An unmanned aerial vehicle inspection system and method for an offshore wind turbine

Abstract

The present invention discloses a drone inspection system and method applied to an offshore wind turbine, which relates to the technical field of drone inspection. The inspection method includes the following steps: constructing an inspection data management system to store the monitoring data and photos collected during each inspection of the drone, and generating a corresponding inspection record; based on the monitoring data stored in any inspection record, performing a risk assessment and wear assessment on the operating state of the wind turbine; arbitrarily selecting an inspection record, analyzing the monitoring data stored in the inspection record regarding the surrounding environment, extracting characteristic factors, and obtaining the proportion of the influence degree of each characteristic factor; setting a performance influence ratio for the two evaluation results of any inspection record to obtain the comprehensive risk value of any inspection record; obtaining the meteorological information of the location where the wind turbine is located in real time, predicting the change in the operating state of the wind turbine, and determining whether to generate a temporary inspection alarm.

Description

Technical Field

[0001] The present invention relates to the technical field of drone inspection, and specifically to a drone inspection system and method applied to offshore wind turbines. Background Art

[0002] Offshore wind turbines are wind power generation devices specifically designed for the marine environment. Compared with onshore wind turbines, offshore wind turbines have higher power generation efficiency and larger power generation capacity; the operation and inspection requirements of offshore wind power are becoming increasingly strong. When maintaining wind turbines in the marine environment using traditional manual inspection methods, the downtime of wind turbines will be long and the power generation loss will be serious; moreover, the usage fees and labor costs for each voyage of maintenance vessels will also account for a large part of the operation and maintenance costs of offshore wind farms.

[0003] Nowadays, unmanned systems represented by drones have been increasingly used in aspects such as the inspection of offshore wind turbine blades and underwater wind power pile foundations year by year, which can help save a large amount of manpower and material resources and can also improve the inspection efficiency to a certain extent; however, using drones for inspection will be affected by environmental factors and the drone equipment itself. Since drones cannot conduct long-term inspections, normally an inspection cycle will be formulated. However, the environment on the sea surface is relatively complex, and it is very easy for the wind turbine to malfunction outside the inspection cycle, thus affecting normal power generation operation. Summary of the Invention

[0004] The purpose of the present invention is to provide a drone inspection system and method applied to offshore wind turbines to solve the problems raised in the prior art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A drone inspection method applied to offshore wind turbines, the inspection method includes the following steps:

[0006] Step S100: Monitor the operating performance and the environment of the wind turbine by installing monitoring equipment on the wind turbine. During each inspection process, take pictures of the wind turbine by the drone and receive the monitoring data monitored by the monitoring equipment; construct an inspection data management system to store the monitoring data and photos collected during each inspection process of the drone, and generate a corresponding inspection record;

[0007] Step S200: Based on the monitoring data stored in any inspection record, conduct a risk assessment on the operating status of the wind turbine; obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment on the wind turbine based on the difference from the adjacent inspection records.

[0008] Step S300: Arbitrarily select an inspection record, analyze the monitoring data stored in the inspection record regarding the surrounding environment, and extract the characteristic factors that affect the operating status and wear condition of the fan; compare the difference in the evaluation results between the inspection record and the adjacent inspection record to obtain the proportion of the influence of each characteristic factor on the two evaluation results.

[0009] Step S400: Based on the proportion of the influence of each characteristic factor, set the performance influence ratio for the two evaluation results of any inspection record to obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; obtain the meteorological information at the location of the fan in real time, predict the change in the operating status of the fan, and determine whether to generate a temporary inspection alarm to send a reminder to the staff.

[0010] Further, step S100 includes the following steps:

[0011] Step S101: When the drone inspects the fan, receive the monitoring data stored in all monitoring devices. The monitoring data includes environmental data and fan performance data. After the drone finishes receiving, each monitoring device resumes monitoring and overwrites the historical monitoring data; each inspection only obtains the monitoring data that has not been collected within a certain period of time because the previous data has lost its validity, and overwriting the invalid data can also reduce the data storage burden of the monitoring devices.

[0012] Step S102: Preset the inspection route and several shooting points of the drone in advance. When the drone flies to any shooting point, take several photos of each part of the fan; when the drone returns, store the received monitoring data and several photos in the inspection data management system and generate an inspection record.

[0013] Further, step S200 includes the following steps:

[0014] Step S201: Obtain the fan performance data stored in any inspection record, and divide the fan performance data into different types according to the data type; preset corresponding evaluation rules for each type of performance data to obtain the risk value presented by the fan in any type of performance data; accumulate the risk values presented by each type of performance data to obtain the first characteristic risk value F1 of the fan; the first characteristic risk value is used to evaluate the risk of the internal operating performance of the fan and is one of the important indicators for judging whether the fan is abnormal, which is beneficial to making the calculation of the subsequent comprehensive risk value more reasonable.

[0015] Step S202: Extract several photos stored in the inspection record, and determine the shooting positions of each photo according to the preset shooting positions; sort all inspection records in the order of record generation, obtain several photos of the previous inspection record of the inspection record, and similarly determine the shooting positions of each photo; respectively identify the photos obtained at the same shooting position in the two inspection records, extract the structural features presented in the photos, compare the structural features of the two photos at the same shooting position, respectively identify several worn areas in the two photos, and respectively obtain the total area of the worn areas presented in the two photos; because environmental factors will have a certain impact on the fan, in addition to the internal performance, it will also cause losses to the external structure, and the losses of the external structure will also affect the performance of the fan;

[0016] Step S203: Respectively obtain the record generation times of the two inspection records to obtain the time interval Δt between the two inspection records; According to the formula:

[0017]

[0018] where A1 is the total area of the worn areas presented in the photos of the previous inspection record, A2 is the total area of the worn areas presented in the photos of the inspection record, k is the area conversion ratio of the photos taken by the drone, and the area conversion ratio is the ratio of the actual area to the area presented in the picture is k; calculate the wear coefficient W presented by the fan in the inspection record; The simple subtraction of the worn areas at different time points is because the worn areas will only go from none to some, from small to large, so simple subtraction is sufficient. Through the worn areas, the wear condition of the fan can be effectively and directly analyzed, and the wear coefficient reflects the degree of wear on the external structure during this period, providing scientific reference data for subsequent wear risk assessment;

[0019] Step S204: Identify the worn areas of each photo stored in the inspection record to obtain the wear area of the fan and several wear coefficients, and calculate the average value of the several wear coefficients to obtain the comprehensive wear coefficient Wav of the inspection record e ; According to the formula:

[0020] F2 = A × k × Waave

[0021] where A is the wear area of the fan, which is obtained by accumulating the total areas of the worn areas presented in each photo; calculate the second characteristic risk value F2 of the fan; The second characteristic risk value represents the wear condition of the external structure of the fan. The larger the worn area or the higher the wear degree within a certain period of time, the more effectively it can reflect the abnormal degree of the fan.

[0022] Further, step S300 includes the following steps:

[0023] Step S301: Set the monitoring data on the environment in the patrol inspection records as environmental data, sort all the patrol inspection records in the order of record generation, and respectively obtain the environmental data in any two adjacent patrol inspection records; divide the environmental data into several categories according to the environmental data type, compare the same-category environmental data in the two patrol inspection records, preset corresponding comparison rules for each category of environmental data, and obtain the difference range of the two patrol inspection records in any category of environmental data.

[0024] Step S302: Preset different safety range limits for each category of environmental data. If the difference range of a certain category of environmental data exceeds the safety range limit, set the said category of environmental data as a characteristic factor.

[0025] Step S303: Set the first characteristic risk ratio of two adjacent patrol inspection records as ΔF1 and the second characteristic risk ratio as ΔF2; obtain the number of characteristic factors between two adjacent patrol inspection records as m, according to the formula:

[0026]

[0027] where C i is the difference range of the i-th characteristic factor between two adjacent patrol inspection records; calculate the influence degree α i on the operating state of the fan and the influence degree β i on the wear condition of the fan, and obtain the proportion of the influence degree of the said i-th characteristic factor on the two evaluation results as P i =|α i | / (|α i |+|β i |) and Q i =|β i | / (|α i |+|β i |); each environmental factor will have a certain impact on the internal operating performance and external structure of the fan. For example, the wind speed will affect the power generation efficiency of the fan and also cause wear on the external structure; however, the degrees of impact on the two are different. Therefore, separate calculations are required. Due to the different degrees of impact, the influence proportions of the two evaluation results on the comprehensive risk value are also different in the end. Therefore, it is necessary to determine the proportion of their influence degrees in order to obtain a more accurate value in the subsequent calculation of the comprehensive risk value.

[0028] Further, step S400 includes the following steps:

[0029] Step S401: Arbitrarily select an inspection record, and obtain the first characteristic risk value F1 and the second characteristic risk value F2 of the inspection record; obtain the proportion of the influence degree of each characteristic factor in the inspection record on the two evaluation results. According to the formula:

[0030]

[0031] where both z1 and z2 are positive integers, and z1 ∈ (1, n), z2 ∈ (1, n), P z1 is the proportion of the influence degree of the z1-th characteristic factor on the evaluation result of the fan operation state, P z2 is the proportion of the influence degree of the z2-th characteristic factor on the evaluation result of the fan wear condition, and n is the number of characteristic factors included in the inspection record; calculate the comprehensive risk value F of the fan presented by the inspection record;

[0032] Step S402: Whenever a maintenance behavior occurs for the fan, an abnormal mark is made on the latest generated inspection record before the maintenance; obtain the comprehensive risk values of all inspection records containing abnormal marks, and select the smallest comprehensive risk value as the risk threshold F for judging whether the fan is abnormal min ;

[0033] Step S403: Divide the meteorological information at the location of the fan according to the environmental data type, obtain the numerical changes of each type of environmental data, extract the environmental data whose numerical change range exceeds the set safe range, and generate several real-time characteristic factors;

[0034] Step S404: Compare the several real-time characteristic factors with the characteristic factors in each inspection record to determine the corresponding characteristic factors of each real-time characteristic factor in each inspection record; obtain the proportion of the influence degree P and Q of any real-time characteristic factor on the two evaluation results in each inspection record respectively, and calculate the average influence degree proportion P ave and Q ave ; obtain the first characteristic risk value (F1) of the latest generated inspection record currently ’ and the second characteristic risk value (F2) ’ of the current fan, and calculate the comprehensive risk value F ’ of the current fan; if F ’ > F min , then generate a temporary inspection alarm to send a reminder to the staff, reminding that it is necessary to dispatch a drone to inspect the fan currently.

[0035] To better implement the above method, a drone inspection system for offshore wind turbines is also proposed. The inspection system includes a fan data acquisition module, a monitoring data analysis module, an influencing factor analysis module, and a real-time inspection judgment module;

[0036] The fan data acquisition module is used to monitor the operating performance and the surrounding environment of the fan by installing monitoring devices on the fan, take pictures of the fan by means of an unmanned aerial vehicle (UAV) during each inspection, and receive the monitoring data detected by the monitoring devices; construct an inspection data management system to store the monitoring data and photos collected during each inspection by the UAV, and generate a corresponding inspection record.

[0037] The monitoring data analysis module is used to conduct a risk assessment on the operating status of the fan based on the monitoring data stored in any inspection record; obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment on the fan based on the differences from adjacent inspection records.

[0038] The influencing factor analysis module is used to arbitrarily select an inspection record, analyze the monitoring data on the surrounding environment stored in the inspection record, and extract the characteristic factors that affect the operating status and wear condition of the fan; compare the evaluation result differences between the inspection record and adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results.

[0039] The real-time inspection judgment module is used to set a performance influence ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; obtain the meteorological information at the location of the fan in real time, predict the change in the operating status of the fan, and judge whether to generate a temporary inspection alarm to send a reminder to the staff.

[0040] Furthermore, the fan data acquisition module includes a monitoring data acquisition unit and an inspection record generation unit;

[0041] The monitoring data acquisition unit is used to monitor the operating performance and the surrounding environment of the fan by installing monitoring devices on the fan, take pictures of the fan by means of an unmanned aerial vehicle (UAV) during each inspection, and receive the monitoring data detected by the monitoring devices; the inspection record generation unit is used to construct an inspection data management system to store the monitoring data and photos collected during each inspection by the UAV, and generate a corresponding inspection record.

[0042] Furthermore, the monitoring data analysis module includes a risk assessment unit and a wear assessment unit;

[0043] The risk assessment unit is used to conduct a risk assessment on the operating status of the fan based on the monitoring data stored in any inspection record; the wear assessment unit is used to obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment on the fan based on the differences from adjacent inspection records.

[0044] Furthermore, the influencing factor analysis module includes a characteristic factor extraction unit and an influence degree calculation unit;

[0045] The characteristic factor extraction unit is used to arbitrarily select an inspection record, analyze the monitoring data on the surrounding environment stored in the inspection record, and extract the characteristic factors that affect the operating state and wear condition of the fan; the influence degree calculation unit is used to compare the evaluation result differences between the inspection record and the adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results.

[0046] Furthermore, the real-time inspection judgment module includes a risk threshold confirmation unit and a temporary inspection generation unit;

[0047] The risk threshold confirmation unit is used to set the performance influence ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; the temporary inspection generation unit is used to obtain the meteorological information at the location of the fan in real time, predict the change in the operating state of the fan, and judge whether to generate a temporary inspection alarm to send a reminder to the staff.

[0048] Compared with the prior art, the beneficial effects of the present invention are:

[0049] 1. By analyzing the complex conditions of the marine environment, the present invention helps the staff to timely judge the abnormal conditions of the fan during the change of the environment where the fan is located, and timely investigate the abnormal conditions to avoid affecting the normal operation of the fan;

[0050] 2. The present invention double-evaluates the operating performance and appearance structure of the fan itself, carefully analyzes the possible abnormal conditions of the fan, and combines the influence of environmental factors on the operating performance and structure to accurately evaluate the actual risk value of the fan, helping the staff to timely investigate the abnormalities;

[0051] 3. Based on the meteorological information, the present invention predicts the real-time situation of the fan, conducts a temporary inspection on the situation with risks in the prediction result, can effectively improve the inspection efficiency, and can also help to timely discover abnormal conditions; avoid the abnormal situation of the fan caused by special circumstances but cannot be repaired in time, thus affecting the overall operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a step schematic diagram of an unmanned aerial vehicle inspection system applied to an offshore wind turbine;

[0053] Figure 2 It is a structural schematic diagram of an unmanned aerial vehicle inspection system applied to an offshore wind turbine. DETAILED DESCRIPTION OF THE INVENTION

[0054] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0055] Embodiment: As Figures 1 to 2 shown, the present invention provides a method for inspecting an unmanned aerial vehicle (UAV) applied to an offshore wind turbine. The inspection method includes the following steps:

[0056] Step S100: Monitor the operating performance and the environment of the wind turbine by installing monitoring devices on the wind turbine. During each inspection process, take pictures of the wind turbine by the UAV and receive the monitoring data monitored by the monitoring devices; construct an inspection data management system to store the monitoring data and photos collected during each inspection process of the UAV, and generate a corresponding inspection record.

[0057] Among them, step S100 includes the following steps:

[0058] Step S101: When the UAV inspects the wind turbine, receive the monitoring data stored in all the monitoring devices. The monitoring data includes environmental data and wind turbine performance data. After the UAV finishes receiving, each monitoring device resumes monitoring and overwrites the historical monitoring data.

[0059] Step S102: Preset the inspection route and several shooting points of the UAV in advance. When the UAV flies to any shooting point, take several pictures of each part of the wind turbine; when the UAV returns, store the received monitoring data and several pictures in the inspection data management system and generate an inspection record.

[0060] Step S200: Based on the monitoring data stored in any inspection record, conduct a risk assessment on the operating state of the wind turbine; obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment on the wind turbine based on the difference from the adjacent inspection records.

[0061] Among them, step S200 includes the following steps:

[0062] Step S201: Obtain the wind turbine performance data stored in any inspection record, and divide the wind turbine performance data into different types according to the data type; preset corresponding evaluation rules for each type of performance data to obtain the risk value presented by the wind turbine in any type of performance data; accumulate the risk values presented by various types of performance data to obtain the first characteristic risk value F1 of the wind turbine.

[0063] Step S202: Extract several photos stored in the inspection record, determine the shooting points of each photo according to the preset shooting points; sort all inspection records according to the record generation order, obtain several photos of the previous inspection record of the inspection record, and also determine the shooting points of each photo; respectively identify the photos obtained at the same shooting point in the two inspection records, extract the structural features presented in the photos, compare the structural features of the two photos at the same shooting point, respectively identify several worn areas in the two photos, and respectively obtain the total area of the worn areas presented in the two photos;

[0064] Step S203: Respectively obtain the record generation times of the two inspection records to obtain the time interval Δt between the two inspection records; according to the formula:

[0065]

[0066] where A1 is the total area of the worn area presented in the photo of the previous inspection record, A2 is the total area of the worn area presented in the photo of the inspection record, k is the area conversion ratio of the photos taken by the drone, and the area conversion ratio is the ratio of the actual area to the area presented in the picture is k; calculate the wear coefficient W presented by the fan in the inspection record;

[0067] Example 1: It is set that three positions are photographed in the two inspection records, and the differences in the worn area areas presented in the photos are 5 cm 2 , 6 cm 2 , 4 cm 2 ; and the worn area areas presented in each photo of the previous inspection record are 10 cm 2 , 10 cm 2 , 10 cm 2 , it is set that the area conversion ratio of the drone is 2 and the time interval between the two inspection records is 1 month. Therefore, the calculated wear coefficients are 5 / (10×1)×2 = 1 cm 2 / month, 6 / (10×1)×2 = 1.2 cm 2 / month, 4 / (10×1)×2 = 0.8 cm 2 / month; the average wear coefficient is obtained as 1 cm 2 / month;

[0068] Step S204: Identify the worn areas of each photo stored in the inspection record to obtain the wear area of the fan and several wear coefficients, calculate the average value of the several wear coefficients to obtain the comprehensive wear coefficient Wav of the inspection record e ; according to the formula:

[0069] F2 = A × k × Waa ve

[0070] Among them, A is the worn area of the fan, which is obtained by accumulating the total area of the worn areas presented in each photo; the second characteristic risk value F2 of the fan is calculated.

[0071] Step S300: Arbitrarily select an inspection record, analyze the monitoring data on the surrounding environment stored in the inspection record, and extract the characteristic factors affecting the operating state and wear condition of the fan; compare the evaluation result differences between the inspection record and the adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results;

[0072] Among them, step S300 includes the following steps:

[0073] Step S301: Set the monitoring data on the surrounding environment in the inspection record as environmental data, sort all inspection records in the order of record generation, and respectively obtain the environmental data in any two adjacent inspection records; divide the environmental data into several categories according to the environmental data type, compare the same-type environmental data in the two inspection records, preset corresponding comparison rules for each type of environmental data, and obtain the difference amplitude of the two inspection records in any type of environmental data;

[0074] Step S302: Preset different safety amplitude ranges for each type of environmental data. If the difference amplitude of a certain type of environmental data exceeds the safety amplitude range, set the certain type of environmental data as a characteristic factor;

[0075] Step S303: Set the first characteristic risk ratio of two adjacent inspection records as ΔF1 and the second characteristic risk ratio as ΔF2; obtain the number of characteristic factors between two adjacent inspection records as m, according to the formula:

[0076]

[0077] Among them, C i is the difference amplitude of the i-th characteristic factor between two adjacent inspection records; calculate the influence degree α i of the i-th characteristic factor on the operating state of the fan and the influence degree β i , and obtain the proportion of the influence degree of the i-th characteristic factor on the two evaluation results as P i = |α i | / (|α i | + |β i |) and Q i = |β i | / (|α i | + |β i |);

[0078] Example 2: The first characteristic risk ratio of two adjacent inspection records is 3.6, the second characteristic risk difference is 2.4, and there are 3 characteristic factors with difference amplitudes of 50%, 40%, and 60% respectively. Taking the first characteristic factor as an example, we get 3.6 / 3 = 1 + 0.5×α i , α i = 0.4; 2.4 / 3 = 1 + 0.5×β i , β i = -0.4; then the influence degree ratios are 50% and 50% respectively.

[0079] Step S400: Based on the influence degree ratios of each characteristic factor, set the performance influence ratio for the two evaluation results of any inspection record, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; obtain the meteorological information at the location of the fan in real time, predict the change in the operating state of the fan, and judge whether to generate a temporary inspection alarm to send a reminder to the staff;

[0080] Among them, step S400 includes the following steps:

[0081] Step S401: Arbitrarily select an inspection record to obtain the first characteristic risk value F1 and the second characteristic risk value F2 of the inspection record; obtain the influence degree ratios of each characteristic factor in the inspection record on the two evaluation results. According to the formula:

[0082]

[0083] where z1 and z2 are both positive integers, and z1 ∈ (1, n), z2 ∈ (1, n), P z1 is the influence degree ratio of the z1-th characteristic factor on the evaluation result of the fan operating state, P z2 is the influence degree ratio of the z2-th characteristic factor on the evaluation result of the fan wear condition, and n is the number of characteristic factors included in the inspection record; calculate the comprehensive risk value F of the fan presented by the inspection record;

[0084] Step S402: Whenever a maintenance behavior occurs for the fan, mark the latest generated inspection record before the maintenance as abnormal; obtain the comprehensive risk values of all inspection records with abnormal marks, and select the smallest comprehensive risk value as the risk threshold F for judging whether the fan is abnormal min ;

[0085] Step S403: Divide the meteorological information at the location of the fan according to the environmental data type, obtain the numerical changes of each type of environmental data, extract the environmental data whose numerical change amplitude exceeds the set safe amplitude range, and generate several real-time characteristic factors;

[0086] Step S404: Compare the several real-time characteristic factors with the characteristic factors in each inspection record to determine the corresponding characteristic factors of each real-time characteristic factor in each inspection record; Obtain the proportion of the influence degree of any real-time characteristic factor on the two evaluation results in each inspection record, namely P and Q, and calculate the average value respectively to obtain the average influence degree proportion P ave and Q ave ; Obtain the first characteristic risk value (F1) ’ and the second characteristic risk value (F2) ’ of the currently newly generated inspection record, and calculate the comprehensive risk value F ’ of the current fan; If F ’ > F min , then generate a temporary inspection alarm to send a reminder to the staff, reminding that it is necessary to dispatch a drone to repair the fan currently.

[0087] An unmanned aerial vehicle (UAV) inspection system applied to an offshore wind turbine, the inspection system includes a fan data acquisition module, a monitoring data analysis module, an influencing factor analysis module, and a real-time inspection judgment module;

[0088] The fan data acquisition module is used to monitor the operating performance and the surrounding environment of the fan by installing monitoring devices on the fan, take pictures of the fan by the UAV during each inspection process, and receive the monitoring data monitored by the monitoring devices; Construct an inspection data management system to store the monitoring data and photos collected by the UAV during each inspection process, and generate a corresponding inspection record;

[0089] The monitoring data analysis module is used to perform a risk assessment on the operating state of the fan based on the monitoring data stored in any inspection record; Obtain the operating conditions presented in the photos of the inspection record, and perform a wear assessment on the fan based on the difference from the adjacent inspection record;

[0090] The influencing factor analysis module is used to arbitrarily select an inspection record, analyze the monitoring data stored in the inspection record regarding the surrounding environment, and extract the characteristic factors that affect the operating state and wear condition of the fan; Compare the evaluation result differences between the inspection record and the adjacent inspection record to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results;

[0091] The real-time inspection and judgment module is used to set the performance impact ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether there is an abnormality in the fan; it also obtains the meteorological information of the fan's location in real time, predicts the change of the fan's operating state, and judges whether to generate a temporary inspection alarm to send a reminder to the staff.

[0092] Among them, the fan data acquisition module includes a monitoring data acquisition unit and an inspection record generation unit;

[0093] The monitoring data acquisition unit is used to monitor the operating performance and the environment of the fan by installing monitoring devices on the fan, take pictures of the fan by an unmanned aerial vehicle during each inspection process, and receive the monitoring data monitored by the monitoring devices; the inspection record generation unit is used to construct an inspection data management system to store the monitoring data and photos collected during each inspection process by the unmanned aerial vehicle, and generate a corresponding inspection record.

[0094] Among them, the monitoring data analysis module includes a risk assessment unit and a wear assessment unit;

[0095] The risk assessment unit is used to conduct a risk assessment of the operating state of the fan based on the monitoring data stored in any inspection record; the wear assessment unit is used to obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment of the fan based on the difference from the adjacent inspection records.

[0096] Among them, the influence factor analysis module includes a characteristic factor extraction unit and an influence degree calculation unit;

[0097] The characteristic factor extraction unit is used to arbitrarily select an inspection record, analyze the monitoring data on the environment stored in the inspection record, and extract the characteristic factors that affect the operating state and wear condition of the fan; the influence degree calculation unit is used to compare the difference in evaluation results between the inspection record and the adjacent inspection records, and obtain the proportion of the influence degree of each characteristic factor on the two evaluation results.

[0098] Among them, the real-time inspection and judgment module includes a risk threshold confirmation unit and a temporary inspection generation unit;

[0099] The risk threshold confirmation unit is used to set the performance impact ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether there is an abnormality in the fan; the temporary inspection generation unit is used to obtain the meteorological information of the fan's location in real time, predict the change of the fan's operating state, and judge whether to generate a temporary inspection alarm to send a reminder to the staff.

[0100] It is obvious to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. An unmanned aerial vehicle inspection method applied to an offshore wind turbine, characterized in that: The inspection method includes the following steps: Step S100: Monitor the operating performance and the environment of the wind turbine by installing monitoring devices on the wind turbine. During each inspection process, take pictures of the wind turbine by using a drone, and receive the monitoring data monitored by the monitoring devices; construct an inspection data management system to store the monitoring data and photos collected during each inspection process by the drone, and generate a corresponding inspection record. Step S200: Based on the monitoring data stored in any inspection record, conduct a risk assessment on the operating status of the wind turbine; obtain the operating conditions presented in the photos of the inspection record, and conduct a wear assessment on the wind turbine based on the difference from the adjacent inspection records. Step S300: Arbitrarily select an inspection record, analyze the monitoring data on the environment stored in the inspection record, and extract the characteristic factors affecting the operating status and wear condition of the wind turbine; compare the evaluation result differences between the inspection record and the adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results. Step S400: Based on the proportion of the influence degree of each characteristic factor, set a performance influence ratio for the two evaluation results of any inspection record to obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the wind turbine is abnormal; obtain the meteorological information at the location of the wind turbine in real time, predict the change of the operating status of the wind turbine, and judge whether to generate a temporary inspection alarm to send a reminder to the staff. The step S200 includes the following steps: Step S201: Obtain the wind turbine performance data stored in any inspection record, and divide the wind turbine performance data into different types according to the data type; preset corresponding evaluation rules for each type of performance data to obtain the risk value presented by the wind turbine in any type of performance data; accumulate the risk values presented by various types of performance data to obtain the first characteristic risk value F1 of the wind turbine. Step S202: Extract several photos stored in the inspection record, and determine the shooting points of each photo according to the preset shooting points; sort all inspection records according to the record generation order, obtain several photos of the previous inspection record of the inspection record, and also determine the shooting points of each photo; respectively identify the photos obtained at the same shooting point in the two inspection records, extract the structural features presented in the photos, compare the structural features of the two photos at the same shooting point, respectively identify several wear areas in the two photos, and respectively obtain the total area of the wear areas presented in the two photos. Step S203: Respectively obtain the record generation times of the two inspection records to obtain the time interval Δt between the two inspection records; according to the formula: where A1 is the total area of the wear areas presented in the photos of the previous inspection record, A2 is the total area of the wear areas presented in the photos of the inspection record, k is the area conversion ratio of the photos taken by the drone, and the area conversion ratio is the ratio of the actual area to the area presented in the picture as k; calculate the wear coefficient W presented by the wind turbine in the inspection record. Step S204: Identify the worn areas in each photo stored in the inspection record to obtain the worn area of the fan and several wear coefficients, and calculate the average value of the several wear coefficients to obtain the comprehensive wear coefficient W of the inspection record ave ; According to the formula: F2 = A × k × W ave ; Among them, A is the worn area of the fan, which is obtained by accumulating the total area of the worn areas presented in each photo; the second characteristic risk value F2 of the fan is calculated; The step S400 includes the following steps: Step S401: Arbitrarily select an inspection record to obtain the first characteristic risk value F1 and the second characteristic risk value F2 of the inspection record; obtain the proportion of the influence degree of each characteristic factor in the inspection record on the two evaluation results. According to the formula: wherein, both z1 and z2 are positive integers, and z1 ∈ (1, n), z2 ∈ (1, n), P z1 is the proportion of the influence degree of the z1-th characteristic factor on the evaluation result of the fan operation state, P z2 is the proportion of the influence degree of the z2-th characteristic factor on the evaluation result of the fan wear condition, and n is the number of characteristic factors included in the inspection record; the comprehensive risk value F of the fan presented by the inspection record is calculated; Step S402: Whenever a maintenance operation occurs on the fan, an abnormal mark is made on the latest generated inspection record before the maintenance; obtain the comprehensive risk values of all inspection records containing abnormal marks, and select the smallest comprehensive risk value as the risk threshold F for judging whether the fan is abnormal min ; Step S403: Divide the meteorological information at the location of the fan according to the environmental data type, obtain the numerical changes of each type of environmental data, extract the environmental data whose numerical change range exceeds the set safety range, and generate several real-time characteristic factors; Step S404: Compare the several real-time characteristic factors with the characteristic factors in each inspection record to determine the corresponding characteristic factors of each real-time characteristic factor in each inspection record; obtain the proportion of the influence degree of any real-time characteristic factor on the two evaluation results in each inspection record, namely P and Q, and calculate the average value respectively to obtain the average influence degree proportion P ave and Q ave ; obtain the first characteristic risk value (F1) ’ and the second characteristic risk value (F2) ’ of the currently newly generated inspection record, and calculate the comprehensive risk value F ’ of the current wind turbine; if F ’ > F min , generate a temporary inspection alarm and send a reminder to the staff, reminding that it is necessary to dispatch a drone to repair the wind turbine currently.

2. The drone inspection method for an offshore wind turbine according to claim 1, characterized in that: The step S100 includes the following steps: Step S101: When the drone inspects the fan, receive the monitoring data stored in all monitoring devices. The monitoring data includes environmental data and fan performance data. After the drone finishes receiving, each monitoring device restarts monitoring and overwrites the historical monitoring data; Step S102: Preset the inspection route of the drone and several shooting points in advance. When the drone flies to any shooting point, take several photos of each part of the fan; when the drone returns, store the received monitoring data and several photos in the inspection data management system and generate an inspection record.

3. The drone inspection method for an offshore wind turbine according to claim 2, wherein: The step S300 includes the following steps: Step S301: Set the monitoring data on the environment in the inspection record as environmental data, sort all inspection records in the order of record generation, and respectively obtain the environmental data in any two adjacent inspection records; divide the environmental data into several categories according to the environmental data type, compare the same type of environmental data in the two inspection records, preset corresponding comparison rules for each type of environmental data, and obtain the difference range of the two inspection records in any type of environmental data; Step S302: Preset different safety range for each type of environmental data. If there is a type of environmental data whose difference range exceeds the safety range, set the type of environmental data as a characteristic factor; Step S303: Set the first characteristic risk ratio of two adjacent inspection records as ΔF1 and the second characteristic risk ratio as ΔF2; obtain the number of characteristic factors between two adjacent inspection records as m. According to the formula: Among them, C i is the difference amplitude of the i-th characteristic factor between two adjacent inspection records; the influence degree α i of the i-th characteristic factor on the operation state of the fan and the influence degree β i on the wear condition of the fan are calculated, and the proportion of the influence degree of the i-th characteristic factor on the two evaluation results is obtained as P i =|α i | / (|α i |+|β i |) and Q i =|β i | / (|α i |+|β i |).

4. An unmanned aerial vehicle inspection system for an offshore wind turbine, which is used to execute an unmanned aerial vehicle inspection method for an offshore wind turbine according to any one of claims 1-3, and is characterized in that: The inspection system includes a fan data acquisition module, a monitoring data analysis module, an influencing factor analysis module, and a real-time inspection judgment module; The fan data acquisition module is used to monitor the operating performance and the environment of the fan by installing monitoring devices on the fan, take photos of the fan by the drone during each inspection process, and receive the monitoring data monitored by the monitoring devices; construct an inspection data management system to store the monitoring data and photos collected during each inspection process of the drone and generate a corresponding inspection record; The monitoring data analysis module is used to perform risk assessment on the operating status of the fan based on the monitoring data stored in any inspection record; Obtain the operating conditions presented in the photos of the inspection records, and conduct a wear assessment of the fan based on the differences from adjacent inspection records; The influencing factor analysis module is used to arbitrarily select an inspection record, analyze the monitoring data on the surrounding environment stored in the inspection record, and extract the characteristic factors that affect the operating state and wear condition of the fan; Compare the difference in the evaluation results between the inspection record and adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results; The real-time inspection judgment module is used to set the performance influence ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; obtain the meteorological information at the location of the fan in real time, predict the change in the operating state of the fan, and judge whether to generate a temporary inspection alarm to send a reminder to the staff.

5. The UAV inspection system for an offshore wind turbine according to claim 4, characterized in that: The fan data acquisition module includes a monitoring data acquisition unit and an inspection record generation unit; The monitoring data acquisition unit is used to monitor the operating performance and surrounding environment of the fan by installing monitoring equipment on the fan, take pictures of the fan by using a drone during each inspection process, and receive the monitoring data monitored by the monitoring equipment; the inspection record generation unit is used to construct an inspection data management system to store the monitoring data and photos collected during each inspection process by the drone, and generate a corresponding inspection record.

6. The UAV inspection system for an offshore wind turbine according to claim 4, characterized in that: The monitoring data analysis module includes a risk assessment unit and a wear assessment unit; The risk assessment unit is used to conduct a risk assessment of the operating state of the fan based on the monitoring data stored in any inspection record; The wear assessment unit is used to obtain the operating conditions presented in the photos of the inspection records, and conduct a wear assessment of the fan based on the differences from adjacent inspection records.

7. The drone inspection system for an offshore wind turbine according to claim 4, characterized in that: The influencing factor analysis module includes a characteristic factor extraction unit and an influence degree calculation unit; The characteristic factor extraction unit is used to arbitrarily select an inspection record, analyze the monitoring data on the surrounding environment stored in the inspection record, and extract the characteristic factors that affect the operating state and wear condition of the fan; The influence degree calculation unit is used to compare the difference in the evaluation results between the inspection record and adjacent inspection records to obtain the proportion of the influence degree of each characteristic factor on the two evaluation results.

8. The drone inspection system for an offshore wind turbine according to claim 4, wherein: The real-time inspection judgment module includes a risk threshold confirmation unit and a temporary inspection generation unit; The risk threshold confirmation unit is used to set the performance influence ratio for the two evaluation results of any inspection record based on the proportion of the influence degree of each characteristic factor, obtain the comprehensive risk value of any inspection record, and confirm the risk threshold for judging whether the fan is abnormal; The temporary inspection generation unit is used to obtain the meteorological information at the location of the fan in real time, predict the change in the operating state of the fan, and judge whether to generate a temporary inspection alarm to send a reminder to the staff.

Images