A power performance assessment intelligent system for a wind turbine generator system

Abstract

The utility model provides a kind of power fast evaluation intelligent system of wind generating set, it includes: data acquisition module, data analysis processing module and forecast control execution module.The utility model constructs the intelligent device system including data acquisition module, data analysis processing module and forecast control execution module by modular design: data acquisition module realizes to the multi-source data real-time acquisition and pre-processing of unit itself and environment;Data analysis processing module extracts key features and predicts short-term power change by spatial correlation and time series depth analysis;Forecast control execution module executes advance yaw control, performance optimization adjustment and protective action based on analysis result, realizes the initiative protection and power optimization of unit, and can overcome the defect that traditional method data is single, analysis lags behind and control is passive, realizes the fast evaluation, forecast control and initiative protection of wind generating set power, to improve wind farm operating efficiency and equipment reliability.

Description

Technical Field

[0001] This utility model relates to an intelligent system for power speed assessment of wind turbine generator sets, belonging to the field of wind turbine generator set operation monitoring, power prediction and intelligent control technology. Background Technology

[0002] With the adjustment of the global energy structure and the increasing demand for renewable energy development, the wind power industry has achieved large-scale development. Along with the expansion of wind farm scale and the increase in the number of turbines, the need for refined and intelligent management of wind farm operation is becoming increasingly urgent. In wind farm operation and management, the output power of wind turbine generators is not only a core indicator for evaluating unit performance and economic benefits, but also an important basis for monitoring the health status of equipment.

[0003] Currently, wind turbine power assessment primarily relies on Supervisory Control and Data Acquisition (SCADA) systems. Traditional technical methods typically include the following: 1. Unit self-data monitoring: Traditional methods mainly involve collecting unit operating parameters, such as wind speed, wind direction, generator power, speed, pitch angle, gearbox oil temperature, and ambient temperature, through local sensors and SCADA systems. This data is used for post-event statistics, trend analysis, and threshold alarms.

[0004] 2. Threshold-based performance evaluation: In existing technologies, the evaluation of unit power performance is usually based on simple threshold judgments, such as triggering an alarm when the power generation is lower than a certain percentage of the rated power or the gearbox temperature exceeds a set limit. This method relies on fixed thresholds and lacks comprehensive analysis of the dynamic operating status of the unit.

[0005] 3. Historical data comparison and statistical analysis: Some wind farms use historical power data comparison methods to assess the trend of unit operation. For example, the current power is compared with the historical average power or standard power curve to determine whether the unit is abnormal. Such methods are mostly linear or simple statistical processing, which makes it difficult to capture the nonlinear laws and complex patterns in power changes.

[0006] However, the aforementioned traditional technical methods have obvious limitations: 1. Data collection is limited to a single dimension. Traditional methods primarily focus on the parameters of individual generating units, neglecting the spatial correlations between units, such as geographical location, wake effects, and local wind field disturbances. This makes it impossible to quantitatively analyze the spatial coupling effects of unit power variations.

[0007] 2. Data analysis is lagging and has limited accuracy. Existing analyses are mostly post-hoc statistics or simple comparisons, lacking in-depth analysis of multi-variable time series data such as power, wind speed, and temperature, making it difficult to accurately predict short-term power trends. Furthermore, the ability to identify abnormal patterns is limited, resulting in insufficient early warning capabilities for potential unit failures or performance degradation.

[0008] 3. Passive control strategies: Traditional control methods mostly trigger maintenance and alarms after a fault or severe performance degradation, lacking predictive and proactive adjustment capabilities. This passive strategy not only leads to power generation losses but may also increase maintenance costs and the risk of equipment fatigue damage, reducing the overall operating efficiency of the wind farm.

[0009] Based on the above technical background, there is an urgent need for a modular device that can integrate multi-source data sensing, intelligent analysis and calculation, and predictive control functions to achieve rapid and accurate assessment and intelligent protection of wind turbine power. Utility Model Content

[0010] In order to solve the above-mentioned technical problems, this utility model provides an intelligent system for evaluating the power speed of wind turbine generator sets.

[0011] This utility model solves the above-mentioned technical problems through the following technical solutions: This utility model provides an intelligent system for power speed assessment of wind turbine generator sets, including a data acquisition module, a data analysis and processing module, and a predictive control execution module. The data acquisition module is connected to the SCADA system and external environmental monitoring system of multiple wind turbine generator sets in the wind farm via wired or wireless communication. The data acquisition module includes a unit data acquisition unit, an environmental data acquisition unit, and a data preprocessing and caching unit; the data analysis and processing module includes a spatial correlation analysis submodule, a temporal correlation analysis submodule, and a feature extraction and decision-making unit; the predictive control execution module includes an instruction receiving and parsing unit, a control signal output unit, a predictive action execution submodule, and a protective action execution submodule.

[0012] In this technical solution, the unit data acquisition unit is communicatively connected to the wind turbine generator set and its adjacent units.

[0013] In this technical solution, the data acquisition unit of the generator set is connected to the wind turbine generator set and its adjacent generator sets via either wired or wireless communication.

[0014] In this technical solution, the environmental data acquisition unit is installed in the wind farm control room or on top of the meteorological tower.

[0015] In this technical solution, the data preprocessing and caching unit is communicatively connected to the unit data acquisition unit and the environmental data acquisition unit, respectively.

[0016] In this technical solution, the feature extraction and decision-making unit is communicatively connected to the spatial correlation analysis submodule and the temporal correlation analysis submodule, respectively.

[0017] In this technical solution, the feature extraction and decision-making unit is communicatively connected to the spatial correlation analysis submodule and the temporal correlation analysis submodule, respectively.

[0018] In this technical solution, the instruction receiving and parsing unit and the data analysis and processing module are communicatively connected.

[0019] In this technical solution, the instruction receiving and parsing unit, the control signal output unit, the predictive action execution submodule, and the protective action execution submodule are interconnected.

[0020] In this technical solution, the data acquisition module, the data analysis and processing module, and the predictive control execution module are sequentially connected in communication.

[0021] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of this utility model.

[0022] The positive and progressive effects of this utility model are as follows: The aforementioned intelligent system for power speed assessment of wind turbine generators utilizes a modular design to construct an intelligent device system comprising a data acquisition module, a data analysis and processing module, and a predictive control execution module. The data acquisition module collects and preprocesses multi-source data from the generator itself and its environment in real time. The data analysis and processing module extracts key features and predicts short-term power changes through spatial correlation and in-depth time series analysis. The predictive control execution module executes advanced yaw control, performance optimization adjustments, and protective actions based on the analysis results, achieving proactive protection and power optimization for the generator. This intelligent device overcomes the shortcomings of traditional methods, such as limited data, delayed analysis, and passive control, enabling rapid assessment, predictive control, and proactive protection of wind turbine generator power, thereby improving wind farm operating efficiency and equipment reliability. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the system flow of this utility model. Detailed Implementation

[0024] The present invention will be further illustrated by way of embodiments below, but the present invention is not limited to the scope of the embodiments described herein.

[0025] like Figure 1As shown, the intelligent system for rapid power assessment of the wind turbine generator set includes: a data acquisition module, a data analysis and processing module, and a predictive control execution module, which are electrically connected in sequence. The data acquisition module is connected to the SCADA system and external environmental monitoring system of multiple wind turbine generator sets in the wind farm via wired or wireless communication to acquire raw operating data. The data analysis and processing module receives the data transmitted by the data acquisition module and performs spatial and temporal correlation analysis using built-in algorithms to extract features and generate predictive and control commands. The predictive control execution module receives the control commands and outputs control signals to the controller of the wind turbine generator set to execute predictive actions and protection.

[0026] The data acquisition module is the basic unit of this utility model device. Its main function is to realize high-precision real-time acquisition, preprocessing and caching of wind farm operation data and environmental data, so as to provide complete, accurate and time-series continuous input information for subsequent data analysis and processing modules.

[0027] This module establishes data exchange channels with the SCADA systems and external environmental monitoring systems of multiple wind turbine generators in the wind farm via wired or wireless communication, forming a multi-source data access network covering both the turbine generator and the wind farm level. Through this module, synchronous acquisition of turbine generator operating status, wind farm meteorological conditions and their dynamic changes can be achieved, providing data support for rapid power assessment.

[0028] The data acquisition module specifically includes the following units: 1. Unit Data Acquisition Unit. This unit is used to acquire real-time operating parameters of the target wind turbine generator set and its adjacent units. To ensure data reliability and timeliness, the unit data acquisition unit communicates with the SCADA system interface of each unit through an industrial-grade Ethernet switching device (preferably Hirschmann MACH 4000 series), and acquires unit operating data using the standard Modbus TCP / IP or IEC 60870-5-104 protocol.

[0029] The parameters collected include, but are not limited to: actual generator power (kW), wind speed (m / s), wind direction (°), generator speed (rpm), pitch angle (°), gearbox oil temperature (°), ambient temperature (°), etc.

[0030] These data reflect the dynamic response characteristics of the unit under different operating conditions, and can provide basic input for subsequent time series prediction models.

[0031] In addition, this unit can synchronously collect key parameters of adjacent units as needed, so that subsequent modules can perform spatial correlation analysis (wake effect modeling).

[0032] 2. Environmental Data Acquisition Unit. This unit collects macroscopic environmental information about the wind farm, providing external supplementation to the turbine's operating conditions. It integrates a Vaisala WMT52 ultrasonic anemometer, installed in the wind farm's control room or on top of the meteorological tower, providing high-precision wind speed, direction, and turbulence intensity data at the farm level. Compared to the turbine's built-in anemometer, this ultrasonic instrument offers higher temporal resolution and better anti-interference capabilities, enabling the identification of overall wind conditions and gust characteristics. By fusing farm-level meteorological data with turbine-level data, the global consistency of power prediction is effectively improved, and a reference benchmark is provided for the spatial analysis module to identify upwind turbines.

[0033] 3. Data Preprocessing and Caching Unit. This unit serves as an intermediate layer between the data acquisition module and the data analysis and processing module, playing a crucial role in data purification, structuring, and time-series alignment. Built on an Intel Atom® E3900 series low-power, high-performance processor and equipped with DDR4 memory and high-speed SSD storage, this unit performs the following processing steps on the acquired raw data: Data cleaning: Remove missing values, duplicate values, and sensor anomalies to ensure data integrity; Filtering: Algorithms such as moving average and Kalman filtering are used to smooth the signal and eliminate short-term noise fluctuations; Normalization: Standardize data of different dimensions to meet the input requirements of subsequent machine learning models; Time synchronization and caching: Align and temporarily store multi-source data based on timestamps to ensure the timing consistency of data with different sampling frequencies.

[0034] The cached data will be transmitted to the data analysis and processing module in a structured format (Parquet or CSV stream) to provide a continuous data stream for its real-time computation.

[0035] Under high system load or network latency, this unit can also maintain stable short-term data input through a ring buffer mechanism to prevent data loss.

[0036] Synergy with other modules: The data acquisition module plays a core role in information input and quality assurance in this invention, and is the first layer of the entire intelligent power assessment system. It not only provides real-time and accurate input data for the analysis model, but also reduces the computational burden on subsequent calculation modules through preprocessing, thereby improving the overall system's response speed and prediction accuracy.

[0037] This module connects to the data analysis and processing module via a standardized data interface (MQTT or OPC UA protocol), forming a highly reliable communication link. Data processed by the data acquisition module is directly input into the spatial correlation and temporal correlation submodules within the analysis module for calculating the upwind turbine influence coefficient and training the LSTM prediction model. The analysis results are then converted into control signals by the predictive control execution module and fed back to the wind turbine's main control system.

[0038] Thus, the data acquisition module, analysis and processing module, and predictive control module together form a closed-loop intelligent evaluation and control system, realizing the entire process relationship of data perception, intelligent analysis, and predictive execution.

[0039] Furthermore, the data analysis and processing module is the core intelligent computing unit of this utility model device, undertaking the key functions of "data understanding, state recognition, trend prediction and strategy generation" in the entire system architecture.

[0040] This module directly receives multi-source preprocessed data transmitted from the data acquisition module. Through the built-in algorithm model, it performs spatial and temporal correlation analysis on the wind turbine operation data, extracts key features reflecting changes in unit performance, and generates predictive control strategies and protection commands based on the analysis results for subsequent predictive control execution modules to call.

[0041] This module is built on the Advantech ARK-3530 series high-performance embedded industrial control computer, with built-in multi-core CPU and GPU acceleration units, enabling simultaneous multi-threaded data computation and deep learning inference tasks. Its highly parallel computing architecture ensures real-time performance and accuracy, making it suitable for deployment in local control systems of wind farms to achieve intelligent computing at the edge.

[0042] This module includes the following functional sub-units: 1. Spatial Correlation Analysis Submodule. This submodule is mainly used to identify the spatial interaction relationships between units within a wind farm, especially the impact of the wake effect on the output power of the target unit.

[0043] In wind farms, airflow disturbances caused by upwind turbines can reduce the wind energy capture capacity of downwind turbines, leading to power attenuation. To quantitatively characterize this impact, the spatial correlation analysis submodule dynamically identifies upwind adjacent turbines based on real-time wind direction, wind speed, and wind farm layout data.

[0044] Its main analytical process includes: I. Upwind direction identification: Based on the real-time wind direction angle data provided by the environmental data acquisition unit, the set of wind farm unit coordinates is matched with the wind farm unit coordinate database to determine the set of units upwind relative to the target unit; II. Correlation Calculation: Using the real-time power data of each unit transmitted by the data acquisition module, calculate the Pearson Correlation Coefficient between the target unit and the upwind unit to determine the intensity of the wake effect. III. Causal Analysis Modeling: The Granger Causality model is further introduced, and the degree of causal influence of the upwind unit power change on the target unit power change is determined by calculating the time lag characteristics. IV. Wake Impact Assessment: Based on the above results, a "wake impact coefficient" is generated as an important input parameter for subsequent decision-making units.

[0045] Through the operation of this module, the system can identify the dynamic coupling relationship between units in the wind farm in real time and provide quantitative basis for predictive yaw or power compensation.

[0046] 2. Time Correlation Analysis Submodule. This submodule is used to characterize the power variation pattern of a single unit over time and identify short-term trends and abnormal fluctuations.

[0047] This module is built on a Long Short-Term Memory (LSTM) model from deep learning, enabling it to model and predict nonlinear, multivariate time series. Its input data primarily comes from historical power, wind speed, and temperature sequences output by the preprocessing and caching units in the data acquisition module; its output includes predicted power values ​​and anomaly deviation indicators for the next short period (e.g., 10 to 30 minutes).

[0048] The main workflow of the model includes: I. Input Data Construction: Input data such as power, wind speed, gearbox oil temperature, and ambient temperature into the network in units of time steps; II. Feature Learning: By using the gating structure of LSTM, we learn the nonlinear dependencies between different variables to capture the trends and abrupt changes in power. III. Short-term forecast: Outputs power forecast values ​​for future time windows; IV. Anomaly Detection: When the deviation between the predicted value and the actual power exceeds the set threshold, it is automatically marked as a potential abnormal event and a "Power Curve Anomaly" index is generated.

[0049] This submodule can not only perform short-term power prediction, but also serve as an auxiliary basis for early fault identification. For example, when abnormal power fluctuations are detected accompanied by an abnormal temperature rise, protective warning commands can be triggered in advance.

[0050] 3. Feature Extraction and Decision-Making Unit. This unit is located at the top layer of the data analysis and processing module and is responsible for comprehensive judgment and strategy generation.

[0051] It receives results from the spatial correlation analysis submodule and the temporal correlation analysis submodule, and integrates multi-dimensional feature information, including wake influence coefficient, short-term power prediction value, power curve anomaly, unit health index and other indicators.

[0052] Based on this, the feature extraction and decision-making unit generates corresponding control strategies using built-in expert knowledge base rules or a fuzzy logic-based decision system. Typical rule examples are as follows: If the wake influence coefficient exceeds the set threshold and the predicted power is less than 80% of the rated power, an "advance yaw control" command will be generated. If the power curve anomaly continues to increase while the gearbox temperature shows an upward trend, an "early fault warning" command will be generated. If the predicted wind speed will exceed the gust threshold and the turbulence intensity increases rapidly, a "preventive load reduction" control signal will be generated.

[0053] The generated control commands are packaged and encapsulated with protocols and then transmitted to the predictive control execution module, which then performs signal output and execution actions.

[0054] Collaboration with other modules: The data analysis and processing module serves as a bridge connecting the data input layer and the control execution layer, playing a core role in intelligent decision-making.

[0055] Its functions are reflected in the following aspects: (1) The multi-source heterogeneous data (generator end and field level end) from the data acquisition module are fused and analyzed to form a global understanding of the wind farm status; (2) By using spatiotemporal coupling modeling, the power decline trend or potential equipment anomalies can be identified in advance, providing the control module with an advanced decision-making basis; (3) Through closed-loop communication with the predictive control execution module, the unit response data is received and the model parameters are dynamically corrected to achieve self-learning and model adaptive updating; (4) When the data acquisition module experiences data delay or missing data, the analysis module can automatically start a compensation algorithm (such as sliding prediction interpolation) to ensure the continuity of system calculation.

[0056] This module is connected to the data acquisition module via a high-speed industrial Ethernet and uses MQTT or OPC UA communication protocols to ensure low-latency and high-reliability data transmission.

[0057] The interface with the predictive control execution module uses a standard control signal format (such as Modbus RTU / TCP) to ensure that control commands can be directly recognized and executed by the unit's main controller.

[0058] The predictive control execution module is the execution and feedback unit of this utility model device. It is located at the forefront of the system architecture and undertakes the core function of transforming the predictive control strategy generated by the data analysis and processing module into actual unit actions.

[0059] This module receives control commands from the analysis module in real time, parses and converts them into standard communication signals that can be recognized by the main controller of the unit, and directly drives the wind turbine to perform predictive optimization and protective actions, thereby achieving unit power optimization and operational safety assurance.

[0060] The main functions of the predictive control execution module include: 1. Command execution: Quickly convert the predictive control and protective control strategies generated by the data analysis and processing module into control signals that the unit can execute.

[0061] 2. Unit protection: Take mechanical or electrical adjustment actions in advance to prevent key components such as gearbox, generator and blades from being overloaded or fatigued due to wake, gusts or abnormal power fluctuations.

[0062] 3. Performance optimization: Based on power prediction results and wake impact assessment, dynamically fine-tune parameters such as pitch angle and torque control to make the unit operate close to the optimal tip speed ratio and improve energy capture efficiency.

[0063] 4. System closed loop: By executing control actions and monitoring unit response, the system provides real-time feedback to the data acquisition and analysis modules, realizing intelligent closed-loop control of "perception-analysis-execution-feedback".

[0064] The predictive control execution module mainly includes the following sub-units: A. Command Receiving and Parsing Unit: This unit is responsible for receiving control commands from the data analysis and processing module, and performing parsing, decoding, and signal conversion. It communicates with the analysis module using standard industrial communication protocols (such as Modbus TCP / RTU, OPC UA, or a custom low-latency protocol). It parses high-level strategy commands (such as 5° advance yaw or 0.3° pitch angle adjustment) into directly executable control quantities. To cope with communication delays or network jitter, the parsing unit can temporarily cache the latest commands and execute them in timestamp order, ensuring the real-time performance and consistency of the unit's actions. This unit ensures that predictive control strategies can be transmitted to the unit's execution end with minimal delay, guaranteeing proactive response capabilities to wake gusts or power anomalies.

[0065] B. Control Signal Output Unit: This unit serves as the interface conversion layer for the module, responsible for converting the parsed control commands into standard control signals recognizable by the wind turbine's main controller. It utilizes digital and analog output modules from a B&R X20 series programmable logic controller (PLC). Depending on the main controller model (e.g., Bachmann MLC or Beckhoff CX series), the control commands are converted into signal formats compatible with its communication protocol, including analog outputs, digital outputs, and CAN or EtherCAT communication frames. The output unit incorporates a redundant monitoring mechanism, triggering alarms or safe load reduction in case of signal anomalies to ensure safe unit operation. Through the control signal output unit, a reliable closed-loop conversion from the data analysis and processing module's predicted commands to the actual actions of the unit is achieved.

[0066] C. Predictive Action Execution Submodule: This module proactively adjusts unit operating parameters based on predictive information provided by the analysis module, achieving power optimization and proactive operational control. Its main functions include: Advanced yaw control, based on the wake influence coefficient and wind direction prediction results provided by the spatial correlation analysis submodule, sends micro-yaw commands to the main controller in advance, causing the unit to yaw away from the strongest wake region. This reduces wake power loss, improves the wind energy capture efficiency of downwind units, and reduces mechanical stress on the unit.

[0067] Performance optimization and adjustment utilizes short-term power forecasts and tip speed ratio assessments provided by the time correlation analysis submodule to fine-tune pitch angle, torque control parameters, or generator load settings. The goal is to maintain optimal operating conditions for the unit under high wind speeds, low wind speeds, and gusts, thereby improving power output stability and generation efficiency.

[0068] D. Protective Action Execution Submodule: This submodule, based on anomaly detection and trend analysis results, proactively executes unit protection measures to reduce the risk of damage to mechanical components. Its main functions include: (1) Early fault warning: When time series analysis identifies abnormal rise in gearbox temperature and increased high-frequency power fluctuations, maintenance warning instructions are generated in advance to notify the central monitoring room to conduct inspections or manual intervention. Early intervention before the absolute temperature threshold is triggered extends the equipment life.

[0069] (2) Preventive load reduction: When the analysis module predicts that extreme gusts or turbulence are about to arrive, the pitch angle is adjusted in advance or the power output is reduced for a short time to achieve short-term load reduction of the unit. This reduces the mechanical stress on the transmission chain and blades and avoids damage caused by instantaneous overload.

[0070] Synergy with other modules: The predictive control execution module, data analysis and processing module, and data acquisition module form a complete closed-loop control system. (1) Connection with the data analysis and processing module: Receive the control strategy generated by the results of spatial correlation analysis and temporal correlation analysis, and feed back the execution status or anomalies to the analysis module for model adaptation and optimization.

[0071] (2) Connection with the data acquisition module: After the action is executed, the actual operating data of the unit (power change, pitch angle, torque, temperature, etc.) is collected and preprocessed in real time by the acquisition module to provide feedback to the analysis module. Through this closed-loop data flow, the cycle of acquisition-analysis-execution-feedback is realized to ensure continuous and reliable power optimization and protection control.

[0072] This utility model is not limited to the above-described embodiments. Any changes in its shape or structure fall within the protection scope of this utility model. The protection scope of this utility model is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of this utility model, but all such changes and modifications fall within the protection scope of this utility model.

Claims

1. A power speed assessment intelligent system for wind turbine generator sets, comprising a data acquisition module, a data analysis and processing module, and a predictive control execution module, characterized in that, The data acquisition module is connected to the SCADA system of multiple wind turbine generators and the external environmental monitoring system in the wind farm via wired or wireless communication. The data acquisition module includes a unit data acquisition unit, an environmental data acquisition unit, and a data preprocessing and caching unit; the data analysis and processing module includes a spatial correlation analysis submodule, a temporal correlation analysis submodule, and a feature extraction and decision-making unit; the predictive control execution module includes an instruction receiving and parsing unit, a control signal output unit, a predictive action execution submodule, and a protective action execution submodule.

2. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The unit's data acquisition unit is communicatively connected to the wind turbine generator set and its adjacent units.

3. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The data acquisition unit of the unit communicates with the wind turbine generator and its adjacent units via either wired or wireless communication.

4. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The environmental data acquisition unit is installed in the wind farm control room or on top of the weather tower.

5. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The data preprocessing and caching unit is communicatively connected to the unit data acquisition unit and the environmental data acquisition unit, respectively.

6. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The feature extraction and decision-making unit is communicatively connected to the spatial correlation analysis submodule and the temporal correlation analysis submodule, respectively.

7. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The feature extraction and decision-making unit is communicatively connected to the spatial correlation analysis submodule and the temporal correlation analysis submodule, respectively.

8. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The instruction receiving and parsing unit and the data analysis and processing module are communicatively connected.

9. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The instruction receiving and parsing unit, control signal output unit, predictive action execution submodule, and protective action execution submodule are interconnected.

10. The intelligent system for power speed assessment of wind turbine generator sets as described in claim 1, characterized in that: The data acquisition module, data analysis and processing module, and predictive control execution module are sequentially connected in communication.

Images