Wind turbine whole unit bolt monitoring system

Abstract

The application provides a wind turbine complete machine bolt monitoring system. The wind turbine complete machine bolt monitoring system comprises a sensor failure monitoring module, a plurality of high-precision flange bolt gap sensors installed at the fan blade flanges and the tower flange, which are used for monitoring the flange gap in real time, and a multi-channel flange bolt gap acquisition instrument, which is used for receiving the monitoring signals output by the flange bolt gap sensors and collecting data. The wind turbine complete machine bolt monitoring system effectively solves the problem of alarm omission caused by sensor failure or failure in the prior art by introducing a sensor failure monitoring module and a fault prediction module. The system uses a model-based fault detection algorithm to monitor the state of the sensor in real time, timely issues a warning signal when an abnormality occurs, thereby avoiding the blind spots of the traditional system and improving the operation safety and reliability of the wind turbine.

Description

Technical Field

[0001] This invention relates to the field of wind turbine monitoring technology, specifically a wind turbine whole machine bolt monitoring system. Background Technology

[0002] The wind turbine bolt monitoring system mainly consists of multiple components, including high-precision sensors, flange bolt gap acquisition devices, data processors, a synchronous acquisition master station, wireless communication equipment, and a host computer system. This system monitors bolt loosening in real time by installing sensors on the blades and tower flanges of the wind turbine. The sensors transmit the monitored flange gap data to the acquisition device, which then transmits the data to the host computer for storage, analysis, and display via wireless communication or a wired network. The entire system operates on the principle that bolt loosening or breakage causes changes in the flange clearance. High-precision displacement sensors monitor these changes in real time, providing fault warnings and analysis to ensure the stable operation of the wind turbine.

[0003] Although the wind turbine bolt monitoring system can effectively monitor changes in flange clearance, it still has some shortcomings. The system requires high sensor installation accuracy; improper installation may lead to measurement errors and affect the accuracy of the monitoring data. The system relies heavily on sensor accuracy; if a sensor malfunctions or fails, alarm signals will be missed. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a wind turbine whole-machine bolt monitoring system, which solves the problem that the system has high requirements for sensor installation accuracy, and improper installation may lead to measurement errors; the system mainly relies on the accuracy of sensors, and if the sensors malfunction or fail, alarm signals will be missed.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a wind turbine generator set bolt monitoring system, comprising:

[0006] Sensor failure monitoring module;

[0007] Multiple high-precision flange bolt gap sensors are installed on the flanges of the wind turbine blades and the tower flanges to monitor the flange gap in real time;

[0008] A multi-channel flange bolt gap acquisition instrument is used to receive the monitoring signal output by the flange bolt gap sensor and to acquire data.

[0009] A data processor is used to receive data from the data acquisition instrument and transmit the data to a host computer via wireless communication or a wired network.

[0010] The host computer system is used to store, analyze, alarm, and display the flange bolt gap data, and to provide fault early warning.

[0011] The system further includes an integrated installation guidance system suitable for ensuring the accurate installation of the flange bolt gap sensor;

[0012] The system further includes a fault prediction module that predicts bolt loosening or breakage faults using the following model:

[0013]

[0014] Where, σ failure This represents the probability of fault prediction, where A, B, C, and λ are preset parameters, and t is a time variable.

[0015] Preferably, the integrated installation guidance system includes:

[0016] Install the tooling to ensure that the sensor mounting surface is parallel to the flange surface to be monitored, and ensure that the installation error does not exceed 2°.

[0017] Positioning blocks are used to fix the sensor in place so that it can be accurately positioned.

[0018] Structural adhesive is used to ensure a reliable connection between the sensor and the flange surface, and the adhesive layer thickness is greater than 1cm to avoid hollow areas.

[0019] Preferably, the flange bolt gap sensor is a displacement sensor, capable of measuring minute displacement changes between flange surfaces with a resolution of 0.5 μm. The data processor includes a wireless communication module for wirelessly transmitting monitoring data to a remote wind farm server.

[0020] Preferably, the data processor further includes an early warning module based on machine learning algorithms, used to automatically adjust the monitoring and early warning threshold according to historical data and issue early warning signals in a timely manner. The host computer system includes a fault diagnosis module, used to provide a detailed fault report when a fault occurs, and to give the specific fault location and type.

[0021] Preferably, the flange bolt gap acquisition instrument is connected to multiple flange bolt gap sensors via an RS485 interface and supports simultaneous acquisition of data from up to 16 sensors. The sensor failure monitoring module adopts a model-based fault detection algorithm, which can issue a failure alarm in real time when the sensor data is abnormal.

[0022] Preferably, the data processor further includes a real-time calculation module based on the flange clearance variation formula, which can convert the data collected by the sensor into a change in preload and analyze the change. The formula is:

[0023]

[0024] Where F is the change in bolt preload, k is the flange stiffness coefficient, Δx is the change in flange clearance, and L is the initial value of flange clearance. The flange clearance change formula is further optimized as follows:

[0025]

[0026] Where ΔP is the percentage change in bolt preload, and F new F is the preload value at the current moment. old This is the historical preload value.

[0027] Preferably, the monitoring system further includes a fault prediction module based on a multi-dimensional optimization model. This fault prediction module utilizes historical data and real-time sensor data to perform fault prediction and alarm functions through the following complex model:

[0028]

[0029] Where, σ failure This represents the probability of fault prediction, where A, B, C, and λ are preset parameters, and t is a time variable.

[0030] Preferably, the data processor includes a real-time feedback mechanism based on a composite dynamics model, which takes into account wind turbine load, climate change, and equipment aging factors, and performs real-time feedback and adjustment through the following formula:

[0031] Feedback = α·(P) load -P target )+β·(ΔT ambient -ΔT target )

[0032] Where Feedback represents the system feedback signal, α and β are adjustment coefficients, and P load For the current wind turbine load, P target For the target load, ΔT ambient and ΔT target These are the current ambient temperature and the target temperature, respectively.

[0033] This invention provides a whole-unit bolt monitoring system for wind turbines. It has the following beneficial effects:

[0034] This wind turbine bolt monitoring system effectively solves the problem of missed alarms caused by sensor failure or malfunction in existing technologies by introducing a sensor failure monitoring module and a fault prediction module. Utilizing a model-based fault detection algorithm, the system can monitor the sensor status in real time and issue timely warning signals when anomalies occur, thus avoiding blind spots in traditional systems and improving the operational safety and reliability of the wind turbine. The integrated high-precision flange bolt gap sensor and multi-channel data acquisition instrument can accurately capture minute changes in flange gap, further improving monitoring accuracy and ensuring data accuracy. These innovative designs solve the problem of high sensor accuracy requirements and ensure high-quality and real-time monitoring data.

[0035] This invention also utilizes an integrated installation guidance system to ensure precise installation of the flange bolt gap sensor, reducing measurement errors caused by improper installation. The design of the installation fixture and positioning block guarantees the parallelism between the sensor and the flange surface and effectively controls installation errors, further improving the system's stability and accuracy. In addition, the machine learning-based early warning module in the data processor enables the monitoring system to adaptively adjust early warning thresholds based on historical data, dynamically responding to changes in the wind turbine. Furthermore, a real-time feedback mechanism based on a composite dynamic model is employed, incorporating adjustments based on multiple factors such as wind turbine load, climate change, and equipment aging, achieving more intelligent and dynamic monitoring and management. This allows the wind turbine to adjust monitoring strategies in real time and accurately, thereby maximizing equipment lifespan and improving system efficiency and safety. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] Example 1

[0039] like Figure 1 As shown, this embodiment of the invention provides a wind turbine whole-machine bolt monitoring system, including a sensor failure monitoring module.

[0040] Multiple high-precision flange bolt gap sensors are installed on the flanges of the wind turbine blades and the tower flanges to monitor the flange gap in real time. The flange bolt gap sensors are displacement sensors that can measure minute displacement changes between the flange surfaces with a resolution of 0.5μm. The data processor includes a wireless communication module to transmit the monitoring data to a remote wind farm server wirelessly.

[0041] The multi-channel flange bolt gap acquisition instrument is used to receive the monitoring signals output by the flange bolt gap sensor and perform data acquisition. The flange bolt gap acquisition instrument connects to multiple flange bolt gap sensors through an RS485 interface and supports simultaneous acquisition of data from up to 16 sensors. The sensor failure monitoring module adopts a model-based fault detection algorithm, which can issue a failure alarm in real time when the sensor data is abnormal.

[0042] The data processor receives data from the acquisition instrument and transmits it to the host computer via wireless communication or a wired network. The data processor further includes a machine learning-based early warning module, which automatically adjusts the monitoring and early warning thresholds based on historical data and issues timely early warning signals. The host computer system includes a fault diagnosis module, which provides detailed fault reports and specifies the fault location and type when a fault occurs. The data processor further includes a real-time calculation module based on a flange clearance variation formula, which converts the data collected by the sensor into a change in preload and analyzes this change. The formula is:

[0043]

[0044] Where F is the change in bolt preload, k is the flange stiffness coefficient, Δx is the change in flange clearance, and L is the initial value of flange clearance. The flange clearance change formula is further optimized as follows:

[0045]

[0046] Where ΔP is the percentage change in bolt preload, and F new F is the preload value at the current moment. old This is the historical preload value.

[0047] The host computer system is used to store, analyze, alarm, and display flange bolt gap data, and to provide fault warnings.

[0048] The system further includes an integrated installation guide system suitable for ensuring accurate installation of the flange bolt gap sensor. The integrated installation guide system includes:

[0049] Install the tooling to ensure that the sensor mounting surface is parallel to the flange surface to be monitored, and ensure that the installation error does not exceed 2°.

[0050] Positioning blocks are used to fix the sensor in place so that it can be accurately positioned.

[0051] Structural adhesive is used to ensure a reliable connection between the sensor and the flange surface, and the adhesive layer thickness is greater than 1cm to avoid hollow areas.

[0052] The system further includes a fault prediction module that predicts bolt loosening or breakage faults using the following model:

[0053]

[0054] Where, σ failure This represents the probability of fault prediction, where A, B, C, and λ are preset parameters, and t is a time variable.

[0055] The monitoring system further includes a fault prediction module based on a multi-dimensional optimization model. This module utilizes historical data and real-time sensor data to predict and issue alarms using the following complex model:

[0056]

[0057] Where, σ failure The probability of failure prediction is represented by A, B, C, and λ, which are preset parameters, and t is a time variable. The data processor includes a real-time feedback mechanism based on a composite dynamics model. The composite dynamics model considers wind turbine load, climate change, and equipment aging factors, and provides real-time feedback and adjustment through the following formula:

[0058] Feedback = α·(P) load -P target )+β·(ΔT ambient -ΔT target )

[0059] Where Feedback represents the system feedback signal, α and β are adjustment coefficients, and P load For the current wind turbine load, P target For the target load, ΔT ambient and ΔT target These are the current ambient temperature and the target temperature, respectively.

[0060] Example 2: Experimental Example:

[0061] Experimental objective:

[0062] This experiment aims to verify the performance of the wind turbine's overall bolt monitoring system, particularly the effectiveness of the flange bolt gap monitoring, sensor failure detection, fault prediction module, and real-time feedback mechanism. We conducted tests and analyses using actual parameters.

[0063] Experimental equipment:

[0064] Wind turbine: The wind turbine adopts a three-bladed design. The flange clearance of the tower and blade flanges is achieved by installing six high-precision flange bolt clearance sensors on each blade flange and tower flange.

[0065] Flange bolt gap sensor: Displacement type sensor with a resolution of 0.5μm;

[0066] Flange bolt gap acquisition instrument: adopts RS485 interface and supports simultaneous acquisition of data from up to 16 sensors;

[0067] Data processor: Includes a wireless communication module that transmits monitoring data to a remote wind farm server;

[0068] Host computer system: used to store, analyze, and display flange bolt clearance data, and to provide fault warnings;

[0069] Sensor failure monitoring module: Model-based fault detection algorithm for issuing failure alarms in real time;

[0070] Fault prediction module: used to predict bolt loosening or breakage faults;

[0071] Composite dynamics model: used for real-time feedback and adjustment of system parameters.

[0072] Experimental steps:

[0073] 1. Sensor Installation:

[0074] Install flange bolt gap sensors at each wind turbine blade flange and tower flange. Use appropriate tooling to ensure the sensor is parallel to the flange face, with installation error controlled within ±2°.

[0075] Positioning blocks and structural adhesive are used to ensure a stable connection between the sensor and the flange surface, with the adhesive layer thickness maintained at more than 1 cm.

[0076] 2. Data Collection:

[0077] Start the wind turbine and operate it under different wind speeds and load conditions. Gradually increase the operating load of the wind turbine from 10% to 100%, and monitor the data including the turbine load P. load (Unit: kW), Wind speed V wind (Unit: m / s), and flange clearance variation Δx (unit: μm).

[0078] The flange bolt gap sensor transmits data to the multi-channel flange bolt gap acquisition instrument via an RS485 interface, allowing up to 16 sensors to acquire data simultaneously.

[0079] The data processor calculates the change in flange clearance and transmits the data to the host computer system.

[0080] 3. Sensor failure detection:

[0081] The system simulates sensor failure scenarios, such as a sensor malfunctioning and unable to transmit data. The failure detection module uses a model-based fault detection algorithm for real-time monitoring; when abnormal sensor data is detected, the system issues an alarm signal.

[0082] The model formula for sensor failure is:

[0083]

[0084] Parameter settings: A = 1, λ = 0.05, B = 2, C = 0.1, t is time (unit: hours).

[0085] 4. Calculation of bolt preload variation:

[0086] The data processor calculates the change in bolt preload force F in real time based on the flange clearance variation formula and analyzes the change. The formula is as follows:

[0087]

[0088] Parameter settings: flange stiffness coefficient k = 1000 N / mm, flange clearance variation Δx = 0.5 μm, initial flange clearance L = 5 mm.

[0089] The change in preload F is calculated based on the data collected by the sensor, and the warning threshold is adjusted in real time.

[0090] 5. Fault prediction and alarm:

[0091] Activate the fault prediction module and use the following model to predict bolt loosening or breakage faults:

[0092]

[0093] Parameter settings: A = 0.8, λ = 0.05, B = 1.5, C = 0.2, t is time (unit: hours).

[0094] 6. Multi-dimensional optimization model:

[0095] By utilizing historical data and real-time sensor data, a multi-dimensional optimization model is used to predict bolt loosening or fracture failures. The model is calculated using the following formula:

[0096]

[0097] Parameter settings: A = 1, λ = 0.05, B = 2, C = 0.1, t is time (unit: hours).

[0098] 7. Real-time feedback of the composite dynamics model:

[0099] The composite dynamics model incorporates real-time feedback and adjustments by considering wind turbine load, climate change, and equipment aging factors. The formula is:

[0100] Feedback = α·(P) load -P target )+β·(ΔT ambient -ΔT target )

[0101] Parameter settings: α = 0.5, β = 0.5, target load P target =500kW, target ambient temperature T target =20℃.

[0102] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A wind turbine system bolt monitoring system, characterized in that, The wind turbine bolt monitoring system comprises the following components: a sensor failure monitoring module; a plurality of high-precision flange bolt gap sensors installed on the flanges of the fan blades and the tower drum for real-time monitoring of the flange gap; a multi-channel flange bolt gap acquisition instrument for receiving the monitoring signals output by the flange bolt gap sensors and collecting data; a data processor for receiving the data from the acquisition instrument and transmitting the data to an upper computer via wireless communication or wired network; an upper computer system for storing, analyzing, alarming and displaying the flange bolt gap data and providing fault early warning; the wind turbine bolt monitoring system further comprises an integrated installation guide system for ensuring accurate installation of the flange bolt gap sensors; the wind turbine bolt monitoring system further comprises a fault prediction module for predicting bolt loosening or fracture failure through the following model: wherein σ failure represents the failure prediction probability, A, B, C, λ are preset parameters, and t is a time variable; the integrated installation guide system comprises: an installation tool to ensure that the sensor installation surface is parallel to the flange surface to be monitored and that the installation error is not more than 2°; a positioning block for fixing the sensor to ensure accurate positioning; structural adhesive for ensuring reliable connection between the sensor and the flange surface, with a glue layer thickness of more than 1 cm to avoid hollowing; the data processor further comprises a real-time calculation module based on a flange gap change formula, which can convert the data collected by the sensor into the change amount of the pre-tightening force and analyze the change amount, the formula is: where F is the change amount of the bolt pre-tightening force, k is the flange stiffness coefficient, Δx is the change amount of the flange gap, and L is the initial value of the flange gap, the flange gap change formula is further optimized as: wherein ΔP is the percentage of change in bolt pre-tightening force, F new is the pre-tightening force value at the current time, F old is the historical pre-tightening force value; the monitoring system further comprises a fault prediction module based on a multi-dimensional optimization model, which uses historical data and real-time sensor data to predict and alarm through the following complex model: wherein σ failure represents the failure prediction probability, A, B, C, λ are preset parameters, and t is a time variable; the data processor comprises a real-time feedback mechanism based on a composite dynamics model, which takes into account the wind turbine load, climate change and equipment aging factors, and performs real-time feedback and adjustment through the following formula: Feedback = a · (P load - P target ) + b · (AT ambient - AT target ) where Feedback represents the system feedback signal, a and β are adjustment factors, P load is the current fan load, P target is the target load, ΔT ambient and ΔT target are the current and target temperatures, respectively.

2. The wind turbine unit bolt monitoring system of claim 1, wherein: the flange bolt gap sensor is a displacement sensor that can measure the small displacement change between the flange surfaces with a resolution of 0.5 μm, and the data processor includes a wireless communication module for transmitting monitoring data to a remote wind farm server through wireless means.

3. The wind turbine unit bolt monitoring system of claim 1, wherein: the data processor further comprises a warning module based on a machine learning algorithm for automatically adjusting the monitoring warning threshold based on historical data and issuing a warning signal in a timely manner, and the upper computer system comprises a fault diagnosis module for providing detailed fault reports and specific fault location and type when a fault occurs.

4. The wind turbine system bolt monitoring system of claim 1, wherein: the flange bolt gap acquisition instrument is connected to a plurality of flange bolt gap sensors through an RS485 interface and supports simultaneous data collection of up to 16 sensors, and the sensor failure monitoring module uses a model-based fault detection algorithm that can issue a failure alarm in real time when the sensor data is abnormal.

Images