning / afternoon (high sensitivity to roll and second highest sensitivity to pitch). The two sets of equations are highly independent, the Jacobian matrix is ​​nonsingular, the iterative process converges quickly and accurately, and the angle offset in both dimensions can be stably solved.

[0042] As an example, and not a limitation, we select two characteristic times with a large difference in solar azimuth angle, such as 10:00 AM and 12:00 PM. Let's assume 10:00 AM: Solar altitude angle... Sun azimuth Corresponding to total irradiance deviation 12:00 noon: Solar altitude angle Sun azimuth Corresponding to total irradiance deviation .

[0043] Design value of photovoltaic panels in photovoltaic power plants , As the initial values ​​for iteration (installed strictly according to design values, with minimal installation error), the actual total irradiance at two characteristic moments is substituted into the irradiance-panel tilt angle mapping model, and continuously corrected through iteration. and The iteration result is output when the relative error between the calculated total irradiance and the actual total irradiance is less than 0.1%, and the single-step angle change is less than a preset angle threshold (e.g., 0.001°). The iteration parameters can be set empirically; for example, if the number of iterations is 50, the iteration step size can be set to... The minimum effective step size is .

[0044] After iterative convergence, the actual plane attitude of the target string is obtained: actual pitch angle. The pitch angle offset value is The negative sign indicates a decrease in tilt angle; actual azimuth angle The roll angle offset value is The negative sign indicates that the normal line is biased to the east.

[0045] After calculating the angular offset of the target string according to the aforementioned scheme, if the angular offset is greater than the preset offset threshold, the area covered by the target string is taken as the target area, and the ground settlement data in the target area is collected by the sensor. The ground settlement data is used to issue a settlement warning.

[0046] For example, the pitch angle threshold is set to 0.3°, and the roll angle threshold is 0.2°. If any angle value in the angular offset of the target string is greater than the corresponding angle threshold, it means that the angular offset is greater than the preset offset threshold.

[0047] In some other preferred embodiments, the settlement risk area within the target string coverage area can also be determined by the angular offset, and foundation settlement data in the settlement risk area can be collected.

[0048] The implementation process of controlling the acquisition of foundation settlement data in the target string based on the angle offset in this invention is as follows: F01: Calculate the settlement value set of the target string based on the angle offset; determine the settlement risk area in the target string based on the settlement value set and mark it as the target area; F02: Collect ground settlement data within the target area using sensors, and issue settlement warnings based on the ground settlement data; the ground settlement data is collected either through deployed sensors or manually.

[0049] After identifying the target area, sensors within that area are used to collect settlement monitoring data, and settlement warnings are issued based on this data. Sensors include hydrostatic levels and MEMS tilt sensors, and settlement monitoring data includes cumulative settlement and differences in settlement between adjacent areas. If the coverage of the target area by deployed sensors is incomplete, data is collected manually.

[0050] For example, if there are 8 pillars in the target area, but only 2 pillars are equipped with sensors, resulting in a coverage rate of only 25%, the sensor coverage is deemed incomplete, and manual supplementary measurements are initiated simultaneously. The remaining 6 pillars are manually measured using a precision electronic level, collecting elevation data for each pillar with second-order leveling accuracy, and converting this data into cumulative settlement. If the cumulative settlement exceeds a preset threshold, an early warning is issued.

[0051] In photovoltaic power plants, micro-sensors are sparsely deployed at benchmark points and high-risk areas to reduce hardware investment costs.

[0052] The benchmarks include permanent off-site benchmarks and working benchmarks on-site. Permanent off-site benchmarks are located in the undisturbed soil area outside the photovoltaic power station's perimeter wall, away from construction disturbances, vehicle traffic, and settlement areas. The base is situated on stable undisturbed soil / bedrock, and two benchmarks can be set up. Working benchmarks on-site are located in the stable area at the edge of each array, adjacent to the array inspection channel, avoiding the disturbance range of the support foundation. One working benchmark can be set up for each array.

[0053] High-risk areas include high-level backfill soil areas, high-risk slope areas, and high-risk soft soil areas. In high-level areas, several sensors can be sparsely and evenly deployed, with the number of sensor deployment points being greater than 50% of the number of columns.

[0054] The implementation process of calculating the settlement values ​​of the target string based on the angular offset in this invention is as follows: G01: Extract the structural design parameters of the target string; among which, the structural design parameters include the pitch direction span and the roll direction column spacing or the number of covered spans; The strings are typically laid continuously along the longitudinal direction (north-south) across multiple support spans, covering... A single support span typically covers two rows of columns horizontally; the support span is the basic load-bearing unit in a photovoltaic power station.

[0055] G02: Settlement numerical data are calculated based on structural design parameters and angular offsets; the settlement numerical data includes average settlement gradient, average settlement value per span, total settlement value of the string, settlement value of columns in the same row, and lateral settlement gradient.

[0056] Uneven foundation settlement causes a vertical height difference between adjacent columns, which in turn causes the photovoltaic panels to rotate around the corresponding axis, ultimately resulting in a shift in pitch / tilt angle. Uneven foundation settlement is also the monitoring object of this invention. The process for obtaining the settlement data set is as follows: Extract the pitch span from the design parameters of the photovoltaic power station Column spacing in the tilt direction The target string covers a number of spans Pitch direction span The horizontal spacing between two adjacent columns in the front-to-back direction (usually north-south) of the support structure is the span of the main beam; the column spacing in the lateral tilt direction is... The horizontal spacing between adjacent columns in the same row along the left-right (usually east-west) direction of the target string; the number of spans covered by the target string. The number of support spans along the pitch direction of the target string is used to calculate the average settlement gradient of the target string.

[0057] The pitch angle offset corresponds to the vertical settlement difference between adjacent columns in the front and rear directions of the column, and the calculation formula is as follows: , This represents the pitch angle offset. If the pitch angle decreases, the photovoltaic panel tilts backward, and the settlement of the north-side support column of the string is greater than that of the south-side support column; if the pitch angle increases, the photovoltaic panel tilts forward, and the settlement of the south-side support column of the string is greater than that of the north-side support column.

[0058] The difference in vertical settlement between adjacent columns in the same row on the left and right sides of the column corresponds to the tilt angle offset. The calculation formula is as follows: , This represents the tilt angle offset. If the photovoltaic panel is tilted to the east, the settlement of the east-side support column will be greater than that of the west-side support column; if the photovoltaic panel is tilted to the west, the settlement of the west-side support column will be greater than that of the east-side support column.

[0059] If the target string covers multiple spans of supports, the inverted angular offset is the average offset within the string range, and the corresponding average settlement gradient of the string region is: , This represents the settlement difference per unit length. Combined with the total span covered by the target string, the total settlement value at both ends of the target string (the product of the number of spans and the average settlement gradient) can be calculated.

[0060] The settlement data set includes the mean settlement gradient. Average settlement per span Total settlement of the string, settlement of the columns in the same row The lateral settlement gradient is calculated in the same way as the average settlement gradient.

[0061] The implementation process for determining the settlement risk area in the target string based on the settlement data set in this invention is as follows: H01: Extract settlement extreme points from the settlement data set; where the settlement extreme points are the points with the largest foundation settlement. H02: Extend several support spans centered on the settlement extreme point, and use the area covered by several support spans as the target area; wherein, the number of support spans extended shall not be less than 2 spans.

[0062] For example, suppose the pitch angle offset of the target string is... The average settlement per span was 33.5 mm, with the north-side column experiencing greater settlement than the south-side column; the lateral tilt angle shifted. The settlement difference of the columns in the same row is 3.8mm, and the settlement of the columns on the east side is greater than that on the west side.

[0063] Spatial overlay was performed using the results of both pitch and tilt settlement assessments: the north side was identified as the extreme pitch settlement point, and the east side as the extreme tilt settlement point. The northeast corner of the target string was determined as the extreme settlement point. The span containing this extreme settlement point was taken as the base span, and several extended spans were determined by extending the base span by two spans. The area covered by the base span and the extended spans was taken as the target area.

[0064] It should be noted that if several strings in a photovoltaic power station are set up adjacently, when an extended span is obtained by expanding from the base span, the extended span can be located in an adjacent string.

[0065] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments.

[0066] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any other combination thereof. When implemented using a software program, it can be implemented entirely or partially in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0067] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data, characterized in that, include: Observation and early warning module: used to identify strings with abnormal output power fluctuations as target strings under standard environmental conditions; The angular offset of the target string is calculated based on the output power deviation; where the impact of scattered irradiance fluctuations on the total irradiance under standard environmental conditions is negligible; and... This is used to control the acquisition of foundation settlement data of target groups based on angular offset, and to provide settlement early warning based on the foundation settlement data; wherein, the foundation settlement data is acquired through sensors.

2. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 1, characterized in that, Under standard operating conditions, strings exhibiting abnormal output power fluctuations are identified as target strings, including: The theoretical reference power of each string under standard environmental conditions is calculated using an irradiation power model; the power reference range of each string is generated based on the theoretical reference power and power fluctuation error. When the measured output power of a string remains outside the power reference range for several consecutive sampling periods, the corresponding string is marked as the target string.

3. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 1, characterized in that, The irradiation power model is as follows: ;in, These are the string numbers in a photovoltaic power station. For the first The theoretical reference power of the string under current operating conditions. For the first The standard test conditions for the total rated power of the string are as follows: The total irradiance of the photovoltaic panel. The standard test total irradiance; The power temperature coefficient of the component; Real-time temperature of the component backplane. Standard test temperature; For the first The string correction factor.

4. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 1, characterized in that, The angular offset of the target string is calculated based on the output power deviation, including: An irradiance-panel tilt angle mapping model is constructed based on the geometric principles of solar radiation; the total irradiance deviation of the target string is estimated based on the output power deviation. The angular offset of the target string is obtained by iteratively solving the combined total irradiance deviation and irradiance-panel tilt mapping model; the angular offset includes pitch angle offset and roll angle offset.

5. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 4, characterized in that, The irradiance-plate tilt angle mapping model is as follows: ;in, The angle of incidence of the sun. The solar altitude angle, The azimuth of the sun. The tilt angle of the photovoltaic panel. The azimuth angle of the photovoltaic panel. Normal direct irradiance, This represents the irradiance scattered by the inclined plane.

6. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 4, characterized in that, The total irradiance deviation of the target string is estimated based on the output power deviation, including: Extract the average output power and theoretical reference power of the target string under standard environmental conditions; The output power deviation is calculated based on the theoretical reference power and the average output power, and the output power deviation is used as the total irradiance deviation.

7. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 4, characterized in that, The angular offset of the target string is obtained by iteratively solving the combined total irradiance deviation and irradiance-plate tilt mapping model, including: Retrieve the model iteration parameters for two characteristic time points; where the difference in solar azimuth angle between the two characteristic time points is greater than... Furthermore, one characteristic time is noon, and the model iteration parameters include solar altitude angle, solar azimuth angle, and total irradiance deviation; Based on the target string design angle value, the irradiance-panel tilt angle mapping model is iteratively solved through model iteration parameters to obtain the angle offset; among which, the design angle value includes the pitch angle and azimuth angle of the photovoltaic panel.

8. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 1, characterized in that, The foundation settlement data in the target string is collected based on the angle offset, including: Calculate the settlement value set of the target string based on the angle offset; determine the settlement risk area in the target string based on the settlement value set and mark it as the target area; The system collects ground settlement data within the target area using sensors and provides early warnings of ground settlement based on this data. The ground settlement data can be collected either through deployed sensors or manually using handheld sensors.

9. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 8, characterized in that, The settlement values ​​of the target string are calculated based on the angular offset, including: Extract the structural design parameters of the target string; among which, the structural design parameters include the pitch direction span and the tilt direction column spacing or the number of cover spans; The settlement numerical set is calculated based on the structural design parameters and angular offset; the settlement numerical set includes the average settlement gradient, the average settlement value of a single span, the total settlement value of the string, the settlement value of the columns in the same row, and the lateral settlement gradient.

10. The photovoltaic foundation settlement observation and early warning system based on multimodal monitoring data according to claim 8, characterized in that, Based on the settlement data set, the settlement risk areas in the target string are determined, including: Extract settlement extreme points from the settlement data set; where the settlement extreme points are the points with the largest foundation settlement. Extend several support spans centered on the settlement extreme point, and use the area covered by these support spans as the target area; the number of support spans extended shall not be less than 2 spans.

Images