A Wind Power Control Method and System Based on Intelligent Robots

Abstract

The present invention discloses a wind power control method and system based on an intelligent robot. The method includes: during the operation of wind power equipment, acquiring historical wind power data; performing a feature extraction operation according to the historical wind power data to obtain data items corresponding to the historical wind power data; performing a data unit quantization operation according to the data items to obtain data vectors; performing a correlation analysis operation according to the data vectors to obtain an abnormal data set corresponding to the wind power equipment; performing a feature vector extraction operation according to the abnormal data set to obtain potential feature vectors; performing a state change metric calculation operation according to the potential feature vectors to obtain a state change metric; performing a risk value calculation operation according to the state change metric to obtain a state change risk value; and performing an alarm reminder operation according to the state change risk value to monitor the wind power equipment. The present invention can improve the accuracy of evaluating the risk situation of wind power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a wind power control method and system based on an intelligent robot. Background Art

[0002] At present, with the development of new energy, the demand for wind power equipment has been continuously rising. During the operation of wind power equipment, its operating state changes due to the influence of climate, environment and mechanical stress. Therefore, it is necessary to comprehensively monitor wind turbines to ensure the safe operation of wind power equipment. The traditional wind power monitoring system collects the state data of wind turbines through multiple sensors to achieve real-time monitoring and fault detection of different components and different moments of wind power equipment. The wind power monitoring system collects and analyzes various information through sensors to monitor the operating state of wind turbines. It can monitor various parameters, including temperature, pressure, rotational speed, vibration and sound, etc., to help detect potential problems and faults. However, the traditional wind power monitoring system has the following defects: The traditional wind power monitoring system mainly relies on hardware devices, and these devices may fail or malfunction, resulting in unreliable data or inability to process data in a timely manner. In addition, the traditional method requires a large number of hardware devices and complex software systems, which increases the cost and complexity of the system. The traditional wind power monitoring system mainly relies on preset thresholds and rules and cannot adaptively adjust the monitoring strategy. When the operating environment and conditions of wind power equipment change, these fixed rules are no longer applicable, resulting in false alarms or missed alarms. At the same time, with the development of artificial intelligence technology, intelligent robot technology has been widely applied in the wind power field. By establishing an intelligent wind power monitoring platform, the performance state of wind power equipment is monitored in real time, and predictive maintenance is carried out on wind power equipment. The intelligent wind power monitoring platform can receive data from various sensors in real time, analyze and give early warnings on the performance of wind power equipment. By applying historical data and artificial intelligence algorithms, the platform can not only identify and process common wind power equipment alarm information in real time, but also predict and give early warnings on potential degradation of wind power equipment performance in advance, thus greatly reducing the downtime and maintenance cost of wind power equipment.

[0003] Since there are many significant defects in the prior art. Firstly, it completely ignores the differences in potential risks at different times. In reality, the environmental, personnel and equipment conditions at different times are constantly changing, and the characteristics, degrees and probabilities of risks also change accordingly. However, the prior art uses a unified model for evaluation in the face of all changes, resulting in results deviating from reality. Secondly, it is extremely unreasonable to define the importance of the collected data using the same index. The data sources are diverse and their natures are different. Data such as key equipment parameters is crucial, while ordinary environmental data is relatively less important. However, the prior art treats them equally, resulting in the key data not playing a prominent role, and the secondary data interfering with the judgment. Eventually, the risk assessment results are unsatisfactory and difficult to be effectively applied.

[0004] Therefore, although the prior art can monitor the performance degradation of wind power equipment in real time, the existing defects greatly reduce the accuracy of the assessment of the risk situation of wind power equipment, making it difficult to meet the requirements of accurate risk control in practical applications. Summary of the Invention

[0005] The present invention provides a wind power control method and system based on an intelligent robot to achieve real-time monitoring of the performance degradation of wind power equipment and improve the accuracy of assessing the risk situation of wind power equipment.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a wind power control method based on an intelligent robot, including:

[0007] During the operation of the wind power equipment, historical wind power data is obtained, where the historical wind power data is obtained by an intelligent robot;

[0008] According to the historical wind power data, a feature extraction operation is performed to obtain data items corresponding to the historical wind power data;

[0009] According to the data items, a data unit vectorization operation is performed to obtain a data vector;

[0010] According to the data vector, a correlation analysis operation is performed to obtain an abnormal data set corresponding to the wind power equipment;

[0011] According to the abnormal data set, a feature vector extraction operation is performed to obtain a potential feature vector;

[0012] According to the potential feature vector, a state change metric calculation operation is performed to obtain a state change metric;

[0013] According to the state change metric, a risk value calculation operation is performed to obtain a state change risk value;

[0014] According to the state change risk value, an alarm reminder operation is performed to monitor the wind power equipment.

[0015] As an optional implementation manner, after obtaining the historical wind power data during the operation of the wind power equipment, the method further includes:

[0016] According to the historical wind power data, a data cleaning operation is performed to obtain cleaned data;

[0017] According to the cleaned data, a preprocessing operation is performed to obtain preprocessed data, where the preprocessing operation formula is:

[0018] ,

[0019] is the preprocessed data, is the cleaned data, is the mean of the cleaned data, is the standard deviation of the cleaned data.

[0020] As an alternative implementation, the operation of extracting features from the historical wind power data to obtain the data items of the historical wind power data includes:

[0021] The formula for the feature extraction operation is:

[0022]

[0023] where, is the data item at time t, is the total number of data features extracted from the preprocessed data, is the weight coefficient of the th data feature, is the value of the th data feature in the preprocessed data at time is the th preprocessed data, is a specific time .

[0024] As an alternative implementation, the operation of vectorizing data units according to the data items to obtain data vectors includes:

[0025] The data unit vectorization operation is to input the obtained data items into a clustering model, where the clustering model uses the K-means algorithm and includes normal features, fluctuation features, and abnormal features;

[0026] The data vectors include abnormal feature vectors, fluctuation feature vectors, and normal feature vectors.

[0027] As an alternative implementation, the operation of performing correlation analysis on the data vectors to obtain an abnormal data set corresponding to the wind power equipment includes:

[0028] Performing a time distance calculation operation on the data vectors to obtain the time distance corresponding to the data vectors, where the time distance calculation formula is:

[0029] ,

[0030] In the formula, and are two different times, is the data vector at The time distance from is and from is the number of dimensions of the time attribute, where are two data vectors in the

[0031] According to the time distance, a similarity calculation operation is performed to obtain the correlation of the data vectors, where the similarity calculation formula is:

[0032] ,

[0033] In the formula, is a preset parameter for controlling the attenuation speed, is the correlation, and are two different moments;

[0034] According to the correlation, a correlation threshold comparison operation is performed to obtain the abnormal data set, where the correlation threshold comparison includes if > v, then it belongs to the abnormal data set, where v is a preset correlation threshold, is the correlation.

[0035] As an optional implementation manner, the operation of calculating the state change metric according to the potential feature vector to obtain the state change metric includes:

[0036] The state change metric formula is:

[0037] ,

[0038] where , , is the state change metric, is the average state change index, is the potential feature vector at time , is the potential feature vector at time , is a preset coefficient, and are time variables, is the state change metric at time.

[0039] As an optional implementation manner, the operation of calculating the risk value according to the state change metric to obtain the state change risk value includes:

[0040] The risk value calculation formula is:

[0041]

[0042] Among them, is the risk value of the state change, z is a preset control coefficient, is a preset control factor, , is the state change measurement index at time is the state change measurement index at time

[0043] Second, the present invention provides a wind power control method based on an intelligent robot, including:

[0044] A data acquisition module, which acquires historical wind power data during the operation of wind power equipment, where the historical wind power data is obtained by an intelligent robot;

[0045] A feature extraction module, which is used to perform feature extraction operations according to the historical wind power data to obtain data items corresponding to the historical wind power data;

[0046] A vector conversion module, which is used to perform data unit vectorization operations according to the data items to obtain data vectors;

[0047] A correlation analysis module, which is used to perform correlation analysis operations according to the data vectors to obtain an abnormal data set corresponding to the wind power equipment;

[0048] A feature vector extraction module, which is used to perform feature vector extraction operations according to the abnormal data set to obtain potential feature vectors;

[0049] A state change measurement calculation module, which is used to perform state change measurement index calculation operations according to the potential feature vectors to obtain state change measurement indexes;

[0050] A risk value calculation module, which is used to perform risk value calculation operations according to the state change measurement indexes to obtain state change risk values;

[0051] A user reminder module, which is used to perform alarm reminder operations according to the state change risk values to monitor the wind power equipment.

[0052] Third, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a wind power control method based on an intelligent robot as described above.

[0053] Fourthly, the present invention also provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute a wind power control method based on an intelligent robot as described above.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention provides a wind power control method based on an intelligent robot, including: during the operation of wind power equipment, obtaining historical wind power data, where the historical wind power data is obtained by an intelligent robot; performing a feature extraction operation according to the historical wind power data to obtain data items of the historical wind power data; performing a data unit quantization operation according to the data items to obtain a vector; performing a correlation analysis operation according to the data vector to obtain an abnormal data set of the wind power equipment; performing a feature vector extraction operation according to the abnormal data set to obtain a potential feature vector; performing a state change metric calculation operation according to the potential feature vector to obtain a state change metric; performing a risk value calculation operation according to the state change metric to obtain a state change risk value; and performing an alarm reminder operation according to the state change risk value to monitor the wind power equipment.

[0056] In the present invention, the wind power control method based on the intelligent robot has the following advantages: First, the integrity and accuracy of the wind power equipment data are improved through data preprocessing. Second, key information is effectively extracted to provide a basis for monitoring. Third, cluster analysis and neural networks are used to accurately identify abnormal equipment states and related information, thereby improving operational safety and reliability. Fourth, real-time monitoring is achieved to promptly handle potential problems and avoid losses caused by equipment failures. The specific invention process of the present invention is as follows: During the operation of the wind power equipment, the relevant historical wind power data must first be obtained. Next, feature extraction is performed on the acquired historical wind power data to obtain data items. Subsequently, the obtained data items are vectorized to convert them into vector form. Afterwards, a correlation analysis operation is performed based on this vector to obtain an abnormal data set of the wind power equipment. Feature vectors are further extracted from the abnormal data set to obtain potential feature vectors. Subsequently, a state change metric is calculated based on the potential feature vector. Then, a risk value is calculated based on the calculated state change metric. Finally, alarm reminders are performed based on the risk value to achieve comprehensive and effective monitoring of wind power equipment, ensure that wind power equipment can operate safely, promptly discover and handle problems, and ensure the normal operation of the entire wind power system. According to the above method, wind power control based on intelligent robots can not only achieve real-time monitoring of wind power equipment performance degradation, but also improve the accuracy of assessing the risk of wind power equipment, ensure that wind power equipment can operate stably and safely, ensure that staff can promptly discover and handle problems, and ensure the normal operation of the entire wind power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of a wind power control method based on an intelligent robot provided by an embodiment of the present invention;

[0058] Figure 2 It is a structural schematic diagram of a wind power control system based on an intelligent robot provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] It should be noted that with the development of new energy, the demand for wind power equipment has been continuously rising. During the operation of wind power equipment, its operating state changes due to the influence of climate, environment and mechanical stress. Therefore, it is necessary to comprehensively monitor wind turbines to ensure the safe operation of wind power equipment. The traditional wind power monitoring system collects the state data of wind turbines through multiple sensors to achieve real-time monitoring and fault detection of different components and different moments of wind power equipment. The wind power monitoring system collects and analyzes various information through sensors to monitor the operating state of wind turbines. It can monitor various parameters, including temperature, pressure, speed, vibration and sound, etc., to help detect potential problems and faults. However, the traditional wind power monitoring system has the following defects: The traditional wind power monitoring system mainly relies on hardware devices, and these devices may fail or malfunction, resulting in unreliable data or inability to process data in a timely manner. In addition, the traditional method requires a large number of hardware devices and complex software systems, which increases the cost and complexity of the system. The traditional wind power monitoring system mainly relies on preset thresholds and rules and cannot adaptively adjust the monitoring strategy. When the operating environment and conditions of wind power equipment change, these fixed rules are no longer applicable, resulting in false alarms or missed alarms. At the same time, with the development of artificial intelligence technology, intelligent robot technology has been widely used in the wind power field. By establishing an intelligent wind power monitoring platform, the performance state of wind power equipment is monitored in real time, and predictive maintenance is carried out on wind power equipment. The intelligent wind power monitoring platform can receive the data from each sensor in real time, analyze and give early warnings on the performance of wind power equipment. Through the application of historical data and artificial intelligence algorithms, this platform can not only identify and process common wind power equipment alarm information in real time, but also predict and give early warnings on potential performance degradation of wind power equipment in advance, thus greatly reducing the downtime and maintenance cost of wind power equipment. However, there are many significant defects in the existing technology. Firstly, it completely ignores the differences in potential risks at different times. In practice, the environmental, personnel and equipment conditions at different times are constantly changing, and the characteristics, degrees and probabilities of risks also change accordingly. However, the existing technology uses a unified model to evaluate with a fixed approach, resulting in results deviating from reality. Secondly, it is extremely unreasonable to define the importance of the collected data using the same index. The data sources are diverse and their natures are different. Data such as key equipment parameters is crucial, while ordinary environmental data is relatively less important. However, the existing technology treats them equally, resulting in the key data not playing a prominent role, and the secondary data interfering with the judgment. Eventually, the risk assessment results are unsatisfactory and difficult to be effectively applied. Therefore, although the existing technology can monitor the performance degradation of wind power equipment in real time, the existing defects greatly reduce the accuracy of the risk assessment of wind power equipment and are difficult to meet the demand for accurate risk control in practical applications.

[0061] In view of the above problems, referring to Figure 1 , the first embodiment of the present invention provides a wind power control method based on an intelligent robot, including the following steps:

[0062] S11. During the operation of the wind power equipment, obtain historical wind power data, where the historical wind power data is obtained by an intelligent robot.

[0063] S12. According to the historical wind power data, perform a feature extraction operation to obtain data items corresponding to the historical wind power data.

[0064] S13. According to the data items, perform a data unit vectorization operation to obtain a data vector.

[0065] S14. According to the data vector, perform a correlation analysis operation to obtain an abnormal data set corresponding to the wind power equipment.

[0066] S15. According to the abnormal data set, perform a feature vector extraction operation to obtain a potential feature vector.

[0067] S16. According to the potential feature vector, perform a state change metric calculation operation to obtain a state change metric.

[0068] S17. According to the state change metric, perform a risk value calculation operation to obtain a state change risk value.

[0069] S18. According to the state change risk value, perform an alarm reminder operation to monitor the wind power equipment.

[0070] In step S11, during the operation of the wind power equipment, obtain historical wind power data, where the historical wind power data is obtained by an intelligent robot.

[0071] In one implementation, after obtaining the historical wind power data during the operation of the wind power equipment, it further includes: according to the historical wind power data, perform a data cleaning operation to obtain cleaned data; according to the cleaned data, perform a preprocessing operation to obtain preprocessed data, where the preprocessing operation formula is:

[0072] ,

[0073] is the preprocessed data, is the cleaned data, is the mean of the cleaned data, is the standard deviation of the cleaned data.

[0074] In the above embodiments, historical wind power data such as wind speed, wind direction, and power generation are collected in real time by intelligent robots and stored in the central control system. After being processed, these data are used to analyze the operation efficiency and health status of the equipment, so as to optimize the operation strategy, improve the power generation efficiency, reduce the maintenance cost, and enhance the reliability of the wind farm. In this way, the data of wind power equipment can be effectively monitored, providing a data basis for evaluating the risk situation of wind power equipment.

[0075] Specifically, based on the wind power equipment working status monitoring platform, data collection work for wind power equipment is carried out. The collected data covers multiple key dimensions such as the temperature, voltage, current, and vibration data of the wind power equipment. Statistical methods are used to screen the collected data, aiming to identify and retain valid data samples. Subsequently, the collected data is preprocessed. Specifically, the principal component analysis (PCA) method is used to normalize the data to reduce the data correlation, so as to provide a more reasonable data basis for subsequent in-depth data analysis and application, and ensure the accuracy and reliability of the evaluation and monitoring of the wind power equipment working status.

[0076] It should also be specifically noted that during the operation of wind power equipment, the acquisition of historical wind power data mainly relies on the cooperation of intelligent robots and sensors. The intelligent robot can automatically patrol in the wind farm and is equipped with multiple sensors to monitor the key parameters of the equipment, such as temperature, voltage, current, and vibration data. For example, a certain type of intelligent robot can be set to automatically go to each wind turbine at fixed times manually. It uses the built-in temperature sensor to monitor the operating temperature of the equipment, and at the same time uses voltage and current sensors to record the power output. On this basis, the vibration sensor configured on the intelligent robot itself can also monitor the vibration level of the wind turbine in real time, so as to detect potential fault signs. The collected data will be transmitted to the central control system through a wireless network for analysis and storage. These historical data obtained by the intelligent robot not only help users evaluate the operation status of the equipment, but also improve the accuracy of the risk situation assessment of wind power equipment through data trend analysis.

[0077] In step S12, according to the historical wind power data, a feature extraction operation is performed to obtain data items corresponding to the historical wind power data.

[0078] In one implementation, the formula for the feature extraction operation is:

[0079]

[0080] where is the data item at time t, is the total number of data features extracted from the preprocessed data, is the weight coefficient of the th data feature, is the value of the th data feature in the preprocessed data at time . is the th preprocessed data, is a specific time .

[0081] In the above embodiment, feature extraction operations are performed based on historical wind power data, aiming to extract key information from the original data for subsequent analysis. This process first cleans and standardizes the data through data preprocessing, then selects features related to wind power generation, such as wind speed, wind direction, and air pressure, and then performs feature transformation and combination to generate new data items. These extracted data items can more accurately reflect the operation law of wind power, provide high-quality input for subsequent prediction models, thereby improving the prediction accuracy and reliability of wind power generation, and optimizing power dispatching.

[0082] In step S13, according to the data items, a data unit vectorization operation is performed to obtain a data vector.

[0083] In one implementation, the data unit vectorization operation is to input the obtained data items into a clustering model, where the clustering model uses the K-means algorithm and includes normal features, fluctuation features, and abnormal features; the data vector includes an abnormal feature vector, a fluctuation feature vector, and a normal feature vector.

[0084] In the above embodiment, key information is extracted from the original data for subsequent analysis. This process first performs preprocessing cleaning and standardization operations on the collected data, then selects features related to wind power generation, such as wind speed, wind direction, and air pressure, and then performs feature transformation and combination to generate new data items. These extracted data items can more accurately reflect the operation law of wind power, provide high-quality input for subsequent prediction models, thereby improving the risk prediction accuracy and reliability of wind power equipment.

[0085] Exemplarily, the clustering model uses the K-means algorithm, with K of the preset clustering algorithm being 3, divided into normal features, fluctuation features, and abnormal features. The clustering model calculates the similarity of the features of the data items of each unit, takes the M items of data at the same time as a unit, and obtains N vectors S, where the vectors include an abnormal feature vector, a fluctuation feature vector, and a normal feature vector.

[0086] In step S14, according to the data vector, a correlation analysis operation is performed to obtain an abnormal data set corresponding to the wind power equipment.

[0087] In the above embodiments, a correlation analysis operation is performed based on the data vectors, aiming to identify abnormal data sets related to the operating state of wind power equipment. By calculating the correlation between data vectors, potential relationships between different data points can be discovered. This analysis helps identify specific data patterns related to equipment performance degradation or failures, and then extract abnormal data sets. This process contributes to timely detection of potential equipment problems, optimization of maintenance strategies, improvement of equipment operating efficiency and reliability, thereby reducing downtime and maintenance costs.

[0088] In step S15, based on the abnormal data set, an operation of extracting feature vectors is performed to obtain potential feature vectors.

[0089] In the above embodiments, these abnormal data sets are obtained from the abnormal data sets resulting from performing a correlation analysis operation on the real-time monitoring data and historical operation data collected during the operation of wind power equipment. By extracting feature vectors, real-time monitoring of equipment performance changes can be achieved, abnormal situations can be detected in a timely manner, thereby improving the accuracy of risk assessment of wind power equipment.

[0090] In step S16, based on the potential feature vectors, an operation of calculating state change measurement indicators is performed to obtain state change measurement indicators.

[0091] In one implementation, the state change measurement formula is:

[0092] ,

[0093] where, , , is the state change measurement indicator, is the average state change indicator, is the potential feature vector at time , is the potential feature vector at time , is a preset coefficient, and are time variables, is the state change measurement indicator at time.

[0094] In the above embodiments, the state change measurement indicator can reflect the degree of change in the system state, specifically including quantifying the changes in each dimension of the feature vector and using mathematical model methods to evaluate the stability and change trend of the system in the time series. By comparing the measurement indicators of the current state and the historical state, potential abnormal states or trend changes can be identified, thereby providing data support and basis for subsequent decision-making.

[0095] Specifically, the state change measurement formula accurately inputs a dataset at a specific moment into a well-trained and performance-stable model. Subsequently, the model deeply analyzes and processes this dataset to extract the potential feature vectors corresponding to the wind power equipment at each moment. Finally, based on these extracted potential feature vectors, algorithms and calculation rules are used to further calculate the measurement indicators that can quantify the state changes of the wind power equipment. It can not only reflect the operating state changes of the wind power equipment at different moments in real time, but also provide important data support and decision-making basis for subsequent equipment state monitoring, fault diagnosis, and operation optimization, thereby ensuring the stable operation and efficient power generation of the wind power equipment and improving the accuracy of risk assessment of the wind power equipment.

[0096] It also needs to be specifically explained that the introduced neural network model is a multi-layer neural network, including various types such as BP neural network, RNN neural network, and CNN neural network. For such a trained neural network model, feature extraction operations need to be performed. Specifically, the wind power equipment data at each moment is input into the model one by one, and then the corresponding feature vectors at each moment are obtained. On this basis, the state difference index and state risk value between the nth moment and the (n + 1)th moment are further obtained. In this way, the operating state and its changes of the wind power equipment at different moments can be analyzed more deeply, providing strong data support and analysis basis for subsequent related research and applications.

[0097] In step S17, according to the state change measurement index, a risk value calculation operation is performed to obtain the state change risk value.

[0098] In one implementation, the risk value calculation formula is:

[0099]

[0100] where is the state change risk value, z is a preset control coefficient, is a preset control factor, , is the state change measurement index at the moment, and

[0101] In the above embodiments, the calculation of the risk value is based on the comprehensive analysis of historical data and real-time monitoring data. The finally obtained risk value of the state change is collected in real time by sensors and monitoring systems, and then obtained through operations such as feature extraction, vector conversion, calculation of state change measurement indicators, and calculation of risk values. Through this method, the downward trend of the equipment performance can be identified in a timely manner, so as to evaluate its risk situation. Realizing the real-time monitoring of the performance degradation of wind power equipment can effectively reduce the occurrence of sudden failures, reduce maintenance costs, and at the same time improve the accuracy of evaluating the risk situation of wind power equipment, ensuring the safe and efficient operation of the wind power system.

[0102] In step S18, according to the risk value of the state change, an alarm reminder operation is performed to monitor the wind power equipment.

[0103] In the above embodiments, the risk value of the state change is obtained by analyzing the real-time operation data of the wind power equipment, such as power generation power, rotation speed, vibration frequency, and temperature, in combination with historical data and machine learning algorithms. These data are used to evaluate the performance degradation of the equipment and judge the health status of the equipment in real time. Once it is detected that the risk value exceeds the preset threshold, the system will automatically issue an alarm reminder to prompt the maintenance personnel to check and maintain the equipment in a timely manner, thereby effectively reducing the probability of failures, extending the service life of the equipment, and improving the overall operation efficiency and reliability of the wind power system. The purpose of this step is to realize the real-time monitoring and early warning of the performance degradation of wind power equipment, and the accuracy of the risk assessment of wind power equipment can be improved by setting accurate risk value thresholds.

[0104] In summary, the present invention provides a wind power control method based on an intelligent robot, including: during the operation of the wind power equipment, obtaining historical wind power data, where the historical wind power data is obtained by the intelligent robot; performing a feature extraction operation according to the historical wind power data to obtain data items of the historical wind power data; performing a vectorization operation on the data unit according to the data items to obtain a vector; performing a correlation analysis operation according to the data vector to obtain an abnormal data set of the wind power equipment; performing a feature vector extraction operation according to the abnormal data set to obtain a potential feature vector; performing a state change measurement index calculation operation according to the potential feature vector to obtain a state change measurement index; performing a risk value calculation operation according to the state change measurement index to obtain a risk value of the state change; and performing an alarm reminder operation according to the risk value of the state change to monitor the wind power equipment.

[0105] An embodiment of the present invention also provides a wind power control system based on an intelligent robot, as Figure 2 shown, which is a structural block diagram of the wind power control system based on an intelligent robot provided by the embodiment of the present invention, including:

[0106] Data acquisition module 1, during the operation of the wind power equipment, acquires historical wind power data, wherein the historical wind power data is obtained by an intelligent robot;

[0107] Feature extraction module 2, used to perform feature extraction operations according to the historical wind power data to obtain data items corresponding to the historical wind power data;

[0108] Vector transformation module 3, used to perform data unit vectorization operations according to the data items to obtain data vectors;

[0109] Correlation analysis module 4, used to perform correlation analysis operations according to the data vectors to obtain an abnormal data set corresponding to the wind power equipment;

[0110] Feature vector extraction module 5, used to perform feature vector extraction operations according to the abnormal data set to obtain potential feature vectors;

[0111] State change metric calculation module 6, used to perform state change metric calculation operations according to the potential feature vectors to obtain state change metrics;

[0112] Risk value calculation module 7, used to perform risk value calculation operations according to the state change metrics to obtain state change risk values;

[0113] User reminder module 8, used to perform alarm reminder operations according to the state change risk values to monitor the wind power equipment.

[0114] In an alternative embodiment, the data acquisition module 1 is further used for:

[0115] After acquiring the historical wind power data during the operation of the wind power equipment, it further includes: performing data cleaning operations according to the historical wind power data to obtain cleaned data; performing preprocessing operations according to the cleaned data to obtain preprocessed data, wherein the preprocessing operation formula is:

[0116] ,

[0117] is the preprocessed data, is the cleaned data, is the mean of the cleaned data, is the standard deviation of the cleaned data.

[0118] In an alternative embodiment, the feature extraction module 2 is further used for:

[0119] Performing a feature extraction operation on the historical wind power data to obtain data items of the historical wind power data, including: The feature extraction operation formula is:

[0120]

[0121] where is the data item at time t, is the total number of data features extracted from the preprocessed data, is the weight coefficient of the th data feature, is at time the th data feature value in the preprocessed data, is the th preprocessed data, is a specific time .

[0122] In an alternative embodiment, the vector conversion module 3 is further configured to:

[0123] Performing a data unit vectorization operation on the data item to obtain a data vector, including:

[0124] The data unit vectorization operation is to input the obtained data item into a clustering model, where the clustering model uses the K-means algorithm and includes normal features, fluctuation features, and abnormal features; the data vector includes an abnormal feature vector, a fluctuation feature vector, and a normal feature vector.

[0125] In an alternative embodiment, the correlation analysis module 4 is further configured to:

[0126] Performing a correlation analysis operation on the data vector to obtain an abnormal data set corresponding to the wind power equipment, including: performing a time distance calculation operation on the data vector to obtain a time distance corresponding to the data vector, where the time distance calculation formula is:

[0127] ,

[0128] In the formula, and are two different times, is the data vector at and the time distance, and are the dimensions of the time attribute, are two at The data vector of the dimension; perform a similarity calculation operation according to the time distance to obtain the correlation of the data vector, where the similarity calculation formula is:

[0129] ,

[0130] In the formula, is a preset parameter for controlling the attenuation speed, is the correlation, and are two different moments;

[0131] Perform a correlation threshold comparison operation according to the correlation to obtain the abnormal data set, where the correlation threshold comparison includes if > v, then it belongs to the abnormal data set, and v is a preset correlation threshold, is the correlation.

[0132] In an alternative embodiment, the state change metric calculation module 6 is further configured to:

[0133] Perform a state change metric calculation operation according to the latent feature vector to obtain a state change metric, including: The state change metric formula is:

[0134] ,

[0135] Wherein, , , is the state change metric, is the average state change metric, is the latent feature vector at time , is the latent feature vector at time , is a preset coefficient, and are time variables, is the state change metric at time.

[0136] In an alternative embodiment, the risk value calculation module 7 is further configured to:

[0137] Perform a risk value calculation operation according to the state change metric to obtain a state change risk value, including: The risk value calculation formula is:

[0138]

[0139] Wherein, is the state change risk value, z is a preset control coefficient, is a preset control factor, , is the state change metric at time is the state change metric at time

[0140] An embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the wind power control method based on an intelligent robot as described in the above embodiment is implemented.

[0141] Exemplarily, the computer program can be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device.

[0142] The electronic device can be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine certain components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0143] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0144] The memory can be used to store the computer program and / or module. By running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory, the processor can implement various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0145] Among them, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, an executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0146] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative work.

[0147] Compared with the prior art, in the present invention, the wind power control method based on an intelligent robot has the following advantages: First, the integrity and accuracy of wind power equipment data are improved through data preprocessing. Second, key information is effectively extracted to provide a basis for monitoring. Third, clustering analysis and neural networks are used to accurately identify abnormal states and related information of equipment, improving operation safety and reliability. Fourth, real-time monitoring is realized to promptly handle potential problems and avoid losses caused by equipment failures. The specific invention process of the present invention is as follows: During the operation of wind power equipment, first, relevant historical wind power data needs to be obtained. Immediately afterwards, feature extraction work is carried out on the obtained historical wind power data to obtain data items. Subsequently, the obtained data items are vectorized and transformed into vector form. Then, based on this vector, a correlation analysis operation is performed to obtain an abnormal data set of wind power equipment. Further, feature vectors are extracted from this abnormal data set to obtain potential feature vectors. Subsequently, a state change metric is calculated based on the potential feature vectors. Then, a risk value is calculated based on the calculated state change metric. Finally, an alarm reminder operation is performed based on the risk value to achieve comprehensive and effective monitoring of wind power equipment, ensure the safe operation of wind power equipment, promptly discover and handle problems that occur, and ensure the normal operation of the entire wind power system. Implementing wind power control based on an intelligent robot according to the above method not only realizes real-time monitoring of the performance degradation of wind power equipment, but also improves the accuracy of evaluating the risk situation of wind power equipment, ensures the stable and safe operation of wind power equipment, enables staff to promptly discover and handle problems that occur, and ensures the normal operation of the entire wind power system.

[0148] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A wind power control method based on an intelligent robot, characterized in that: include: During the operation of the wind power equipment, historical wind power data is obtained, wherein the historical wind power data is obtained by an intelligent robot; Performing a feature extraction operation according to the historical wind power data to obtain data items corresponding to the historical wind power data; According to the data item, a data unit vectorization operation is performed to obtain a data vector; Performing a correlation analysis operation according to the data vector to obtain an abnormal data set corresponding to the wind power equipment; Performing a feature vector extraction operation according to the abnormal data set to obtain a potential feature vector; Performing a state change metric calculation operation according to the potential feature vector to obtain a state change metric; Performing a risk value calculation operation according to the state change measurement indicator to obtain a state change risk value; According to the state change risk value, an alarm reminder operation is performed to monitor the wind power equipment; The step of performing a state change metric index calculation operation according to the potential feature vector to obtain the state change metric index includes: The state change measurement formula is: StateChangeMetric(t)=S(t)-E(Sm), in, S(t) is the state change metric, E(Sm) is the average state change metric, P(k) is the potential feature vector at time k, P(k-1) is the potential feature vector at time k-1, α is the preset coefficient, t and k are time variables, and S(k) is the state change metric at time k; The step of performing a risk value calculation operation according to the state change measurement indicator to obtain a state change risk value includes: The risk value calculation formula is: Perr(t)=z·ΔP(t)+β Among them, Perr(t) is the state change risk value, z is the preset control coefficient, β is the preset control factor, ΔP(t)=P(t+1)-P(t), P(t+1) is the state change measurement index at time t+1, and P(t) is the state change measurement index at time t.

2. The wind power control method based on intelligent robots according to claim 1 is characterized in that: In the process of wind power equipment operation, after obtaining historical wind power data, the method further includes: Performing a data cleaning operation according to the historical wind power data to obtain cleaned data; According to the cleaned data, a preprocessing operation is performed to obtain preprocessed data, wherein the preprocessing operation formula is: x″ is the preprocessed data, x is the cleaned data, μ is the mean of the cleaned data, and σ is the standard deviation of the cleaned data.

3. The wind power control method based on intelligent robots according to claim 2 is characterized in that: The performing of feature extraction operation according to the historical wind power data to obtain data items of the historical wind power data includes: The feature extraction operation formula is: Among them, W CM (t) is the data item at time t, n is the total number of data features extracted from the preprocessed data, w i is the weight coefficient of the i-th data feature, CM i (t) is the value of the i-th data feature in the preprocessed data at time t, i is the i-th preprocessed data, and t is a specific time t.

4. The wind power control method based on intelligent robots according to claim 1 is characterized in that: The step of performing a data unit vectorization operation according to the data item to obtain a data vector includes: The data unit vectorization operation is to input the obtained data items into a clustering model, wherein the clustering model adopts a K-means algorithm and includes normal features, fluctuation features and abnormal features; The data vectors include abnormal feature vectors, fluctuation feature vectors and normal feature vectors.

5. The wind power control method based on intelligent robots according to claim 1, characterized in that: The performing a correlation analysis operation according to the data vector to obtain an abnormal data set corresponding to the wind power equipment includes: According to the data vector, a time distance calculation operation is performed to obtain the time distance corresponding to the data vector, wherein the time distance calculation formula is: In the formula, t1 and t2 are two different moments, d(t1, t2) is the time distance between the data vector at t1 and t2, n and i are the dimensions of the time attribute, and x 1i With x 2i are two data vectors in dimension i; According to the time distance, a similarity calculation operation is performed to obtain the correlation of the data vector, wherein the similarity calculation formula is: Wherein, σ is a preset parameter for controlling the decay speed, sim is the correlation, and t1 and t2 are two different moments; According to the correlation, a correlation threshold comparison operation is performed to obtain the abnormal data set, wherein the correlation threshold comparison includes if sim(t1, t2)>v, it belongs to an abnormal data set, v is a preset correlation threshold, and sim is the correlation.

6. A wind power control system based on an intelligent robot, characterized in that: The method for controlling wind power based on an intelligent robot according to any one of claims 1 to 5 comprises: A data acquisition module, which acquires historical wind power data during the operation of the wind power equipment, wherein the historical wind power data is acquired by an intelligent robot; A feature extraction module, used to perform a feature extraction operation based on the historical wind power data to obtain data items corresponding to the historical wind power data; A vector conversion module, used for performing a data unit vectorization operation according to the data item to obtain a data vector; A correlation analysis module, used to perform a correlation analysis operation according to the data vector to obtain an abnormal data set corresponding to the wind power equipment; A feature vector extraction module, used to perform a feature vector extraction operation according to the abnormal data set to obtain a potential feature vector; A state change metric calculation module, used to perform a state change metric index calculation operation according to the potential feature vector to obtain a state change metric index; A risk value calculation module, used to perform a risk value calculation operation according to the state change measurement index to obtain a state change risk value; The user reminder module is used to perform an alarm reminder operation according to the state change risk value to monitor the wind power equipment.

7. An electronic device, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the wind power control method based on the intelligent robot as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the wind power control method based on the intelligent robot according to any one of claims 1 to 5.

Images