Online monitoring and fault warning system for comprehensive health status of fans

Abstract

The present invention provides an online monitoring and fault warning system for the comprehensive health status of wind turbines, including a wind turbine data acquisition unit for collecting comprehensive data of wind turbine groups and uploading it to a wind power monitoring cloud platform; a wind power monitoring cloud platform for analyzing and processing the comprehensive data and managing the corresponding wind turbines; a data fusion control unit for providing data transmission and aggregation between wind turbine groups and the wind power monitoring cloud platform, the data fusion control unit including a centralized aggregator and several node controllers, the node controllers communicating with the centralized aggregator via an Ethernet wireless communication module; and several wind turbine groups for providing comprehensive data, each wind turbine group being equipped with one of the node controllers, and each of the wind turbine groups including several monitoring points. The present invention can realize local edge computing applications, and build a wind turbine health management cloud platform in combination with cloud-edge collaborative applications.

Description

Technical Field

[0001] The present invention relates to the technical field, and in particular to an online monitoring system for the comprehensive health status of a fan and a fault early warning system. Background Art

[0002] Since the wind turbine monitoring systems produced in the early stage were all independent systems, most of them did not have the ability to monitor the real-time status of the operation and evaluate the operating performance of the wind turbine, it was difficult to timely understand the real-time operating status of the unit and to provide accurate intelligent early warning of the operating health of the unit. The main problems currently faced by the construction of the health management system for in-service wind turbines are as follows:

[0003] 1. There are many monitoring subsystems: they are all isolated subsystems, the manufacturer interfaces are cumbersome and not unified, and the effective integration of data cannot be achieved; 2. There are many types of sensors, there is no standardized definition, and the interfaces are not unified, which is not conducive to installation and operation and maintenance; 3. There are many monitoring points and types of monitoring, and there is no standard guidance. The amount of data collected by each monitoring system is large, which occupies a large amount of network bandwidth. There is a lack of classification, calculation, and aggregation optimization of edge-side data, which brings a huge load to the platform; 4. When an unforeseen downtime occurs, the wind turbine power generation is seriously affected due to the inadequate supervision of wind turbine downtime, causing further damage to the internal equipment. The single maintenance means makes it difficult to detect faults early, and the maintenance method is relatively passive and preventive maintenance cannot be achieved, which is easy to cause accidents. The life safety of operators is threatened and serious economic losses are caused to the enterprise. In addition, there is a lack of standardized indicator models for data such as fault diagnosis, fault warning, and failure model. When analyzing faults, there is a lack of data on the unit's own operating status at the time of the fault. Only part of the data in the SCADA system can be used, which is not conducive to knowledge accumulation. The expert knowledge system cannot play a role, and it also directly causes the software platform to be unable to play a role.

[0004] Therefore, in order to better apply various monitoring data and optimize the wind turbine health management system, an intelligent collection device that is compatible with the data interface and communication protocol of various sensors and equipment and has edge computing functions is needed to maximize the collection, fusion and calculation of multiple signals on the wind turbine side. Summary of the invention

[0005] In view of this, the problem to be solved by the present invention is to provide an online monitoring system for the comprehensive health status of a wind turbine and a fault warning system.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a comprehensive health status online monitoring and fault warning system for a fan, comprising

[0007] Wind turbine data acquisition unit, used to collect comprehensive data of wind turbine groups and upload it to the wind power monitoring cloud platform;

[0008] A wind power monitoring cloud platform is used to analyze and process the comprehensive data and manage the corresponding wind turbines;

[0009] A data fusion control unit, used to provide data transmission and aggregation between the wind turbine group and the wind power monitoring cloud platform, the data fusion control unit includes a centralized aggregator and a plurality of node controllers, the node controllers communicate with the centralized aggregator through an Ethernet wireless communication module; and

[0010] A plurality of wind turbine groups are used to provide comprehensive data, each wind turbine group is equipped with a node controller, and each wind turbine group includes a plurality of monitoring points.

[0011] In the present invention, preferably, the centralized collector realizes data interconnection with the wind power monitoring cloud platform through any of optical fiber, 4G, data link, microwave communication or Beidou.

[0012] In the present invention, preferably, the wind turbine data acquisition unit includes an oil monitoring module, an equipment wear monitoring module, a vibration monitoring module, an inclination monitoring module, an arc monitoring module and a blade load detection module.

[0013] In the present invention, preferably, the vibration monitoring module is configured as a biaxial vibration sensor, the biaxial vibration sensor is installed inside the blade, and the distance between the biaxial vibration sensor and the blade root is one third of the length of the blade.

[0014] In the present invention, preferably, the inclination monitoring module is configured as a dual-axis inclination sensor, and the dual-axis inclination sensor is arranged at the tower base of the wind turbine.

[0015] In the present invention, preferably, the blade load detection module comprises a stress monitoring sensor, and the stress monitoring sensor is arranged at the blade root to detect the bending moment in the swing direction and radial direction of the fan blade root.

[0016] In the present invention, preferably, the monitoring points include common parts, wind rotor system, nacelle tower, yaw system, gearbox, generator, converter, hydraulic and brake system, wind turbine transformer, alarm code, fault code, set value, control command and safety chain.

[0017] In the present invention, preferably, the wind power monitoring cloud platform includes a preprocessing module, an analysis module, a fault warning module, a work order management module and a report management module. The preprocessing module is used to screen and denoise the data according to the uploaded comprehensive data using an information classification strategy. The analysis module is configured with an association rule strategy. The analysis module obtains fault point information according to the association rule strategy. The fault warning module is used to indicate the fault point status information of the wind turbine group according to the fault classification decision algorithm to achieve the purpose of monitoring the wind turbine group. The report management module is used to comprehensively summarize the operating status and fault points of the wind turbine group and generate report data.

[0018] In the present invention, preferably, the information classification strategy is specifically a feature selection algorithm, and the corresponding monitoring point position of the wind turbine group is judged by information classification to obtain the corresponding fault point.

[0019] In the present invention, preferably, the fault classification decision algorithm adopts any one of the nearest neighbor algorithm, K-nearest neighbor algorithm, decision tree, rough set algorithm or Bayesian classification method.

[0020] The advantages and positive effects of the present invention are:

[0021] (1) Through the cooperation between the wind turbine data acquisition unit, wind power monitoring cloud platform, data fusion control unit and wind turbine group, the data is collected through the data fusion control unit, and local edge computing applications are realized. Combined with the application of cloud-edge collaboration, a wind turbine health management cloud platform is built to achieve integrated online monitoring and data analysis of the vibration, settlement, arcing, gearbox oil and load of the wind turbine's impeller, tower and foundation, effectively discover hidden faults, determine the causes of blade damage, tower vibration and foundation hidden dangers, evaluate its fatigue life and safety, provide load basis for the optimization of wind turbine control strategy, and verify the optimization control effect.

[0022] (2) As the comprehensive data of wind turbines is complex and huge, the amount of fault alarm information data is correspondingly large. A large amount of alarm information will be generated in a short period of time. Too frequent warnings will cause operators to be unable to obtain truly effective information. By setting up a wind power monitoring cloud platform including a preprocessing module, a research and analysis module, a fault warning module, a work order management module and a report management module, the preprocessing module can obtain the correlation between the actual fault of the wind turbine and the preset parameters, and then obtain the correlation between the faults of the wind turbine, and finally generate query and trend analysis functions including tower, tower base, blades and other monitoring report data through the report management module. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0024] Figure 1 It is a schematic diagram of the overall structure of the wind turbine comprehensive health status online monitoring and fault early warning system of the present invention;

[0025] Figure 2 It is a schematic diagram of data transmission between the data fusion control unit and the wind power monitoring cloud platform of the wind turbine comprehensive health status online monitoring and fault warning system of the present invention;

[0026] Figure 3 It is a schematic diagram of the collection structure of the oil monitoring module of the fan comprehensive health status online monitoring and fault warning system of the present invention;

[0027] Figure 4 It is a schematic diagram of the installation position of the vibration monitoring module of the online monitoring and fault early warning system for the comprehensive health status of the fan of the present invention;

[0028] Figure 5 It is a schematic diagram of the real-time sway angles in each direction of the tower top sway monitoring by the inclination monitoring module of the wind turbine comprehensive health status online monitoring and fault warning system of the present invention, as well as the real-time value of the displacement value;

[0029] Figure 6 It is a schematic diagram of the real-time settlement angles and settlement displacement values ​​of the inclination monitoring module of the wind turbine comprehensive health status online monitoring and fault warning system of the present invention for monitoring the tower foundation shaking in all directions;

[0030] Figure 7 It is a schematic diagram of the installation of the inclination monitoring module of the wind turbine comprehensive health status online monitoring and fault warning system of the present invention;

[0031] Figure 8 It is a spectrum simulation diagram of a blade load detection module of the wind turbine comprehensive health status online monitoring and fault early warning system of the present invention;

[0032] Fig. 9 It is a blade waveform diagram of a blade load detection module of the wind turbine comprehensive health status online monitoring and fault early warning system of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] It should be noted that when a component is referred to as being "fixed to" another component, it may be directly on the other component or there may also be a component centered. When a component is considered to be "connected to" another component, it may be directly connected to the other component or there may also be a component centered. When a component is considered to be "set on" another component, it may be directly set on the other component or there may also be a component centered. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0036] like Figure 1 and Figure 2 As shown, the present invention provides an online monitoring system for the comprehensive health status of a fan and a fault warning system, comprising

[0037] Wind turbine data acquisition unit, used to collect comprehensive data of wind turbine groups and upload it to the wind power monitoring cloud platform;

[0038] Wind power monitoring cloud platform, used to analyze and process comprehensive data and manage corresponding wind turbines;

[0039] A data fusion control unit is used to provide data transmission and aggregation between the wind turbine group and the wind power monitoring cloud platform. The data fusion control unit includes a centralized aggregator and several node controllers. The node controllers communicate with the centralized aggregator through an Ethernet wireless communication module; and

[0040] A plurality of wind turbine groups are used to provide comprehensive data, each wind turbine group is equipped with a node controller, and each wind turbine group includes a plurality of monitoring points.

[0041] In this embodiment, further, the centralized collector realizes data interconnection with the wind power monitoring cloud platform through any method such as optical fiber, 4G, data link, microwave communication or Beidou.

[0042] In this embodiment, further, the wind turbine data acquisition unit includes an oil monitoring module, an equipment wear monitoring module, a vibration monitoring module, an inclination monitoring module, an arc monitoring module and a blade load detection module.

[0043] In this embodiment, further, the vibration monitoring module is configured as a biaxial vibration sensor, the biaxial vibration sensor is installed inside the blade, and the distance between the biaxial vibration sensor and the blade root is one third of the length of the blade.

[0044] In this embodiment, further, the inclination monitoring module is configured as a dual-axis inclination sensor, and the dual-axis inclination sensor is arranged at the tower base of the wind turbine.

[0045] In this embodiment, further, the blade load detection module includes a stress monitoring sensor, which is arranged at the blade root and is used to detect the bending moment in the swing direction and radial direction of the fan blade root.

[0046] In this embodiment, further, the monitoring points include common parts, wind rotor system, nacelle tower, yaw system, gearbox, generator, converter, hydraulic and brake system, wind turbine transformer, alarm code, fault code, set value, control command and safety chain.

[0047] In this embodiment, further, the wind power monitoring cloud platform includes a preprocessing module, an analysis module, a fault warning module, a work order management module and a report management module. The preprocessing module is used to screen and denoise the data according to the uploaded comprehensive data using an information classification strategy. The analysis module is configured with an association rule strategy. The analysis module obtains fault point information according to the association rule strategy. The fault warning module is used to indicate the fault point status information of the wind turbine group according to the fault classification decision algorithm to achieve the monitoring purpose of the wind turbine group. The report management module is used to comprehensively summarize the operating status and fault points of the wind turbine group and generate report data.

[0048] In this embodiment, further, the information classification strategy is specifically a feature selection algorithm, which determines the corresponding monitoring point position of the wind turbine group through information classification to obtain the corresponding fault point.

[0049] In this embodiment, further, the fault classification decision algorithm adopts any one of the nearest neighbor algorithm, K-nearest neighbor algorithm, decision tree, rough set algorithm or Bayesian classification method.

[0050] The fan data acquisition unit includes an oil monitoring module, an equipment wear monitoring module, a vibration monitoring module, an inclination monitoring module, an arc monitoring module and a blade load detection module.

[0051] The oil monitoring module is an integrated online monitoring sensor for lubricant quality and wear. It is connected to the oil circuit of the gearbox lubrication system by bypassing the oil. After testing indicators such as gearbox wear particles, lubricant viscosity, moisture, and dielectric constant, each parameter is transmitted to the collector through an RS485 interface for easy external collection. The oil monitoring module includes a viscosity sensor, a moisture sensor, a dielectric constant sensor, and a wear particle sensor. For details, please refer to the attached Figure 3 Among them, the viscosity sensor uses fluid vibration sensing technology to monitor the viscosity changes of gear oil and monitors the 4K-11KHz frequency band. The system is equipped with a PT1000 temperature sensor for temperature acquisition. The temperature signal corrects the calibration curve in the MCU, thereby improving the viscosity detection accuracy; the moisture sensor uses polymer film capacitor sensing technology to monitor the changes in trace moisture in gear oil. The convenient local calibration function can adapt to various harsh use environments; the dielectric constant sensor is used to detect the dielectric constant value of various non-conductive liquids such as oil. The dielectric constant can comprehensively reflect the quality performance of lubricating oil and other oils, and reflect the effects of water ingress, acidification, oxidation, and additive failure. The wear particle sensor uses abrasive particle detection technology to monitor the changes of wear particles in the oil. The size and quantity of metal abrasive particles generated in different wear stages are also different. The system classifies the abrasive particles according to the size, and divides them into coarse abrasive particles and fine abrasive particles. The smallest ferromagnetic particles that can be monitored are 1μm in size. By analyzing the size and quantity growth trend of the abrasive particles, the state of the working surface can be judged, and the wear of the friction pair can be discovered in time, so that the equipment can be inspected and repaired in a targeted manner.

[0052] The equipment wear monitoring module uses a wear monitoring sensor to monitor the operating status of the equipment. It uses the principle of electromagnetic induction to generate an opposite magnetic field in the sensor oil circuit. When the oil passes through the section, the magnetic field disturbance generated by the metal wear debris in the oil can be accurately identified to generate an induced voltage in the center sensor coil. The amplitude and phase of the output signal are used to identify the size type of the particles. The abrasive particles in the lubricating oil are an important information carrier about the wear state. By monitoring, measuring and analyzing the wear particles generated by the mechanical device, useful information such as the state of the working surface and whether it can withstand further wear can be obtained. The characteristic size of the abrasive particles at the failure period of most machines is mostly between 20um and 200um. This information is an important basis for giving full play to the working life of the wear parts.

[0053] The vibration monitoring module includes a composite dual-axis vibration sensor installed inside the blade at one-third of the distance from the blade root. Figure 4As shown in the figure, the dual-axis vibration sensor is fixed to the blade with 3M glue and uses an S-shaped wiring method, which is more secure. It is used to monitor the vibration state of the blade, such as the swing array and torsion. By analyzing the vibration signal, faults that cause changes in the aerodynamic characteristics of the blade, such as blade icing, pitch angle deviation or blade tearing, are found. By analyzing the changes in the blade's natural frequency, the damage that causes the change in blade stiffness is found. The expert system and prior data are used to display the analysis results in multiple dimensions. The analysis method uses multi-bar value diagrams, spectrum waterfall diagrams, waveform spectrum analysis, and envelope diagram analysis. The vibration analysis states include normal, misalignment, eccentricity, pitting, and wear.

[0054] The tilt monitoring module uses a dual-axis tilt sensor, as shown in the attached Figure 5 , Figure 6 and Figure 7 As shown, it is used to monitor the dynamic form of the tower, sway and foundation settlement monitoring based on a high-precision dual-axis inclination sensor, which is arranged at the tower base of the wind turbine. The dual-axis inclination sensor needs to be parallel to the measurement platform during measurement, and the dual axes of the sensor need to be perpendicular to each other. It is used to monitor the uneven settlement amount and settlement direction of the wind turbine foundation. The collected signal is directly transmitted to the data fusion controller through the communication cable. By comparing and analyzing the sensor data at the bottom and top of the tower, it can be found that the tower is unevenly stressed and deformed locally. When the uneven settlement value of the foundation of the unit exceeds the standard, an alarm can be given in time to avoid damage to the main structure of the wind turbine due to uneven settlement of the foundation, ensure the safe and stable operation of the wind turbine unit, and escort the safety of the tower.

[0055] Harsh working environment makes electrical insulation layer very easy to age. Harsh working environment easily forms contamination and conductive film between phases or between phases and ground. Long-term vibration of fans causes bolts at the connection of electrical equipment to loosen. The running current of generator sets is large and the equipment structure is compact, which easily causes arc. The materials of components do not meet the relevant national standards. Long-term high temperature overheating causes the continuous strength and physical properties of electrical equipment to decrease. The above reasons may cause electrical failures, so it is necessary to monitor the arc when a failure occurs. By setting an arc monitoring module, the specific arc monitoring module can use arc unit SAQEA12 or arc temperature unit SAQEAP3 0-T, the arc unit SAQEA12 can receive 12 arc sensors and interconnect through the RJ45 port, trip within 10ms, and support RS485 communication; the arc temperature unit SAQEAP30-T can receive 2 arc sensors, collect 3*4 wireless temperature sensor signals, and the trip logic can be set to "arc" or "temperature" or "arc + temperature" mode, supporting RS485 communication, so that different monitoring modules can be installed on the generator, inverter, grid cabinet, control cabinet, power module cabinet, and fan bottom box substation to protect the safe, stable and reliable operation of the fan.

[0056] The blade load detection module consists of stress monitoring sensors, front-end control cabinets, etc., and uses wired power supply. Figure 8 and attached Fig. 9 , 4 stress sensors are installed at the root of each blade. The stress sensors complete the stress monitoring of the blade root, monitor and evaluate the static and dynamic loads of the blade root. In addition, a signal acquisition instrument is installed in the hub to collect and preprocess the signal, communicate with the cabin through the slip ring (or wireless communication module), and transmit the signal to the data fusion edge controller in the cabin. By monitoring the strain of the blade root under load, combined with the force characteristics and stress-strain state analysis of the blade root, the blade load state of the wind turbine under various operating conditions is obtained, and the data can be used for accurate independent pitch and yaw control. The data is transmitted to the front-end control cabinet by wire, and the data is collected to the data fusion controller through a dedicated Ethernet wireless communication module to realize the communication between the cabin data fusion controller and the front-end control cabinet in the hub. A single wind turbine requires a total of 12 stress sensors, 1 front-end control cabinet, and 2 wireless communication modules.

[0057] By establishing monitoring modules for wind turbine impeller status, wind turbine load, tower sway, foundation settlement, and arc protection, and developing corresponding numerical calculation models, fault diagnosis and early warning models, the data is collected through the data fusion control unit, and local edge computing applications are realized. Combined with cloud-edge collaborative applications, a wind turbine health management cloud platform is built to achieve integrated online monitoring and data analysis of the vibration, settlement, arc, gearbox oil, and wind turbine load of the wind turbine impeller, tower, and foundation, effectively discover hidden faults, determine the causes of blade damage, tower vibration, and foundation hidden dangers, evaluate their fatigue life and safety, provide load basis for wind turbine control strategy optimization, and verify the optimization control effect. Carry out health assessment of in-service wind turbines in wind farms, provide technical support for fault analysis and safety early warning, realize intelligent operation and maintenance, and improve economic benefits.

[0058] The embodiments of the present invention are described in detail above, but the contents described are only preferred embodiments of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of this patent.

Claims

1. The wind turbine comprehensive health status online monitoring and fault warning system is characterized by: include Wind turbine data acquisition unit, used to collect comprehensive data of wind turbine groups and upload it to the wind power monitoring cloud platform; A wind power monitoring cloud platform is used to analyze and process the comprehensive data and manage the corresponding wind turbines; A data fusion control unit is used to provide data transmission and aggregation between the wind turbine group and the wind power monitoring cloud platform. The data fusion control unit includes a centralized aggregator and a plurality of node controllers. The node controllers communicate with the centralized aggregator through an Ethernet wireless communication module. as well as A plurality of wind turbine groups for providing comprehensive data, each wind turbine group is equipped with a node controller, and each wind turbine group includes a plurality of monitoring points; The fan data acquisition unit includes an oil monitoring module, an equipment wear monitoring module, a vibration monitoring module, an inclination monitoring module, an arc monitoring module and a blade load detection module; The vibration monitoring module is configured as a biaxial vibration sensor, which is installed inside the blade, and the distance between the biaxial vibration sensor and the blade root is one third of the length of the blade; The wind power monitoring cloud platform includes a preprocessing module, an analysis module, a fault warning module, a work order management module and a report management module. The preprocessing module is used to screen and denoise the data according to the uploaded comprehensive data using an information classification strategy. The analysis module is configured with an association rule strategy. The analysis module obtains fault point information according to the association rule strategy. The fault warning module is used to indicate the fault point status information of the wind turbine group according to the fault classification decision algorithm to achieve the monitoring purpose of the wind turbine group. The report management module is used to comprehensively summarize the operating status and fault points of the wind turbine group and generate report data.

2. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The centralized collector realizes data interconnection with the wind power monitoring cloud platform through any method such as optical fiber, 4G, data link, microwave communication or Beidou.

3. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The inclination monitoring module is configured as a dual-axis inclination sensor, and the dual-axis inclination sensor is arranged at the tower base of the wind turbine.

4. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The blade load detection module includes a stress monitoring sensor, which is arranged at the blade root and is used to detect the bending moment in the swing direction and radial direction of the fan blade root.

5. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The monitoring points include common parts, wind rotor system, nacelle tower, yaw system, gearbox, generator, converter, hydraulic and brake system, wind turbine transformer, alarm code, fault code, set value, control command and safety chain.

6. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The information classification strategy is specifically a feature selection algorithm, which determines the corresponding monitoring point position of the wind turbine group through information classification to obtain the corresponding fault point.

7. The wind turbine comprehensive health status online monitoring and fault warning system according to claim 1 is characterized in that: The fault classification decision algorithm adopts any one of the nearest neighbor algorithm, K-nearest neighbor algorithm, decision tree, rough set algorithm or Bayesian classification method.

Images