Multi-source data integration method and system for wind power load test system

Abstract

The invention discloses a multi-source data integration method and system for a wind power load test system, and relates to the technical field of data processing. The method comprises the following steps: determining N load scenes; traversing N load scenes, and constructing a Markov load network; a data integration module is developed in a wind power load test system, supervised training is performed by taking a Markov load network as a benchmark, and direct connection between the data integration module and a data interface of a distributed sensor array is established; the method comprises the following steps: generating a test target through a wind power load test system, controlling a distributed sensing array to perform load test sampling according to a difference frequency sampler configured at the front end, and determining multi-source sampling data; according to the data interface, multi-source sampling data is transmitted back, data integration processing based on load scene transition is carried out in combination with the data integration module, and load integration data is determined and stored. The technical problem of low multi-source data integration efficiency of a wind power load test system in the prior art is solved, and the technical effect of improving the data integration efficiency is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a multi-source data integration method and system for a wind power load test system. Background Art

[0002] During the operation and maintenance of wind turbines, load testing is a key link in evaluating their structural reliability and operational safety. Existing wind power load testing systems usually rely on distributed sensors to collect load data from different parts of the unit. However, due to the complex structure of wind turbines and the changing load conditions, the data collected by different sensors have problems such as time and space asynchrony and inconsistent formats, which makes it difficult to integrate multi-source data. In addition, the existing data processing methods lack effective modeling of the dynamic changes of load scenarios, making it difficult to accurately reflect the load response characteristics of the unit under different working conditions, limiting the in-depth analysis and application of test data. Especially in complex wind environments, the load changes of wind turbines are not only related to external factors such as wind speed and wind direction, but also closely related to the control posture and operation mode of the unit. Traditional data processing methods are difficult to fully capture the dynamic correlation between these multi-dimensional factors, resulting in insufficient reliability and accuracy of test results. Summary of the invention

[0003] The present application provides a multi-source data integration method and system for a wind power load test system, which solves the technical problem of low efficiency of multi-source data integration of a wind power load test system in the prior art.

[0004] In view of the above problems, the present application provides a multi-source data integration method and system for a wind power load test system.

[0005] In a first aspect of the present application, a multi-source data integration method for a wind power load test system is provided, the method comprising: For wind turbines, N load scenarios are determined according to the key load-unit control posture; the N load scenarios are traversed to construct a Markov load network, wherein the load scenarios are nodes and the connection relationship is the integration mode transition based on the scene transition between nodes; a data integration module is developed in the wind power load test system, and supervised training is performed based on the Markov load network, and a data interface direct connection between the data integration module and the distributed sensor array is established; a test target is generated through the wind power load test system, and the distributed sensor array is controlled to perform load test sampling according to the difference frequency sampler configured at the front end, and multi-source sampling data is determined; according to the data interface, the multi-source sampling data is transmitted back and combined with the data integration module to perform data integration processing based on the load scenario transition, and the load integration data is determined and stored, wherein the transition analysis is performed based on the upper scene node.

[0006] A second aspect of the present application provides a multi-source data integration system for a wind power load test system, the system comprising: A scenario determination unit is used to determine N load scenarios for a wind turbine according to the load of key parts and the control posture of the unit; a load network construction unit is used to traverse the N load scenarios and construct a Markov load network, wherein the load scenarios are nodes and the connection relationship is based on the integration mode transition between the nodes; a training unit is used to develop a data integration module in the wind power load test system, perform supervised training based on the Markov load network, and establish a data interface direct connection between the data integration module and the distributed sensor array; a sampling unit is used to generate a test target through the wind power load test system, control the distributed sensor array to perform load test sampling according to the difference frequency sampler configured at the front end, and determine multi-source sampling data; a data processing unit is used to transmit the multi-source sampling data back according to the data interface and perform data integration processing based on the load scenario transition in combination with the data integration module, determine the load integration data and store it, wherein the transition analysis is performed based on the upper scene node.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, for the wind turbine, according to the key load-unit control posture, N load scenarios are determined. Then, the N load scenarios are traversed to construct a Markov load network, in which the load scenario is used as a node and the integration mode based on the scene transition between nodes is used as the connection relationship. Further, a data integration module is developed in the wind power load test system, and supervised training is performed based on the Markov load network, and a data interface direct connection between the data integration module and the distributed sensor array is established. Then, the test target is generated through the wind power load test system, and the distributed sensor array is controlled to perform load test sampling according to the difference frequency sampler configured at the front end to determine the multi-source sampling data. Finally, according to the data interface, the multi-source sampling data is transmitted back and combined with the data integration module to perform data integration processing based on the load scenario transition, and the load integration data is determined and stored, wherein the transition analysis is performed based on the upper scene node. The technical problem of low efficiency of multi-source data integration of the wind power load test system in the prior art is solved, and the technical effect of improving data integration efficiency is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1A schematic flow chart of a multi-source data integration method for a wind power load test system provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a multi-source data integration system for a wind power load test system provided in an embodiment of the present application.

[0010] Explanation of the reference numerals: scenario determination unit 11 , load network construction unit 12 , training unit 13 , sampling unit 14 , data processing unit 15 . DETAILED DESCRIPTION

[0011] The present application solves the technical problem of low efficiency of multi-source data integration of wind power load test system in the prior art by providing a multi-source data integration method and system for wind power load test system.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a multi-source data integration method for a wind power load test system, wherein the method includes: For wind turbines, N load scenarios are determined based on the loads at key locations and the control attitude of the turbine.

[0015] In wind turbines, different operating conditions (such as wind speed, wind direction, blade angle, unit attitude, etc.) will have a significant impact on the load distribution of key parts (such as blades, hubs, towers, main shafts, gearboxes, etc.). Therefore, in order to comprehensively analyze the load characteristics of the unit under different working conditions, it is necessary to determine N representative load scenarios based on the loads of key parts and the control attitude of the unit. Specifically, firstly, the load data of key parts of the wind turbine are collected through a distributed sensor array, including blade bending moment, hub torque, tower axial force and gearbox shear force, and the control attitude parameters of the unit, such as yaw angle, pitch angle, speed, etc., are obtained synchronously; the collected data are preprocessed, outliers are removed and time synchronization is performed to ensure the accuracy and consistency of the data; then, a multidimensional feature vector is constructed based on the relationship between load and control attitude, and the load parameters are combined with attitude parameters to form a complete data set; then, a clustering algorithm (such as K-means or DBSCAN) is used to perform cluster analysis on the multidimensional data, and N load scenarios are automatically divided, covering typical working conditions as shown in Table 1, including low wind speed partial load conditions, medium wind speed high load conditions, high wind speed extreme yaw conditions, brake shutdown conditions, and wind shear rapid response conditions.

[0016] Table 1 (typical load conditions): .

[0017] The N load scenarios are traversed to construct a Markov load network, wherein the load scenarios are nodes and the connection relationships are formed by integrating the changes in the scenarios between the nodes.

[0018] By traversing N load scenarios, a Markov load network is constructed. Specifically, each load scenario of the wind turbine is defined as a network node, and each load scenario represents the operating state of the unit under specific working conditions; based on N load scenarios, a network graph containing N nodes is constructed, and the connection relationship between nodes is based on the transition between scenarios, that is, the transition between load scenarios forms the edges connecting these nodes, where the transition of each scenario can be caused by changes in the external environment (such as wind speed, wind direction, etc.), unit control strategy (such as yaw angle, pitch angle adjustment) or unit operating state (such as fault, start and stop), for example, transitioning from low wind speed state to medium wind speed state, or switching from normal operation state to brake stop state.

[0019] Furthermore, a Markov load network is constructed, including: The communication protocol and data format of interactive multi-source data are used to set a unified data standard, wherein the unified data standard is any communication protocol and data format; the unified data standard is used as a universal integration standard, the N load scenarios are traversed, and the transition analysis between scenario groups and the integration mode transition analysis are performed respectively to construct the Markov load network.

[0020] In the process of building a Markov load network, it is necessary to first define a unified data standard to ensure that multi-source data can communicate and integrate smoothly. This unified data standard involves the communication protocol and data format of the data, ensuring that the data generated by sensors, control systems, simulation models and other devices from different sources can be exchanged and processed in a consistent manner. Specifically, common communication protocols (such as Modbus, CAN, etc.) and standardized data formats (such as JSON, XML, CSV, Protocol Buffers, etc.) can be used. Based on the unified data standard, N load scenarios are traversed to perform transition analysis between scenarios. Each load scenario represents the state of the unit under specific working conditions, and the transition between different scenarios reflects the influence of the unit control strategy, external wind force changes or other factors. For example, factors such as wind speed changes and blade angle adjustments may cause the wind turbine to transfer from one load scenario to another. By analyzing historical data or processing real-time data, the system can identify the transfer rules between these load scenarios and form a transfer matrix to represent the change trend and transition probability between scenarios. Based on the load scenario transition, it is necessary to further analyze the integration mode transition. Integration mode transition analysis aims to dynamically integrate multi-source data to ensure that the data can accurately reflect the changes in load characteristics when the state transitions between different scenarios. On the basis of load scenario transitions, integration mode transition analysis needs to be further performed. Integration mode transition analysis aims to dynamically integrate multi-source data to ensure that the data can accurately reflect the changes in load characteristics when the state transitions between different scenarios. The selection of integration methods can be based on a variety of methods, such as weighted average method, interpolation method or more advanced machine learning methods, to ensure that the load characteristics of the unit can be accurately reflected in different scenarios. Finally, based on the transition analysis between scenarios and the integration mode transition analysis, a Markov load network is constructed. The nodes of the Markov load network represent different load scenarios, and the connection relationship between nodes is determined by the transition probability between scenarios. These connection relationships not only reflect the dynamic transformation between different load scenarios, but also improve the accuracy and efficiency of state transition between scenarios through data integration analysis. The weight of each edge of the network is the transition probability between scenarios, reflecting the possibility of transitioning from one load scenario to another.

[0021] Furthermore, the change analysis between scenario groups and the change analysis of integration methods are carried out separately, including: Taking the unified data standard as the universal integration standard, starting with distributed source data, and guided by data requirements based on test objectives, the data integration method of the first load scenario is determined by introducing integration constraints, wherein the data integration method at least includes unified integration-differential relaxation integration-key management integration; for the first load scenario and the remaining N-1 load scenarios, the scenario change characteristics are identified and the integration change characteristics based on the data integration method are mined, and the first load network is determined by integration.

[0022] Specifically, unified data standards are used as universal integration standards to ensure that multi-source data can interact efficiently between different systems and devices. Unified data standards refer to the use of consistent protocols and structures in data communication and format processing, such as standardized data formats such as JSON and XML, and communication protocols such as CAN, ETH, modbus, and fiber optic communications, to ensure that data remains consistent during collection, transmission, and storage. Data integration starts with distributed source data. Distributed source data refers to multi-source sensor data from different parts of wind turbines, such as blade stress sensors, main shaft torque sensors, tower vibration sensors, etc. These data have diversity in time, space, and physical quantities.

[0023] In the process of data integration, data requirements are determined based on the test objectives of the wind power load test system. The test objectives may include load response analysis of wind turbines under specific wind speeds, loads or control strategies. Based on this data requirement, integration constraints are introduced. Integration constraints refer to specific requirements that need to be met during data processing and analysis, such as data consistency, real-time, integrity and security, to ensure that the integrated data has high reliability and validity. On this basis, the data integration method for the first load scenario is determined. Data integration methods include at least the following three methods: unified integration, differential relaxation integration and key management integration. Unified integration refers to standardizing data from different sensors or systems to solve problems such as inconsistent data formats and different sampling frequencies. For example, align the time series data of different sensors to ensure comparison and analysis on the same timeline. Differential relaxation integration is a dynamic adjustment strategy based on data differences. It is mainly used to deal with subtle differences between multi-source data. The sensitivity of data fusion is adjusted by introducing relaxation factors to adapt to dynamic changes in different load scenarios. For example, when the wind speed changes drastically, differential relaxation integration can automatically relax the difference tolerance of certain data to avoid data anomalies caused by instantaneous fluctuations. Key management integration focuses on the security of data transmission and storage. It uses key encryption technology to ensure the confidentiality and integrity of data during transmission and prevent data from being illegally tampered with or leaked. It is especially suitable for scenarios involving remote data transmission or distributed systems.

[0024] After determining the data integration method for the first load scenario, further analyze the transition relationship between this scenario and the remaining N-1 load scenarios. Scenario transition characteristics refer to the key factors that cause a wind turbine to transfer from one operating state to another. These characteristics may include drastic changes in wind speed, adjustment of the yaw angle of the unit, changes in the pitch angle of the blades, etc. For example, a sudden increase in wind speed from 8m / s to 15m / s may cause the unit to switch from a partial load state to a full load state. This change belongs to the scenario transition feature. On the basis of identifying these features, further explore the integration transition features based on the data integration method. The integration transition features refer to how the data integration method is adjusted to adapt to new scenario requirements during the conversion of different load scenarios. For example, when the wind speed changes drastically, the system may switch from a standard unified integration mode to a more flexible differential relaxation integration mode to better cope with the uncertainty caused by data fluctuations.

[0025] Finally, by integrating the scene change features and the integrated change features, the system determines the structure of the first load network. The first load network is a directed graph network based on the Markov model. The nodes in the network represent different load scenarios, the connection relationship between the nodes represents the change path between the scenarios, and the connection weight represents the transition probability between the scenarios. In this network, the scene change is not only based on the physical state change of the wind turbine, but also combined with the dynamic adjustment of the data integration method to ensure that the data processing strategy can flexibly adapt to the scene changes.

[0026] Furthermore, the data integration method at least includes unified integration-differential relaxation integration-key management integration, including: Guided by the test goal, the differential relaxation is determined, wherein the differential relaxation is determined based on the data value; based on the differential relaxation, the standard multi-source data determined after the unified integration is differentially transformed to determine the valid data; the valid data is subjected to sensitivity determination, the sensitive data portion is located, and key encryption is performed.

[0027] Specifically, the differential relaxation is determined based on the test objectives. The test objectives usually include load response analysis of wind turbines under specific working conditions, structural safety assessment under extreme environments, etc. Differential relaxation is a parameter that measures the tolerance range of data differences and is used to adjust the sensitivity to differences during data integration. The differential relaxation is determined based on data value. Data value refers to the importance of data in achieving test objectives, which is usually quantified by indicators such as data contribution, change frequency, and anomaly detection probability. For example, for key load data (such as spindle torque or tower stress data), the differential relaxation will be set lower to improve the accuracy of data integration; while for auxiliary data (such as ambient temperature or humidity data), the differential relaxation can be appropriately relaxed to improve processing efficiency. After determining the differential relaxation, it is used as a benchmark to perform differential conversion processing on the standard multi-source data after unified integration to screen out effective data with practical significance. The purpose of unified integration is to standardize data from different sensors, solve problems such as inconsistent formats and different sampling frequencies, and ensure that the data has a unified structure and timestamp. The differential conversion process dynamically adjusts the data based on the differential relaxation. It mainly calculates the difference between the data points and compares them with the relaxation threshold, filters out invalid data with fluctuations below the relaxation threshold, and retains only data with significant changes or key features. This process can not only effectively reduce data redundancy and improve data processing efficiency, but also highlight the characteristics of data changes under critical working conditions. After obtaining valid data, further sensitivity determination is performed to locate the sensitive data part. Sensitive data determination is mainly based on the importance of data, sensitivity level classification, and potential security risk assessment. For sensitive data involving the core control strategy of wind turbines, extreme load values ​​of key components, or interaction with external communications, the system will automatically mark them; after locating sensitive data, the system will enable key management integration and perform key encryption processing on the sensitive data part. The key encryption processing adopts a dynamic key distribution mechanism, combined with symmetric encryption (such as AES) or asymmetric encryption (such as RSA) algorithms to ensure the confidentiality and integrity of data during transmission and storage. Encryption processing not only effectively prevents data from being tampered with or leaked during transmission, but also realizes refined control of data access rights to ensure that sensitive data is only available within the authorized scope.

[0028] Furthermore, the introduction of integration constraints includes: Divide the load types, wherein the load types include static loads, dynamic loads, steady-state loads and non-steady-state loads; divide the data access types, wherein the data access types include sensitive data and non-sensitive data; traverse the load types and the data access types to determine integration constraints, wherein the integration constraints are determined based on personalized integration requirements of the load types and data access types.

[0029] Introducing integrated constraints is a key link to ensure data processing efficiency and security. First, it is necessary to scientifically divide the load types based on the operating characteristics of wind turbines, mainly including static loads, dynamic loads, steady-state loads and unsteady-state loads. Among them, static loads refer to the constant force that the unit bears under unchanged or slowly changing conditions, such as the weight of the tower or the support load of the fixed structure; dynamic loads fluctuate with time, wind speed, rotation speed and other factors, such as the aerodynamic changes of the blades during rotation; steady-state loads usually refer to loads that remain relatively constant or fluctuate regularly during the stable operation of the unit, and typical scenarios include torque loads at constant wind speeds; and unsteady-state loads are manifested as loads that fluctuate violently in a short period of time, often occurring in sudden wind speed changes, emergency braking or fault conditions, and are highly uncertain and sudden. At the same time, it is also necessary to divide the data access types, mainly including sensitive data and non-sensitive data. Sensitive data usually involves key information such as the unit's core operating parameters, control instructions, fault diagnosis results, etc. Once such data is leaked or tampered with, it may cause serious safety hazards; while non-sensitive data includes routine monitoring data such as ambient temperature and wind speed. Even if anomalies occur, the impact on system safety is relatively limited.

[0030] After completing the division of load types and data access types, the system will traverse all possible combinations of loads and data to determine personalized integration constraints. The formulation of integration constraints aims to meet the specific data processing requirements under different combinations, ensuring that the data can be processed efficiently and have high security and reliability during the integration process. For example, for the combination of dynamic loads and sensitive data, since the data changes frequently and involves the core control information of the unit, the integration process must meet the requirements of high real-time, low latency and strong consistency, and encryption and permission management mechanisms must be introduced to prevent data leakage and illegal access. For the combination of static loads and non-sensitive data, the integration constraints can be relatively loose, allowing data batch processing and delayed transmission to reduce system resource consumption and improve data processing efficiency. In this process, the integration constraints not only consider the characteristics of the data itself, but also need to be dynamically adjusted according to the real-time operating status of the wind turbine to achieve adaptive optimization of data processing.

[0031] In addition, the formulation of integration constraints also needs to take into account the data processing priority and security strategy. Sensitive data usually has a higher processing priority and must be encrypted and integrity checked at the beginning of data integration, while non-sensitive data can be supplemented and integrated in the subsequent processing stage. For non-steady-state load scenarios, the system needs to have a rapid response capability and adjust the integration strategy in time to cope with sudden data fluctuations to ensure the real-time and reliability of the data; in steady-state load scenarios, data smoothing or redundant data compression strategies can be used to further optimize data storage and transmission efficiency. Finally, through a comprehensive traversal and analysis of load types and data access types, the system can form a complete set of integration constraint rule systems to provide accurate, efficient and secure technical support for data collection, processing and analysis of wind turbines under different operating conditions.

[0032] A data integration module is developed in the wind power load test system, supervised training is performed based on the Markov load network, and a data interface direct connection between the data integration module and the distributed sensor array is established.

[0033] Developing a data integration module in the wind power load test system is a key step to achieve efficient fusion and intelligent analysis of multi-source data. The data integration module is supervised and trained based on the constructed Markov load network to improve the accuracy and robustness of load scenario recognition and data processing. The Markov load network is the core model. Based on the change law of wind turbines under different load scenarios, the dynamic transfer relationship of wind turbines under different working conditions is established by defining each load scenario as a node and the change probability between scenarios as the edge weight. On this basis, the data integration module is trained through a supervised learning algorithm, using historical multi-source load data as training samples, matching the actual collected load data with known load scenarios, identifying the mapping relationship between data features and scenario change patterns, and gradually optimizing the classification accuracy of the integrated model and the prediction ability of scenario switching.

[0034] In order to ensure that the data integration module can process multi-source data from various components of the wind turbine in real time and efficiently, the system further establishes a direct data interface connection between the data integration module and the distributed sensor array. The distributed sensor array includes stress sensors, acceleration sensors, temperature sensors, vibration monitoring devices, etc. arranged on key structures such as blades, hubs, main shafts, towers, etc., which are used to collect load data of the unit under different working conditions in real time. The direct connection design of the data interface ensures that the sensor data can be directly transmitted to the data integration module without the need for intermediate conversion links, reducing data delays and transmission losses, and improving the real-time and stability of data processing. The data interface supports multiple communication protocols, such as CAN, Modbus, Ethernet, etc., to meet the access requirements of different sensor devices. At the same time, it has high bandwidth and low latency characteristics to ensure the reliable transmission of high-frequency sampling data.

[0035] The test target is generated by the wind power load test system, and the distributed sensor array is controlled to perform load test sampling according to the difference frequency sampler configured at the front end to determine the multi-source sampling data.

[0036] In the wind power load test system, the test target is first generated through the preset rules built into the system. The test target is usually determined based on the operating status, environmental conditions and historical load data analysis results of the wind turbine, and is intended to evaluate the load response characteristics and structural safety of the unit under specific working conditions. These test targets can include load assessment for specific wind speed ranges, stress testing during the start-up and shutdown of the unit, load change analysis under extreme weather conditions, etc., to ensure that various critical working conditions that the unit may face are covered.

[0037] Once the test target is generated, the system will automatically configure the difference frequency sampler at the front end to achieve accurate collection of load data in different frequency ranges. The difference frequency sampler is a key device used to optimize signal sampling efficiency and quality. It can dynamically adjust the sampling frequency according to the needs of the test target, capture high-frequency and low-frequency signal components, and avoid frequency aliasing during data sampling. For example, in a scenario where the unit experiences severe load fluctuations or transient shocks, the difference frequency sampler will automatically increase the sampling frequency to capture rapidly changing load data in a short period of time; in a steady-state operation scenario, the sampling frequency can be appropriately reduced to reduce data redundancy and storage pressure.

[0038] After the difference frequency sampler is configured, the system controls the distributed sensor array to perform load test sampling. The distributed sensor array is deployed at key locations of wind turbines, including the blade root, hub, main shaft, middle section of the tower and foundation, covering various types of monitoring equipment such as stress strain sensors, accelerometers, torque sensors, and temperature sensors. The system efficiently links with these sensors through the difference frequency sampler to collect load data of the unit under different working conditions in real time to ensure wide coverage and high accuracy of the data. During the sampling process, the system will also automatically calibrate the sensor data, perform noise filtering and anomaly detection, and ensure the validity and reliability of the collected data. Finally, the system will perform preliminary processing on the multi-source sampling data obtained from the distributed sensor array, including time synchronization, data format standardization, and preliminary data integrity verification, to form a structured load data set.

[0039] Further, controlling the distributed sensor array to perform load test sampling includes: Guided by the test goal, the sampling requirements based on the target load scenario are determined; the timestamp constraints of distributed sampling are introduced, and the difference frequency sampler generates a sensor sampling instruction in combination with the sampling requirements; the sensor sampling instruction responds to the target sensor to obtain the multi-source sampling data, wherein the target sensor belongs to the distributed sensor array.

[0040] Specifically, according to the generated test objectives, the sampling requirements based on the target load scenario are determined. The test objectives may include structural response under a specific wind speed range, stress distribution under extreme load conditions, or dynamic load changes during the start-up and shutdown of the unit. Based on these objectives, the system will analyze the data requirements under different load scenarios and determine parameters such as sampling frequency, sampling duration, data accuracy, and target monitoring points. For example, when performing extreme load tests under high wind speed conditions, the system will require higher sampling frequencies and finer time resolution to capture rapidly changing stress and vibration signals; while in steady-state operation scenarios, the sampling frequency can be appropriately reduced to focus on long-term trend data. After determining the sampling requirements, the system introduces timestamp constraints for distributed sampling to ensure that data from different sensors can be accurately synchronized in time. Timestamp constraints are the core of the distributed sampling system, which aims to solve the problem of inconsistent timing of multi-source data and ensure that data collected by all sensors are comparable under the same time base. Based on global clock or GPS time synchronization technology, the system will stamp each sampled data with a high-precision timestamp to achieve time alignment across sensors, which is especially suitable for complex dynamic system environments such as wind turbines. Based on the timestamp constraints and sampling requirements, the difference frequency sampler generates specific sensor sampling instructions. The difference frequency sampler dynamically adjusts the sampling strategy according to the working characteristics and sampling frequency requirements of different sensors, and generates instructions including parameters such as sampling frequency, sampling duration, and data format. These sampling instructions can flexibly adapt to different types of sensors, such as stress sensors, accelerometers, torque sensors, temperature sensors, etc., to ensure the integrity and diversity of the sampled data. For example, when the system detects a rapid change in wind speed, the difference frequency sampler will automatically adjust the sampling frequency to increase the sampling density of key sensors (such as blade root stress sensors) and capture the details of transient load changes. The generated sensor sampling instructions will be directly sent to the target sensors in the distributed sensor array. The selection of these target sensors depends on the test objectives and sampling requirements. After receiving the sampling instructions, the sensors will collect data according to the preset parameters, record multi-source data such as load, vibration, temperature, etc. in real time, and attach timestamp information to ensure accurate synchronization of data.

[0041] According to the data interface, the multi-source sampling data is transmitted back and combined with the data integration module to perform data integration processing based on load scenario changes, and the load integration data is determined and stored, wherein the change analysis is performed based on the upper scenario node.

[0042] In the wind power load test system, the multi-source sampling data transmission and its integrated processing based on the data interface are the key links to achieve efficient management and intelligent analysis of load data. First, the multi-source sampling data collected by the distributed sensor array will be transmitted back to the system's data integration module through an efficient data interface. This data interface not only supports high-speed data transmission, but also has multi-protocol compatibility, and can adapt to different types of sensor data, including stress, strain, acceleration, torque, temperature and other physical quantity data, while ensuring the integrity and real-time nature of the data. During the data transmission process, the system automatically attaches timestamps, sensor identifiers and sampling parameters to ensure that the data can maintain accurate timing consistency and traceability in subsequent processing.

[0043] After the multi-source sampling data is transmitted back into the data integration module, the system will combine the existing Markov load network model to carry out data integration processing based on load scenario changes. Specifically, the data integration module will first pre-process the transmitted data, including data cleaning (removing outliers and noise), time alignment (multi-source data synchronization based on timestamps), and format standardization. Subsequently, based on the scene change path of the Markov load network, the system identifies the load scene to which the current data belongs and its position in the historical scene chain, and analyzes the dynamic change rules between scenes. In this process, the system will use the upper scene node as a reference to carry out change analysis. The so-called upper scene node refers to the previous scene of the current load scene in the Markov chain, that is, the state node of the unit under the previous operating conditions. By comparing the data characteristics of the current data with those of the upper scene, the system can capture subtle changes in load characteristics during scene changes and identify potential abnormal patterns or trends, such as stress concentration and fatigue accumulation.

[0044] Based on the scenario change analysis, the system deeply integrates the sampling data from different sources and different time periods to generate high-quality load integration data. Data integration processing not only includes simple data aggregation, but also involves advanced processing methods such as data dimension reduction, feature extraction, and differential analysis to ensure that the integrated data can fully reflect the actual operating status of the unit under different load scenarios. For example, when the wind turbine switches from a medium wind speed scenario to a high wind speed scenario, the system will automatically adjust the data integration strategy to increase the weight of the blade root stress and tower vibration data to more accurately reflect the response characteristics of the structure under extreme conditions. After completing the data integration, the system will store the load integration data in a standardized manner, and the storage format includes structured databases and distributed data warehouses to support subsequent rapid retrieval and in-depth analysis.

[0045] Furthermore, the data integration process based on load scenario changes includes: The Markov load network is used to locate the upper scene node of the upper load scene, and the lower scene node is determined by the target load scene; the connection relationship between the upper scene node and the lower scene node is determined to determine the target integration method based on the scene change; the multi-source sampling data is integrated and processed in the target integration method, wherein the data integration module has a short-term memory based on the upper scene node.

[0046] Specifically, according to the current target load scenario, the upper and lower scenario nodes in the Markov load network are located. The upper scenario node represents the operating state experienced by the unit before the current scenario, reflecting the path and characteristics of the historical evolution of the load, while the lower scenario node represents the next stage of operation to which the unit may transfer, indicating the trend of future load changes. After completing the positioning of the upper and lower scenario nodes, the system further analyzes the connection relationship between the two nodes to determine the target integration method based on scenario changes. This connection relationship not only reflects the transition probability between load scenarios, but also includes the dynamic characteristics of scenario changes, such as load fluctuation amplitude, change rate, stress concentration changes, etc. Based on these characteristics, the system intelligently matches the most suitable data integration strategy. For example, when the scenario change involves drastic load fluctuations (such as switching from low wind speed to extremely high wind speed conditions), the system will choose a highly sensitive data integration method, focusing on capturing instantaneous load changes and local stress concentration phenomena; in the scenario switching of steady-state operation, the system may adopt a smoothing or trend analysis type of integration method to reduce data noise and improve long-term trend recognition capabilities.

[0047] After determining the target integration method, the system uses this as a basis for deep integration processing of multi-source sampling data. In this process, the data integration module makes full use of its built-in short-term memory capabilities based on the upper-level scene nodes to enhance the temporal relevance and historical sensitivity of data processing. The short-term memory mechanism allows the system to retain and refer to key data features and load change patterns related to the upper-level scene when performing current data integration, ensuring that data integration not only focuses on current transient data, but also captures deep information related to historical working condition changes. Finally, the data processed by the target integration method is normalized into load integration data, including the fusion results of multi-source information such as stress, strain, vibration, and temperature, supplemented by metadata such as scene change paths, timestamps, and processing logs to ensure data traceability and reusability.

[0048] In summary, the embodiments of the present application have at least the following technical effects: First, for the wind turbine, according to the key load-unit control posture, N load scenarios are determined. Then, the N load scenarios are traversed to construct a Markov load network, in which the load scenario is used as a node and the integration mode based on the scene transition between nodes is used as the connection relationship. Further, a data integration module is developed in the wind power load test system, and supervised training is performed based on the Markov load network, and a data interface direct connection between the data integration module and the distributed sensor array is established. Then, the test target is generated through the wind power load test system, and the distributed sensor array is controlled to perform load test sampling according to the difference frequency sampler configured at the front end to determine the multi-source sampling data. Finally, according to the data interface, the multi-source sampling data is transmitted back and combined with the data integration module to perform data integration processing based on the load scenario transition, and the load integration data is determined and stored, wherein the transition analysis is performed based on the upper scene node. The technical problem of low efficiency of multi-source data integration of the wind power load test system in the prior art is solved, and the technical effect of improving data integration efficiency is achieved.

[0049] Embodiment 2, based on the same inventive concept as the multi-source data integration method for a wind power load test system in the above embodiment, Figure 2 As shown, the present application provides a multi-source data integration system for a wind power load test system, wherein the system includes: A scenario determination unit 11 is used to determine N load scenarios for a wind turbine according to the load of key parts and the control posture of the unit; a load network construction unit 12 is used to traverse the N load scenarios and construct a Markov load network, wherein the load scenarios are nodes and the connection relationship is based on the integration mode transition between the nodes; a training unit 13 is used to develop a data integration module in the wind power load test system, perform supervised training based on the Markov load network, and establish a data interface direct connection between the data integration module and the distributed sensor array; a sampling unit 14 is used to generate a test target through the wind power load test system, control the distributed sensor array to perform load test sampling according to the difference frequency sampler configured at the front end, and determine multi-source sampling data; a data processing unit 15 is used to transmit the multi-source sampling data back according to the data interface and perform data integration processing based on the load scenario transition in combination with the data integration module, determine the load integration data and store it, wherein the transition analysis is performed based on the upper scene node.

[0050] Furthermore, the load network construction unit 12 is used to execute the following method: The communication protocol and data format of interactive multi-source data are used to set a unified data standard, wherein the unified data standard is any communication protocol and data format; the unified data standard is used as a universal integration standard, the N load scenarios are traversed, and the transition analysis between scenario groups and the integration mode transition analysis are performed respectively to construct the Markov load network.

[0051] Furthermore, the load network construction unit 12 is used to execute the following method: Taking the unified data standard as the universal integration standard, starting with distributed source data, and guided by data requirements based on test objectives, the data integration method of the first load scenario is determined by introducing integration constraints, wherein the data integration method at least includes unified integration-differential relaxation integration-key management integration; for the first load scenario and the remaining N-1 load scenarios, the scenario change characteristics are identified and the integration change characteristics based on the data integration method are mined, and the first load network is determined by integration.

[0052] Furthermore, the load network construction unit 12 is used to execute the following method: Guided by the test goal, the differential relaxation is determined, wherein the differential relaxation is determined based on the data value; based on the differential relaxation, the standard multi-source data determined after the unified integration is differentially transformed to determine the valid data; the valid data is subjected to sensitivity determination, the sensitive data portion is located, and key encryption is performed.

[0053] Furthermore, the load network construction unit 12 is used to execute the following method: Divide the load types, wherein the load types include static loads, dynamic loads, steady-state loads and non-steady-state loads; divide the data access types, wherein the data access types include sensitive data and non-sensitive data; traverse the load types and the data access types to determine integration constraints, wherein the integration constraints are determined based on personalized integration requirements of the load types and data access types.

[0054] Furthermore, the sampling unit 14 is used to perform the following method: Guided by the test goal, the sampling requirements based on the target load scenario are determined; the timestamp constraints of distributed sampling are introduced, and the difference frequency sampler generates a sensor sampling instruction in combination with the sampling requirements; the sensor sampling instruction responds to the target sensor to obtain the multi-source sampling data, wherein the target sensor belongs to the distributed sensor array.

[0055] Furthermore, the data processing unit 15 is used to execute the following method: The Markov load network is used to locate the upper scene node of the upper load scene, and the lower scene node is determined by the target load scene; the connection relationship between the upper scene node and the lower scene node is determined to determine the target integration method based on the scene change; the multi-source sampling data is integrated and processed in the target integration method, wherein the data integration module has a short-term memory based on the upper scene node.

[0056] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0058] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A multi-source data integration method for a wind power load test system, characterized in that: The method comprises: For wind turbines, determine N load scenarios based on the loads at key locations and the control attitude of the turbines; Traversing the N load scenarios, constructing a Markov load network, wherein the load scenarios are nodes, and the integration mode transition based on the scenario transition between nodes is the connection relationship; Developing a data integration module in the wind power load test system, performing supervised training based on the Markov load network, and establishing a direct data interface connection between the data integration module and the distributed sensor array; Generate test targets through the wind power load test system, control the distributed sensor array to perform load test sampling according to the difference frequency sampler configured at the front end, and determine the multi-source sampling data; According to the data interface, the multi-source sampling data is transmitted back and combined with the data integration module to perform data integration processing based on load scenario changes, and the load integration data is determined and stored, wherein the change analysis is performed based on the upper scenario node.

2. The multi-source data integration method for a wind power load test system according to claim 1, characterized in that: Construct a Markov load network, including: Communication protocols and data formats for interacting with multiple source data, setting a unified data standard, wherein the unified data standard is any communication protocol and data format; Taking the unified data standard as the universal integration standard, the N load scenarios are traversed, and the transition analysis between scenario groups and the integration mode transition analysis are performed respectively to construct the Markov load network.

3. The multi-source data integration method for a wind power load test system according to claim 2, characterized in that: Conduct change analysis between scenario groups and integration mode change analysis respectively, including: Taking the unified data standard as the universal integration standard, starting with the distributed source data, and guided by the data requirements based on the test objectives, the data integration method of the first load scenario is determined by introducing integration constraints, wherein the data integration method at least includes unified integration-differential relaxation integration-key management integration; For the first load scenario and the remaining N-1 load scenarios, the scenario transition features are identified and the integrated transition features based on the data integration method are mined to integrate and determine the first load network.

4. The multi-source data integration method for a wind power load test system according to claim 3, characterized in that: The data integration method at least includes unified integration-differential relaxation integration-key management integration, including: Determining a differential relaxation based on a test goal, wherein the differential relaxation is determined based on data value; Taking the differential relaxation as a benchmark, performing differential conversion processing on the standard multi-source data determined after unified integration to determine valid data; Perform sensitivity determination on the valid data, locate the sensitive data portion and perform key encryption processing.

5. The multi-source data integration method for a wind power load test system according to claim 3, characterized in that: The introduction of integration constraints includes: Classify load types, wherein the load types include static loads, dynamic loads, steady-state loads and unsteady-state loads; Classify data access types, wherein the data access types include sensitive data and non-sensitive data; The load types and the data access types are traversed to determine integration constraints, wherein the integration constraints are determined based on personalized integration requirements of the load types and the data access types.

6. The multi-source data integration method for a wind power load test system according to claim 1, characterized in that: Control distributed sensor arrays for load test sampling, including: Guided by the test objectives, determine the sampling requirements based on the target load scenario; Introducing the timestamp constraint of distributed sampling, and combining the sampling requirements, the difference frequency sampler generates a sensing sampling instruction; The sensor sampling instruction acquires the multi-source sampling data in response to a target sensor, wherein the target sensor belongs to the distributed sensor array.

7. The multi-source data integration method for a wind power load test system according to claim 6, characterized in that: The data integration process based on load scenario changes includes: Using the Markov load network, locating the upper scenario node of the upper load scenario, and determining the lower scenario node using the target load scenario; For the connection relationship between the upper scene node and the lower scene node, determine a target integration method based on scene transition; The multi-source sampling data are integrated and processed in the target integration manner, wherein the data integration module has a short-term memory based on the upper scene node.

8. A multi-source data integration system for a wind power load test system, characterized in that: A method for integrating multi-source data for a wind power load test system according to any one of claims 1 to 7, the system comprising: A scenario determination unit is used to determine N load scenarios for a wind turbine according to the loads at key parts and the control attitude of the turbine; A load network construction unit, used to traverse the N load scenarios and construct a Markov load network, wherein the load scenarios are nodes and the integration mode based on the scenario transition between nodes is the connection relationship; A training unit, used for developing a data integration module in the wind power load test system, performing supervised training based on the Markov load network, and establishing a data interface direct connection between the data integration module and a distributed sensor array; The sampling unit is used to generate the test target through the wind power load test system, control the distributed sensor array to perform load test sampling according to the difference frequency sampler configured at the front end, and determine the multi-source sampling data; The data processing unit is used to transmit the multi-source sampling data back according to the data interface and perform data integration processing based on load scenario changes in combination with the data integration module, determine the load integration data and store it, wherein the change analysis is performed based on the upper scenario node.

Images