Motor compensation control method and system for dual-identification network anomaly prediction diving

Abstract

The invention provides a compensation control method and system for a diving motor for abnormal prediction of a double-recognition network, and relates to the technical field of motor control. Equipment monitoring of the underwater special power equipment is executed, and a time sequence monitoring data set is established; executing model self-inspection, and completing dynamic weight configuration of the first and second anomaly identification networks according to a self-inspection result; after a time sequence monitoring data set is preprocessed, an anomaly prediction model is input, an anomaly prediction result is generated, anomaly evaluation is carried out on the result, a step compensation strategy is established, and therefore optimization control is carried out; and verification identification of optimization control is carried out through a verification window to complete optimization control, so that the technical problems of low equipment operation efficiency, poor stability and adaptability and incapability of timely and accurately carrying out anomaly prediction and optimization in the control optimization process of the submersible motor are solved; the technical effects of accurate anomaly prediction and optimal control of the submersible motor and improvement of the operation efficiency and safety of the equipment are achieved.

Description

Technical Field

[0001] This invention relates to the field of motor control technology, specifically to a motor compensation control method and system for anomaly prediction using a dual-identification network. Background Technology

[0002] Submersible power equipment, especially submersible motors, plays a vital role in various fields such as marine engineering, water resource development, and fisheries. As specialized underwater power equipment, submersible motors operate in complex and variable environments, with factors such as water temperature, pressure, and quality all potentially affecting their operation. Furthermore, they must contend with challenges such as load variations and power fluctuations during operation. Therefore, optimizing the control of submersible motors is crucial for improving their operational efficiency, stability, and safety.

[0003] Existing technologies for optimizing submersible motor control often suffer from low operating efficiency, poor stability and adaptability, and the inability to predict and optimize anomalies in a timely and accurate manner, which increases the failure rate and maintenance costs. Summary of the Invention

[0004] This application provides a dual-identification network anomaly prediction submersible motor compensation control method and system, which is used to address the technical problems of low equipment operating efficiency, poor stability and adaptability, inability to predict and optimize anomalies in a timely and accurate manner during the submersible motor control optimization process, thereby increasing the failure rate and maintenance costs.

[0005] In view of the above problems, this application provides a method and system for motor compensation control under submersible with dual identification network anomaly prediction.

[0006] In a first aspect, this application provides a motor compensation control method for underwater vehicles using dual-identification network anomaly prediction. The method is applied to a motor compensation control system for underwater vehicles using dual-identification network anomaly prediction, and includes: An anomaly prediction model is established for predicting anomalies in underwater dedicated power equipment. The model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is constructed using a usage dataset of the underwater dedicated power equipment, and the second anomaly identification network is constructed using big data. Equipment monitoring of the underwater dedicated power equipment is performed to establish a time-series monitoring dataset. A model self-check of the anomaly prediction model is performed, and dynamic weight configuration of the first and second anomaly identification networks is completed based on the self-check results. The time-series monitoring dataset is preprocessed and input into the dynamically weighted anomaly prediction model to generate anomaly prediction results. Anomaly evaluation is performed on the anomaly prediction results, and a tiered compensation strategy is established based on the evaluation results. Optimization control is performed using the tiered compensation strategy, and a verification window is set. Verification and identification of the optimization control are performed through the verification window, and optimization control is completed based on the verification and identification results.

[0007] Secondly, this application provides a motor compensation control system for underwater vehicles using dual-identification network anomaly prediction, the system comprising: The system includes a model building module for establishing an anomaly prediction model for underwater dedicated power equipment. This model comprises a first anomaly identification network and a second anomaly identification network. The first network is constructed using a usage dataset of the underwater dedicated power equipment, while the second network is constructed using big data. A device monitoring module is used to perform device monitoring of the underwater dedicated power equipment and establish a time-series monitoring dataset. A model self-checking module is used to perform a self-check of the anomaly prediction model and dynamically configure the weights of the first and second anomaly identification networks based on the self-check results. An anomaly prediction module is used to preprocess the time-series monitoring dataset and input it into the dynamically weighted anomaly prediction model to generate anomaly prediction results. A strategy compensation module is used to evaluate the anomaly prediction results, establish a tiered compensation strategy based on the evaluation results, perform optimization control using the tiered compensation strategy, and set a verification window. A verification and identification module is used to perform verification and identification of the optimization control through the verification window and complete the optimization control based on the verification and identification results.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: The dual-identification network anomaly prediction submersible motor compensation control method provided in this application establishes an anomaly prediction model, which is a model for predicting anomalies in underwater dedicated power equipment. The anomaly prediction model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is constructed using the usage dataset of the underwater dedicated power equipment, and the second anomaly identification network is constructed using big data. The method involves: performing equipment monitoring of the underwater dedicated power equipment to establish a time-series monitoring dataset; performing a model self-check of the anomaly prediction model and configuring the dynamic weights of the first and second anomaly identification networks based on the self-check results; and preprocessing the time-series monitoring dataset before inputting it into... An anomaly prediction model with dynamic weight configuration generates anomaly prediction results. Anomaly evaluation is performed on these results, and a tiered compensation strategy is established based on the evaluation. Optimization control is then implemented using this strategy, and a verification window is set. The verification window is used to verify and identify the optimized control, and the optimized control is completed based on the verification results. This approach solves the technical problems of low equipment operating efficiency, poor stability and adaptability, and the inability to perform timely and accurate anomaly prediction and optimization during the optimization of submersible motor control, which increases the failure rate and maintenance costs. It achieves the technical effects of accurate anomaly prediction and optimized control of submersible motors, improving equipment operating efficiency and safety, and reducing the failure rate and maintenance costs. Attached Figure Description

[0009] Figure 1 This application provides a schematic flowchart of the submersible motor compensation control method for dual-identification network anomaly prediction.

[0010] Figure 2 This application provides a schematic diagram of the structure of a motor compensation control system for underwater vehicles using a dual-identification network for anomaly prediction.

[0011] Figure labeling: Model building module 11, Equipment monitoring module 12, Model self-test module 13, Anomaly prediction module 14, Strategy compensation module 15, Verification and identification module 16. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. Example

[0013] like Figure 1 As shown, this application provides a motor compensation control method for underwater vehicles using dual-identification network anomaly prediction. The method is applied to a motor compensation control system for underwater vehicles using dual-identification network anomaly prediction, and includes: Step S100: Establish an anomaly prediction model. The anomaly prediction model is a model for predicting anomalies in underwater dedicated power equipment. The anomaly prediction model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is constructed using the usage dataset of underwater dedicated power equipment, and the second anomaly identification network is constructed using big data.

[0014] Underwater specialized power equipment is electrical equipment designed and optimized for underwater environments. These devices typically operate in harsh marine environments, thus requiring key characteristics such as corrosion resistance, sealing, pressure resistance, and waterproofing. Furthermore, since underwater power equipment is commonly used in marine engineering and scientific research, it may also need to meet specific application requirements, such as high torque, high power output, and high-precision control. Therefore, research on the optimized control of underwater specialized power equipment is a necessary topic. Underwater power equipment generally includes underwater cables, submersible motors, underwater transformers, frequency converters, etc. The method proposed in this application embodiment is mainly used for the control optimization of submersible motors, providing a reliable control scheme for underwater kinetic energy applications.

[0015] First, to predict anomalies in submersible motors, an anomaly prediction model is established. This model includes a first anomaly detection network and a second anomaly detection network. The first anomaly detection network is constructed using a dataset of submersible motor usage, while the second anomaly detection network is constructed using big data. Specifically, historical usage data of the submersible motors is collected, including but not limited to parameters such as motor operating status, temperature, pressure, voltage, and current, as well as any relevant log information or fault records. Duplicate, incomplete, or erroneous data is removed from the acquired data, and the data is standardized to ensure consistency and comparability. Meaningful features, such as statistical features and time series features, are extracted from the raw data. These features will be used to train the first anomaly detection network. Similarly, big data related to submersible motors is collected from various sources, including but not limited to historical data from other similar equipment, industry reports, and public datasets. Similar data preprocessing as that used for the first anomaly detection network dataset is performed to obtain the dataset for the second anomaly detection network.

[0016] Next, the anomaly prediction model is constructed. First, a first anomaly detection network is built. This involves selecting a suitable machine learning or deep learning algorithm, such as Support Vector Machine (SVM), Random Forest, or Neural Network, based on the characteristics of the submersible motor dataset. The model is trained using extracted features and corresponding labels (normal / abnormal). These labels can be generated automatically through expert annotation or based on historical fault records. After training, cross-validation and other methods are used to evaluate the model's performance, and parameters are fine-tuned as needed. Finally, a model with the first anomaly detection network is obtained. Similarly, a second anomaly detection network is built using a similar method. However, due to the complexity and diversity of big data, a more powerful deep learning algorithm, such as Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN), is selected. A pre-trained model is obtained using a large dataset, enabling the model to learn a wider range of anomaly patterns in submersible motors. The parameters of the pre-trained model are then transferred to the first anomaly detection network and fine-tuned using the submersible motor dataset to improve the model's predictive ability for specific equipment. The above steps complete the construction of the entire anomaly prediction model. Finally, the trained anomaly prediction model can be deployed to the actual application environment to interact with the submersible motor in real time, thereby monitoring its operating status in real time and inputting real-time data into the anomaly prediction model to obtain the anomaly prediction results of the equipment.

[0017] By building an anomaly prediction model from both the user and big data perspectives, the personalized anomaly detection and response needs of specialized equipment are met, while also incorporating in-depth big data analysis, thereby ensuring the comprehensiveness and accuracy of the model's predictions.

[0018] Step S200: Perform equipment monitoring of the underwater dedicated power equipment and establish a time-series monitoring dataset.

[0019] Optionally, clearly define the monitoring needs and objectives, i.e., with performance optimization as the goal, determine the parameters of the submersible motor that need to be monitored, such as current, voltage, temperature, vibration, and insulation performance. Select appropriate sensors based on the monitoring needs, such as temperature sensors, vibration sensors, and current sensors, and employ a real-time online monitoring system to continuously and in real-time monitor the submersible motor. Design a specific monitoring scheme, including monitoring points, monitoring frequency, and data acquisition design. Determine the specific locations on the submersible motor where sensors need to be installed to ensure comprehensive and accurate monitoring of the motor's operating status. Based on the motor's operating conditions and monitoring needs, set an appropriate monitoring frequency, such as collecting data once per minute, hour, or day. Design a data acquisition system to ensure real-time and accurate acquisition of sensor data. During data acquisition, design a specific data format, including fields such as timestamp, sensor type, sensor location, and monitoring data, and upload the collected data in real-time to a time-series database to form a time-series monitoring dataset.

[0020] By establishing a time-series monitoring dataset, the changing trends of various parameters of the submersible motor can be analyzed using methods such as time series analysis, and potential problems can be predicted. At the same time, statistical methods or machine learning algorithms can be used to detect outliers in the data, identify possible faults or anomalies in the submersible motor, and evaluate the performance status of the submersible motor, such as operating efficiency and energy consumption, thus laying a data analysis foundation for subsequent optimization control.

[0021] Step S300: Perform a model self-check of the anomaly prediction model, and complete the dynamic weight configuration of the first anomaly identification network and the second anomaly identification network based on the model self-check results.

[0022] For example, to verify the accuracy, stability, and generalization ability of anomaly prediction models and ensure their effective operation on new, unseen data, a self-test can be performed. Specifically, a subset of known labeled datasets (different from the training set) is used as a validation set for model self-testing. The validation set should include both normal and anomalous data. Suitable metrics for evaluating anomaly detection models, such as accuracy and recall, are selected, and the model is evaluated using the validation set data. The evaluation results are recorded. Based on the self-test results, the model's performance on different types of data is analyzed to identify potential problems and areas for improvement. The model's evaluation results on the validation set are compared with the expected goals to determine if the model meets the requirements. False positives (identifying normal data as anomalous) and false negatives (identifying anomalous data as normal) are examined and their causes analyzed. The model's dependence on various features is also analyzed to identify features that significantly impact model performance. Finally, the weights of the first and second anomaly detection networks are dynamically adjusted based on the self-test results to optimize overall model performance. When adjusting weights, if the first anomaly detection network performs well in self-testing, its weight can be appropriately increased to make its contribution to the overall model greater. Similarly, if the second anomaly detection network performs better on specific types of data, its weight can be increased on those types of data. Specifically, a weight allocation mechanism can be designed to collect data and evaluate model performance in real time during model operation, and then dynamically adjust the weights of the two networks based on performance changes. In addition to adjusting weights, ensemble learning methods can be considered to fuse the prediction results of the two networks to improve the overall model performance. Specific fusion methods can be achieved through simple weighted averaging or more complex ensemble learning algorithms (such as stacked ensemble).

[0023] By dynamically configuring the weights of the anomaly detection network based on model self-checking, we can ensure that the model maintains optimal performance in practical applications, thereby improving the accuracy of optimization control.

[0024] Step S400: After preprocessing the time-series monitoring dataset, input it into the anomaly prediction model with dynamically configured weights to generate anomaly prediction results.

[0025] Furthermore, before inputting time-series monitoring datasets into anomaly prediction models, data preprocessing is typically required to improve data quality and model performance. Data preprocessing includes data cleaning, data transformation, feature selection, and feature engineering. This involves removing noise, missing values, outliers, or duplicates from the data, and standardizing, normalizing, or discretizing the data as needed to enable better model processing. It also involves selecting the most relevant features from the original dataset to simplify the model and improve prediction accuracy, and creating new features or modifying existing features to capture hidden patterns or relationships within the data.

[0026] After data preprocessing, the dynamically weighted anomaly prediction model needs to be loaded and configured. Specifically, the pre-trained anomaly prediction model is loaded from storage, and the weights of each component in the model (such as the first and second anomaly detection networks) are set according to the previous dynamic weight configuration results. Finally, the model's parameters and state are checked to ensure that the model is correctly loaded and ready for prediction. Once the model is loaded and configured, the preprocessed time-series monitoring dataset can be input into the model to generate anomaly prediction results. Specifically, if the dataset is large, it may need to be divided into smaller batches for processing to improve computational efficiency. The preprocessed data is passed as input to the anomaly prediction model, which uses the input data to perform calculations and generate anomaly prediction results, which may include anomaly scores, anomaly labels (normal / abnormal), or anomaly probabilities.

[0027] Step S500: Perform anomaly evaluation on the anomaly prediction results, establish a tiered compensation strategy based on the anomaly evaluation results, perform optimization control through the tiered compensation strategy, and set a verification window.

[0028] Next, the metrics used to evaluate the anomaly prediction results are determined, including accuracy, recall, and false positive rate. Specific metrics depend on actual needs and the characteristics of the prediction model. The selected evaluation metrics are used to assess the anomaly prediction results, comparing them with the actual situation and calculating the corresponding metric values. Based on the assessment results, the anomaly prediction results are categorized according to severity, for example, into minor, moderate, and severe anomalies. This categorization lays the foundation for developing different levels of compensation strategies. Next, based on the anomaly severity classification, compensation measures for different levels of anomalies are determined. These measures may include adjusting control parameters, sending alarm notifications, and activating backup equipment. The compensation measures are set in a tiered manner according to the anomaly severity. For example, for minor anomalies, only minor parameter adjustments may be needed; while for severe anomalies, immediate activation of backup equipment and notification of relevant personnel may be required. Simultaneously, trigger conditions are set for each tier of compensation measures. These conditions can be set based on factors such as the confidence level of the prediction results and the duration of the anomaly.

[0029] During the optimization control process, the system is monitored in real time to promptly detect new anomaly predictions. When a certain trigger condition is met, corresponding compensation measures are automatically triggered for optimization control. Simultaneously, the execution status of these compensation measures is tracked and recorded for subsequent analysis and optimization. Finally, verification indicators closely related to the control objectives and system performance are selected to validate the effectiveness of the optimization control. The size of the verification window is determined based on control requirements and data update frequency. It is important that the verification window be large enough to contain sufficient data to evaluate the effectiveness of the optimization control. Data is collected within the verification window, and the selected verification indicators are used to evaluate the effectiveness of the optimization control. The differences in performance indicators before and after optimization are compared to generate evaluation results. Finally, based on the evaluation results, the tiered compensation strategy and optimization control parameters are adjusted and optimized to improve system performance and anomaly handling.

[0030] Step S600: Perform verification and identification for optimization control through the verification window, and complete optimization control based on the verification and identification results.

[0031] Specifically, within the verification window, data related to optimized control is collected in real time, including system status, performance indicators, anomaly prediction results, and the execution status of compensation measures. Verification and identification are performed based on the collected data to verify whether the optimized control strategy works as expected and achieves the desired effect. Specifically, data before and after implementing the optimized control strategy are compared and analyzed to observe whether performance indicators show significant improvement; the trend of data changes is analyzed to determine whether the optimized control strategy has a positive impact on the system; and it is checked whether new anomalies still occur after the implementation of the optimized control strategy, and whether these anomalies have been effectively handled. The verification and identification results are then evaluated, and the success of the optimized control strategy is determined according to preset evaluation criteria (such as performance improvement percentage, anomaly reduction rate, etc.). If the evaluation results meet expectations, the optimized control strategy is effective; otherwise, further analysis of the causes and adjustments are needed. Furthermore, the optimized control strategy is adjusted based on the evaluation of the verification and identification results. If the strategy is successful, further optimization can be considered to improve performance; if the strategy fails, the causes of failure need to be analyzed and corresponding improvement measures taken. The direction of the adjustments can include parameter adjustment, algorithm improvement, and optimization of compensation measures. Since adjusting the optimization control strategy is an iterative process, the verification window needs to be reset and data collected for verification and identification after each adjustment. Through continuous verification and adjustment, the optimization control strategy can be gradually improved to better meet actual needs and achieve optimal performance, thereby ensuring the effectiveness and reliability of the optimization control strategy in practical applications, while also improving the performance and stability of the system.

[0032] Furthermore, the step S300 of this application further includes: performing a model self-check on the anomaly prediction model and configuring the dynamic weights of the first and second anomaly recognition networks based on the model self-check results. Step S310: Call the self-checking processing layer to perform data richness evaluation of the first anomaly identification network, and construct the first self-checking result based on the data richness evaluation result. The self-checking processing layer is a sub-processing layer of the anomaly prediction model.

[0033] Step S320: Invoke the self-check processing layer, perform data adaptation evaluation of the second abnormal network, construct the second self-check result based on the adaptation evaluation result, integrate the first self-check result and the second self-check result, and complete the dynamic weight configuration.

[0034] Optionally, when dynamically configuring the weights of the anomaly detection network based on model self-checking, the self-checking processing layer is first initialized. This confirms that the self-checking processing layer is a defined and configured sub-processing layer in the anomaly prediction model, and that this processing layer contains the algorithms and logic required to perform data richness evaluation and data fit evaluation. Next, the dataset required by the first anomaly detection network is loaded, and the data richness evaluation algorithm in the self-checking processing layer is called to evaluate the dataset of the first anomaly detection network. This data richness evaluation considers multiple aspects such as data diversity, completeness, and novelty. After the evaluation is completed, the evaluation result is recorded as part of the first self-checking result. Then, the dataset required by the second anomaly detection network is loaded, and the data fit evaluation algorithm in the self-checking processing layer is called to evaluate the dataset of the second anomaly detection network. This data fit evaluation mainly focuses on the degree of matching between the data and the network model, such as whether the data distribution is consistent with the model assumptions and whether it contains sufficient anomaly samples. Similarly, after the evaluation is completed, the evaluation result is recorded as part of the second self-checking result. Furthermore, key information such as data richness scores and data fit scores are extracted from the two self-check results. These results are then integrated using strategies such as weighted averaging and maximum / minimum scaling to form a comprehensive self-check report or score. Based on the integrated self-check results, the weights of the first and second anomaly detection networks in the final anomaly prediction model are determined. That is, if a network has higher data richness and data fit scores, its weight in the model should be increased accordingly. Dynamic weight configuration ensures that the model can fully utilize the advantages of each network during runtime, improving prediction performance.

[0035] Furthermore, the step S500 of this application further includes: performing anomaly evaluation on the anomaly prediction results and establishing a tiered compensation strategy based on the anomaly evaluation results. Step S510: Analyze the abnormal evaluation results and obtain the maximum abnormal deviation value.

[0036] Step S520: Establish the compensation extreme value of the step compensation strategy based on the maximum abnormal deviation value.

[0037] Step S530: Perform equipment stability analysis on the underwater special power equipment, and configure the number and length of steps based on the equipment stability analysis results.

[0038] Step S540: Establish the step compensation strategy by using the compensation extreme value, number of steps, and step length.

[0039] For example, firstly, the results of the anomaly evaluation (i.e., a set of one or more indicators or metrics) are analyzed, reflecting the submersible motor's performance under abnormal conditions. Parameters representing the degree of abnormal deviation are identified from these indicators or metrics, and the maximum value is determined, i.e., the maximum abnormal deviation value. This value reflects the maximum performance deviation or degree of deviation from the expected operating range of the equipment under abnormal conditions. Then, a compensation extreme value for a stepped compensation strategy is established based on the maximum abnormal deviation value. Here, the compensation extreme value refers to the maximum range or limit value requiring compensation. Typically, the compensation extreme value is set based on the maximum abnormal deviation value to ensure that the compensation strategy can cover the maximum deviation that the equipment may exhibit under abnormal conditions.

[0040] Simultaneously, equipment stability analysis is performed on the submersible motor. Equipment stability analysis is the process of evaluating the stability and reliability of equipment under normal operating conditions and potential abnormal conditions. This analysis includes examining the equipment's operating history data, performance parameters, fault records, etc. Equipment stability analysis reveals the equipment's performance under different operating conditions, as well as its sensitivity to abnormal situations and its recovery capability. Furthermore, based on the equipment stability analysis results, the number and length of steps are configured. The number of steps determines the granularity of the compensation strategy, i.e., how many steps or intervals the abnormal deviation range covered by the compensation strategy is divided into. The step length represents the size of each step or interval, i.e., the difference between two adjacent steps. The appropriate number and length of steps can be determined based on the equipment stability analysis results. For example, if the equipment performs relatively stably under abnormal conditions, a smaller number of steps and a larger step length can be configured; conversely, if the equipment is more sensitive to abnormal situations, a larger number of steps and a smaller step length are required.

[0041] Finally, after determining the compensation extreme values, the number of steps, and the step length, a complete stepped compensation strategy can be established. This strategy is typically a table or function that defines the compensation measures or values ​​to be taken under different abnormal deviations. The compensation measures or values ​​can be set according to actual needs, such as adjusting equipment operating parameters, triggering fault alarms, or activating backup equipment. The establishment of this stepped compensation strategy should ensure that appropriate compensation measures can be taken promptly and accurately when abnormal conditions occur in the equipment, thereby guaranteeing the normal operation of the equipment and the stability of the system.

[0042] By implementing the above steps, abnormal equipment conditions can be accurately addressed, avoiding potential equipment failures or system instability. At the same time, this flexibility allows the compensation strategy to adapt to abnormal conditions of varying severity, thereby providing more refined and effective compensation measures and improving control optimization efficiency.

[0043] Furthermore, the verification and identification process performed through the verification window, and the optimization control completed based on the verification and identification results, further includes step S600 of this application: Step S610: Construct node deviation based on the verification and identification results. The node deviation is the mapping deviation between the verification window and each step node in the step compensation strategy.

[0044] Step S620: Analyze the unit deviation controlled by the node deviation to generate optimized compensation.

[0045] Step S630: Reconstruct the uncompensated step compensation strategy based on the node deviation, and continue to optimize the control based on the reconstruction result and optimized compensation.

[0046] Furthermore, to quantitatively evaluate the accuracy and effectiveness of the tiered compensation strategy in practical applications, node deviations are constructed based on the verification and identification results. The node deviation refers to the mapping deviation between the verification window (i.e., actual equipment operating data or performance indicators) and each tier node in the tiered compensation strategy. By comparing the actual data in the verification window with the expected data or state in the tiered compensation strategy, the deviation value at each tier node can be determined, serving as a reference for subsequent optimization. Subsequently, unit deviation analysis is performed using the node deviations. Unit deviation analysis involves a detailed analysis of the node deviations to determine the specific causes, impact, and possible solutions. By analyzing the deviation of each unit (e.g., each tier node), deficiencies or problems in the compensation strategy can be identified, providing direction for subsequent optimization. Finally, optimized compensations are generated based on the results of the unit deviation analysis. These optimized compensations may involve fine-tuning certain parameters in the tiered compensation strategy or introducing new compensation measures to compensate for existing deficiencies. Through targeted optimization measures, the accuracy and effectiveness of the tiered compensation strategy are improved, further enhancing equipment stability and system performance.

[0047] If significant deviations are found at certain step nodes during node deviation analysis that cannot be eliminated through simple optimization compensation, the entire step compensation strategy needs to be reconstructed. This reconstruction process includes redesigning the number and length of steps and compensation measures. Comprehensive reconstruction fundamentally solves the problems in the step compensation strategy, providing a more stable and reliable control strategy for the equipment. Finally, after completing the reconstruction of the step compensation strategy and generating optimized compensation, these optimization measures are applied to the actual control system, and the control effect continues to be monitored and analyzed. If new problems are found or further optimization is needed, the above steps can be repeated for iterative improvement. This achieves a closed-loop optimization process, continuously improving the performance of the control system and the stability of the equipment.

[0048] Furthermore, this application also includes step S700, which further includes: Step S710: Extract historical data from underwater dedicated power equipment and establish a denoised historical dataset.

[0049] Step S720: Configure comprehensive performance evaluation indicators, evaluate the performance of underwater dedicated power equipment using the comprehensive performance evaluation indicators and the historical dataset, and generate an optimized control strategy based on the performance evaluation results.

[0050] Step S730: Optimize the task response of the underwater dedicated power equipment using the optimized control strategy.

[0051] Optionally, besides optimizing strategies from a compensation perspective, optimization can also be performed from a basic control optimization perspective to ensure accuracy. Specifically, historical data is extracted from the submersible motor's operating records, monitoring systems, or other data sources. This data includes the equipment's operating parameters, performance indicators, and fault records. The extracted raw historical data is then denoised to eliminate noise and outliers, resulting in the final historical dataset. Next, the submersible motor's performance is evaluated. Before the performance evaluation, a series of comprehensive performance evaluation indicators are configured. These indicators comprehensively reflect various aspects of the equipment's performance, such as operating efficiency, stability, reliability, and energy consumption. Then, the configured comprehensive performance evaluation indicators and the denoised historical dataset are used to evaluate the submersible motor's performance, generating evaluation results. For example, scores for various indicators of the submersible motor are calculated, performance curves are plotted, and statistical analysis is performed. The performance evaluation results objectively reflect the equipment's performance status, providing an important basis for developing optimized control strategies.

[0052] Furthermore, based on the equipment performance evaluation results, the problems and deficiencies in the equipment's performance are analyzed, and corresponding optimization control strategies are formulated. These strategies include adjusting the equipment's operating parameters, improving the control algorithm, and increasing maintenance measures. Optimized control strategies can propose solutions to specific equipment problems, helping to improve equipment performance and stability, and extend the equipment's service life. Finally, the generated optimized control strategies are applied to the actual operation of the submersible motor to optimize the equipment's task response, including improving task response speed, reducing task execution time, and improving task completion quality. Task response optimization can improve the equipment's working efficiency and reliability, enabling the equipment to better meet the needs of actual work; at the same time, it can also reduce the equipment's operating costs and failure rate, improving the overall performance of the equipment.

[0053] Furthermore, regarding the configuration of comprehensive performance evaluation indicators, step S720 of this application also includes: The efficiency index is established using the following formula: .

[0054] in, For efficiency indicators, For power efficiency, For power stability, and These are the weighting factors for power efficiency and power stability, respectively.

[0055] The reliability index is established using the following formula: .

[0056] in, As a reliability indicator, Characterizing the failure rate, Characterizing fault-free uptime, and These are the weighting factors for failure rate and fault-free uptime, respectively.

[0057] Establish an adaptation index, with the following formula: .

[0058] in, Characteristic of adaptability To enhance the adaptability of underwater-specific power equipment to the underwater environment, To enhance the adaptability of underwater specialized power equipment to load changes, and It is an adaptive weighting factor for changes in underwater environment and load.

[0059] By integrating the aforementioned efficiency, reliability, and adaptability indicators, a comprehensive performance evaluation index is established.

[0060] Specifically, the comprehensive performance evaluation indicators proposed in this application embodiment include efficiency indicators, reliability indicators, and adaptability indicators. The calculation formula for the efficiency indicator is as follows: ;in, For efficiency indicators, For power efficiency, For power stability, and These are weighting factors for power efficiency and power stability, respectively. The formula for calculating the reliability index is as follows: ;in, As a reliability indicator, Characterizing the failure rate, Characterizing fault-free uptime, and These are the weighting factors for failure rate and fault-free uptime, respectively. The formula for calculating the adaptability index is as follows: ;in, Characteristic of adaptability To enhance the adaptability of underwater-specific power equipment to the underwater environment, To enhance the adaptability of underwater specialized power equipment to load changes, and The adaptive weighting factors are determined based on changes in the underwater environment and load. Finally, the efficiency, reliability, and adaptability indicators are integrated to establish a comprehensive performance evaluation index.

[0061] By constructing a comprehensive performance evaluation index that includes efficiency, reliability, and adaptability indicators, the performance of submersible motors can be comprehensively and objectively evaluated. This will guide corresponding optimization decisions, improve the operating efficiency, stability, and adaptability of the equipment, and ultimately reduce operating costs and risks.

[0062] Furthermore, the preprocessing of the time-series monitoring dataset, step S400 of this application further includes: Step S410: Perform data correlation analysis on the time-series monitoring dataset to construct independent data identifiers.

[0063] Step S420: Use the data corresponding to the independent data identifier as sensitive data and perform global similarity verification within the time series monitoring dataset.

[0064] Step S430: Perform corresponding data removal based on the global similarity verification results, and complete the time series monitoring dataset preprocessing based on the removal results.

[0065] Specifically, before preprocessing the time-series monitoring dataset, correlation analysis needs to be performed on the data. Correlation analysis aims to discover whether there are relationships or patterns between data items. In time-series datasets, such correlations may manifest as temporal correlations, value trends, or cross-referencing with other data sources. Specifically, statistical methods (such as correlation analysis and regression analysis), data mining techniques (such as cluster analysis and association rule mining), or machine learning algorithms (such as deep learning models) can be used for correlation analysis. Based on the correlation analysis, data items with unique characteristics or attributes are identified, and independent data identifiers are constructed for them. These independent data identifiers can be unique identifiers (such as ID numbers), specific time-series patterns, or outliers above a certain threshold. The construction of independent data identifiers facilitates the subsequent identification and management of sensitive data.

[0066] Furthermore, data with independent data identifiers are designated as sensitive data, and global similarity verification is performed within the time-series monitoring dataset. Global similarity verification aims to check whether there are data items in the entire dataset that are similar to or related to the sensitive data. This can be achieved by calculating the similarity between data items (such as cosine similarity, Euclidean distance, etc.) or using machine learning algorithms (such as the K-nearest neighbors algorithm). The results of global similarity verification determine which data items are sufficiently similar to the sensitive data to warrant being treated as sensitive data. Based on the results of global similarity verification, data items similar to or related to the sensitive data are identified, and based on preset thresholds or rules, which data items need to be removed or processed. Finally, data removal operations are performed, removing these data items from the dataset or anonymizing or encrypting them. The preprocessed dataset will no longer contain sensitive data or data items highly related to sensitive data, thus reducing the risk of data leakage. Simultaneously, the preprocessed dataset is more suitable for subsequent data analysis and mining operations.

[0067] Through the technical solutions of the above embodiments, the dual-identification network anomaly prediction submersible motor compensation control method provided in this application solves the technical problems of low equipment operating efficiency, poor stability and adaptability, inability to timely and accurately predict and optimize anomalies during the submersible motor control optimization process, which increases the failure rate and maintenance cost. It achieves the technical effect of accurate anomaly prediction and optimized control of submersible motors, improving equipment operating efficiency and safety, and reducing failure rate and maintenance cost. Example

[0068] Based on the same inventive concept as the dual-identification network anomaly prediction and under-submersible motor compensation control method in the aforementioned embodiments, such as Figure 2 As shown, this application provides a motor compensation control system for underwater vehicles using a dual-identification network anomaly prediction system. The system includes: The model building module 11 is used to establish an anomaly prediction model. The anomaly prediction model is a model for predicting anomalies in underwater special power equipment. The anomaly prediction model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is constructed using the usage dataset of underwater special power equipment, and the second anomaly identification network is constructed using big data.

[0069] The equipment monitoring module 12 is used to perform equipment monitoring of the underwater special power equipment and establish a time-series monitoring dataset.

[0070] The model self-checking module 13 is used to perform a model self-check of the anomaly prediction model and to complete the dynamic weight configuration of the first anomaly identification network and the second anomaly identification network based on the model self-checking results.

[0071] The anomaly prediction module 14 is used to preprocess the time-series monitoring dataset and input it into the anomaly prediction model with dynamically configured weights to generate anomaly prediction results.

[0072] The strategy compensation module 15 is used to evaluate the anomaly prediction results, establish a tiered compensation strategy based on the anomaly evaluation results, perform optimization control through the tiered compensation strategy, and set a verification window.

[0073] The verification and identification module 16 is used for verification and identification through the verification window to optimize control, and to complete the optimization control based on the verification and identification results.

[0074] Furthermore, the model self-test module 13 is also used to perform the following steps: The self-checking processing layer is invoked to perform a data richness evaluation of the first anomaly identification network. Based on the data richness evaluation results, a first self-checking result is constructed. The self-checking processing layer is a sub-processing layer of the anomaly prediction model.

[0075] The self-check processing layer is invoked to perform data adaptation evaluation of the second abnormal network. Based on the adaptation evaluation results, a second self-check result is constructed. The first and second self-check results are integrated to complete the dynamic weight configuration.

[0076] Furthermore, the strategy compensation module 15 is also used to perform the following steps: Analyze the anomaly evaluation results to obtain the maximum anomaly deviation value.

[0077] The compensation extreme value of the step compensation strategy is established based on the maximum abnormal deviation value.

[0078] A stability analysis was performed on the underwater dedicated power equipment, and the number and length of steps were configured based on the results of the stability analysis.

[0079] The step compensation strategy is established by defining the compensation extreme value, the number of steps, and the step length.

[0080] Furthermore, the verification and identification module 16 is also used to perform the following steps: Based on the verification and identification results, a node deviation is constructed, which is the mapping deviation between the verification window and each step node in the step compensation strategy.

[0081] Unit deviation analysis, which controls the node deviation, generates optimized compensation.

[0082] Based on the node deviation, the uncompensated step compensation strategy is reconstructed, and the control is further optimized based on the reconstruction results and optimized compensation.

[0083] Furthermore, this application also includes a performance evaluation module, which is further configured to perform the following steps: Historical data was extracted from underwater specialized power equipment to create a denoised historical dataset.

[0084] Configure comprehensive performance evaluation indicators, evaluate the performance of underwater dedicated power equipment using the comprehensive performance evaluation indicators and the historical dataset, and generate optimized control strategies based on the performance evaluation results.

[0085] The optimized control strategy is used to optimize the task response of underwater dedicated power equipment.

[0086] Furthermore, the performance evaluation module is also used to perform the following steps: The efficiency index is established using the following formula: .

[0087] in, As an efficiency indicator, For power efficiency, For power stability, and These are the weighting factors for power efficiency and power stability, respectively.

[0088] The reliability index is established using the following formula: .

[0089] in, As a reliability indicator, Characterizing the failure rate, Characterizing fault-free uptime, and These are the weighting factors for failure rate and fault-free uptime, respectively.

[0090] Establish an adaptation index, with the following formula: .

[0091] in, Characteristic of adaptability To enhance the adaptability of underwater-specific power equipment to the underwater environment, To enhance the adaptability of underwater specialized power equipment to load changes, and It is an adaptive weighting factor for changes in underwater environment and load.

[0092] By integrating the aforementioned efficiency, reliability, and adaptability indicators, a comprehensive performance evaluation index is established.

[0093] Furthermore, the anomaly prediction module 14 is also used to perform the following steps: Perform data correlation analysis on the time-series monitoring dataset to construct independent data identifiers.

[0094] The data corresponding to the independent data identifier is used as sensitive data, and global similarity verification is performed within the time-series monitoring dataset.

[0095] Based on the global similarity verification results, corresponding data are removed, and the time series monitoring dataset is preprocessed based on the removal results.

[0096] Through the foregoing detailed description of the dual-identification network anomaly prediction submersion motor compensation control method, those skilled in the art can clearly understand that the dual-identification network anomaly prediction submersion motor compensation control system in this embodiment is relatively simple to describe as it corresponds to the method disclosed in the embodiment. For relevant parts, please refer to the method section description.

[0097] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A motor compensation control method for undersea drilling using dual-identification network anomaly prediction, characterized in that, The method includes: An anomaly prediction model is established, which is a model for predicting anomalies in underwater dedicated power equipment. The anomaly prediction model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is constructed using the usage dataset of underwater dedicated power equipment, and the second anomaly identification network is constructed using big data. Perform equipment monitoring of the underwater dedicated power equipment and establish a time-series monitoring dataset; Perform a model self-check of the anomaly prediction model, and complete the dynamic weight configuration of the first anomaly identification network and the second anomaly identification network based on the model self-check results; After preprocessing the time-series monitoring dataset, it is input into the anomaly prediction model with dynamic weight configuration to generate anomaly prediction results. Anomaly evaluation is performed on the anomaly prediction results, a tiered compensation strategy is established based on the anomaly evaluation results, optimization control is performed through the tiered compensation strategy, and a verification window is set. The verification window is used to perform verification and identification for optimization control, and the optimization control is completed based on the verification and identification results.

2. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 1, characterized in that, The step of performing a model self-check on the anomaly prediction model, and configuring the dynamic weights of the first and second anomaly detection networks based on the model self-check results, further includes: The self-checking processing layer is invoked to perform data richness evaluation of the first anomaly identification network, and the first self-checking result is constructed based on the data richness evaluation result. The self-checking processing layer is a sub-processing layer of the anomaly prediction model. The self-check processing layer is invoked to perform data adaptation evaluation of the second abnormal network. Based on the adaptation evaluation results, a second self-check result is constructed. The first and second self-check results are integrated to complete the dynamic weight configuration.

3. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 1, characterized in that, The step of evaluating the anomaly prediction results and establishing a tiered compensation strategy based on the anomaly evaluation results further includes: Analyze the anomaly evaluation results to obtain the maximum anomaly deviation value; The compensation extreme value of the step compensation strategy is established based on the maximum abnormal deviation value; A stability analysis was performed on the underwater dedicated power equipment, and the number and length of steps were configured based on the stability analysis results. The step compensation strategy is established by defining the compensation extreme value, the number of steps, and the step length.

4. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 3, characterized in that, The verification and identification process for optimization control via the verification window, and the completion of optimization control based on the verification and identification results, further includes: Based on the verification and identification results, a node deviation is constructed, which is the mapping deviation between the verification window and each step node in the step compensation strategy. Unit deviation analysis, which controls the node deviation, generates optimized compensation. Based on the node deviation, the uncompensated step compensation strategy is reconstructed, and the control is further optimized based on the reconstruction results and optimized compensation.

5. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 1, characterized in that, The method further includes: Historical data of underwater specialized power equipment was extracted to establish a denoised historical dataset; Configure comprehensive performance evaluation indicators, evaluate the performance of underwater dedicated power equipment using the comprehensive performance evaluation indicators and the historical dataset, and generate an optimized control strategy based on the performance evaluation results. The optimized control strategy is used to optimize the task response of underwater dedicated power equipment.

6. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 5, characterized in that, The overall performance evaluation indicators for the configuration also include: The efficiency index is established using the following formula: ； in, As an efficiency indicator, For power efficiency, For power stability, and These are the weighting factors for power efficiency and power stability, respectively. The reliability index is established using the following formula: ； in, As a reliability indicator, Characterizing the failure rate, Characterizing fault-free uptime, and These are the weighting factors for failure rate and fault-free uptime, respectively. Establish an adaptation index, with the following formula: ； in, Characteristic of adaptability To enhance the adaptability of underwater-specific power equipment to the underwater environment, To enhance the adaptability of underwater specialized power equipment to load changes, and For adaptive weighting factors of underwater environment and load changes; By integrating the aforementioned efficiency, reliability, and adaptability indicators, a comprehensive performance evaluation index is established.

7. The motor compensation control method for undersea drilling using dual-identification network anomaly prediction as described in claim 1, characterized in that, The preprocessing of the time-series monitoring dataset further includes: Perform data correlation analysis on the time-series monitoring dataset to construct independent data identifiers; The data corresponding to the independent data identifier is used as sensitive data, and global similarity verification is performed within the time-series monitoring dataset. Based on the global similarity verification results, corresponding data are removed, and the time series monitoring dataset is preprocessed based on the removal results.

8. A dual-identification network anomaly prediction and submersion motor compensation control system, characterized in that, The system for implementing the method according to any one of claims 1-7 comprises: The model building module is used to build an anomaly prediction model. The anomaly prediction model is a model for predicting anomalies in underwater special power equipment. The anomaly prediction model includes a first anomaly identification network and a second anomaly identification network. The first anomaly identification network is built using the usage dataset of underwater special power equipment, and the second anomaly identification network is built using big data. The equipment monitoring module is used to perform equipment monitoring of the underwater dedicated power equipment and establish a time-series monitoring dataset; The model self-checking module is used to perform a model self-check of the anomaly prediction model and to complete the dynamic weight configuration of the first anomaly identification network and the second anomaly identification network based on the model self-checking results. The anomaly prediction module is used to preprocess the time-series monitoring dataset and input it into the anomaly prediction model with dynamically configured weights to generate anomaly prediction results. The strategy compensation module is used to evaluate the anomaly prediction results, establish a tiered compensation strategy based on the anomaly evaluation results, optimize and control the system through the tiered compensation strategy, and set a verification window. The verification and identification module is used for verification and identification through the verification window to optimize control, and to complete the optimization control based on the verification and identification results.

Images