Wind power plant operation and maintenance data interaction and use method

Abstract

The invention relates to a wind power plant operation and maintenance data interaction and use method, and aims to improve the operation and maintenance efficiency and equipment reliability of a wind power plant. According to the method, sensors are installed on wind power plant equipment, data such as wind speed, temperature and humidity, power output, vibration frequency and voltage are collected in real time, and the data are transmitted to a data fusion platform to be integrated and cleaned. The platform performs fault prediction and health assessment by adopting a machine learning algorithm, generates intelligent early warning in combination with equipment operation data, external meteorological information and power grid load, pushes the intelligent early warning to operation and maintenance personnel, identifies potential faults in advance, and optimizes a maintenance plan. Operation and maintenance personnel check real-time data, fault positions and maintenance steps through augmented reality equipment, field support is provided, and fault response efficiency and maintenance precision are improved. The wind power plant operation scheduling is optimized through big data analysis, the fan operation mode is automatically adjusted, the power generation efficiency is maximized, and the equipment fault and shutdown time is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of wind farm construction and operation and maintenance, and particularly to a method for data interaction and use in wind farm operation and maintenance. Background Art

[0002] With the continuous expansion of the scale of wind farms, wind power generation has gradually become an important renewable energy source. However, wind farm equipment faces frequent failures and maintenance requirements during long-term operation, which not only affects power generation efficiency but also may lead to high maintenance costs. Existing wind farm operation and maintenance management mostly relies on manual inspections and regular maintenance, with problems such as data islands, inability to monitor the health status of equipment in real time, and inaccurate fault warnings, resulting in difficult-to-predict faults and untimely responses, thus affecting the overall efficiency of wind farms.

[0003] In the prior art, although some wind farms have introduced intelligent monitoring systems, these systems generally have certain limitations in data interaction, fault prediction, and operation and maintenance optimization, and cannot achieve efficient dynamic adjustment and remote real-time support. Therefore, how to improve the operation efficiency and equipment reliability of wind farms through effective data interaction and intelligent operation and maintenance methods has become an urgent technical problem to be solved. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for data interaction and use in wind farm operation and maintenance. By integrating advanced data acquisition, intelligent analysis, fault prediction, and operation and maintenance support technologies, it solves problems such as data islands, lagging fault response, and low operation and maintenance efficiency in the prior art, optimizes the operation and maintenance management of wind farms, reduces the incidence of equipment failures, improves the operation stability of equipment, and reduces operation and maintenance costs.

[0005] A method for data interaction and use in wind farm operation and maintenance, characterized in that the method includes the following steps:

[0006] Step 1: Install sensors on each device in the wind farm to collect data such as wind speed, temperature and humidity, vibration frequency, power output, voltage, and current;

[0007] Step 2: Transmit the data to the data fusion platform in real time through either wireless communication or wired communication for data cleaning, deduplication, and integration;

[0008] Step 3: Based on the real-time data and historical data of the equipment, use machine learning algorithms to generate a fault prediction model and conduct a fault risk assessment;

[0009] Step 4: According to the fault prediction results, push warning information to the operation and maintenance personnel in real time to support the monitoring of the equipment health status and operation and maintenance decision-making;

[0010] Step 5: The operation and maintenance personnel view the real-time data, fault diagnosis, and maintenance suggestions of the equipment through augmented reality devices for on-site support.

[0011] Furthermore, the machine learning algorithm adopts any one of support vector machines, random forests, or decision trees to train a fault prediction model based on vibration, power output, and temperature data.

[0012] Furthermore, the fault warning signals are divided into three categories: green (normal), yellow (warning), and red (urgent), and are pushed to the operation and maintenance personnel through mobile devices or PC terminals.

[0013] Furthermore, the AR device is any one of an AR headset or smart glasses. The operation and maintenance personnel can view the running status, fault location, and maintenance steps of the equipment in real time through this device.

[0014] Furthermore, the method further includes combining external meteorological data and grid load data with equipment operation data, and using big data analysis to optimize the operation scheduling and maintenance strategies of the wind farm.

[0015] Furthermore, the data fusion platform conducts data interaction with the monitoring system and control system of the wind farm through a standardized interface to ensure the real-time and consistency of data.

[0016] Furthermore, the dynamic scheduling algorithm automatically adjusts the operation mode of the wind farm based on real-time wind speed, equipment health status, and load demand to maximize power generation efficiency.

[0017] Furthermore, the fault prediction model dynamically evaluates the health status of the equipment by analyzing the historical fault data of the equipment and combining real-time monitoring data, and adjusts the maintenance plan according to the health index.

[0018] Furthermore, the method also includes automatically generating and adjusting the operation and maintenance plan by calculating the health index of the equipment, giving priority to repairing equipment with poor health, and extending the equipment life.

[0019] A system for realizing the operation and maintenance data interaction and usage method of a wind farm

[0020] The system includes:

[0021] a. A data acquisition module for collecting real-time data of wind farm equipment;

[0022] b. A data fusion platform for receiving, cleaning, storing, and analyzing data;

[0023] c. A fault prediction module for generating a fault prediction model and pushing warning information;

[0024] d. An AR assistance module for providing real-time fault diagnosis and maintenance support for operation and maintenance personnel;

[0025] e. A scheduling optimization module for optimizing the operation scheduling of a wind farm according to the device health status and external conditions

[0026] Advantages of the present invention: By implementing technologies such as real-time data collection of wind farm devices, fault prediction and health assessment, intelligent early warning, augmented reality (AR) assisted operation and maintenance, and big data scheduling optimization, the present invention significantly improves the operation and maintenance efficiency and reliability of the wind farm. Through the optimization of intelligent fault prediction and maintenance plans, not only the device fault downtime is reduced, but also the service life of the device is extended, and the operation and maintenance costs are reduced. Operation and maintenance personnel can view the device status in real time through AR technology and obtain maintenance guidance, which improves the fault response speed and maintenance accuracy, and overall enhances the economic benefits and operation stability of the wind farm. Description of the Drawings

[0027] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0028] Figure 1 Schematic diagram of the wind farm operation and maintenance data collection and transmission process for the embodiment of the present invention;

[0029] Figure 2 Schematic diagram of the fault prediction and early warning system process for the embodiment of the present invention;

[0030] Figure 3 Schematic diagram of the augmented reality assisted operation and maintenance process for the embodiment of the present invention;

[0031] Figure 4 Schematic diagram of the structure of the wind farm data fusion and optimized scheduling system for the embodiment of the present invention;

[0032] Figure 5 Schematic diagram of the wind farm dynamic optimization scheduling process for the embodiment of the present invention. Detailed Embodiments

[0033] The present invention will be described in detail below in combination with the drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; moreover, the drawing part is only for more specifically describing the embodiments, and is not intended to specifically limit the present invention.

[0034] It should be noted that in the specification, references to "an embodiment", "embodiments", "exemplary embodiments", "some embodiments", etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not every embodiment necessarily includes such specific features, structures, or characteristics. Additionally, when describing a specific feature, structure, or characteristic in connection with an embodiment, implementing such feature, structure, or characteristic in connection with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0035] Generally, terms can be understood at least in part from their use in context. For example, at least in part depending on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or can be used to describe a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but rather can alternatively, at least in part depending on the context, allow for the existence of other factors that may not be explicitly described.

[0036] Embodiment 1

[0037] 1. Construction of the data fusion platform

[0038] Refer to Figure 1

[0039] 1.1 Data collection and transmission

[0040] Install sensors and monitoring devices at key equipment such as each fan, substation, and meteorological monitoring station to collect the following key data:

[0041] Wind speed (m / s)

[0042] Fan speed (rpm)

[0043] Power generation (kW)

[0044] Temperature (°C)

[0045] Vibration frequency (Hz)

[0046] Current, voltage (A, V)

[0047] Humidity, air pressure (%RH, Pa)

[0048] Transmit this data to the cloud data center in real time via optical fiber or 5G communication.

[0049] 1.2 Data integration and processing

[0050] After all the data is uploaded, a unified data format (JSON, CSV) is adopted, and the connection with the platform system is realized through the data access interface (API). All the original data will be cleaned (duplicate removal, missing value filling, noise removal, etc.) to ensure data accuracy.

[0051] 1.3 Data Storage and Database Management

[0052] Data storage structure: A distributed database (Hadoop HDFS) is used to store massive data. The historical data and real-time data of all devices are stored in different tables and indexed by device, time, and data type.

[0053] Data access: The data is accessed through a distributed query system (Presto or Apache Hive) to support multi-dimensional query analysis.

[0054] Data archiving: For historical data over one year old, the storage space is optimized through compressed storage and archiving.

[0055] 1.4 Visualization Interface

[0056] Develop a web-based real-time data monitoring interface to display information such as the overall status of the wind farm, the operation of each wind turbine, and equipment health data. The charts include:

[0057] Comparison chart of wind turbine power output and target power curve

[0058] Relationship chart of wind speed and power generation

[0059] Equipment health status (good, warning, failure)

[0060] Fault and early warning information

[0061] Refer to Figure 2

[0062] 2. Intelligent Early Warning and Dynamic Analysis System

[0063] 2.1 Data Analysis and Feature Extraction

[0064] Using statistical methods and signal processing techniques, characteristic indicators are extracted from the sensor data of the wind turbines:

[0065] Vibration acceleration of the wind turbine:

[0066] Temperature change rate:

[0067] Power output fluctuation:

[0068] 2.2 Fault Prediction Model

[0069] Machine learning model: Combine support vector machine (SVM), decision tree or random forest algorithm, and train a fault prediction model based on the historical data and characteristics of the device. For example, based on the vibration frequency and power output fluctuation of the fan, the model determines whether there is an impending bearing fault.

[0070] Training formula of the model (using SVM here):

[0071]

[0072] subject to:y i (w·x i +b)≥1-ξ i ,ξ i ≥0

[0073] Predicted output: The probability of device fault prediction, with the threshold set at over 80% as a high-risk fault.

[0074] 2.3 Real-time anomaly detection

[0075] Normalization analysis: Compare the real-time data with the historical operation data of the device and calculate the anomaly index:

[0076]

[0077] where X 当前 is the real-time data collected currently, X 历史 is the historical average data, and σ 历史 is the standard deviation of the historical data. If I 异常 >2, it is considered that the data is abnormal and a warning is triggered.

[0078] 2.4 Fault warning mechanism

[0079] Set three warning levels:

[0080] 1. Green (normal): The device indicators are within the normal range, I anomaly < 1;

[0081] 2. Yellow (warning): The device has a slight anomaly, 1 ≤ I anomaly ≤ 2;

[0082] 3. Red (urgent): The device has a serious anomaly, I anomaly > 2.

[0083] Refer to Figure 3

[0084] 3. AR-assisted operation and maintenance system

[0085] 3.1 AR device and software functions

[0086] Use AR devices for on-site operation and maintenance support. Operation and maintenance personnel can view the operation data of the equipment in real time through AR glasses, and the data is superimposed on the actual scene map of the equipment.

[0087] 3.2 Real-time Data Overlay and Equipment Status Identification

[0088] When operation and maintenance personnel are performing on-site repairs, the AR device displays the real-time operation data and health status of the equipment. For example, if the temperature of the fan is too high, the AR interface will display the comparison between the current temperature and the normal temperature range in a red warning box.

[0089] 3.3 Remote Collaboration and Problem Solving

[0090] Operation and maintenance personnel can start the remote collaboration mode through the AR device, share the on-site picture in real time with remote experts, and the experts guide the repair operation through voice and marked diagrams.

[0091] Technical Index Overlay: For the faulty parts of the fan (such as bearings, gears), operation and maintenance personnel can see the inspection parts and steps marked by virtual arrows through AR glasses.

[0092] 4. Big Data Analysis and Decision Support System

[0093] 4.1 Big Data Analysis Framework

[0094] Use a big data platform (Hadoop or Apache Spark) to deeply analyze the operation data of the wind farm and identify potential optimization spaces. Predict the long-term health status of the equipment through regression analysis:

[0095] P 预测 = β0 + β1X1 + β2X2 + … + β n X n

[0096] where P 预测 is the predicted performance of the fan, X i are the characteristics affecting the performance (such as wind speed, temperature, vibration frequency, etc.), and β i are the regression coefficients.

[0097] Refer to Figure 4

[0098] 4.2 Fan Optimization Scheduling

[0099] Scheduling optimization based on wind speed and load demand: Dynamically adjust the operation strategy of the wind farm through algorithms to maximize the power generation efficiency. For example, when the wind speed is low and the load demand is low, the operation speed of some fans can be reduced to reduce the power output and optimize the overall power generation efficiency.

[0100] Fan Power Adjustment Formula:

[0101] P adjusted = P actual × (1 - α)

[0102] where α is the adjustment coefficient (based on factors such as wind speed and load).

[0103] 4.3 Optimization of Wind Farm Maintenance Cycle

[0104] Optimization algorithm: Use genetic algorithm or simulated annealing algorithm to optimize the maintenance plan. For example, according to the health status and historical failure data of the wind turbines, dynamically adjust the inspection and maintenance cycles of each wind turbine.

[0105] Fitness function: The objective function for measuring the optimization of the maintenance cycle:

[0106]

[0107] Refer to Figure 5

[0108] 5. Adaptive Operation and Maintenance and Intelligent Optimization Scheduling

[0109] 5.1 Adaptive Operation and Maintenance Scheme

[0110] Equipment health assessment and scheduling adjustment: Automatically adjust the operation and maintenance plan according to the real-time health status of the equipment (such as vibration frequency, power output, etc.). Health assessment formula:

[0111]

[0112] where w i is the weight of each health parameter, and X i is the value of each health index.

[0113] 5.2 Fault Location and Repair Priority

[0114] Fault location based on data analysis: Through real-time data analysis, the system can automatically determine the location of equipment faults. For example, if a wind turbine shows a power drop and abnormal vibration, the system will determine that the fault location is the motor or gear and give priority to arranging repairs.

[0115] 5.3 Intelligent Scheduling System

[0116] According to the load demand, wind speed prediction, equipment health status, etc. of the wind farm, the intelligent scheduling system dynamically adjusts the operation mode of the wind turbines. For example, if the wind speed in a certain area is low and the equipment health status is good, the system will transfer the load to another part of the wind turbines with high efficiency operation.

[0117] This implementation combines multiple innovative technologies such as data fusion, big data analysis, intelligent early warning, AR assistance technology, and optimized scheduling, achieving the intelligence and automation of wind farm operation and maintenance. Through precise data collection, real-time analysis, fault prediction, and equipment optimized scheduling, the operation efficiency of the wind farm and the reliability of equipment are significantly improved.

[0118] The present invention covers any alternatives, modifications, equivalent methods, and solutions made on the essence and scope of the present invention. To enable the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, and those skilled in the art can fully understand the present invention without the description of these details. In addition, well-known methods, processes, flows, components, and circuits are not described in detail to avoid unnecessary confusion to the essence of the present invention.

[0119] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for wind farm operation and maintenance data interaction and use, characterized in that The method includes the following steps: Step 1, install sensors on each device in the wind farm to collect data such as wind speed, temperature and humidity, vibration frequency, power output, voltage, and current; Step 2, transmit the data to the data fusion platform in real time through either wireless communication or wired communication for data cleaning, duplicate removal, and integration; Step 3, generate a fault prediction model and conduct a fault risk assessment based on the real-time data and historical data of the device using machine learning algorithms; Step 4, according to the fault prediction results, push warning information to the operation and maintenance personnel in real time to support device health status monitoring and operation and maintenance decision-making; Step 5, the operation and maintenance personnel view the real-time data, fault diagnosis, and repair suggestions of the device through augmented reality devices for on-site support.

2. The method for wind farm operation and maintenance data interaction and use according to claim 1, wherein The machine learning algorithm adopts any one of support vector machines, random forests, or decision trees to train the fault prediction model based on vibration, power output, and temperature data.

3. The method for wind farm operation and maintenance data interaction and use according to claim 1, wherein, The fault warning signals are divided into three categories: green (normal), yellow (warning), and red (emergency), and are pushed to the operation and maintenance personnel through mobile devices or the PC side.

4. The method for wind farm operation and maintenance data interaction and use according to claim 1, characterized in that, The AR device is any one of an AR headset or smart glasses. The operation and maintenance personnel can view the operation status, fault location, and repair steps of the device in real time through this device.

5. The method for wind farm operation and maintenance data interaction and use according to claim 1, characterized in that The method further includes combining external meteorological data and grid load data with device operation data and using big data analysis to optimize the operation scheduling and maintenance strategies of the wind farm.

6. The method for wind farm operation and maintenance data interaction and usage according to claim 1, characterized in that The data fusion platform conducts data interaction with the monitoring system and control system of the wind farm through a standardized interface to ensure the real-time and consistency of the data.

7. The method for wind farm operation and maintenance data interaction and use according to claim 1, characterized in that The dynamic scheduling algorithm automatically adjusts the operation mode of the wind farm based on real-time wind speed, device health status, and load demand to maximize power generation efficiency.

8. The method for wind farm operation and maintenance data interaction and use according to claim 1, wherein The fault prediction model dynamically evaluates the health status of the device by analyzing the historical fault data of the device and combining real-time monitoring data, and adjusts the maintenance plan according to the health index.

9. The method for wind farm operation and maintenance data interaction and use according to claim 1, characterized in that The method also includes automatically generating and adjusting the operation and maintenance plan by calculating the health index of the device, giving priority to repairing devices with poor health, and extending the service life of the device.

10. A system for implementing the method according to any one of claims 1 to 9, characterized in that the system includes: a. A data acquisition module for collecting real-time data of wind farm devices; b. A data fusion platform for receiving, cleaning, storing, and analyzing data; c. A fault prediction module for generating a fault prediction model and pushing warning information; d. An AR assistance module for providing real-time fault diagnosis and repair support for operation and maintenance personnel; e. A scheduling optimization module for optimizing the operation scheduling of the wind farm according to the device health status and external conditions.

Images