A digital twin holography monitoring and self-healing system for wind turbines

Abstract

A digital twin holography monitoring and self-healing system for wind turbines, characterized by the fact that it comprises sequentially connected components: a physical plane, including the body of the wind turbines (1), the multi-source sensor network deployed on it and the execution device; a data acquisition and transmission layer for acquiring and transmitting data from the aforementioned multi-source sensor network; Digital twin model layer for creating and running virtual twins that are synchronized with the physical entity layer, including geometric models, physical models, behavioral models, rule models, and data model fusion modules; Holographic monitoring and diagnostic layers for visualizing plant condition, health assessment, fault prediction, and diagnosis based on the digital twin layer; and A fault self-healing regulation layer serves to independently generate and execute regulation strategies based on diagnostic results, which are returned to the execution organization of the physical entity layer.

Description

TECHNICAL AREA

[0001] The invention belongs to the field of wind turbine technology and relates in particular to a digital twin holographic monitoring and fault self-healing control system for wind turbines. BACKGROUND

[0002] With the rapid development of the wind energy industry, the increasing capacity of wind turbines, and their remote locations, the high operating costs and high risk of failure are becoming increasingly apparent. Existing troubleshooting methods are mostly based on data drivers or physical models, but often have the following shortcomings: Information silos: Vibration monitoring, SCADA, and systems such as video surveillance operate independently, data convergence is low, and it is difficult to gain a comprehensive understanding of the plant's condition. Lack of predictability: Most systems can only trigger a fault alarm and cannot accurately predict component power input or assess remaining service life. Lack of active control capability: After detecting a fault, the system can typically only operate during downtime; the operating strategy cannot be adapted to the type and severity of the fault to suppress its development or maintain limited power operation, and there is no "self-healing" capability. SUMMARY OF THE INVENTION

[0003] The purpose of the present invention is to overcome the shortcomings of existing technology and to provide a digital twin hologram monitoring and fault control system for wind turbines. The system is capable of creating virtual twins that are fully mapped to physical wind turbines and enable monitoring, prediction, diagnosis, and proactive control throughout their entire life cycle.

[0004] The invention relates to a digital twin holographic monitoring and disturbance control system for wind turbines, comprising a physical layer including the wind turbine, a multi-source sensor network deployed thereon, and an execution device; a data acquisition and transmission layer for acquiring and transmitting data from said multi-source sensor network;

[0005] Digital twin model layers for creating and running virtual twins that are synchronized with the physical entity layer, including geometric models, physical models, behavioral models, rule models, and data model fusion modules; holographic monitoring and diagnostic layers for visualizing asset health, health assessment, fault prediction, and diagnosis based on the digital twin layer; and the fault self-healing regulation layer, which is used to autonomously generate and execute regulation strategies based on diagnostic results that are returned to the execution organization of the physical entity layer.

[0006] Further description of the above-mentioned technology: The multi-source sensor network includes at least three types of vibration, temperature, angle, voltage, acoustic sensors, video surveillance equipment and SCADA system interfaces.

[0007] As a further description of the above-mentioned technical option: The behavioral model in the digital twin layer is trained on the historical operating data of the plant by a machine learning algorithm in order to simulate the normal operating state of the plant at certain wind speeds and environmental conditions.

[0008] As a further detail of the technology described above: An angle sensor is provided at the shaft end of the wind turbine blade, and another angle sensor is installed at the end of the main shaft. The angle sensor is centrally attached to the shaft end of the blade / main shaft using a fixing assembly. The fixing assembly comprises a clamping component and a central alignment assembly. The clamping component serves to secure the angle sensor, while the central alignment assembly ensures that the connection at the shaft end of the blade / main shaft is centered.

[0009] As a further description of the aforementioned technical solution: The holding component comprises the base, the claw, and the drive train; the base is circular; the claw has three claws, each with a circular circumference and set at equal intervals at the end of the base; the base is fixed at the end with a rotary seat; the claw is set on the rotary seat; the turbine is set on the axis of rotation of the claw; the drive rod is a worm gear; the drive rods are set in the base; the drive rod is set coaxially with the base; the turbine is connected to the worm gear; the claw is set on the side of the forward holder with a rubber seal.

[0010] As a further description of the above-mentioned technical solution: The coaxial positioning components comprise support rods and drive housing; the support rods are set at uniform intervals along the base's direction of rotation; the axis of rotation of the support rods is set on the base; the axis of rotation of the support rods is perpendicular to the axis of the base; the threaded reservoirs of the drive housing are set on the base; the axis of rotation of the support rods is provided with a return spring; and the support rod is always in contact with the end of the drive housing under the action of the return spring.

[0011] As a further description of the aforementioned technical option: The function of the holographic monitoring and diagnostics layer's fault prediction and diagnosis is to perform residual analysis to detect early anomalies by comparing the real-time monitoring data of the physical entity layer with the predictions of the behavioral model. The holographic monitoring and diagnostics layer also includes a remaining lifetime prediction module for modeling and predicting performance degradation of key components.

[0012] In further detail regarding the aforementioned technical solutions: Parameter adjustment, active power limits, dynamic optimization of the maintenance strategy, and one or more emergency and alarm strategies relate to the automatic adjustment of the control reference values ​​for the torque of the paddle, travel path, or generator. Dynamic optimization of the maintenance strategy is based on the predicted remaining service life and automatically generates or updates a preventive maintenance plan.

[0013] In summary, the advantageous effect of the invention due to the application of the above-mentioned technique is: (1) Redundant data fusion and a highly reliable model that achieves comprehensive and transparent monitoring of the physical unit from macro-spectrative operation to microscopic condition, is able to detect faulty devices in advance, predict the lifetime of components, transform the operating model from "reactive maintenance" to "predictive maintenance", significantly reducing unplanned downtime, the system can not only detect problems but also adopt an active regulatory strategy to achieve the "suppression" or "disease operation" of faults, improve unit availability and power generation performance, provides operating personnel with intuitive visualization and detailed decision data support, significantly improves operational efficiency and science. (2) The angle sensor can be stably fixed coaxially by means of a fixing device on the axis of rotation and the main shaft of the blade, and installation and disassembly are simple and quick to ensure the accuracy of the measurement, to guarantee the accuracy of the data acquisition and to improve the real-time accuracy of the data model. BRIEF DESCRIPTION OF THE DRAWINGS Fig. demonstrates a system principle of the invention; Fig. shows the wind turbine structure according to the invention; Fig. shows the construction of the solid body according to the invention; Fig. shows the internal structure of a wind turbine of the invention; Fig. is an enlargement at point A in Fig. ; Fig. is an enlargement of Fig. at B.

[0014] Description: 1. Wind turbine; 2. Blade; 3. Shaft of rotation; 4. Main shaft; 5. Angle sensor; 6. Base; 7. Gripping jaws; 8. Drive rod; 9. Turbine; 11. Support rod; 12. Drive housing; 13. Mounting holes; 14. Rubber seal; 15. Knob; 16. Anti-slip rubber layer; 17. Swivel seat. DETAILED DESCRIPTION OF THE INVENTION

[0015] The following section clearly and completely describes the technical embodiments of this embodiment in conjunction with the accompanying figures of the present invention, whereby it is obvious that the described embodiments represent only a part and not the entire embodiment of the present invention. Based on the embodiments of the present invention, all other embodiments that a person skilled in the art in this field obtains without any creative work fall within the scope of protection of the present invention.

[0016] See Fig.The invention offers a technical solution for a digital twin holography monitoring and self-healing system for wind turbines: A digital holographic twin monitoring and disturbance control system for wind turbines 1, comprising the following interconnection: the physical plane, including the body of the wind turbines 1, a multi-source sensor network installed on the key components of the plant, and an execution device; the multi-source sensors include vibration sensors, temperature sensors, voltage change sensors, noise sensors, angle sensor 5, oil condition sensors, video surveillance devices, and SCADA system data interfaces.

[0017] Data acquisition and transmission layer, responsible for the real-time acquisition of heterogeneous data from the physical layer and uploading the data to the cloud or local servers via industrial Ethernet or wireless networks; The digital twin layer, deployed in the cloud or on a local server, is the core of the system and includes: Geometric model modules: highly accurate, scalable 3D models based on 3D modeling software that accurately reflect the structure of a physical unit.Physical model module: Integration of a highly reliable mechanism model describing the dynamics, thermodynamics, electromagnetics, and materials mechanics of the wind turbines; Behavioral model module: Training of historical operating data through machine learning algorithms to simulate the normal operating behavior of the machine under various operating conditions; Control model module: Storage of troubleshooting rules, operational knowledge bases, and regulatory policy bases; Data-driven model fusion module: Fusion of real-time data with the above-mentioned models to enable virtual twins, synchronized operations, and status mappings with physical entities.

[0018] Holographic monitoring and diagnostics layer, based on the digital twin model layer, implements the following functions: Holographic visualization: In a virtual three-dimensional environment, the plant's operating status is displayed in real time, intuitively showing the distribution and changes of key parameters (e.g., temperature, vibrations) in the form of colors, animations, etc. Health status assessment: Aggregation of data from multiple sources to calculate the health index of individual components; Failure prediction and diagnosis: Leverage the predictive capabilities of the digital twin to compare actual data with model predictions and detect early anomalies through residual analysis; Combination of rule libraries and AI diagnostic algorithms for precise diagnosis of failure types, locations, and severity; Remaining service life prediction: Model the performance degradation of key components (e.g.,4-bearings, gearbox) and predict their remaining service life.

[0019] The fault self-healing regulation layer receives instructions from the diagnostic layer and executes one or more of the following self-healing strategies: Parameter adjustment: If slight anomalies are detected, automatically change the parameters of the alternating bath, deflection, or torque control to modify the system's operating conditions and slow down component deterioration; Active power limits: Actively limit the plant's output to the safety margin to prevent failures while maintaining power generation when a potential failure risk is predicted; Dynamically optimize the maintenance strategy: Recommendation of optimal maintenance times and programs based on RUL forecasts, dynamic generation and maintenance plan; Emergency and alarm messages: Immediately send a high alert and perform safe shutdown procedures when a critical fault is diagnosed.

[0020] To ensure the accurate recording of the rotation angle data of the main axis 4 and the blade 2, the angle sensor 5 is attached by a fixing.

[0021] The fastening device comprises a holding device and a coaxial positioning device used to fix the angle sensor 5, and a coaxial positioning device used for the end / main axis 4 connected to the axis of rotation.The holding elements comprise the base 6, the gripping jaws 7 and the drive train 8, the base 6 being circular, the gripping jaws 7 having three, the three gripping jaws 7 with a circular circumference are set at equal intervals at the end of the base 6, the base 6 is fixed at its end with a rotary seat 17, the gripping jaws 7 are set with a rotary seat 17, the turbine 9 is set on the axis of rotation 3 of the gripping jaws 7, the drive rod 8 with a worm that rotates the drive rod 8 in the base 6, the drive rod 8 is set coaxially with the base 6, the drive rod 8 is set with a knob 15 at the end, the turbine 9 engages with the worm, and the gripping jaws 7 with a rubber seal 14 on the lateral adjustment in the direction of the holding elements.

[0022] The described coaxial positioning component comprises a support rod 11 and a drive housing 12. The support rod 11 is arranged evenly around the base 6 and is rotatably mounted on the base 6. The axis of rotation 3 of the support rod 11 is perpendicular to the axis of the base 6. The drive housing 12 is screwed to the base 6. A return spring (not shown) is attached to the axis of rotation 3 of the support rod 11 to ensure that the rod is always in contact with the end of the drive housing 12. A non-slip rubber layer 16 is attached to the end of the support rod 11, away from its axis of rotation 3.

[0023] During the installation of the angle sensor 5, the angle sensor 5 is first placed between the gripping jaws 7. By rotating the worm, all three gripping jaws 7 are rotated simultaneously until the sensor is held in place. This ensures that the sensor is aligned with the base material 6 on the same axis. The angle sensor 5 remains stable under the action of the rubber seal 14 on the side edge of the gripping jaws 7. Both ends of the rotation axis of the blade 2 and the main axis 4 are equipped with mounting holes 13. By rotating the drive housing 12, the drive housing 12 is used to rotate all support rods 11, thereby securing and holding the ends of the support rods 11 against the inner wall of the mounting holes 13. This allows the base material 6 and the angle sensor 5 to be aligned with the rotation axis 3 on a common axis to obtain precise rotation angle data.

[0024] The above-mentioned are only the better specific embodiments of the invention, but the scope of protection of the invention is not limited to them, and technicians familiar with the invention should be replaced or changed within the scope of the technology disclosed in the invention in accordance with the technical program of the invention and its invention concepts, all falling within the scope of protection of the invention.

Claims

[1] A digital twin holography monitoring and self-healing system for wind turbines, characterized in that it comprises sequentially connected: a physical plane, including the body of the wind turbines (1), the multi-source sensor network deployed on it and the execution device; a data acquisition and transmission layer for acquiring and transmitting data from the aforementioned multi-source sensor network; Digital twin model layer for creating and running virtual twins that are synchronized with the physical entity layer, including geometric models, physical models, behavioral models, rule models, and data model fusion modules; Holographic monitoring and diagnostic layers for visualizing plant condition, health assessment, fault prediction, and diagnosis based on the digital twin layer; and A fault self-healing regulation layer serves to independently generate and execute regulation strategies based on diagnostic results, which are returned to the execution organization of the physical entity layer. [2] A digital twin holography monitoring and self-healing system for wind turbines according to claim 1, characterized in that the multi-source sensor network comprises at least three of the vibration, temperature, angle, voltage, acoustic sensors, video surveillance devices and SCADA system interfaces. [3] A digital twin holography monitoring and self-healing system for wind turbines according to claim 2, which characterized byThe behavioral model at the level of the digital twin model is trained on the historical operating data of the plant using machine learning algorithms in order to simulate the normal operating state of the plant at certain wind speeds and environmental conditions. [4] A digital twin holography monitoring and self-healing system for wind turbines according to claim 2, characterized bythe provision of an angle sensor (5) on the rotating disc of the blade (2) of the wind turbine (1), an angle sensor (5) at the end of the main axis (4) of the wind turbine (1), the angle sensor (5) via a fastening device which is coaxially attached to the end of the rotating disc / main axis (4) of the blade (2) and the fastening device comprising a retaining element and a coaxial positioning element which is used to fasten the angle sensor (5), and the coaxial positioning element being used to the end of the end which is coaxially connected to the rotating disc / main axis (4). [5] A digital twin holography monitoring and self-healing system for wind turbines according to claim 4, characterized by : The holding elements comprise a base (6), a gripping jaw (7) and a drive train (8). The base (6) is circular. The gripping jaws (7) are arranged in threes. The three gripping jaws (7) are arranged equally at the end of the base (6). The base (6) is fixed at its end by a rotary seat (17). The gripping jaws (7) are adjusted on the rotary seat (17). The turbine (9) is adjusted on the axis of rotation (3) of the gripping jaws (7). The drive train (8) is a worm gear. The drive train (8) is rotated in the base (6). The drive train (8) is adjusted coaxially with the base (6). The turbine (9) is engaged with the worm gear. The gripping jaws (7) are adjusted with a rubber seal (14) on the forward holding side. [6] A digital twin holography monitoring and self-healing system for wind turbines according to claim 5, characterized by : The coaxial positioning component comprises a support rod (11) and a drive shell (12), the support rod (11) is offset along the base (6) at a uniform distance, the support rod (11) rotates on the base (6), the axis of rotation (3) of the support rod (11) is perpendicular to the axis direction of the base (6), the drive shell (12) is offset on the base (6), the axis of rotation (3) of the support rod (11) is provided with a return spring and the support rod (11) is always in contact with the end of the drive shell (12) under the action of the return spring. [7] A digital twin holography monitoring and self-healing system for wind turbines according to claim 3, characterized byThe following: The function of fault prediction and diagnosis of the holographic monitoring and diagnostic layer is to perform residual analysis to detect anomalies early by comparing the real-time monitoring data of the physical entity layer with the predictive data of the behavioral model, and the holographic monitoring and diagnostic layer also includes a remaining lifetime prediction module for modeling and predicting performance losses of key components. [8] A digital twin holography monitoring and self-healing system for wind turbines according to claim 7, characterized by : one or more of the parametric adjustments, active power limits, dynamic optimization of the maintenance strategy and emergency and alarm signals relating to the automatic adjustment of the control reference values ​​for the variable paddle, the descent or the generator torque, which relates to the dynamic optimization strategy of the maintenance strategy, which is based on the forecast results of the remaining service life to automatically generate or update a preventive maintenance plan.

Images