Automatic adjustment method, system and storage medium for monitoring strategy of offshore wind turbines

Abstract

The present application relates to the technical field of offshore wind power, and discloses an automatic adjustment method, system, and storage medium for an offshore wind turbine monitoring strategy. The method collects data through electrochemical sensors on the same unit, calculates corrosion data through a first algorithm, and issues a first or second unmanned monitoring mark accordingly. A first inspection path is constructed based on the first unmanned monitoring mark, and a first unmanned device inspection is called to obtain first monitoring data. The first corrosion monitoring data is calculated, and if it exceeds the standard, an early warning is issued and the monitoring mark is modified. A second inspection path is constructed based on the second unmanned monitoring mark, and a second unmanned device inspection is called to obtain second monitoring data. The second corrosion monitoring data is calculated, and if it exceeds the standard, an alarm is issued. This method realizes intelligent monitoring of the corrosion status of offshore wind turbines and effectively improves monitoring efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of offshore wind power, and in particular to an automatic adjustment method, system and storage medium for an offshore wind turbine monitoring strategy. Background Art

[0002] With the surge in global demand for clean energy, offshore wind power, with its abundant resources, high power generation efficiency, and lack of land occupation, has become a hotspot in the new energy sector. In recent years, the scale of offshore wind farms has continued to expand, the capacity of individual turbines has continued to increase, and their development has gradually moved from nearshore to offshore locations.

[0003] The marine environment, characterized by high salt spray, high humidity, and strong winds and waves, is highly corrosive. The connection structures of offshore wind turbines, exposed to such conditions for long periods, are highly susceptible to corrosion. This corrosion can reduce the bearing capacity of the connection structures, leading to loosening and increased vibration, seriously threatening the structural safety of the wind turbines.

[0004] Currently, corrosion assessment of offshore wind turbine connection structures primarily relies on manual inspections. Workers must travel by boat to the turbine site and assess corrosion using visual inspections and simple measurement tools. This method is significantly affected by weather and sea conditions, resulting in low efficiency and highly subjective results. With the growing scale of offshore wind farms, existing assessment methods are no longer sufficient to meet operational and maintenance needs, and intelligent monitoring strategies are urgently needed. Summary of the Invention

[0005] In order to improve the working efficiency of offshore wind turbine monitoring, the present application provides an automatic adjustment method, system and storage medium for offshore wind turbine monitoring strategy.

[0006] In a first aspect, the present application provides a method for automatically adjusting an offshore wind turbine monitoring strategy, which employs the following technical solutions:

[0007] A method for automatically adjusting an offshore wind turbine monitoring strategy comprises the following steps:

[0008] Acquiring multiple electrochemical data based on multiple electrochemical sensors disposed on the same unit, and calculating corrosion data using a first algorithm based on the multiple electrochemical data;

[0009] If the corrosion data is less than a preset reference corrosion value, a first unmanned monitoring flag is issued; otherwise, a second unmanned monitoring flag is issued; both the first unmanned monitoring flag and the second unmanned monitoring flag include the address data of the unit;

[0010] Obtaining the first unmanned monitoring mark and the corresponding address data, and constructing a first inspection path for the unit corresponding to the address data according to a preset first sorting strategy;

[0011] Invoking the calibrated first unmanned equipment to perform inspection based on the first inspection path to obtain first monitoring data of the corresponding unit;

[0012] Calculating first corrosion monitoring data using a first algorithm based on the first monitoring data; and if the first corrosion monitoring data is greater than the reference corrosion value, issuing an on-site electrochemical sensor warning prompt and changing the first unmanned monitoring flag of the unit to the second unmanned monitoring flag;

[0013] Obtaining the second unmanned monitoring mark and the corresponding address data, and constructing a second inspection path for the unit corresponding to the address data according to a preset second sorting strategy;

[0014] Invoking the calibrated second unmanned equipment to perform inspection based on the second inspection path to obtain second monitoring data of the corresponding unit;

[0015] Second corrosion monitoring data is calculated using a second algorithm based on the second monitoring data. If the second corrosion monitoring data is greater than a preset standard corrosion value, an on-site electrochemical sensor alarm prompt is issued.

[0016] By adopting the above technical solution, electrochemical data is collected in real time with the help of multiple electrochemical sensors installed on the same unit, and corrosion data is calculated by a first algorithm. If the corrosion data is less than a preset reference corrosion value, a first unmanned monitoring mark containing the unit address data is issued, otherwise a second unmanned monitoring mark is issued; then, according to a preset first sorting strategy, a first inspection path is constructed for the unit address data corresponding to the first unmanned monitoring mark, and a calibrated first unmanned device (unmanned ship) is called to inspect along this path and obtain first monitoring data, and then the first corrosion monitoring data is calculated by the first algorithm. If the first corrosion monitoring data is greater than the reference corrosion value, an on-site electrochemical sensor early warning prompt is triggered, and the unmanned monitoring mark of the unit is upgraded to a second mark; according to the preset second sorting strategy, a second inspection path is constructed for the unit address data corresponding to the second unmanned monitoring mark, and a calibrated second unmanned device (unmanned aerial vehicle) is called to inspect along the path to obtain second monitoring data, and the second corrosion monitoring data is calculated by a second algorithm. If the second corrosion monitoring data exceeds the preset standard corrosion value, an on-site electrochemical sensor alarm prompt is issued. This solution uses electrochemical sensors to initially screen the corrosion status, combined with graded inspections by unmanned ships and unmanned aerial vehicles, to improve monitoring efficiency and accuracy, and reduce reliance on manual inspections.

[0017] Optionally, the step of calculating the corrosion data further includes the following sub-steps:

[0018] The electrochemical sensor sends the collected electrochemical data to a transfer station via a first communication protocol, and the transfer station sends the collected or received electrochemical data to a remote backend via a second communication protocol;

[0019] The first algorithm is a weighted average algorithm, wherein the weight of the electrochemical data is positively correlated with the usage time of the electrochemical sensor; the longer the usage time of the electrochemical sensor, the greater the weight of the electrochemical data; the shorter the usage time of the electrochemical sensor, the smaller the weight of the electrochemical data;

[0020] If the first corrosion monitoring data is less than the reference corrosion value, the difference between the first corrosion monitoring data and the corresponding corrosion data is calculated; the weight and value of the electrochemical data are adjusted according to the positive correlation of the difference; the larger the difference, the larger the weight and value of the electrochemical data; the smaller the difference, the smaller the weight and value of the electrochemical data.

[0021] By adopting the above technical solution, the first communication protocol is used for short-distance communication, the second communication protocol is used for long-distance communication, the weight of the electrochemical data is adjusted according to the usage time of the electrochemical sensor, and the proportion of data from the electrochemical sensor with a longer usage time in the first corrosion monitoring data is increased. The weight and value of the electrochemical data are adjusted according to the positive correlation of the difference, which can amplify the final value of the first corrosion monitoring data.

[0022] Optionally, the step of calling the calibrated first unmanned device further includes the following sub-steps:

[0023] The first unmanned equipment is an unmanned boat equipment equipped with the electrochemical sensor;

[0024] The first corrosion monitoring data is calculated using the first algorithm based on the first monitoring data. If the first corrosion monitoring data is smaller than the corrosion data, the gain corresponding to the corrosion data in the first algorithm is adjusted in a positive correlation based on the difference between the first corrosion monitoring data and the corresponding corrosion data, or the reference corrosion value is adjusted in an anti-correlation based on the difference between the first corrosion monitoring data and the corresponding corrosion data. The greater the difference between the first corrosion monitoring data and the corresponding corrosion data, the greater the gain corresponding to the corrosion data in the first algorithm, and the smaller the reference corrosion value. The smaller the difference between the first corrosion monitoring data and the corresponding corrosion data, the smaller the gain corresponding to the corrosion data in the first algorithm, and the larger the reference corrosion value.

[0025] The first sorting strategy is the shortest path strategy after removing duplicate units: based on the address data corresponding to the first unmanned monitoring mark, duplicate units are eliminated; starting from the unmanned boat equipment, all units are traversed and the driving path length is calculated; the shortest driving path length is selected as the first shortest path.

[0026] By adopting the above technical solution, the first unmanned equipment uses an unmanned boat equipped with an electrochemical sensor to monitor the flange connecting bolts at the root of the tower. When the first corrosion monitoring data is less than the original corrosion data, the system adjusts the algorithm gain through the positive correlation of the difference to improve the sensitivity to weak corrosion signals; or the larger the difference, the greater the dynamic downward adjustment of the reference corrosion value, avoiding false negative results caused by environmental fluctuations (such as ocean current erosion) and improving the alarm accuracy; the shortest path strategy after deduplication is adopted to compress the inspection distance and improve the average daily inspection capacity.

[0027] Optionally, the step of calling the calibrated first unmanned device further includes the following sub-steps:

[0028] Within a preset time period, calculating the number of times the first unmanned monitoring mark is modified to the second unmanned monitoring mark;

[0029] According to the difference between the number of modifications and the preset reference number, the calling cycle of the first unmanned device is adjusted in an anti-correlation manner; the greater the difference between the number of modifications and the preset reference number, the shorter the calling cycle of the first unmanned device; the smaller the difference between the number of modifications and the preset reference number, the longer the calling cycle of the first unmanned device.

[0030] By adopting this technical solution, adjusting the call cycle based on the number of modifications automatically adapts the monitoring strategy of the first unmanned device to the actual situation of the on-site unit. A large difference between the number of modifications and the reference number indicates an increased corrosion risk for the on-site unit, and the system automatically shortens the unmanned vessel call cycle to promptly detect rapidly developing corrosion hazards. Conversely, a small difference indicates stable corrosion conditions, and the call cycle can be extended to avoid redundant monitoring.

[0031] Optionally, the step of calling the calibrated second unmanned device further includes the following sub-steps:

[0032] The second unmanned equipment is an unmanned aerial vehicle equipped with an image sensor;

[0033] The second algorithm is an image recognition algorithm, and a template matching value between the second monitoring data and a preset corrosion template is used as the second corrosion monitoring data;

[0034] The second sorting strategy includes the following sub-steps:

[0035] Calculating the shortest path strategy after removing duplicate units: Based on the address data corresponding to the second unmanned monitoring mark, after removing duplicate units; starting from the unmanned aerial vehicle device, traversing all units and calculating the travel path length; selecting the shortest travel path length as the second shortest path;

[0036] The current wind direction is obtained, the upwind path and the upwind path in the shortest path strategy are calculated, and the second unmanned device is controlled to first perform detection according to the upwind path and then perform detection according to the upwind path.

[0037] By adopting the above technical solution, the drone tests the flange connection bolts above the tower root, the blade root bolts, etc., first testing against the wind and then against the wind. If an emergency situation of insufficient power is encountered in the second half of the route, more units can be tested on the return journey.

[0038] Optionally, the step of calling the calibrated second unmanned device further includes the following sub-steps:

[0039] Calculating a distance ratio between the upwind path and the upwind path;

[0040] The calling period of the second unmanned device is adjusted inversely according to the distance ratio; the larger the distance ratio, the shorter the calling period of the second unmanned device; the smaller the distance ratio, the longer the calling period of the second unmanned device.

[0041] By adopting the above technical solution, adjusting the calling cycle according to the distance ratio can enable the monitoring strategy of the second unmanned device to automatically adapt to the actual situation of the on-site environment.

[0042] Optionally, the step of calling the calibrated second unmanned device further includes the following sub-steps:

[0043] If the first unmanned device and the second unmanned device are working at the same time;

[0044] Then, a plurality of first shortest paths of the first unmanned equipment and a plurality of second shortest paths of the second unmanned equipment are obtained;

[0045] Among the plurality of first shortest paths and the plurality of second shortest paths, the first shortest path and the second shortest path that are closest to each other are selected.

[0046] By adopting the above technical solution, the planning of the drone detection path - multiple drones work together (including unmanned ships), selecting the first shortest path and the second shortest path that are closest to each other, can make the control stations of the first unmanned equipment and the second unmanned equipment as close as possible or located in the same position.

[0047] Optionally, the method further comprises the following steps:

[0048] Obtaining the current location, current energy consumption, and remaining power of the second unmanned device;

[0049] Calculating the remaining mileage based on the current energy consumption and the remaining power;

[0050] Calculating return mileage based on the current location and a preset key location;

[0051] If the remaining mileage is less than the return mileage, a collaborative operation request is issued;

[0052] Based on the collaborative operation request, scanning the first unmanned equipment within an area with the current location as the center and a radius of the remaining mileage;

[0053] Sending the real-time coordinates of the first unmanned device closest to the current position to the second unmanned device;

[0054] The second unmanned device moves to the first unmanned device according to the real-time coordinates.

[0055] By adopting the above technical solution, if the unmanned aerial vehicle is low on power, it will fly to the nearest unmanned ship without affecting the operation of the unmanned ship.

[0056] In a second aspect, the present application provides an automatic adjustment system for monitoring strategies of offshore wind turbines, which adopts the following technical solutions:

[0057] A system for automatically adjusting a monitoring strategy for an offshore wind turbine generator set comprises a processor, wherein the processor executes the steps of any one of the above-mentioned methods for automatically adjusting a monitoring strategy for an offshore wind turbine generator set.

[0058] In a third aspect, the present application provides a storage medium that adopts the following technical solution:

[0059] A storage medium stores a program, which, when executed by a processor, implements the steps of the automatic adjustment method of the offshore wind turbine monitoring strategy described in any one of the above.

[0060] In summary, the present application includes at least one of the following beneficial technical effects: electrochemical data is collected using multiple electrochemical sensors on the same unit, corrosion data is calculated using a weighted average algorithm, and if the corrosion data is less than a reference value, a first unmanned monitoring flag is issued, and the unmanned ship conducts an inspection along the shortest path after eliminating duplicate units; if the corrosion data is greater than or equal to the reference value, a second unmanned monitoring flag is issued, and the unmanned aerial vehicle conducts an inspection along the shortest path taking into account wind direction. During the inspection process, if the first corrosion monitoring data of the unmanned ship is less than the corrosion data, the algorithm gain is adjusted in a positive correlation or the reference corrosion value is adjusted in an anti-correlation according to the difference between the number of times the first flag is modified to the second flag and the reference number, and the unmanned ship call cycle is adjusted in an anti-correlation according to the ratio of the windward and upwind path distances of the unmanned aerial vehicle inspection. When the unmanned ship and the unmanned aerial vehicle are working simultaneously, the closest path is selected to centralize the control stations, and when the unmanned aerial vehicle is low on power, it can fly to the nearest unmanned ship. This method realizes hierarchical monitoring, dynamic algorithm, path optimization, and equipment coordination, improving monitoring efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A step diagram of an automatic adjustment method for an offshore wind turbine monitoring strategy.

[0062] Figure 2 It is a diagram of the sub-steps for calculating corrosion data.

[0063] Figure 3 This is a sub-step diagram for calling the first unmanned device after calibration. DETAILED DESCRIPTION

[0064] Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings.

[0065] Throughout this specification, reference to the terms "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0066] The present application discloses an automatic adjustment method for monitoring strategy of offshore wind turbines, referring to Figure 1 , including the following steps:

[0067] Corrosion data is calculated using a first algorithm based on multiple electrochemical data collected from multiple electrochemical sensors installed on the same wind turbine. A three-electrode system (Ag / AgCl reference electrode, working electrode of the monitored metal material, and platinum auxiliary electrode) is used to form an electrochemical sensor array. This array is deployed on susceptible corrosion sites, such as tower flange bolts and blade root bolts, on the same wind turbine. The extent of corrosion is quantified through real-time measurements of parameters such as corrosion potential and current density. Temperature, humidity, and salt spray concentration sensors, such as capacitive humidity sensors and conductive salt spray sensors, are also integrated to collect environmental parameters (e.g., the corrosion rate increases by 2-3 times when humidity exceeds 80% RH and salt spray concentration exceeds 30 mg / m³), providing multi-dimensional input for the corrosion model.

[0068] If the corrosion data is less than the preset reference corrosion value, indicating that the corrosion situation of the current unit is relatively safe, a first unmanned monitoring flag is issued. Otherwise, when the corrosion data is greater than or equal to the preset reference corrosion value, indicating that the unit's corrosion situation may be at risk, a second unmanned monitoring flag is issued. Both the first and second unmanned monitoring flags include the unit's address data, which can be obtained through the Global Positioning System (GPS) or other positioning technologies and is subsequently used to accurately identify units requiring inspection.

[0069] A first unmanned monitoring tag and its corresponding address data are obtained, and a first inspection route is constructed for the units corresponding to the address data according to a preset first sorting strategy. The first sorting strategy may comprehensively consider multiple factors, such as the distance between units, their geographic location, and current sea conditions. For example, a distance-prioritized sorting algorithm may be employed to prioritize closely spaced units on the same route, thereby reducing inspection time and energy consumption for unmanned equipment.

[0070] The calibrated first unmanned equipment (such as an unmanned boat) is called to conduct an inspection based on the first inspection path to obtain the first monitoring data of the corresponding unit. During the inspection process, the unmanned boat obtains the first monitoring data of the corresponding unit through various sensors carried by it. These data include not only electrochemical data, but also environmental data and other information.

[0071] Based on the first monitoring data, a first corrosion monitoring value is calculated using a first algorithm. If the first corrosion monitoring value is greater than the reference corrosion value, indicating that the unit's corrosion situation has exceeded the initially assessed safety range, an on-site electrochemical sensor early warning prompt is issued. This warning prompt can be implemented in various ways, such as audible and visual alarms, and text message notifications to operation and maintenance personnel, to promptly attract the attention of relevant personnel. At the same time, the unit's first unmanned monitoring flag is changed to a second unmanned monitoring flag, indicating that the unit's corrosion risk level has increased, requiring closer monitoring and higher-level inspections.

[0072] The second unmanned monitoring flag and its corresponding address data are obtained, and a second inspection route is constructed for the units corresponding to the address data according to the preset second sorting strategy. Compared to the first sorting strategy, the second sorting strategy places greater emphasis on rapid response and comprehensive monitoring of high-risk units. The units are sorted based on factors such as corrosion severity and importance, with units with more severe corrosion conditions prioritized for inspection.

[0073] A calibrated second unmanned device (e.g., an unmanned aerial vehicle) is called upon to conduct an inspection along the second inspection route to obtain the second monitoring data of the corresponding unit. The unmanned aerial vehicle has advantages such as high maneuverability and is not restricted by sea conditions. It can quickly reach the target unit and obtain the second monitoring data of the corresponding unit.

[0074] A second corrosion monitoring value is calculated based on the second monitoring data using a second algorithm. If the second corrosion monitoring value exceeds a preset standard corrosion value, indicating that the unit's corrosion has reached a relatively serious level, an on-site electrochemical sensor alarm will be triggered. Alarm notification methods have been further upgraded, including voice broadcasts and alerts to multiple levels of operation and maintenance management personnel.

[0075] By using electrochemical sensors to initially screen for corrosion conditions, combined with tiered inspections using unmanned vessels and drones, this method improves monitoring efficiency and accuracy, reducing reliance on manual inspections. In practical applications, this method can promptly identify corrosion risks in offshore wind turbines, providing a scientific basis for turbine maintenance and management, and ensuring the safe and stable operation of offshore wind farms.

[0076] Reference Figure 2 , the step of calculating the corrosion data also includes the following sub-steps:

[0077] Electrochemical sensors transmit collected electrochemical data to a transfer station via the ZigBee protocol. The ZigBee protocol is suitable for short-range, low-power, and low-rate wireless communication scenarios. Its advantages include self-organizing networks, low cost, and strong anti-interference capabilities, making it ideal for transmitting data from electrochemical sensors dispersed across offshore wind turbines. Each electrochemical sensor, acting as a node in the network, automatically joins the ZigBee network and establishes a communication connection with the transfer station. In practice, the ZigBee network boasts an effective transmission range of over 100 meters, meeting the communication requirements between sensors within the wind turbine and the transfer station.

[0078] The transfer station transmits the collected or received electrochemical data to a remote backend server via a second communication protocol (LoRa). The LoRa protocol is a long-range, low-power, wide-area IoT communication technology with transmission distances of several kilometers or even tens of kilometers, making it ideal for geographically dispersed scenarios like offshore wind farms. The transfer station, acting as a data aggregation node, collects data from multiple electrochemical sensors and transmits it to a remote backend server via a LoRa gateway. This two-tiered communication architecture (ZigBee + LoRa) leverages the advantages of both short-range and long-range communications, ensuring reliable data transmission while reducing system cost and power consumption.

[0079] The first algorithm is a weighted average algorithm, in which the weight of the electrochemical data is positively correlated with the usage time of the electrochemical sensor. Over the long-term use of electrochemical sensors, the stability and reliability of their measurement data gradually improve. Therefore, giving higher weights to sensors with longer usage time can improve the accuracy of corrosion data calculation. The specific weight calculation formula is as follows:

[0080] wi=α×ti+β;

[0081] Where wi is the weight of the i-th electrochemical sensor data, ti is the usage time of the sensor (unit: month), α and β are preset parameters that can be adjusted according to the sensor characteristics and actual application scenarios.

[0082] The calculation formula of corrosion data C is:

[0083] C=∑wi×di / ∑wi, ∑ is the summation function;

[0084] Among them, di is the data collected by the i-th electrochemical sensor.

[0085] If the first corrosion monitoring data is less than the reference corrosion value, the difference ΔC between the first corrosion monitoring data and the corresponding corrosion data is calculated; the weights and values ​​of the electrochemical data are adjusted according to the positive correlation of the difference; the larger the difference, the larger the weights and values ​​of the electrochemical data; the smaller the difference, the smaller the weights and values ​​of the electrochemical data. The initial weights and values ​​default to 1, or a fixed value. There are two specific methods for adjusting the weights and values: one is to amplify all weights in equal proportion, that is, the sum of the weights becomes larger. The weights of all sensors are amplified in the same proportion to increase the sum of the weights. The new weight calculation formula is:

[0086] wi′=wi×(1+k×ΔC); where k is the adjustment coefficient and can be set according to actual conditions.

[0087] For example, when ΔC = 0.2 and k = 0.5, the new weight of the electrochemical sensor with the original weight of 0.8 is:

[0088] w′=0.8×(1+0.5×0.2)=0.88.

[0089] The other method is to amplify the weights non-proportionally, in which the weights are amplified non-proportionally according to the importance or stability of the electrochemical sensor, but the sum of the weights increases.

[0090] For example, a larger amplification ratio is given to electrochemical sensors with longer service life and higher stability:

[0091] wi′=wi×(1+k1×ΔC×ti / t);

[0092] Wherein, k1 is the adjustment coefficient, and t is the average usage time of all electrochemical sensors.

[0093] The initial weight and value default to 1. Through the above adjustment method, the weight can be dynamically adjusted according to the actual monitoring situation, the final value of the first corrosion monitoring data can be amplified, and the sensitivity of the system to corrosion changes can be improved.

[0094] In actual application at a certain offshore wind farm, when the first corrosion monitoring data was found to be less than the reference corrosion value, the difference ΔC was calculated to be 0.3. After adjusting the weight using the proportional amplification method (k=0.4), the recalculated corrosion data increased from the original 2.5 to 2.8, which was closer to the actual corrosion situation and effectively avoided the risk of missed detection.

[0095] Reference Figure 3 The step of calling the calibrated first unmanned device further includes the following sub-steps:

[0096] The first unmanned device is an unmanned boat device equipped with an electrochemical sensor.

[0097] First corrosion monitoring data is calculated using a first algorithm based on the first monitoring data. If the first corrosion monitoring data is less than the corrosion data, the gain corresponding to the corrosion data in the first algorithm is adjusted in a positive correlation, or in an anti-correlation, based on the difference ΔC between the first corrosion monitoring data and the corresponding corrosion data. The greater the difference between the first corrosion monitoring data and the corresponding corrosion data, the greater the gain corresponding to the corrosion data in the first algorithm and the smaller the reference corrosion value; the smaller the difference between the first corrosion monitoring data and the corresponding corrosion data, the smaller the gain corresponding to the corrosion data in the first algorithm and the larger the reference corrosion value.

[0098] Specifically, the adjusted gain = the initial gain + k2×ΔC, where k2 is the adjustment coefficient; the adjusted reference corrosion value = the initial reference corrosion value - k3×ΔC, where k3 is the adjustment coefficient.

[0099] The first sorting strategy is the shortest path strategy after removing duplicate units: Unit address data is extracted from the first unmanned monitoring tag, and a coordinate set P = {p1, p2, …, pn} containing longitude and latitude information is constructed. By setting a spatial distance threshold (e.g., 10 meters), the DBSCAN density clustering algorithm is used to automatically identify and remove duplicate coordinate points, generating a unique unit set P′. Starting from the UAV's mooring point p0, the Euclidean distance between p0 and each unit in set P′ is calculated to construct an initial distance matrix. The 2-opt algorithm is then used to solve the traveling salesman problem: the path is initialized as a simple traversal sequence S = [p0, p1, …, pn, p0]. Edges are repeatedly swapped and path lengths are evaluated until no shorter path can be found. The optimal path output is the first shortest path.

[0100] An unmanned vessel equipped with an electrochemical sensor is used to monitor the flange connecting bolts at the root of the tower. When the first corrosion monitoring data is less than the original corrosion data, the system adjusts the algorithm gain through a positive correlation adjustment of the difference to improve sensitivity to weak corrosion signals, or dynamically lowers the reference corrosion value according to the size of the difference, thereby reducing false negative results caused by environmental fluctuations and improving alarm accuracy. At the same time, the first sorting strategy eliminates duplicate units based on unit address data, uses the traveling salesman problem algorithm to construct the shortest inspection path, and combines it with sea condition optimization to reduce redundant inspections and lower energy consumption, ultimately achieving improved corrosion monitoring accuracy, effectively extending the service life of bolts, and reducing high-risk manual operations.

[0101] The step of calling the calibrated first unmanned device further includes the following sub-steps:

[0102] During a preset time period, such as a monthly cycle, which can also be set to a quarterly or semi-annual cycle based on the actual needs of the wind farm, the number of times the first unmanned monitoring mark is modified to the second unmanned monitoring mark is continuously counted. The number of modifications intuitively reflects the frequency of the unit's corrosion status transforming from low risk to high risk.

[0103] A reference number is preset, which is determined based on multiple factors such as the historical corrosion data of the wind farm, the operating conditions of the unit and environmental conditions, and serves as a baseline for measuring whether the current corrosion risk changes are abnormal.

[0104] When the calculated actual number of modifications differs from a preset reference number, the call cycle of the first unmanned device, an unmanned vessel equipped with an electrochemical sensor, is inversely adjusted based on this difference. If the actual number of modifications is significantly greater than the preset reference number, it indicates that the corrosion risk of the on-site units is rapidly increasing, with the corrosion conditions of a large number of units exceeding the safety threshold determined in the initial assessment. In this case, the system automatically shortens the call cycle of the first unmanned device. For example, the inspection schedule could be adjusted from monthly to bi-monthly, ensuring that the unmanned vessel can monitor the units more frequently, promptly detecting rapidly developing corrosion risks and saving valuable time for subsequent maintenance measures. Conversely, if the difference between the actual number of modifications and the reference number is small, indicating that the corrosion status of the wind farm units is generally stable and there is no risk of widespread corrosion escalation, the system will accordingly extend the call cycle of the first unmanned device, for example, from monthly to bi-monthly. This ensures monitoring effectiveness while avoiding the waste of resources caused by redundant monitoring and reducing operation and maintenance costs.

[0105] A mechanism that adjusts the call cycle based on the number of modifications achieves deep adaptation of the first unmanned equipment monitoring strategy to the actual conditions of the on-site units. By sensing the dynamic changes in the unit's corrosion risk in real time and intelligently adjusting the unmanned vessel inspection frequency, it can respond promptly to increased corrosion risks and ensure the safe and stable operation of the wind turbine.

[0106] The step of calling the calibrated second unmanned device further includes the following sub-steps:

[0107] The second unmanned equipment is an unmanned aerial vehicle equipment equipped with an image sensor.

[0108] The second algorithm, an image recognition algorithm, compares the second monitoring data with a pre-set corrosion template library to generate a quantitative template matching value as the second corrosion monitoring data. The corrosion template library covers image samples of various levels, ranging from mild rust to severe corrosion. Each sample is verified by electrochemical testing to ensure that the template matching value accurately reflects the actual corrosion level.

[0109] The second sorting strategy calculates the shortest path after removing duplicate units: Based on the address data corresponding to the second unmanned monitoring marker, the system similarly constructs a coordinate set Q = {q1, q2, …, qm} and uses the same DBSCAN algorithm to remove duplicate units, generating a pure target set Q′. Starting from the UAV's takeoff point q0, the Euclidean distance between q0 and each unit in set Q′ is calculated to construct a distance matrix. An ant colony algorithm is used to find the optimal path: an ant population is initialized and randomly assigned to a path. Each ant selects the next node to visit based on pheromone concentration and a heuristic function, and updates its pheromone after completing the traversal. After multiple rounds of iteration, the pheromone concentration converges to the optimal path, which is the second shortest path.

[0110] Current wind direction data is acquired in real time. Using an anemometer from a wind farm weather station or the drone itself, the second shortest path is decomposed into an upwind and a headwind path. During inspections, the UAV prioritizes the upwind path. While this approach reduces flight speed due to headwinds, the image sensor's stability is enhanced, resulting in clearer monitoring images. After completing the upwind path inspection, the UAV returns along the headwind path, effectively conserving power by flying with the wind. This approach takes into account the drone's energy consumption characteristics, ensuring maximum coverage of inspected units in the event of a battery shortage in the latter half of the route, improving monitoring efficiency.

[0111] The step of calling the calibrated second unmanned device further includes the following sub-steps:

[0112] The system acquires meteorological data in real time, accurately determines the current wind direction, and combines it with the generated second shortest path to deconstruct the path in terms of wind direction using Geographic Information System (GIS) technology. Specifically, based on wind direction, the second shortest path is subdivided into an upwind flight segment and an upwind flight segment. A spatial analysis algorithm is used to calculate the actual distances of the two segments, and the distance ratio between the upwind and upwind paths is derived. This ratio intuitively reflects the energy consumption characteristics of the UAV performing inspection tasks under the current wind direction conditions. The larger the ratio, the higher the proportion of upwind flight segments, and the UAV needs to overcome greater wind resistance during the inspection process, which significantly increases flight energy consumption. Conversely, the smaller the ratio, the higher the proportion of upwind flight segments, and downwind flight can effectively reduce energy consumption.

[0113] Based on this distance ratio, an anti-correlation regulation logic is constructed to dynamically adjust the second UAV's call cycle. When the distance ratio between the upwind and downwind paths is large, it indicates that the UAV's energy consumption for performing inspections has increased significantly under the current wind conditions. With limited power, it may be difficult to complete all inspections or there is a high risk of returning to flight. In this case, the system automatically shortens the second UAV's call cycle. For example, the inspection frequency can be increased from once every two weeks to once a week. This increased frequency ensures that key components are monitored before wind conditions change, allowing potential corrosion hazards to be detected in a timely manner. Conversely, if the distance ratio is small, indicating that the current wind conditions are favorable for the UAV to perform inspections and flight energy consumption is low, the system will accordingly extend the second UAV's call cycle, for example, adjusting the inspection cycle from once a week to once every three weeks. While ensuring monitoring effectiveness, this reduces ineffective flight operations, equipment wear, and operational costs.

[0114] By sensing the impact of wind direction changes on the energy consumption of the inspection path in real time and intelligently adjusting the inspection frequency, it can not only ensure the timeliness and accuracy of monitoring under complex wind direction conditions, but also optimize resource allocation under favorable environments, achieving a balance between the efficiency and economy of offshore wind turbine corrosion monitoring.

[0115] The step of calling the calibrated second unmanned device further includes the following sub-steps:

[0116] According to the first sorting strategy and the second sorting strategy, a plurality of first shortest paths of the first unmanned device and a plurality of second shortest paths of the second unmanned device are generated respectively.

[0117] The path matching algorithm is initiated to systematically analyze the multiple first shortest paths and multiple second shortest paths obtained. Based on the Geographic Information System (GIS), this algorithm abstracts each path into a set of spatial line segments containing a starting point, a passing point, and an end point. The proximity between paths is quantitatively assessed by calculating multi-dimensional indicators such as the average Euclidean distance between nodes and the spatial distance between path centers of gravity. Specifically, for any set of first shortest paths and second shortest paths, the system first extracts the key coordinate nodes in the path, such as selecting a node every 50 meters, calculates the three-dimensional spatial distance between the corresponding nodes, and sums them. A correction coefficient for the path direction angle is also introduced. The smaller the angle, the closer the coefficient is to 1, and the larger the angle, the smaller the coefficient is. This results in a comprehensive path matching score. After traversing all path combinations, the first shortest path and the second shortest path with the highest score, that is, the closest distance, are selected.

[0118] The first shortest path and the second shortest path that are closest to each other are selected. In actual application scenarios, when the inspection paths of the two types of equipment are highly close, the control stations of the unmanned ship and the unmanned aerial vehicle can be set in a similar position or integrated into the same control terminal to reduce the cost of laying the control network and the complexity of operation and maintenance. At the same time, this strategy helps to optimize the communication links between devices, reduce signal transmission delays and interference, and realize real-time collaboration between unmanned ships and unmanned aerial vehicles in data collection, task scheduling, fault warning and other links. For example, when an unmanned ship discovers suspected severe corrosion in a unit during an inspection, it can immediately send a task instruction to the unmanned aerial vehicle on the adjacent path through the shared control station, and quickly call its high-resolution image sensor for accurate review, greatly improving the response speed and handling efficiency of abnormal situations, thereby building an efficient and intelligent collaborative monitoring system for offshore wind turbines.

[0119] In offshore wind turbine monitoring operations, to ensure the safe operation of the UAV under complex working conditions, the method further includes the following steps:

[0120] Through the high-precision sensors and navigation modules carried by the unmanned aerial vehicle, the current location, current energy consumption and remaining power of the device can be obtained in real time.

[0121] The remaining mileage is calculated based on the current energy consumption and the remaining power. The remaining mileage is calculated using a dynamic energy consumption prediction model, combined with the UAV's current flight attitude (pitch angle, roll angle, heading angle), environmental parameters (wind speed, wind direction, atmospheric density), and equipment operating mode (cruise mode, detection mode, hovering mode) to establish a nonlinear mapping relationship between energy consumption and flight distance. This technology is an existing technology and will not be described in detail.

[0122] Based on preset key locations, such as the center of the unmanned ship operation area and the coordinates of the operation and maintenance base, combined with the path planning algorithm of the geographic information system (GIS), the return mileage from the current position of the unmanned aerial vehicle to the key location is calculated.

[0123] If the remaining range is less than the return range, indicating that the UAV's battery power is insufficient to independently return to the target location, the system immediately triggers a collaborative operation request mechanism. This request includes the device's current status data, location, battery level, fault code, etc., and is broadcast to the control center and surrounding devices via 5G / satellite communication dual channels.

[0124] Based on the collaborative operation request, the control center initiates a scan for the first UAV. A search area is defined, centered on the UAV's current location and with the remaining range as its radius. The IoT positioning system monitors and screens the first UAV within the area in real time. The screening algorithm prioritizes the device's operating status (idle / busy), flight direction (whether it's approaching the UAV), and distance weighting (a comprehensive evaluation of Euclidean distance and flight time). Ultimately, the first UAV closest to the current location and eligible for collaboration is identified and its real-time coordinates (longitude, latitude, and heading angle) are encrypted and transmitted to the UAV.

[0125] Upon receiving the coordinate information, the UAV immediately switches to emergency navigation mode. This mode utilizes a reinforcement learning-based path optimization algorithm, combined with real-time meteorological data and information about surrounding obstacles, to plan an energy-efficient flight path. Upon approaching the unmanned vessel, the UAV uses its visual positioning system to locate a dedicated landing area on the vessel's deck and achieves a precise landing through a collaborative drone-vessel positioning protocol. During the landing process, the vessel automatically adjusts its navigation attitude to maintain a stable heading and a level surface with the deck, ensuring the UAV's safe docking. This process does not affect the vessel's normal inspection operations.

[0126] An embodiment of the present application further discloses an automatic adjustment system for an offshore wind turbine monitoring strategy, comprising a processor, wherein the processor executes the steps of the automatic adjustment method for an offshore wind turbine monitoring strategy as described in any one of the above.

[0127] An embodiment of the present application further discloses a storage medium, wherein a program is stored in the storage medium. When the program is executed by a processor, the steps of the automatic adjustment method of the offshore wind turbine monitoring strategy described in any one of the above are implemented.

[0128] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for automatically adjusting monitoring strategies for offshore wind turbines, characterized in that: The steps include: Acquiring multiple electrochemical data based on multiple electrochemical sensors disposed on the same unit, and calculating corrosion data using a first algorithm based on the multiple electrochemical data; If the corrosion data is less than a preset reference corrosion value, a first unmanned monitoring flag is issued; otherwise, a second unmanned monitoring flag is issued; both the first unmanned monitoring flag and the second unmanned monitoring flag include the address data of the unit; Obtaining the first unmanned monitoring mark and the corresponding address data, and constructing a first inspection path for the unit corresponding to the address data according to a preset first sorting strategy; Invoking the calibrated first unmanned equipment to perform inspection based on the first inspection path to obtain first monitoring data of the corresponding unit; Calculating first corrosion monitoring data using a first algorithm based on the first monitoring data; and if the first corrosion monitoring data is greater than the reference corrosion value, issuing an on-site electrochemical sensor warning prompt and changing the first unmanned monitoring flag of the unit to the second unmanned monitoring flag; Obtaining the second unmanned monitoring mark and the corresponding address data, and constructing a second inspection path for the unit corresponding to the address data according to a preset second sorting strategy; Invoking the calibrated second unmanned equipment to perform inspection based on the second inspection path to obtain second monitoring data of the corresponding unit; Calculating second corrosion monitoring data using a second algorithm based on the second monitoring data, and triggering an on-site electrochemical sensor alarm if the second corrosion monitoring data is greater than a preset standard corrosion value; The first algorithm is a weighted average algorithm, wherein the weight of the electrochemical data is positively correlated with the usage time of the electrochemical sensor; the longer the usage time of the electrochemical sensor, the greater the weight of the electrochemical data; the shorter the usage time of the electrochemical sensor, the smaller the weight of the electrochemical data; If the first corrosion monitoring data is less than the reference corrosion value, the difference between the first corrosion monitoring data and the corresponding corrosion data is calculated; the weight and value of the electrochemical data are adjusted according to the positive correlation of the difference; the larger the difference, the larger the weight and value of the electrochemical data; the smaller the difference, the smaller the weight and value of the electrochemical data.

2. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 1, characterized in that: It also includes the following sub-steps: The electrochemical sensor sends the collected electrochemical data to the transfer station through a first communication protocol, and the transfer station sends the collected or received electrochemical data to a remote background through a second communication protocol.

3. The automatic adjustment method of the monitoring strategy of an offshore wind turbine according to claim 1, characterized in that: The step of calling the calibrated first unmanned device further includes the following sub-steps: The first unmanned equipment is an unmanned boat equipment equipped with the electrochemical sensor; The first sorting strategy is the shortest path strategy after removing duplicate units: based on the address data corresponding to the first unmanned monitoring mark, duplicate units are eliminated; starting from the unmanned boat equipment, all units are traversed and the driving path length is calculated; the shortest driving path length is selected as the first shortest path.

4. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 3, characterized in that: The step of calling the calibrated first unmanned device further includes the following sub-steps: Within a preset time period, calculating the number of times the first unmanned monitoring mark is modified to the second unmanned monitoring mark; According to the difference between the number of modifications and the preset reference number, the calling cycle of the first unmanned device is adjusted in an anti-correlation manner; the greater the difference between the number of modifications and the preset reference number, the shorter the calling cycle of the first unmanned device; the smaller the difference between the number of modifications and the preset reference number, the longer the calling cycle of the first unmanned device.

5. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 3, characterized in that: The step of calling the calibrated second unmanned device further includes the following sub-steps: The second unmanned equipment is an unmanned aerial vehicle equipped with an image sensor; The second algorithm is an image recognition algorithm, and a template matching value between the second monitoring data and a preset corrosion template is used as the second corrosion monitoring data; The second sorting strategy is a strategy for calculating the shortest path after removing duplicate units: based on the address data corresponding to the second unmanned monitoring mark, after removing duplicate units; starting from the unmanned aerial vehicle device, traversing all units and calculating the travel path length; selecting the shortest travel path length as the second shortest path; The current wind direction is obtained, the upwind path and the upwind path in the shortest path strategy are calculated, and the second unmanned device is controlled to first perform detection according to the upwind path and then perform detection according to the upwind path.

6. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 5, characterized in that: The step of calling the calibrated second unmanned device further includes the following sub-steps: Calculating a distance ratio between the upwind path and the upwind path; The calling period of the second unmanned device is adjusted inversely according to the distance ratio; the larger the distance ratio, the shorter the calling period of the second unmanned device; the smaller the distance ratio, the longer the calling period of the second unmanned device.

7. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 5, characterized in that: The step of calling the calibrated second unmanned device further includes the following sub-steps: If the first unmanned device and the second unmanned device are working at the same time; Then, a plurality of first shortest paths of the first unmanned equipment and a plurality of second shortest paths of the second unmanned equipment are obtained; Among the plurality of first shortest paths and the plurality of second shortest paths, the first shortest path and the second shortest path that are closest to each other are selected.

8. The automatic adjustment method for monitoring strategy of offshore wind turbines according to claim 5, characterized in that: The method further comprises the steps of: Obtaining the current location, current energy consumption, and remaining power of the second unmanned device; Calculating the remaining mileage based on the current energy consumption and the remaining power; Calculating return mileage based on the current location and a preset key location; If the remaining mileage is less than the return mileage, a collaborative operation request is issued; Based on the collaborative operation request, scanning the first unmanned equipment within an area with the current location as the center and a radius of the remaining mileage; Sending the real-time coordinates of the first unmanned device closest to the current position to the second unmanned device; The second unmanned device moves to the first unmanned device according to the real-time coordinates.

9. An automatic adjustment system for monitoring strategy of offshore wind turbines, characterized in that: The method comprises a processor, wherein the processor executes the steps of the automatic adjustment method of the offshore wind turbine monitoring strategy according to any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium stores a program, and when the program is executed by the processor, the steps of the automatic adjustment method of the offshore wind turbine monitoring strategy according to any one of claims 1 to 8 are implemented.

Images