Wind power generation operation monitoring and control system and method

Abstract

The invention discloses a wind power generation operation monitoring and control system and method, and belongs to the technical field of new energy power generation. The system comprises an analysis module, an optimization module, a control module and a feedback module. Wherein the analysis module is used for receiving operation data of wind power equipment, and realizing cross-domain data mapping, time sequence prediction and state evaluation based on a generative adversarial network and a Transform structure; the optimization module combines the prediction result and the equipment damage change, introduces uncertainty constraint and a multi-target weighting mechanism, and generates operation control parameters; the control module performs equipment operation control according to the parameters, and executes local emergency response under abnormal conditions; and the feedback module dynamically updates model parameters by using a control result, and optimizes a maintenance strategy through cloud reinforcement learning. The method can improve the operation stability, the operation and maintenance efficiency and the fault pre-judgment capability, and is suitable for the intelligent operation management of the wind power equipment.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy power generation, and in particular relates to a wind power generation operation monitoring and control system and method. Background Art

[0002] As a key component of clean and renewable energy, wind power generation has seen rapid global deployment in recent years. Wind turbines are typically located in wind farms with complex geographical environments and variable climatic conditions. Their operating conditions are susceptible to multiple factors, including wind speed fluctuations, temperature and humidity variations, and terrain disturbances. This results in highly nonlinear and uncertain operating conditions. In this context, operational status monitoring, health assessment, predictive control, and fault warning for wind turbine equipment are crucial for ensuring the safe, stable, and efficient operation of wind power systems.

[0003] Existing state monitoring and predictive control of wind turbines suffer from the following common challenges: Traditional data-driven models often rely on data from specific wind turbines or wind farms for training, lacking adaptability to multiple wind turbines and across wind farms. This makes it difficult to effectively handle differences in operating characteristics between wind turbines, impacting model transferability and generalizability. Some methods rely solely on static features or short-term windows for state assessment, ignoring long-term dependencies, seasonal fluctuations, and underlying trends in wind turbine operation. This results in limited prediction accuracy and anomaly detection. Fixed threshold or rule-driven anomaly identification methods struggle to adapt to complex environmental changes and lack mechanisms for dynamically adjusting thresholds based on prediction uncertainty and state assessment, leading to false positives and missed detections. Traditional operational control parameters are often optimized based on a single objective (such as optimal power generation efficiency), failing to fully consider multi-dimensional constraints and objectives such as equipment fatigue damage, load balancing, and uncertain disturbances, making it difficult to achieve system-level optimal operation. Some solutions focus solely on forward prediction and offline optimization, failing to establish a collaborative feedback learning path between the device and the cloud, and lacking the ability to update models and control strategies online as operations evolve.

[0004] Therefore, there is an urgent need to provide an intelligent monitoring and decision-making system that integrates cross-wind turbine operation data feature extraction, time series modeling in complex environments, dynamic anomaly detection, multi-objective robust control and end-cloud adaptive feedback optimization capabilities to improve the predictive operation and maintenance level and full life cycle operation efficiency of wind power equipment. Summary of the Invention

[0005] The present invention is proposed based on the above-mentioned needs, and aims to provide a wind power generation operation monitoring and control system and method with cross-device data adaptation, time series prediction and intelligent decision-making capabilities, so as to improve the overall wind farm operation and maintenance intelligence level and operation efficiency.

[0006] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0007] In one aspect, the present invention provides a wind power generation operation monitoring and control system, comprising:

[0008] An analysis module is configured to receive operating data of wind turbines and perform time series prediction, anomaly detection, and equipment status assessment on the operating data; wherein: the analysis module includes a domain adaptation submodule based on a generative adversarial network that introduces fault-sensitive feature preservation constraints, which is configured to map the operating data of a target wind turbine across domains into data that is consistent in distribution with the operating data of a reference wind turbine and retains fault-sensitive features; the analysis module includes a time series prediction submodule based on a Transformer decoder structure and combined with a dynamic attention allocation mechanism, which is configured to perform time series feature encoding, dynamically increase the weight of anomaly-related features during the decoding process, output a prediction result, and generate equipment status assessment information based on the prediction result;

[0009] The optimization module receives prediction results and equipment status assessment information, introduces uncertainty constraints into the multi-objective optimization solution, and dynamically adjusts the multi-objective weights according to the equipment fatigue damage change rate to solve and output the operation control parameters;

[0010] a control module, configured to receive the operation control parameters, perform operation control operations on the wind power generation equipment, and execute local control actions when the abnormality detection result indicates the presence of an emergency abnormality, and output execution result data;

[0011] The feedback module is used to update the analysis module and the optimization module based on the execution result data, and dynamically optimize the maintenance time window and maintenance resource scheduling based on reinforcement learning through the cloud maintenance decision module.

[0012] Preferably, the domain adaptation submodule adopts a cycle-consistent generative adversarial network that introduces fault-sensitive feature weighted constraints in the loss function, which is used to perform cross-domain mapping on the operating data of the target wind turbine and output time series data that is consistent with the distribution of the reference wind turbine operating data and retains fault-sensitive features.

[0013] Preferably, the time series prediction submodule receives the time series data and wind farm environmental condition vector output by the domain adaptation submodule, performs time series feature encoding on the time series data based on the model parameters pre-trained in the fields of energy, meteorology and environment and fine-tuned by migrating the historical data of the target wind farm, and in the decoding process, uses the dynamic attention allocation mechanism to increase the weight of the abnormal correlation feature, outputs the prediction result within the prediction time window, and the prediction result includes the generated power, bearing temperature rise and power curve deviation information, and generates equipment status assessment information based on the prediction result.

[0014] Preferably, the analysis module includes a dynamic threshold updating step when performing anomaly detection, and the dynamic threshold updating step includes:

[0015] Calculate the mean error between the predicted value and the measured value within the sliding time window;

[0016] Calculate the standard deviation of the error between the predicted value and the measured value in the same time window;

[0017] Perform weighted operations based on the prediction uncertainty measure, error mean, and error standard deviation output by the model to generate a new anomaly detection threshold;

[0018] The new anomaly detection threshold is compared with the real-time prediction value, an anomaly detection result is output, and the device status evaluation information is updated based on the anomaly detection result.

[0019] Preferably, the optimization module includes the following steps when solving the operation control parameters:

[0020] Receive the prediction results and corresponding equipment status assessment information output by the analysis module;

[0021] Introducing the confidence interval of the prediction result as an uncertainty constraint in the optimization solution, and using the equipment operating state parameters in the equipment state evaluation information as operating constraints;

[0022] Based on the uncertainty constraints and the equipment operation constraints, a multi-objective weighted optimization operation is performed on the power generation power, the unit load distribution and the equipment fatigue damage, and the weight coefficient of each optimization objective is dynamically adjusted according to the equipment fatigue damage change rate;

[0023] The operation control parameters corresponding to each wind turbine generator are output.

[0024] Preferably, the optimization module includes the following steps when calculating the operation control parameters of the entire wind farm:

[0025] Periodically receive prediction results and corresponding equipment status assessment information for all wind power generation equipment in the wind farm;

[0026] In the optimization solution, the confidence interval of the prediction result of each wind power generation equipment is used as an uncertainty constraint, and the equipment operation status parameter in the equipment status evaluation information is used as the equipment operation constraint;

[0027] The total power generation of the entire wind farm, the load distribution balance of the units, and the equipment fatigue damage index are introduced into the optimization target, and the weight coefficient of the optimization target of the entire wind farm is dynamically adjusted according to the change rate of the load distribution balance of each unit;

[0028] Output the operation control parameters corresponding to each wind power generation equipment in the entire wind farm.

[0029] Preferably, the control module comprises the following steps after executing the operation control parameters:

[0030] Receiving execution result data corresponding to the operation control parameters;

[0031] Calculating the deviation between the operation control parameter and the actual response of the device based on the execution result data;

[0032] When the deviation exceeds the preset abnormal threshold, the local control rule parameters are updated and new operation control parameters are generated;

[0033] The updated operation control parameters are applied to subsequent operation control operations.

[0034] Preferably, the feedback module is connected to a cloud maintenance decision module, and the cloud maintenance decision module includes the following steps:

[0035] Periodically receive long-term operation data and equipment status assessment information;

[0036] Calculate equipment performance degradation trends and failure probability based on long-term operating data;

[0037] Select maintenance time windows based on performance degradation trends and failure probability, and generate maintenance tasks;

[0038] Based on the execution results of the maintenance tasks, the maintenance time window and maintenance resource scheduling parameters are adaptively adjusted through reinforcement learning.

[0039] Preferably, the prediction result, the anomaly detection result and the device status assessment information are transmitted via a message queue, and the communication and feedback mechanism includes the following steps:

[0040] The message queue is managed based on the device identification, event type and timestamp. The optimization module and the control module obtain messages and perform corresponding operations according to the preset subscription rules, and dynamically adjust the subscription priority of the message according to the confidence interval of the prediction result;

[0041] After completing the control or optimization operation, the control module and the feedback module feed back the execution result data to the analysis module through the message queue for model parameter update. When the anomaly detection result exceeds the dynamic threshold, the update of the model parameter or anomaly detection threshold is triggered.

[0042] In another aspect, the present invention provides a wind power generation operation monitoring and control method, which is applied to the wind power generation operation monitoring and control system described above, and includes the following steps:

[0043] Step 1: Receive operating data of wind power generation equipment and perform time series prediction, anomaly detection and equipment status assessment on the operating data;

[0044] Step 2: Through the domain adaptation submodule, based on the adversarial generative network and introducing the fault-sensitive feature preservation constraint, the operating data of the target wind turbine is mapped to data with a distribution consistent with the reference wind turbine operating data and retaining the fault-sensitive features, and the prediction results, anomaly detection results and equipment status assessment information are output;

[0045] Step 3: receiving the prediction result and the device status evaluation information, solving the operation control parameters based on the confidence interval of the prediction result and the device status evaluation information, and outputting the operation control parameters;

[0046] Step 4: receiving the operation control parameters, performing operation control operations on the wind power generation equipment, and when an abnormality detection result indicates that an emergency abnormality exists, executing a local control action and outputting the execution result data;

[0047] Step 5: Receive the execution result data, and use the execution result data to update the model parameters or anomaly detection thresholds in the analysis module and the optimization module.

[0048] The beneficial effects of the present invention are as follows: the present invention realizes refined management of the operating status of wind power equipment through a modular architecture. Among them, the analysis module integrates the adversarial generative network and the Transformer structure, which can extract cross-domain time series features from multi-source heterogeneous data, realize fidelity modeling of fault-sensitive features and accurate state evaluation, and significantly improve the accuracy and robustness of anomaly identification and time series prediction. The optimization module introduces a dynamic weight mechanism of prediction confidence constraints and fatigue drive, coordinates power output, load distribution and fatigue accumulation control in multi-objective optimization, and improves the safety and operation efficiency of the system under complex working conditions. The control module triggers a local rapid response based on the anomaly detection results, has active safety protection capabilities, and can effectively deal with sudden abnormal risks. The feedback module realizes adaptive updating of the model based on the control execution results, and optimizes the maintenance and inspection rhythm and resource scheduling through cloud-based reinforcement learning, realizing the dynamic evolution of operation and maintenance strategies and full life cycle management of equipment. The overall system integrates prediction, optimization and control functions, with a clear closed-loop technical path and close coordination between modules. It has significant advantages such as improving the operational stability of wind power equipment, reducing operation and maintenance costs, and enhancing fault prediction capabilities. It has good engineering promotion value and industrial application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort. Among them: Figure 1Schematic diagram of the overall structure of the system of the present invention; Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0051] like Figure 1 FIG. 1 is an embodiment of the present invention, which provides a wind power generation operation monitoring and control system, including:

[0052] (1) Analysis module

[0053] Used to receive operating data of wind power generation equipment, perform time series prediction, anomaly detection and equipment status assessment on the operating data; wherein: the analysis module includes a domain adaptation submodule based on a generative adversarial network that introduces fault-sensitive feature preservation constraints, which is used to map the operating data of the target wind turbine across domains into data that is consistent with the distribution of the reference wind turbine operating data and retains fault-sensitive features; the analysis module includes a time series prediction submodule based on a Transformer decoder structure and combined with a dynamic attention allocation mechanism, which is used to perform time series feature encoding, dynamically increase the weight of anomaly-related features during the decoding process, output prediction results, and generate equipment status assessment information based on the prediction results;

[0054] Furthermore, to improve the generalization performance and sensitive feature retention of cross-domain data alignment, the domain adaptation submodule uses a cycle-consistent generative adversarial network that introduces fault-sensitive feature weighting constraints in the loss function to perform cross-domain mapping on the operating data of the target wind turbine and output time series data that is consistent with the reference wind turbine operating data distribution and retains fault-sensitive features. The loss function can be expressed as:

[0055] ;

[0056] in: is the standard adversarial loss, is the cycle consistency loss, is a KL divergence weighted constraint for the distribution difference of sensitive channels, which is in the form of:

[0057] ;

[0058] in: 、 The reference fan and the mapping fan are Probability distribution on fault-related feature dimensions, is the experience weighting factor, 、 It is a hyperparameter that can be set dynamically based on the performance of the validation set.

[0059] Specifically, the time series prediction submodule receives the time series data and wind farm environmental condition vector output by the domain adaptation submodule, performs time series feature encoding on the time series data based on the model parameters pre-trained in the fields of energy, meteorology and environment and fine-tuned by migrating the historical data of the target wind farm, and in the decoding process, uses the dynamic attention allocation mechanism to increase the weight of the abnormal correlation features, outputs the prediction results within the prediction time window, and the prediction results include the generated power, bearing temperature rise and power curve deviation information, and generates equipment status assessment information based on the prediction results.

[0060] In the attention allocation stage, an abnormal saliency enhancement factor is introduced to weight the standard attention mechanism, which is expressed as follows:

[0061] ;

[0062] in: For the standard attention score (calculated by the inner product of the query vector and the key vector), is an abnormality scoring factor extracted based on statistics such as prediction residuals and variance. To amplify the weight coefficient, it can be obtained through data-driven learning.

[0063] The analysis module includes a dynamic threshold updating step when performing anomaly detection, and the dynamic threshold updating step includes:

[0064] Calculate the mean error between the predicted value and the measured value within the sliding time window;

[0065] Calculate the standard deviation of the error between the predicted value and the measured value in the same time window;

[0066] Perform weighted operations based on the prediction uncertainty measure, error mean, and error standard deviation output by the model to generate a new anomaly detection threshold;

[0067] The calculation method of the new threshold can be expressed as:

[0068] ;

[0069] in: is the mean prediction error within the sliding window; is the standard deviation of error; uncertainty measures for the model's predicted outputs, such as those based on Bayesian output or Dropout estimates; 、 、 Weighting coefficient, satisfying , which can be adjusted dynamically according to the scene.

[0070] The new anomaly detection threshold is compared with the real-time prediction value, an anomaly detection result is output, and the device status evaluation information is updated based on the anomaly detection result.

[0071] (2) Optimization module

[0072] Used to receive prediction results and equipment status assessment information, introduce uncertainty constraints in the multi-objective optimization solution, and dynamically adjust the multi-objective weights according to the equipment fatigue damage change rate to solve and output operation control parameters; the optimization module preferably includes an operation submodule and a weight adjustment submodule to jointly achieve collaborative operation control optimization based on safety, economy and reliability.

[0073] Specifically, the optimization module includes the following steps when solving the operation control parameters:

[0074] First, the prediction results and corresponding equipment status assessment information output by the analysis module are received. The prediction results include the generated power, bearing temperature rise and power curve deviation in the future time window. The equipment status assessment information includes the fatigue life level, load variation index and operating temperature rise redundancy margin of the current unit. The prediction results are based on the decoding output of the large time series model and include corresponding confidence intervals to express the degree of uncertainty of the prediction results.

[0075] In the optimization solution, the confidence interval of the prediction result is introduced as an uncertainty constraint to limit the stability of the control solution within the range of the predicted value. The confidence constraint can be defined as:

[0076] ;

[0077] in: To predict the expected power generation, is the half-width of its confidence interval, is the confidence level threshold; at the same time, the equipment operating status parameters in the equipment status assessment information are used as operating constraint items, such as the maximum temperature rise threshold, the fatigue cumulative damage rate not exceeding the limit, etc., and are embedded in the optimization model in the form of inequalities to ensure that the optimization results meet the safe operation requirements of the wind turbine.

[0078] In the construction of the optimization objective function, the following multi-objective weighted form is adopted:

[0079] ;

[0080] in: The predicted power generation under the current optimization control strategy; is the unit load fluctuation; is the equipment fatigue damage index; weight coefficient 、 、 It can be dynamically adjusted according to the current equipment fatigue damage change rate to meet the following requirements: ;

[0081] To achieve dynamic weight adjustment, the weight adjustment submodule introduces the following update rules:

[0082] ;

[0083] in: This is a weighted sensitivity coefficient used to enhance the impact of fatigue damage changes on control strategies. When the equipment enters the fatigue acceleration phase, the weight of safety-related objectives is automatically increased.

[0084] In the calculation of the control parameters of the entire wind farm, the optimization module is further expanded to a centralized scheduling mode, which periodically receives the prediction results and status assessment information of all wind power generation equipment in the wind farm; at this time, the following is introduced in the optimization objective:

[0085] Maximize the total power generation of the entire wind farm;

[0086] Minimize the load distribution balance of units within the wind farm (such as the difference between maximum and minimum loads);

[0087] Goals such as controlling the average fatigue index of the entire field.

[0088] The overall objective function is:

[0089] ;

[0090] The corresponding full wind farm operation control parameters are output through distributed solution or cloud-based centralized optimization strategy.

[0091] Finally, the optimization module outputs the operating control parameters corresponding to the wind power generation equipment for execution by the control module, thereby ensuring that the power generation efficiency is improved and the equipment life cycle is extended while ensuring the safety of the equipment.

[0092] (3) Control module

[0093] for receiving the operation control parameters, performing operation control operations on the wind power generation equipment, and executing local control actions when the abnormality detection result indicates that an emergency abnormality exists, and outputting execution result data;

[0094] Specifically, after executing the operation control parameters, the control module includes the following steps:

[0095] Receiving execution result data corresponding to the operation control parameters;

[0096] The execution result data includes operational feedback information of the wind turbine, such as active power feedback value, rotation speed, pitch angle, operating status of the cooling system, etc. The data is collected by underlying sensors and transmitted to the control module in real time through the industrial field bus to ensure the accuracy of the control response evaluation.

[0097] Calculating the deviation between the operation control parameter and the actual response of the device based on the execution result data;

[0098] The deviation calculation adopts a dynamic error analysis method based on a sliding time window, compares the target control value with the real-time feedback value, and establishes a multi-layer tolerance evaluation interval in combination with the environmental disturbance factor to dynamically identify the degree of operation deviation.

[0099] When the deviation exceeds the preset abnormal threshold, the local control rule parameters are updated and new operation control parameters are generated;

[0100] The update of local control rule parameters includes adjustments to proportional-integral-derivative (PID) parameters, adaptive lag compensation coefficients or fuzzy control weights. The control module automatically selects the adjustment strategy based on the deviation type to enhance the response capability to sudden anomalies and improve control stability.

[0101] The updated operation control parameters are applied to subsequent operation control operations.

[0102] This parameter will be incorporated into the next control cycle through a real-time scheduling mechanism, covering key aspects of the wind turbine's pitch control, excitation, and power output, while continuously monitoring its feedback effects. Furthermore, the control module will record control instructions, feedback results, and deviation information and push them to the feedback module for subsequent model updates and status evaluation.

[0103] (4) Feedback module

[0104] It is used to update the analysis module and optimization module based on the execution result data, and dynamically optimize the maintenance time window and maintenance resource scheduling based on reinforcement learning through the cloud maintenance decision module.

[0105] The feedback module is connected to a cloud maintenance decision module, and the cloud maintenance decision module includes the following steps:

[0106] Periodically receive long-term operation data and equipment status assessment information;

[0107] Calculate equipment performance degradation trends and failure probability based on long-term operating data;

[0108] Among them, the probability of failure can be modeled based on the exponential decay model, and the formula is as follows:

[0109] ;

[0110] in: Indicates time The predicted failure probability of the equipment at the moment; is the prior historical failure probability; is the performance trend attenuation factor; For the device at time performance degradation rate.

[0111] Select maintenance time windows based on performance degradation trends and failure probability, and generate maintenance tasks;

[0112] When generating a maintenance task scheduling plan, the feedback module can construct the scheduling objective function as follows:

[0113] ;

[0114] in: Scheduling objective function for maintenance tasks; For the The time cost corresponding to each maintenance task; For the task The urgency weighting coefficient of The total number of maintenance tasks that need to be scheduled.

[0115] Based on the execution results of the maintenance tasks, the maintenance time window and maintenance resource scheduling parameters are adaptively adjusted through reinforcement learning.

[0116] In reinforcement learning scheduling optimization, the system uses the following reward function for strategy training:

[0117] ;

[0118] in: Refers to the improvement in equipment reliability after performing maintenance tasks; The resource overhead corresponding to the current scheduling; 、 A proportionality factor that adjusts the trade-off between reliability gain and cost.

[0119] Through the above methods, the feedback module can dynamically update the analysis and optimization modules based on execution data, and realize intelligent scheduling of maintenance resources based on long-term trend prediction, thereby effectively improving the operation and maintenance efficiency and safety of wind power equipment throughout its life cycle.

[0120] The prediction results, the anomaly detection results, and the device status assessment information are transmitted via a message queue. The communication and feedback mechanism includes the following steps:

[0121] The message queue is managed based on the device identification, event type and timestamp. The optimization module and the control module obtain messages and perform corresponding operations according to the preset subscription rules, and dynamically adjust the subscription priority of the message according to the confidence interval of the prediction result;

[0122] After completing the control or optimization operation, the control module and the feedback module feed back the execution result data to the analysis module through the message queue for model parameter update. When the anomaly detection result exceeds the dynamic threshold, the update of the model parameter or anomaly detection threshold is triggered.

[0123] like Figure 2 The above is another embodiment of the present invention, which provides a wind power generation operation monitoring and control method, which is applied to the wind power generation operation monitoring and control system described above, including the following steps:

[0124] Step 1: Receive operating data of wind power generation equipment and perform time series prediction, anomaly detection and equipment status assessment on the operating data;

[0125] Step 2: Through the domain adaptation submodule, based on the adversarial generative network and introducing the fault-sensitive feature preservation constraint, the operating data of the target wind turbine is mapped to data with a distribution consistent with the reference wind turbine operating data and retaining the fault-sensitive features, and the prediction results, anomaly detection results and equipment status assessment information are output;

[0126] Step 3: receiving the prediction result and the device status evaluation information, solving the operation control parameters based on the confidence interval of the prediction result and the device status evaluation information, and outputting the operation control parameters;

[0127] Step 4: receiving the operation control parameters, performing operation control operations on the wind power generation equipment, and when an abnormality detection result indicates that an emergency abnormality exists, executing a local control action and outputting the execution result data;

[0128] Step 5: Receive the execution result data, and use the execution result data to update the model parameters or anomaly detection thresholds in the analysis module and the optimization module.

[0129] In summary, the wind power generation operation monitoring and control system and method provided by the present invention builds an end-cloud collaborative overall solution around the four core links of wind power equipment operating status perception, intelligent prediction, optimization control and adaptive feedback. The system introduces a number of advanced technologies in data cross-domain adaptation, time series modeling, anomaly identification and operation decision-making, including the introduction of a generative adversarial network with fault-sensitive feature preservation constraints, a Transformer decoder structure combined with a dynamic attention mechanism, a multi-objective optimization strategy that integrates uncertainty and fatigue driving factors, and a maintenance resource scheduling method based on reinforcement learning, forming a complete closed-loop intelligent operation and maintenance system. The system not only significantly improves the wind farm's recognition accuracy of the operating status and the reliability of the control response, but also has good self-learning and generalization capabilities, and is suitable for wind power scenarios of various scales and working conditions. This technology has significant advantages in ensuring power generation efficiency, extending equipment life and reducing operation and maintenance costs, and has broad promotion value and application prospects.

[0130] In the description of this specification, the reference terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0131] Any process or method description in the flowchart or otherwise described herein can be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process. The scope of the preferred embodiments of the present application includes additional implementations in which the functions may be performed in a different order than shown or discussed, including in a substantially simultaneous manner or in a reverse order depending on the functions involved.

[0132] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. Wind power generation operation monitoring and control system, characterized in that: include: An analysis module is configured to receive operating data of wind turbines and perform time series prediction, anomaly detection, and equipment status assessment on the operating data; wherein: the analysis module includes a domain adaptation submodule based on a generative adversarial network that introduces fault-sensitive feature preservation constraints, which is configured to map the operating data of a target wind turbine across domains into data that is consistent in distribution with the operating data of a reference wind turbine and retains fault-sensitive features; the analysis module includes a time series prediction submodule based on a Transformer decoder structure and combined with a dynamic attention allocation mechanism, which is configured to perform time series feature encoding, dynamically increase the weight of anomaly-related features during the decoding process, output a prediction result, and generate equipment status assessment information based on the prediction result; The optimization module is used to receive prediction results and equipment status assessment information, introduce uncertainty constraints into the multi-objective optimization solution, dynamically adjust the multi-objective weights according to the equipment fatigue damage change rate, and solve and output the operation control parameters; a control module configured to receive the operation control parameters, perform operation control operations on the wind power generation equipment, and execute local control actions when an abnormality detection result indicates the presence of an emergency abnormality, and output execution result data; The feedback module is used to update the analysis module and the optimization module based on the execution result data, and dynamically optimize the maintenance time window and maintenance resource scheduling based on reinforcement learning through the cloud maintenance decision module.

2. The wind power generation operation monitoring and control system according to claim 1, characterized in that: The domain adaptation submodule adopts a cycle-consistent generative adversarial network that introduces fault-sensitive feature weighted constraints in the loss function to perform cross-domain mapping on the operating data of the target wind turbine and output time series data that is consistent with the distribution of the reference wind turbine operating data and retains fault-sensitive features.

3. The wind power generation operation monitoring and control system according to claim 1, characterized in that: The time series prediction submodule receives the time series data and wind farm environmental condition vector output by the domain adaptation submodule, performs time series feature encoding on the time series data based on model parameters pre-trained in the fields of energy, meteorology, and environment and fine-tuned by migrating historical data of the target wind farm, and uses a dynamic attention allocation mechanism to increase the weight of abnormal correlation features during the decoding process, outputs the prediction results within the prediction time window, and the prediction results include generated power, bearing temperature rise, and power curve deviation information, and generates equipment status assessment information based on the prediction results.

4. The wind power generation operation monitoring and control system according to claim 3, characterized in that: The analysis module includes a dynamic threshold updating step when performing anomaly detection, and the dynamic threshold updating step includes: Calculate the mean error between the predicted value and the measured value within the sliding time window; Calculate the standard deviation of the error between the predicted value and the measured value in the same time window; Perform weighted operations based on the prediction uncertainty measure, error mean, and error standard deviation output by the model to generate a new anomaly detection threshold; The new anomaly detection threshold is compared with the real-time prediction value, an anomaly detection result is output, and the device status evaluation information is updated based on the anomaly detection result.

5. The wind power generation operation monitoring and control system according to claim 1, characterized in that: The optimization module includes the following steps when solving the operation control parameters: Receive the prediction results and corresponding equipment status assessment information output by the analysis module; Introducing the confidence interval of the prediction result as an uncertainty constraint in the optimization solution, and using the equipment operating state parameters in the equipment state evaluation information as operating constraints; Based on the uncertainty constraints and the equipment operation constraints, a multi-objective weighted optimization operation is performed on the power generation power, the unit load distribution and the equipment fatigue damage, and the weight coefficient of each optimization objective is dynamically adjusted according to the equipment fatigue damage change rate; The operation control parameters corresponding to each wind turbine generator are output.

6. The wind power generation operation monitoring and control system according to claim 5, characterized in that: The optimization module includes the following steps when calculating the operation control parameters of the entire wind farm: Periodically receive prediction results and corresponding equipment status assessment information for all wind power generation equipment in the wind farm; In the optimization solution, the confidence interval of the prediction result of each wind power generation equipment is used as an uncertainty constraint, and the equipment operation status parameter in the equipment status evaluation information is used as the equipment operation constraint; The total power generation of the entire wind farm, the load distribution balance of the units, and the equipment fatigue damage index are introduced into the optimization target, and the weight coefficient of the optimization target of the entire wind farm is dynamically adjusted according to the change rate of the load distribution balance of each unit; Output the operation control parameters corresponding to each wind power generation equipment in the entire wind farm.

7. The wind power generation operation monitoring and control system according to claim 1, characterized in that: After receiving the operation control parameters, the control module includes the following steps: Receiving execution result data corresponding to the operation control parameters; Calculating the deviation between the operation control parameter and the actual response of the device based on the execution result data; When the deviation exceeds the preset abnormal threshold, the local control rule parameters are updated and new operation control parameters are generated; The updated operation control parameters are applied to subsequent operation control operations.

8. The wind power generation operation monitoring and control system according to claim 1, characterized in that: The feedback module is connected to a cloud maintenance decision module, and the cloud maintenance decision module includes the following steps: Periodically receive long-term operation data and equipment status assessment information; Calculate equipment performance degradation trends and failure probability based on long-term operating data; Select maintenance time windows based on performance degradation trends and failure probability, and generate maintenance tasks; Based on the execution results of the maintenance tasks, the maintenance time window and maintenance resource scheduling parameters are adaptively adjusted through reinforcement learning.

9. The wind power generation operation monitoring and control system according to claim 1, characterized in that: The prediction results, the anomaly detection results, and the device status assessment information are transmitted via a message queue. The communication and feedback mechanism includes the following steps: The message queue is managed based on the device identification, event type and timestamp. The optimization module and the control module obtain messages and perform corresponding operations according to the preset subscription rules, and dynamically adjust the subscription priority of the message according to the confidence interval of the prediction result; After completing the control or optimization operation, the control module and the feedback module feed back the execution result data to the analysis module through the message queue for model parameter update. When the anomaly detection result exceeds the dynamic threshold, the update of the model parameter or anomaly detection threshold is triggered.

10. A wind power generation operation monitoring and control method, applied to the wind power generation operation monitoring and control system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: Receive operating data of wind power generation equipment and perform time series prediction, anomaly detection and equipment status assessment on the operating data; Step 2: Through the domain adaptation submodule, based on the adversarial generative network and introducing the fault-sensitive feature preservation constraint, the operating data of the target wind turbine is mapped to data with a distribution consistent with the reference wind turbine operating data and retaining the fault-sensitive features, and the prediction results, anomaly detection results and equipment status assessment information are output; Step 3: receiving the prediction result and the device status evaluation information, solving the operation control parameters based on the confidence interval of the prediction result and the device status evaluation information, and outputting the operation control parameters; Step 4: receiving the operation control parameters, performing operation control operations on the wind power generation equipment, and when an abnormality detection result indicates that an emergency abnormality exists, executing a local control action and outputting the execution result data; Step 5: Receive the execution result data, and use the execution result data to update the model parameters or anomaly detection thresholds in the analysis module and the optimization module.

Images