A tilt angle sensing detection method and system for wind power tower posture monitoring

Abstract

The application relates to the technical field of wind power, and provides a tilt angle sensing detection method and system for wind power tower posture monitoring, which comprises the following steps: acquiring inclination data of a target tilt angle sensor on a wind power tower of a wind power device, environmental meteorological data of an environment where the wind power tower is located, and operation data of a wind power generator of the wind power device; when the inclination data exceeds a preset normal inclination data range, determining icing risk information of the wind power tower according to the environmental meteorological data; determining a reason why the inclination data exceeds the preset normal inclination data range according to the icing risk information of the wind power tower and the operation data; and determining an intervention strategy of the wind power tower according to the reason. Through the scheme, the efficiency of tilt angle sensing detection in wind power tower posture monitoring can be improved.

Description

Technical Field

[0001] This application relates to the field of wind power technology, and in particular to a tilt sensing detection method and system for monitoring the attitude of wind turbine towers. Background Technology

[0002] In the field of wind power generation, monitoring the attitude of wind turbine towers is a crucial step in ensuring the safe and stable operation of wind turbines. Typically, tilt sensors are installed at different locations on the tower to acquire real-time tilt data. However, under certain climatic conditions, such as winter in high-latitude or high-altitude regions, harsh environmental factors can cause the monitoring system to misjudge situations, thereby affecting the normal operation and maintenance efficiency of the wind turbines.

[0003] Under specific winter weather conditions, when strong winds carrying supercooled water droplets arrive, the droplets freeze rapidly upon impact with the windward side of the tower due to the stable wind direction and strong force, forming an ice shell. This icing process is uneven, mainly concentrated on the windward side of the tower, resulting in a significant increase in weight on that side. This uneven additional load causes a slight, but real, physical tilt of the entire tower. As an integral part of the tower, the tilt sensor on the tower naturally tilts accordingly, accurately measuring this actual structural tilt caused by the ice load.

[0004] The key issue is that icing doesn't only occur on the tower surface. The sensor protective cover and its mounting brackets, installed on the outer wall of the tower, are also exposed to the harsh environment. Due to their protruding structure, they are more likely to become focal points for ice buildup. Ice accumulates asymmetrically on the sensor brackets and cover, especially on the windward side, potentially forming an ice wedge. The force generated by this ice wedge acts directly on the sensor body, causing it to tilt additionally and locally on its mounting base. This localized tilt is unrelated to the tower's structural tilt; it is purely due to icing on the sensor's own components.

[0005] In this way, the sensor located on the windward side no longer simply measures and reports the tower's tilt angle, but rather the superposition of two angles: the true tilt angle of the tower caused by ice load and the local tilt angle of the sensor itself caused by icing. Upon receiving this significantly different set of data, the monitoring center's system would determine that a severe local deformation had occurred somewhere on the tower, as the abnormal reading far exceeded the set danger threshold. The system could not distinguish that the main source of this abnormal reading was local interference from the sensor itself, rather than a structural risk to the tower. Therefore, the system would immediately classify it as a major structural safety incident, triggering the highest level alarm, forcing the wind turbine to shut down, and requiring maintenance personnel to be dispatched urgently in severe weather conditions for investigation. After arriving at the site at great risk, the maintenance personnel, after careful inspection, found that the tower structure was intact. Only upon checking the sensor did they discover that icing on its outer shell had caused this costly false alarm. Summary of the Invention

[0006] This application provides a tilt sensing detection method and system for wind turbine tower attitude monitoring, aiming to solve the problem that existing wind turbine tower tilt monitoring systems, under severe weather conditions, suffer from abnormal sensor data due to icing, which leads to false alarms, unnecessary shutdowns and maintenance, and cannot accurately distinguish between the tilt of the tower itself and the tilt caused by local icing of the sensor, thus affecting the safe operation and maintenance efficiency of the wind turbine.

[0007] To achieve the above objectives, this application adopts the following technical solution: Firstly, a tilt sensing detection method for monitoring the attitude of wind turbine towers is provided, specifically including: acquiring tilt data from a target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator set; wherein, the target tilt sensor is any tilt sensor on the wind turbine tower; when the tilt data exceeds a preset normal tilt data range, determining the icing risk information of the wind turbine tower based on the environmental meteorological data; wherein, the icing risk information includes low icing risk, medium icing risk, and high icing risk; determining the cause of the tilt data exceeding the preset normal tilt data range based on the icing risk information and operating data of the wind turbine tower; and determining the intervention strategy for the wind turbine tower based on the cause.

[0008] This technical solution enables the application to comprehensively consider tilt data, environmental meteorological data, and wind turbine operation data to conduct multi-dimensional analysis of abnormal tilt of wind turbine towers. This effectively distinguishes between different causes such as tower tilt itself, tower icing, and sensor icing, avoiding false alarms caused by errors in judgment due to a single data source in traditional methods. It improves the accuracy and reliability of monitoring and provides a more accurate basis for judging the attitude of wind turbine towers.

[0009] Furthermore, based on the above method, the environmental meteorological data includes air humidity and air temperature. The icing risk information of the wind turbine tower is determined according to the environmental meteorological data, specifically including: when the air humidity is less than or equal to a preset humidity threshold, or when the air humidity is greater than the preset humidity threshold and the air temperature is greater than or equal to 5°C, the icing risk information of the wind turbine tower is determined to be low icing risk; when the air humidity is greater than the preset humidity threshold and the air temperature is less than or equal to 0°C, the icing risk information of the wind turbine tower is determined to be high icing risk; when the air humidity is greater than the preset humidity threshold and the air temperature is greater than 0°C and less than 5°C, the icing risk information of the wind turbine tower is determined to be medium icing risk.

[0010] Through this technical solution, this application can refine the icing risk level of wind turbine towers based on specific air humidity and air temperature conditions, making the assessment of icing risk more scientific and accurate, providing a more reliable basis for subsequent cause judgment, and avoiding misjudgments caused by rough assessment of icing risk.

[0011] Based on this, according to the icing risk information and operating data of the wind turbine tower, the reasons for the tilt data exceeding the preset normal tilt data range are determined. Specifically, this includes: determining whether the icing risk information of the wind turbine tower is low icing risk; when the icing risk information of the wind turbine tower is not low icing risk, determining the reasons for the tilt data exceeding the preset normal tilt data range based on the operating data; when the icing risk information of the wind turbine tower is low icing risk, determining that the reason for the tilt data exceeding the preset normal tilt data range is the tilt of the wind turbine tower itself.

[0012] Through this technical solution, this application can perform preliminary screening based on icing risk information, and directly determine that the tower itself is tilting when the icing risk is low, which simplifies the judgment process and improves efficiency; while when there is an icing risk, it further combines operational data for in-depth analysis to ensure the accuracy of the judgment.

[0013] Furthermore, when the icing risk information of the wind turbine tower is not low icing risk, the reason for the tilt data exceeding the preset normal tilt data range is determined based on the operating data. Specifically, this includes: determining whether the icing risk information of the wind turbine tower is high icing risk; when the icing risk information of the wind turbine tower is high icing risk, determining whether the decrease in operating data within a preset time period is greater than a preset second threshold; if the decrease in operating data within a preset time period is greater than the preset second threshold, determining whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower; if the decrease in operating data within a preset time period is less than or equal to the preset second threshold... The reason why the tilt data exceeds the preset normal tilt data range is determined to be icing of the target tilt sensor. When the icing risk information of the wind turbine tower is medium icing risk, it is determined whether the decrease in the operating data within a preset time period is greater than a preset first threshold. If the decrease in the operating data within a preset time period is greater than the preset first threshold, the reason why the tilt data exceeds the preset normal tilt data range is determined to be icing of the wind turbine tower. If the decrease in the operating data within a preset time period is less than or equal to the preset first threshold, the reason why the tilt data exceeds the preset normal tilt data range is determined to be icing of the target tilt sensor. The preset first threshold is less than the preset second threshold.

[0014] Through this technical solution, this application can more precisely determine whether the abnormal tilt is due to icing of the wind turbine tower or icing of the target tilt sensor, based on different icing risk levels (high icing risk or medium icing risk) and the magnitude of the decrease in operating data. Different thresholds are used to distinguish different degrees of icing impact, thereby improving the accuracy of cause determination.

[0015] In some preferred embodiments, determining whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower specifically includes: acquiring the pitch angle of the wind turbine blades; determining whether the pitch angle is greater than a preset pitch angle threshold; if the pitch angle is greater than the preset pitch angle threshold, acquiring and determining, based on the image data of the blades, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower; if the pitch angle is less than or equal to the preset pitch angle threshold, determining that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor.

[0016] Through this technical solution, this application can introduce the blade pitch angle as an auxiliary judgment basis. When the pitch angle is large, image data can be used for further verification, avoiding unnecessary image data analysis, improving judgment efficiency, and providing a new perspective for distinguishing between tower icing and sensor icing.

[0017] As a further improvement, the method of acquiring and determining whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower is based on the image data of the blade. Specifically, this includes: acquiring and identifying whether there is ice in the image data of the blade; if there is ice in the image data of the blade, acquiring and determining whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower is based on the infrared data of the blade; if there is no ice in the image data of the blade, determining that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor.

[0018] Through this technical solution, this application can directly determine whether there is ice on the blades through image recognition, providing intuitive evidence of tower icing. When image data is insufficient to make a complete determination, infrared data is further introduced for auxiliary judgment, which improves the accuracy and reliability of the judgment.

[0019] To improve the solution, the method involves acquiring and determining, based on the infrared data of the blade, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. Specifically, this includes: acquiring and determining a first region in the blade based on the infrared data of the blade; wherein the temperature of the first region is less than 0°C; when the similarity between the contour of the first region and the contour of the ice layer in the image data of the blade is greater than a preset similarity threshold, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the wind turbine tower; otherwise, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the target tilt sensor.

[0020] This technical solution allows the application to combine infrared data and image data, and through contour similarity comparison, more accurately determine whether the low-temperature area on the blade corresponds to the ice layer in the image, thereby further confirming whether the wind turbine tower is icy, avoiding misjudgments that may be caused by a single data source, and improving the accuracy of the judgment.

[0021] Based on the above, the intervention strategy for the wind turbine tower is determined according to the cause, specifically including: when the cause is icing of the target tilt sensor, the intervention strategy for the wind turbine tower is determined based on the tilt data of the target tilt sensor; when the cause is icing of the wind turbine tower, the intervention strategy for the wind turbine tower is to control the wind power equipment to shut down; when the cause is the tilt of the wind turbine tower itself, maintenance personnel are notified to carry out maintenance and repair of the wind turbine tower.

[0022] Through this technical solution, this application can formulate targeted intervention strategies based on different causes of anomalies, avoiding unnecessary downtime and maintenance, improving the operating efficiency and safety of wind power equipment, and achieving precise management.

[0023] As a technological improvement, tilt sensors are installed at multiple locations at the same height as the target tilt sensor on the wind turbine tower. The intervention strategy for the wind turbine tower is determined based on the tilt data from the target tilt sensor. Specifically, this includes: acquiring tilt data from the tilt sensor at each of the multiple locations; using the minimum value among the tilt data from the multiple locations as the target tilt data; determining whether the difference between the tilt data from the target tilt sensor and the target tilt data is greater than a preset difference threshold; if the difference is greater than the preset difference threshold, the intervention strategy for the wind turbine tower is to notify de-icing personnel to remove the ice layer on the target tilt sensor.

[0024] This technical solution allows for the comparison of data from multiple tilt sensors at the same height. By analyzing the differences, it can determine whether the target tilt sensor is partially iced, thereby accurately identifying icing problems and taking targeted de-icing measures. This avoids unnecessary shutdowns or repairs of the entire wind turbine tower and improves maintenance efficiency.

[0025] Secondly, this application also discloses a tilt sensing detection system for monitoring the attitude of wind turbine towers, specifically including: an acquisition device and a processing device; the acquisition device is used to acquire tilt data from a target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator of the wind turbine; wherein, the target tilt sensor is any tilt sensor on the wind turbine tower; the processing device is used to determine the icing risk information of the wind turbine tower based on the environmental meteorological data when the tilt data exceeds a preset normal tilt data range; wherein, the icing risk information includes low icing risk, medium icing risk, and high icing risk; the processing device is used to determine the reason why the tilt data exceeds the preset normal tilt data range based on the icing risk information and operating data of the wind turbine tower; the processing device is used to determine the intervention strategy for the wind turbine tower based on the reason.

[0026] Beneficial effects The tilt sensing detection method for wind turbine tower attitude monitoring disclosed in this application comprehensively acquires tilt data from the target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator of the wind power equipment. When the tilt data exceeds the preset normal tilt data range, the icing risk information of the wind turbine tower is determined based on the environmental meteorological data. Then, the reason for the tilt data exceeding the preset normal tilt data range is determined by combining the icing risk information and the operating data. Finally, the intervention strategy for the wind turbine tower is determined based on the reason.

[0027] This method effectively solves the problem in existing technologies where icing causes abnormal tilt sensor data, making it impossible for the system to accurately distinguish between the tilt of the tower itself and the tilt caused by localized icing of the sensor. By introducing environmental meteorological data and operational data for multi-dimensional analysis, this application can accurately determine the true cause of abnormal tilting, avoiding false alarms caused by errors in judgment due to a single data source in traditional methods. This effectively avoids unnecessary wind turbine shutdowns and maintenance, significantly improves the accuracy and reliability of wind turbine tower attitude monitoring, ensures the safe and stable operation of wind power equipment, and reduces operation and maintenance costs. Attached Figure Description

[0028] Figure 1 A schematic flowchart of a tilt sensing detection method for wind turbine tower attitude monitoring provided in this application; Figure 2 A schematic flowchart of a tilt sensing detection method for wind turbine tower attitude monitoring provided in this application; Figure 3 This is a schematic diagram of the architecture of another tilt sensing detection system for wind turbine tower attitude monitoring provided in this application. Detailed Implementation

[0029] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0030] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] Traditional wind turbine tower attitude monitoring methods typically acquire real-time tower tilt data by installing tilt sensors at different locations on the tower. However, under extreme climatic conditions, such as winters in high-latitude or high-altitude regions, harsh environmental factors can lead to misjudgments by the monitoring system. Specifically, when strong winds carrying supercooled water droplets arrive, the windward side of the tower may ic up unevenly, causing a slight tilt of the entire tower. Simultaneously, the sensor protective covers and their mounting brackets installed on the outer wall of the tower may also ic up, forming ice wedges and causing additional local tilting of the sensor itself. This results in the sensor measuring a superposition of the tower's true tilt angle and the sensor's own local tilt angle, leading the monitoring system to misjudge it as a major structural safety incident, triggering alarms and forcing the wind turbine to shut down, causing unnecessary maintenance costs and risks.

[0032] In this regard, such as Figure 1 As shown, this application proposes a tilt sensing detection method for wind turbine tower attitude monitoring, including: S101. Acquire the tilt data of the target tilt sensor on the wind turbine tower, the environmental meteorological data of the environment where the wind turbine tower is located, and the operating data of the wind turbine generator of the wind turbine; the target tilt sensor can be any tilt sensor on the wind turbine tower.

[0033] S102. When the tilt data exceeds the preset normal tilt data range, determine the icing risk information of the wind turbine tower based on the environmental meteorological data; the icing risk information includes low icing risk, medium icing risk and high icing risk.

[0034] S103. Based on the icing risk information and operation data of the wind turbine tower, determine the reason why the tilt data exceeds the preset normal tilt data range.

[0035] S104. Determine the intervention strategy for wind turbine towers based on the cause.

[0036] This application, through comprehensive analysis of tilt data, environmental meteorological data, and wind turbine operation data, can effectively distinguish between various abnormal situations such as the tilt of the wind turbine tower itself, icing of the wind turbine tower, and icing of the target tilt sensor, thereby avoiding misjudgments, improving the accuracy and reliability of wind turbine tower attitude monitoring, and providing strong protection for the safe and stable operation of wind power equipment.

[0037] To better understand the technical solution of this application, it is necessary to explain some key terms involved. The wind turbine tower is the structure supporting the wind turbine unit, and its attitude monitoring is crucial for the safe operation of the turbine. A tilt sensor is a device used to measure the tilt angle of an object. In this application, the target tilt sensor refers to any tilt sensor installed on the wind turbine tower, used to monitor the tilt state of the tower in real time. The tilt data is the tilt angle data measured by the tilt sensor. Environmental meteorological data refers to the meteorological information of the environment in which the wind turbine tower is located, such as air humidity and air temperature. This data is of great significance for assessing the risk of icing.

[0038] Wind turbine operating data refers to various parameters of the wind power equipment during operation, such as power generation, wind speed, and blade pitch angle. This data reflects the turbine's operating status and load conditions. Icing risk information assesses the likelihood of icing on the wind turbine tower, categorized as low, medium, and high icing risk, to guide subsequent fault diagnosis. Intervention strategies are the corresponding measures taken for different fault causes, aiming to ensure the safe operation of the wind turbine tower.

[0039] The tilt sensing detection method of this application first requires acquiring tilt data from a target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operational data of the wind turbine generator. The target tilt sensor can be any tilt sensor on the wind turbine tower. Tilt data can be acquired by directly measuring with the tilt sensor and uploading it to a data acquisition system. For example, the tilt sensor can be a MEMS (Micro-Electro-Mechanical Systems) tilt sensor, which outputs a tilt angle signal by sensing changes in the direction of gravity. Environmental meteorological data can be acquired by setting up a meteorological station near the wind turbine tower or by obtaining regional meteorological information through a meteorological data interface. For example, temperature and humidity sensors can be deployed to monitor air temperature and humidity in real time. The wind turbine generator's operational data can be obtained through the wind turbine's SCADA (Supervisory and Data Acquisition) system. For example, the SCADA system can provide real-time operating parameters such as wind speed, power generation, and blade pitch angle.

[0040] After obtaining the above data, when the tilt data exceeds the preset normal tilt data range, it is necessary to determine the icing risk information of the wind turbine tower based on environmental meteorological data. The preset normal tilt data range can be determined through historical data analysis, expert experience, or structural mechanics calculations. Icing risk information includes low icing risk, medium icing risk, and high icing risk. For example, icing risk can be judged based on environmental meteorological data such as air humidity and air temperature. When air humidity is low or air temperature is high, the icing risk is low; when air humidity is high and air temperature is close to or below freezing point, the icing risk is high. Specifically, a series of thresholds can be set. For example, when air humidity is less than or equal to a preset humidity threshold, or when air humidity is greater than a preset humidity threshold but air temperature is greater than or equal to 5°C, it can be determined as low icing risk. When air humidity is greater than a preset humidity threshold and air temperature is less than or equal to 0°C, it can be determined as high icing risk. When air humidity is greater than a preset humidity threshold and air temperature is between 0°C and 5°C, it can be determined as medium icing risk.

[0041] Subsequently, based on the icing risk information and operational data of the wind turbine tower, the cause of the tilt data exceeding the preset normal tilt range is determined. For example, it can be first determined whether the icing risk information of the wind turbine tower is low icing risk. If the icing risk information is low icing risk, it can be preliminarily determined that the abnormal tilt data is caused by the tilt of the wind turbine tower itself, because under low icing risk conditions, the possibility of sensors or the tower icing is relatively small. If the icing risk information is not low icing risk, further analysis of operational data is needed to determine the cause. For example, it can be determined based on the decrease in the power generation of the wind turbine. When the wind turbine is icy, its power generation efficiency will decrease significantly, causing a significant drop in operational data (such as power generation) within a certain period of time.

[0042] Finally, the intervention strategy for the wind turbine tower is determined based on the identified cause. For example, if the cause is icing on the target tilt sensor, de-icing of the target tilt sensor is required. If the cause is icing on the wind turbine tower, it may be necessary to shut down the wind turbine equipment to prevent further damage to the turbine structure due to icing. If the cause is tilting of the wind turbine tower itself, maintenance personnel need to be notified to perform maintenance and repairs on the wind turbine tower to eliminate structural safety hazards.

[0043] The overall technical solution of this application aims to solve the problem of false alarms caused by icing in existing wind turbine tower attitude monitoring. Traditional methods often struggle to distinguish between structural tower tilt and misreading caused by localized icing of the sensor when tilt data is abnormal, leading to unnecessary downtime and maintenance costs. This application constructs a multi-dimensional, intelligent judgment mechanism by introducing environmental meteorological data and wind turbine operation data.

[0044] Specifically, when the system detects that the tilt data from the target tilt sensor exceeds the preset normal range, it will not immediately trigger the highest level alarm. Instead, it will conduct a preliminary assessment of the icing risk of the wind turbine tower based on environmental meteorological data of the environment in which the wind turbine tower is located, such as air humidity and air temperature. This step can effectively filter out tilt anomalies that occur under non-icing conditions, avoiding misjudgments caused by sensor misreading when there is no risk of icing.

[0045] If there is a risk of icing, the system will further combine the operating data of the wind turbine generator, such as power generation and wind speed, to more precisely determine the true cause of the tilt data anomaly. For example, if the icing risk is high and the operating data shows a significant decrease, it is likely that the wind turbine tower or blades have experienced severe icing, leading to a decline in turbine performance and causing tower tilting. Conversely, if the icing risk is high but the operating data does not change significantly, it is more likely to be judged as localized icing of the target tilt sensor rather than a structural problem with the entire tower. Through this comprehensive analysis, this application can accurately distinguish between three different anomalies: wind turbine tower tilting itself, wind turbine tower icing, and target tilt sensor icing.

[0046] Once the specific cause of the tilt data anomaly is determined, this application will determine the corresponding intervention strategy based on that cause. For example, if the target tilt sensor is iced, de-icing personnel can be notified to perform local de-icing without shutting down the system. If the wind turbine tower is iced, it may be necessary to shut down the wind turbine equipment and initiate the de-icing procedure. If the wind turbine tower itself is tilted, maintenance personnel need to be notified for on-site inspection. This differentiated intervention strategy avoids a "one-size-fits-all" approach and significantly improves the accuracy of wind turbine tower attitude monitoring and operational efficiency.

[0047] Compared with existing technologies, the core innovation of this application lies in its multi-dimensional data fusion and intelligent judgment mechanism. Traditional methods often rely solely on tilt sensor data, making it difficult to distinguish between localized sensor misreadings caused by icing and the actual structural tilt of the tower. This application constructs a more comprehensive judgment model by introducing environmental meteorological data and wind turbine operating data. For example, in existing technologies, when sensor readings are abnormal, the system may directly determine it as a major structural safety accident, forcing the wind turbine to shut down and requiring maintenance personnel to be dispatched urgently in severe weather to investigate, only to find that it was a false alarm caused by icing on the sensor casing. This application, however, can assess the icing risk by analyzing environmental meteorological data (such as air humidity and temperature) and combine it with operating data (such as the decrease in power generation) to determine whether the icing is on the tower or the sensor, thus avoiding such costly misjudgments. This method not only improves the accuracy of monitoring and reduces unnecessary downtime, but also reduces the risk to maintenance personnel in severe weather, significantly improving the intelligence level and economic benefits of wind turbine tower attitude monitoring.

[0048] In some embodiments described above in this application, it is proposed to determine the icing risk information of wind turbine towers based on environmental meteorological data. Specifically, the steps for determining the icing risk information of wind turbine towers based on environmental meteorological data can be further refined as follows.

[0049] Environmental meteorological data includes air humidity and air temperature. For example... Figure 2 As shown, based on environmental meteorological data, the icing risk information of wind turbine towers is determined, specifically including: S201. When the air humidity is less than or equal to the preset humidity threshold, or when the air humidity is greater than the preset humidity threshold and the air temperature is greater than or equal to 5°C, the icing risk information of the wind turbine tower is determined to be low icing risk.

[0050] S202. When the air humidity is greater than the preset humidity threshold and the air temperature is less than or equal to 0°C, the icing risk information of the wind turbine tower is determined to be high icing risk.

[0051] S203. When the air humidity is greater than the preset humidity threshold and the air temperature is greater than 0℃ or less than 5℃, the icing risk information of the wind turbine tower is determined to be medium icing risk.

[0052] Among them, environmental meteorological data refers to the real-time or predicted meteorological conditions of the area where the wind turbine tower is located, and its main parameters are air humidity and air temperature. Air humidity can be understood as the water vapor content in the air, usually expressed as a percentage of relative humidity; air temperature refers to the actual temperature of the ambient air. The preset humidity threshold is an empirical value or a critical humidity value set based on historical data, geographical location, and other factors, used to distinguish whether the water vapor content in the air has reached the conditions that may lead to icing. 5℃ and 0℃ are two key temperature thresholds, which are of great significance in the formation of icing. Specifically, when the air temperature is above 5℃, even if the humidity is high, icing is usually not easy to form; when the air temperature is below 0℃, water vapor is more likely to condense into ice.

[0053] This application's solution, by comprehensively considering two key environmental meteorological parameters—air humidity and air temperature—can more accurately assess the icing risk of wind turbine towers. Specifically, when air humidity is low or air temperature is high (e.g., greater than or equal to 5°C), icing conditions are insufficient, and therefore it is judged as having a low icing risk. When air humidity is high and air temperature is below or equal to 0°C, water vapor easily condenses into ice, resulting in an extremely high icing risk, and thus it is judged as having a high icing risk. When air humidity is high but air temperature is between 0°C and 5°C, the possibility of icing exists but is not as certain as below 0°C, and therefore it is judged as having a medium icing risk. This tiered assessment mechanism makes the assessment of wind turbine tower icing risk more detailed and accurate, avoiding misjudgments that may result from relying on a single parameter.

[0054] The aforementioned technical solution enables a tiered assessment of icing risk on wind turbine towers based on more refined environmental meteorological data (air humidity and air temperature), thereby improving the accuracy and reliability of icing risk assessment. This tiered mechanism allows for more targeted subsequent intervention strategies, avoiding unnecessary interventions or delays in necessary interventions, thus enhancing the safety and economy of wind power equipment operation.

[0055] This application further proposes the following steps for determining the reasons why the tilt data exceeds the preset normal tilt data range based on the icing risk information and operating data of the wind turbine tower: Determine whether the icing risk information of the wind turbine tower is low; if the icing risk information of the wind turbine tower is not low, determine the reason why the tilt data exceeds the preset normal tilt data range based on the operating data; if the icing risk information of the wind turbine tower is low, determine that the reason why the tilt data exceeds the preset normal tilt data range is the tilt of the wind turbine tower itself.

[0056] Specifically, icing risk information is categorized into low icing risk, medium icing risk, and high icing risk. Low icing risk refers to situations where, based on environmental meteorological data (such as air humidity and air temperature), the probability of icing on the wind turbine tower or target tilt sensor is extremely low. For example, when the air humidity is less than or equal to a preset humidity threshold, or when the air humidity is greater than the preset humidity threshold but the air temperature is greater than or equal to 5°C, it is typically considered a low icing risk.

[0057] In this low-icing-risk scenario, if the tilt data still exceeds the preset normal tilt range, icing is not the primary cause. In this case, the anomaly can be directly attributed to the tilt of the wind turbine tower itself, indicating that the wind turbine tower structure may have experienced deformation or foundation settlement. Conversely, when the icing risk information is not low (i.e., medium or high icing risk), further analysis combining the wind turbine's operating data is needed to distinguish whether the data anomaly is caused by icing of the wind turbine tower or icing of the target tilt sensor.

[0058] This application's solution optimizes the diagnostic process for abnormal tilt data by introducing a priority assessment of icing risk information. When the system detects tilt data exceeding the preset normal tilt data range, it first evaluates the icing risk level of the wind turbine tower. If the icing risk is determined to be low, it means that environmental conditions (such as high temperature or low humidity) are insufficient to cause icing on the wind turbine tower or the target tilt sensor.

[0059] In this scenario, anomalies in the tilt data are more likely to stem from structural issues within the wind turbine tower itself, such as foundation settlement, structural fatigue, or external impacts. By directly identifying the cause as the wind turbine tower's own tilt, unnecessary operational data analysis is avoided in low-icing-risk scenarios, thus simplifying the diagnostic process. However, when the icing risk is not low—that is, when there is a medium or high icing risk—the likelihood of icing increases significantly. In this case, it is necessary to further integrate wind turbine operational data, such as power generation, wind speed, and blade angle, to more precisely determine whether the tilt is caused by overall tower icing or by localized icing of the target tilt sensor. This tiered assessment mechanism makes the diagnostic process more efficient and accurate.

[0060] Through the above technical solution, this application can significantly improve the efficiency and accuracy of diagnosing the causes of abnormal wind turbine tower tilt data. By prioritizing the assessment of icing risk information, icing factors can be quickly eliminated in low-icing-risk scenarios, directly pinpointing the root cause as the wind turbine tower's own tilt. This avoids unnecessary complex operational data analysis, thus saving diagnostic time and computational resources. Simultaneously, when icing risk exists, this solution guides the system to conduct more in-depth and targeted operational data analysis, ensuring accurate identification of wind turbine tower icing or target tilt sensor icing, providing a reliable basis for subsequent intervention strategy formulation. This not only optimizes the fault diagnosis process but also provides more precise decision support for the operation and maintenance management of wind power equipment.

[0061] The reasons why the tilt data exceeded the preset normal tilt data range, as determined by the icing risk information and operational data of the wind turbine tower, specifically include: The system determines whether the icing risk information of the wind turbine tower is high. If the icing risk information is high, it determines whether the decrease in operating data within a preset time period is greater than a preset second threshold. If the decrease in operating data within a preset time period is greater than the preset second threshold, it determines whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in operating data within a preset time period is less than or equal to the preset second threshold, it determines that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. If the icing risk information of the wind turbine tower is medium, it determines whether the decrease in operating data within a preset time period is greater than a preset first threshold. If the decrease in operating data within a preset time period is greater than the preset first threshold, it determines that the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in operating data within a preset time period is less than or equal to the preset first threshold, it determines that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. The preset first threshold is less than the preset second threshold.

[0062] "Operating data" can be understood as parameters reflecting the operating status of wind turbines, such as power generation and rotational speed. When the wind turbine tower or target tilt sensor becomes icy, it usually leads to a decrease in the operating efficiency of the wind turbine, which is reflected in a decrease in operating data. In practical applications, "preset duration" refers to a time window used to observe the magnitude of the decrease in operating data, such as several hours or days, with the aim of capturing the trend of changes in operating data caused by icing.

[0063] "Decrease magnitude" refers to the degree of decrease in operating data relative to its normal value or previous value within a preset time period, which can be calculated as a percentage decrease or an absolute value decrease. As a preferred implementation, the "preset first threshold" and "preset second threshold" are empirical or pre-set values ​​used to distinguish different icing conditions. The preset first threshold is smaller than the preset second threshold, reflecting the different sensitivities to the decrease magnitude of operating data under different icing risk levels. Specifically, when the icing risk is medium, even a small decrease in operating data (greater than the preset first threshold) may indicate a risk of wind turbine tower icing; while when the icing risk is high, a larger decrease in operating data (greater than the preset second threshold) is required to confirm wind turbine tower icing. This may be because under high icing risk, environmental conditions themselves have a certain impact on operating data, requiring a more significant decrease to confirm tower icing.

[0064] This application's solution, by introducing a judgment on the magnitude of the drop in operating data and combining it with different icing risk information, can more precisely identify the causes of tilt data anomalies. Specifically, when the icing risk is not low, i.e., when there is a medium or high icing risk, icing may occur on the wind turbine tower or the target tilt sensor. By monitoring the wind turbine's operating data and analyzing its drop magnitude within a preset time period, the impact of icing on the wind turbine's operating performance can be indirectly reflected. When the drop in operating data is large, it usually means that the icing area is large or the impact is deep, and it is more likely to be judged as overall icing of the wind turbine tower; while when the drop in operating data is small or not obvious, it may only be local icing of the target tilt sensor, with limited impact on overall operation. By setting different thresholds (preset first threshold and preset second threshold) and corresponding them to the icing risk level, the severity and location of icing can be more accurately judged under different risk scenarios.

[0065] Through the above technical solution, this application overcomes the ambiguity that may exist in existing technologies when determining the cause of tilt data anomalies, especially under conditions of high icing risk. By combining the icing risk information determined by environmental meteorological data with the decline in wind turbine operating data, it is possible to effectively distinguish between icing on the wind turbine tower and icing on the target tilt sensor. This distinction helps avoid misjudgment, thereby providing a more accurate basis for subsequent targeted intervention strategies, improving the accuracy and reliability of wind turbine tower attitude monitoring, and reducing unnecessary maintenance costs and downtime.

[0066] This application further proposes steps for determining whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower, including: Obtain the pitch angle of the wind turbine blades; determine whether the pitch angle is greater than the preset pitch angle threshold; if the pitch angle is greater than the preset pitch angle threshold, obtain and determine, based on the image data of the blades, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower; if the pitch angle is less than or equal to the preset pitch angle threshold, determine that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor.

[0067] Specifically, the pitch angle refers to the angle of the wind turbine blades relative to their plane of rotation, which can be obtained in real time through the pitch system sensors integrated into the wind turbine. The preset pitch angle threshold is an empirical value or a critical value set according to the wind turbine's design parameters. It distinguishes between the small adjustment angles of the blades under normal operating conditions and the large pitch angles made in response to external environmental factors (such as icing, strong winds, etc.). When the pitch angle exceeds the preset pitch angle threshold, it usually indicates that the blades are undergoing significant adjustments, which may be related to the wind turbine tower or the blades themselves bearing large loads (such as icing).

[0068] In this situation, to further confirm whether the wind turbine tower is icing, it is necessary to obtain image data of the blades. Blade image data can be obtained by capturing images using a high-definition camera installed on the wind turbine or a drone, and image recognition technology can be used to detect ice layers in the images. If the pitch angle is less than or equal to a preset pitch angle threshold, it indicates that the blades are in a relatively normal operating state or are only undergoing minor adjustments. In this case, the possibility of the entire wind turbine tower being iced is low, therefore the cause is determined to be icing of the target tilt sensor.

[0069] This application's solution effectively solves the problem of misjudgment that may arise from relying solely on the magnitude of the drop in operating data by introducing blade pitch angle and image data as auxiliary judgment criteria. When the drop in operating data is significant, initially indicating possible icing on the wind turbine tower, the severity and extent of icing can be preliminarily screened by analyzing the blade pitch angle. If the pitch angle is small, it indicates that the blades have not made significant adjustments due to icing. In this case, even if the operating data drops, it is more likely to be attributed to localized icing at the target tilt sensor rather than structural icing of the entire wind turbine tower.

[0070] Conversely, a large pitch angle strongly suggests that the blades may be under additional loads, such as severe icing, which aligns with the assessment of overall icing of the wind turbine tower. Based on this, further acquisition and analysis of blade image data, through visual identification of the ice layer, provides direct and reliable evidence, ultimately accurately determining whether the anomaly in tilt data is caused by icing of the wind turbine tower or icing of the target tilt sensor. This multi-dimensional, progressive judgment mechanism significantly improves the accuracy and reliability of cause determination.

[0071] Through the above technical solution, this application can more accurately identify the true cause of abnormal wind turbine tower tilt data and effectively distinguish between overall icing of the wind turbine tower and localized icing of the target tilt sensor. This avoids potential misdiagnosis that may occur if judgment is made solely based on the magnitude of the drop in operating data, thereby ensuring the targetedness and effectiveness of subsequent intervention strategies. For example, when it is determined that the target tilt sensor is icing, only the sensor can be de-iced, avoiding unnecessary downtime or large-scale maintenance of the entire wind turbine tower; while when it is determined that the wind turbine tower is icing, timely major intervention measures such as shutdown can be taken to effectively prevent potential safety risks and equipment damage caused by structural icing of the tower, significantly improving the safety and maintenance efficiency of wind power equipment operation.

[0072] This application further proposes steps for acquiring and determining, based on image data of the blades, whether the reason for tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower, including: Acquire and identify whether there is ice in the image data of the blade; if there is ice in the image data of the blade, acquire and determine whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower based on the infrared data of the blade; if there is no ice in the image data of the blade, determine that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor.

[0073] Specifically, acquiring and identifying the presence of ice in image data of wind turbine blades involves obtaining visual images of the blades using image acquisition equipment and then analyzing the acquired image data using image processing techniques or machine learning algorithms to detect the presence of ice on the blade surface. For example, image processing algorithms such as edge detection, texture analysis, and color recognition can be used to identify ice features, or deep learning models can be trained to classify and locate ice in the images. The aim is to make a preliminary and intuitive judgment on the icing status of the blades.

[0074] If no ice layer is found in the image data of the blades, the reason for the tilt data exceeding the preset normal tilt data range is directly determined to be icing of the target tilt sensor. This means that when no ice layer is visually detected on the blades, the system tends to consider that the possibility of the entire wind turbine tower being iced is low, and that the data anomaly is more likely caused by localized icing or malfunction of the target tilt sensor itself.

[0075] Furthermore, if ice is present in the blade image data, it is necessary to acquire and determine, based on the blade's infrared data, whether the tilt data exceeding the preset normal tilt range is due to icing on the wind turbine tower. This is because while image data can provide visual information about ice, its accuracy and the extent of its impact on the overall wind turbine tower still need further confirmation. Infrared data reflects the temperature distribution of an object's surface. Ice layers are typically below 0°C. By acquiring infrared data from the blade using an infrared thermal imager, it is possible to accurately detect whether low-temperature areas exist on the blade surface, thereby verifying whether the ice layer identified in the image data is indeed real ice.

[0076] The solution proposed in this application effectively addresses the limitations of relying solely on a single image data to determine the cause of icing on wind turbine towers by introducing a step for identifying ice layers in blade image data and combining it with infrared data for secondary verification.

[0077] Specifically, firstly, by identifying blade image data, the presence of ice on the blade surface can be quickly and preliminarily determined. If no ice layer is identified in the image data, the possibility of overall icing of the wind turbine tower can be quickly ruled out, attributing the cause to icing of the target tilt sensor, thus simplifying subsequent judgment processes. Secondly, when ice is identified in the image data, infrared data is further introduced for verification. This is because infrared data can provide temperature information about the ice layer, and the temperature characteristics of ice are key to distinguishing it from other visual artifacts. By comparing image data and infrared data, the actual existence and physical characteristics of the ice layer can be more accurately confirmed, thus avoiding misjudgments caused by non-ice factors such as sunlight and water mist. This multimodal data fusion judgment method makes the judgment of the cause of icing on wind turbine towers more scientific, rigorous, and reliable.

[0078] Through the above technical solution, this application can significantly improve the accuracy and reliability of determining the cause of tilt data anomalies in wind turbine tower attitude monitoring. By combining image data and infrared data for multi-dimensional verification, it can more accurately distinguish between overall icing of the wind turbine tower and localized icing of the target tilt sensor, effectively avoiding misjudgments that may arise from a single data source. This not only helps to take timely and correct intervention strategies, such as targeted de-icing or sensor maintenance, but also reduces unnecessary downtime and maintenance, thereby lowering operating costs and improving the operating efficiency and safety of wind power equipment.

[0079] This application further proposes to obtain and determine, based on the infrared data of the blades, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower, including: The first region in the blade is determined based on the infrared data of the blade; the temperature of the first region is less than 0℃; when the similarity between the outline of the first region and the outline of the ice layer in the image data of the blade is greater than a preset similarity threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is that the wind turbine tower is icing; otherwise, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is that the target tilt sensor is icing.

[0080] Specifically, acquiring and determining the first region within the blade based on infrared data involves collecting temperature distribution data of the blade surface using equipment such as infrared thermal imagers, and identifying areas with temperatures below 0°C based on this data. This first region typically corresponds to areas on the blade surface where ice has formed or is in a critical state for freezing, because ice formation requires temperatures below the freezing point. The temperature of the first region being below 0°C can be understood as the actual temperature of this region being below the freezing point of water, a necessary condition for freezing. The purpose is to initially pinpoint potential freezing areas through temperature characteristics.

[0081] Furthermore, when the similarity between the contour of the first region and the contour of the ice layer in the image data of the blade exceeds a preset similarity threshold, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the wind turbine tower. This means that if the shape and range of the low-temperature region (first region) identified by infrared data highly match the shape and range of the ice layer region identified in the visible light image, it can be inferred that this large-area icing phenomenon related to ambient temperature is more likely due to icing of the wind turbine tower as a whole or its key components, thus affecting the tower's attitude. The preset similarity threshold can be set according to actual application scenarios and experience, for example, it can be calculated using image similarity measurement methods such as Jaccard similarity coefficient, Dice coefficient, or Hausdorff distance.

[0082] Conversely, if the similarity between the contour of the first region and the contour of the ice layer in the image data of the blade is less than or equal to a preset similarity threshold, then the reason why the tilt data exceeds the preset normal tilt data range is determined to be icing of the target tilt sensor. This indicates that if the low-temperature region does not match the ice layer region well, or if the ice layer region is relatively isolated, it is more likely that the tilt data is abnormal due to local icing of the target tilt sensor rather than overall icing of the wind turbine tower.

[0083] This application's solution addresses the accuracy limitations of traditional methods in distinguishing between icing on wind turbine towers and icing on target tilt sensors by introducing the identification of low-temperature regions in blade infrared data and comparing their contours with the contours of ice layers in image data. Since icing is typically accompanied by a significant temperature drop, infrared data can accurately capture the low-temperature regions on the blade surface, i.e., the first region.

[0084] Meanwhile, ice layers possess specific visual characteristics in visible light images. By comparing the contour similarity of the icing areas revealed by these two different dimensions of data, the extent and nature of the icing can be determined more reliably. When the two are highly similar, it indicates that the ice layer on the blades is caused by a large-scale low-temperature environment, which is more likely to cause the entire wind turbine tower to freeze, thus affecting its attitude. Conversely, if the similarity is low, it suggests that the ice layer may only be a local phenomenon, such as only covering the target tilt sensor without affecting the entire wind turbine tower. This mechanism of multi-source data fusion and feature comparison makes the determination of the cause of abnormal tilt data more accurate.

[0085] Through the above technical solution, this application can significantly improve the accuracy of diagnosing the causes of abnormal tilt data in wind turbine towers. Compared with relying solely on image recognition of ice layers or simply acquiring infrared data, this application effectively avoids misjudgments by comparing the contour similarity between the low-temperature areas identified by infrared data and the ice layer areas identified by image data. This accurate judgment helps distinguish whether the abnormal attitude is caused by overall icing of the wind turbine tower or by reading deviations caused only by localized icing of the target tilt sensor. Therefore, it can provide a more reliable basis for subsequent intervention strategies. For example, when it is confirmed that the wind turbine tower is icing, timely shutdown measures can be taken to avoid equipment damage; when it is confirmed that the target tilt sensor is icing, targeted de-icing can be carried out, thereby avoiding unnecessary downtime, reducing operating costs, and improving the operating efficiency and safety of wind power equipment.

[0086] This application further proposes intervention strategies for determining the cause of wind turbine towers, including: When the cause is icing of the target tilt sensor, the intervention strategy for the wind turbine tower is determined based on the tilt data of the target tilt sensor; when the cause is icing of the wind turbine tower, the intervention strategy for the wind turbine tower is to control the wind power equipment to shut down; when the cause is the tilt of the wind turbine tower itself, maintenance personnel are notified to carry out maintenance and repair of the wind turbine tower.

[0087] Specifically, when the system determines that the tilt data exceeds the preset normal tilt data range because the target tilt sensor is iced, targeted treatment of the sensor is required. In this case, the intervention strategy will be determined based on the tilt data of the target tilt sensor. For example, this tilt data can be compared with data from other uniced or normally functioning tilt sensors to assess the severity of the icing and its impact on measurement accuracy, thereby deciding whether de-icing is necessary.

[0088] When the system determines that the tilt data exceeds the preset normal tilt range due to icing on the wind turbine tower, this typically means that a large layer of ice has adhered to the tower, potentially posing a serious threat to the tower's structural stability and the wind turbine's operational safety. In this situation, as an emergency and necessary intervention, the wind turbine tower's intervention strategy is determined to be shutting down the wind turbine equipment. This measure aims to avoid risks such as structural overload, abnormal vibration, or blade ice ejection caused by icing, ensuring the safety of equipment and personnel.

[0089] In practical applications, when the system determines that the tilt data exceeds the preset normal tilt range due to the wind turbine tower's own tilt, this indicates a potential structural problem with the tower, such as foundation settlement, structural fatigue, or loose connections. In this case, the intervention strategy is to notify maintenance personnel to perform maintenance and repairs on the wind turbine tower. This notification can include specific tilt data, the time of occurrence, and the possible direction of tilt, enabling maintenance personnel to quickly locate the problem and take appropriate inspection, reinforcement, or repair measures to prevent further structural damage or safety hazards.

[0090] This application's solution precisely distinguishes between different causes of tilt data exceeding the preset normal tilt range and matches the most appropriate intervention strategy for each cause, thus solving the problems of generalized and unspecific intervention strategies in the aforementioned basic solutions. Specifically, when icing is identified on the target tilt sensor, the solution avoids unnecessary shutdowns or large-scale maintenance of the entire wind turbine, focusing instead on the sensor itself and developing a refined de-icing or calibration plan by analyzing its tilt data. When icing is determined to be on the wind turbine tower itself, the solution can quickly initiate a shutdown mechanism due to its potentially serious hazards, fundamentally eliminating safety risks. When the tilt is determined to be structural tilt of the wind turbine tower itself, the solution can promptly trigger professional maintenance and repair procedures, ensuring a comprehensive assessment and repair of structural safety. This differentiated handling mechanism ensures that each abnormal situation receives the most direct and effective response, avoiding the inefficiency and resource waste that may result from a "one-size-fits-all" approach.

[0091] Through the above technical solutions, this application can significantly improve the intervention efficiency and safety of wind turbine tower attitude monitoring. By adopting differentiated intervention strategies for different causes, over-intervention or under-intervention can be avoided, thereby effectively reducing operating costs and minimizing unplanned downtime. For example, for localized problems such as sensor icing, the entire wind turbine equipment does not need to be shut down; only the sensor needs to be addressed, greatly improving equipment availability. Simultaneously, for major safety hazards such as icing of the wind turbine tower itself or its tilting, decisive measures such as timely shutdown or maintenance can be taken, thereby maximizing the operational safety of the wind turbine equipment and the safety of personnel, and avoiding serious accidents that may result from misjudgment or untimely intervention.

[0092] This application further proposes a more precise method for determining intervention strategies, aiming to improve the accuracy of identifying icing conditions on the target tilt sensor through multi-sensor collaborative judgment.

[0093] Inclination sensors are installed at multiple locations at the same height as the target inclination sensor on the wind turbine tower. Intervention strategies for the wind turbine tower are determined based on the tilt data from the target inclination sensor, including: Acquire tilt data from tilt sensors at each of multiple locations; take the minimum tilt data from the multiple locations as the target tilt data; determine whether the difference between the tilt data from the target tilt sensor and the target tilt data is greater than a preset difference threshold; if the difference between the tilt data from the target tilt sensor and the target tilt data is greater than the preset difference threshold, determine the intervention strategy for the wind turbine tower as notifying de-icing personnel to remove the ice layer on the target tilt sensor.

[0094] Specifically, to more accurately determine whether the target tilt sensor is iced, this application installs additional tilt sensors at multiple locations on the wind turbine tower at the same height as the target tilt sensor. These tilt sensors can be evenly distributed around the circumference at that height, or non-uniformly arranged as needed. The purpose is to provide tilt data from multiple points at the same height to form a reference system. Acquiring the tilt data from the tilt sensor at each of the multiple locations means that the system periodically, or when an anomaly is detected, collects the current tilt measurement value from these tilt sensors located at the same height. This tilt data can be angle values, acceleration values, or other physical quantities that reflect the tilt state.

[0095] Furthermore, the minimum value among the tilt data corresponding to multiple locations is used as the target tilt data. This is to obtain a reference value at that height that most closely approximates the actual tilt state. In cases of localized icing, the readings of icy sensors are typically closer to the actual tilt of the wind turbine tower, while the readings of icy sensors may deviate. Selecting the minimum value effectively eliminates abnormally high readings caused by localized icing, thus obtaining a relatively reliable benchmark. Based on this, it is determined whether the difference between the tilt data of the target tilt sensor and the target tilt data exceeds a preset difference threshold. This aims to identify whether the target tilt sensor is abnormal by comparing its relative differences with other sensors at that height.

[0096] The preset difference threshold is an empirical or calibrated value used to define the boundary between normal fluctuations and abnormal deviations. If the difference between the tilt data from the target tilt sensor and the target tilt data exceeds the preset difference threshold, it is highly likely that the target tilt sensor itself has a problem, such as localized icing. In this case, the intervention strategy for the wind turbine tower is to notify de-icing personnel to remove the ice layer on the target tilt sensor. The purpose is to perform precise local intervention and avoid unnecessary downtime or large-scale maintenance of the entire wind turbine.

[0097] This application's solution, by installing multiple tilt sensors at the same height on the wind turbine tower and comparing and analyzing the data from these sensors, can effectively distinguish between tilt anomalies caused by the overall tilt of the wind turbine tower and tilt data anomalies caused by localized icing of a single tilt sensor. When the tilt data of the target tilt sensor exceeds the preset normal tilt data range, if it differs significantly from the minimum value of other tilt sensors at the same height (serving as the reference tilt data for that height), it indicates that the data anomaly of the target tilt sensor is not caused by the overall tilt of the wind turbine tower, but is more likely caused by localized icing within the tower itself. This comparison mechanism allows the system to more accurately locate the source of the problem, thereby avoiding inappropriate intervention measures due to misjudgment. It is precisely because of this multi-sensor collaborative judgment that the identification of icing on the target tilt sensor is more accurate, thus enabling targeted de-icing strategies to be implemented.

[0098] Through the above technical solution, this application effectively solves the problem of misjudgment that may occur when relying solely on data from a single target tilt sensor to determine its own icing. By introducing data from multiple tilt sensors at the same height for comparison, and using the minimum value as a reference benchmark, it is possible to more accurately identify whether the target tilt sensor's data anomalies are caused by localized icing. This precise identification capability avoids incorrect assessment of the overall condition of the wind turbine tower, thereby reducing unnecessary wind turbine downtime or large-scale maintenance, and improving the operating efficiency and safety of wind turbines. Furthermore, targeted notification of de-icing personnel to remove ice from the target tilt sensor makes intervention measures more precise and efficient, reducing operating costs.

[0099] This application also discloses a tilt sensing detection system for monitoring the attitude of wind turbine towers, comprising: an acquisition device and a processing device; the acquisition device is used to acquire tilt data from a target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator of the wind turbine; the target tilt sensor is any tilt sensor on the wind turbine tower; the processing device is used to determine the icing risk information of the wind turbine tower based on the environmental meteorological data when the tilt data exceeds a preset normal tilt data range; the icing risk information includes low icing risk, medium icing risk, and high icing risk; the processing device is used to determine the reason why the tilt data exceeds the preset normal tilt data range based on the icing risk information and operating data of the wind turbine tower; the processing device is used to determine the intervention strategy for the wind turbine tower based on the reason.

[0100] Specifically, the acquisition device can be configured to include multiple sensor interface modules for connecting to target tilt sensors, temperature and humidity sensors, wind speed sensors, etc., to collect tilt data, environmental meteorological data, and wind turbine operating data in real time. As an optional implementation, the acquisition device can also be an integrated communication module that interacts with external data sources such as the wind turbine's SCADA system and weather stations via wireless or wired network interfaces to acquire the required data. For example, the acquisition device can be a data acquisition unit that integrates a data converter and communication interface, capable of converting analog signals into digital signals and uploading them to a processing device.

[0101] Specifically, the processing device can be a central processing unit or a microcontroller, which internally stores program instructions for executing the aforementioned detection methods. After the acquisition device transmits data to the processing device, the processing device analyzes the data according to preset logic and algorithms.

[0102] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A tilt angle sensing detection method for monitoring the attitude of wind turbine towers, characterized in that, include: Acquire tilt data from the target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator; the target tilt sensor can be any tilt sensor on the wind turbine tower. When the tilt data exceeds the preset normal tilt data range, the icing risk information of the wind turbine tower is determined based on the environmental meteorological data. Icing risk information includes low icing risk, medium icing risk, and high icing risk; Based on the icing risk information of the wind turbine tower and the aforementioned operating data, determine the reason why the tilt data exceeds the preset normal tilt data range; Determine the intervention strategy for wind turbine towers based on the cause; Based on the aforementioned environmental meteorological data, the icing risk information of the wind turbine tower is determined, including: If the air humidity is less than or equal to the preset humidity threshold, or if the air humidity is greater than the preset humidity threshold and the air temperature is greater than or equal to 5°C, the icing risk information of the wind turbine tower is determined to be low icing risk. When the air humidity is greater than the preset humidity threshold and the air temperature is less than or equal to 0°C, the icing risk information of the wind turbine tower is determined to be high icing risk. When the air humidity is greater than the preset humidity threshold and the air temperature is greater than 0℃ or less than 5℃, the icing risk information of the wind turbine tower is determined to be medium icing risk. Based on the icing risk information of the wind turbine tower and the aforementioned operational data, the reasons why the tilt data exceeds the preset normal tilt data range are determined, including: Determine whether the icing risk information of the wind turbine tower is low icing risk; When the icing risk information of the wind turbine tower is not low icing risk, determine the reason why the tilt data exceeds the preset normal tilt data range based on the operation data; When the icing risk information of the wind turbine tower is low, the reason for the tilt data exceeding the preset normal tilt data range is determined to be the tilt of the wind turbine tower itself. When the icing risk information for the wind turbine tower is not low icing risk, the reasons for the tilt data exceeding the preset normal tilt data range should be determined based on the operating data, including: Determine whether the icing risk information of the wind turbine tower is high icing risk; When the icing risk information of the wind turbine tower is high icing risk, it is determined whether the decrease in the operating data within a preset time period is greater than a preset second threshold. If the decrease in the operating data within a preset time period is greater than the preset second threshold, it is determined whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in the operating data within a preset time period is less than or equal to the preset second threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. When the icing risk information of the wind turbine tower is medium icing risk, it is determined whether the decrease in operating data within a preset time period is greater than a preset first threshold. If the decrease in operating data within a preset time period is greater than the preset first threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in operating data within a preset time period is less than or equal to the preset first threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. The preset first threshold is less than the preset second threshold.

2. The tilt sensing detection method for wind turbine tower attitude monitoring according to claim 1, characterized in that, Determine whether the reason for the tilt data exceeding the preset normal tilt data range is icing on the wind turbine tower, including: Obtain the pitch angle of the wind turbine blades; Determine if the pitch angle is greater than the preset pitch angle threshold. If the pitch angle is greater than the preset pitch angle threshold, acquire and determine, based on the image data of the blade, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the pitch angle is less than or equal to the preset pitch angle threshold, determine that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor.

3. The tilt sensing detection method for wind turbine tower attitude monitoring according to claim 2, characterized in that, Acquire and determine, based on the image data of the blades, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower, including: Acquire and identify whether ice layer exists in the image data of the leaves; If ice is present in the image data of the blade, the infrared data of the blade is used to determine whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If there is no ice in the image data of the blade, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the target tilt sensor.

4. The tilt sensing detection method for wind turbine tower attitude monitoring according to claim 3, characterized in that, Acquire and determine, based on the infrared data of the blades, whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower, including: Acquire and determine the first region in the leaf based on the infrared data of the leaf; the temperature of the first region is less than 0℃; If the similarity between the contour of the first region and the contour of the ice layer in the image data of the blade is greater than a preset similarity threshold, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the wind turbine tower; otherwise, the reason for the tilt data exceeding the preset normal tilt data range is determined to be icing of the target tilt sensor.

5. The tilt sensing detection method for wind turbine tower attitude monitoring according to claim 1, characterized in that, Based on the cause, determine the intervention strategy for wind turbine towers, including: When the cause is icing of the target tilt sensor, the intervention strategy for the wind turbine tower is determined based on the tilt data of the target tilt sensor. When the cause is icing on the wind turbine tower, the intervention strategy for the wind turbine tower is to shut down the wind power equipment. If the cause is the tilting of the wind turbine tower itself, notify maintenance personnel to carry out maintenance and repair on the wind turbine tower.

6. The tilt sensing detection method for wind turbine tower attitude monitoring according to claim 5, characterized in that, Inclination sensors are installed at multiple locations at the same height as the target inclination sensor on the wind turbine tower. Intervention strategies for the wind turbine tower are determined based on the tilt data from the target inclination sensor, including: Acquire tilt data from tilt sensors at each of multiple locations; The minimum value among the tilt data corresponding to multiple locations is taken as the target tilt data; Determine whether the difference between the tilt data from the target tilt sensor and the target tilt data is greater than a preset difference threshold; If the difference between the tilt data of the target tilt sensor and the target tilt data is greater than the preset difference threshold, the intervention strategy for the wind turbine tower is to notify de-icing personnel to remove the ice layer on the target tilt sensor.

7. A tilt sensing and detection system for monitoring the attitude of wind turbine towers, characterized in that, include: Acquisition device and processing device; The acquisition device is used to acquire tilt data from a target tilt sensor on the wind turbine tower, environmental meteorological data of the environment where the wind turbine tower is located, and operating data of the wind turbine generator set of the wind turbine equipment; the target tilt sensor can be any tilt sensor on the wind turbine tower. The processing device is used to determine the icing risk information of the wind turbine tower based on the environmental meteorological data when the tilt data exceeds the preset normal tilt data range. Icing risk information includes low icing risk, medium icing risk, and high icing risk; The processing device is also used to determine the reason why the tilt data exceeds the preset normal tilt data range based on the icing risk information of the wind turbine tower and the operating data. The processing device is also used to determine the intervention strategy for the wind turbine tower based on the cause; Processing device, used to determine icing risk information of wind turbine towers based on the environmental meteorological data, including: If the air humidity is less than or equal to the preset humidity threshold, or if the air humidity is greater than the preset humidity threshold and the air temperature is greater than or equal to 5°C, the icing risk information of the wind turbine tower is determined to be low icing risk. When the air humidity is greater than the preset humidity threshold and the air temperature is less than or equal to 0°C, the icing risk information of the wind turbine tower is determined to be high icing risk. When the air humidity is greater than the preset humidity threshold and the air temperature is greater than 0℃ or less than 5℃, the icing risk information of the wind turbine tower is determined to be medium icing risk. The processing device is also used to determine, based on the icing risk information of the wind turbine tower and the operating data, the reasons why the tilt data exceeds the preset normal tilt data range, including: Determine whether the icing risk information of the wind turbine tower is low icing risk; When the icing risk information of the wind turbine tower is not low icing risk, determine the reason why the tilt data exceeds the preset normal tilt data range based on the operation data; When the icing risk information of the wind turbine tower is low, the reason for the tilt data exceeding the preset normal tilt data range is determined to be the tilt of the wind turbine tower itself. The processing device is also used to determine, based on operating data, the reasons why the tilt data exceeds the preset normal tilt data range when the icing risk information of the wind turbine tower is not low icing risk, including: Determine whether the icing risk information of the wind turbine tower is high icing risk; When the icing risk information of the wind turbine tower is high icing risk, it is determined whether the decrease in the operating data within a preset time period is greater than a preset second threshold. If the decrease in the operating data within a preset time period is greater than the preset second threshold, it is determined whether the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in the operating data within a preset time period is less than or equal to the preset second threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. When the icing risk information of the wind turbine tower is medium icing risk, it is determined whether the decrease in operating data within a preset time period is greater than a preset first threshold. If the decrease in operating data within a preset time period is greater than the preset first threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the wind turbine tower. If the decrease in operating data within a preset time period is less than or equal to the preset first threshold, it is determined that the reason for the tilt data exceeding the preset normal tilt data range is icing of the target tilt sensor. The preset first threshold is less than the preset second threshold.

Images