Underwater magnetic induction type non-contact current detection control system and method

Abstract

The invention discloses an underwater magnetic induction type non-contact current detection control system and method, and relates to the technical field of underwater non-contact electric energy transmission and detection. The method comprises the following steps: constructing an underwater electromagnetic model, and carrying out adaptive detection area division by combining deep learning with a genetic algorithm; magnetic field attribute parameter characteristics are obtained based on the division result, and a detection priority value is automatically calibrated for each detection area; introducing a hybrid algorithm to construct a data processing model, and performing optimization simulation training according to parameters such as region detection priority; constructing and training a control model based on a reinforcement learning algorithm and the control parameters; the magnetic field change, the interference condition and the equipment state in the detection process are monitored in real time, the analysis result is fed back to the data processing model and the control model, and self-correction and real-time adjustment are carried out. The accuracy and stability of underwater magnetic induction type non-contact current detection control are improved, and an efficient and reliable method is provided for underwater current detection.

Description

Technical Field

[0001] The invention belongs to the technical field of underwater non-contact electric energy transmission and detection, and specifically relates to an underwater magnetic induction non-contact current detection control system and method. Background Art

[0002] Accurate detection of underwater currents is of vital importance in the fields of marine resource development, underwater facility maintenance, and marine scientific research. The distribution and changes of underwater currents are not only closely related to the physical and chemical processes in the marine environment, but also have a direct impact on the operating status and safety of underwater equipment. For example, in deep-sea oil and gas production, the normal operation of underwater equipment depends on the precise grasp of the surrounding current environment to avoid equipment failures due to electromagnetic interference; in marine ecological research, the monitoring of underwater currents helps to understand the habitats of marine organisms and the dynamic changes of the ecosystem. Traditional underwater current detection methods have many limitations. Some contact detection methods require the detection equipment to be in direct contact with the object being measured, which is not only easy to cause damage to underwater equipment and the environment, but may also be affected by the complex underwater environment, such as the corrosiveness of seawater, water pressure, etc., resulting in inaccurate detection results or equipment damage. Moreover, traditional detection methods often lack the ability to adapt to complex underwater environments and are difficult to flexibly adjust according to different underwater environmental parameters and detection requirements. For example, in different sea areas, the conductivity, temperature, salinity and other parameters of seawater are different, and traditional methods are difficult to achieve efficient and accurate detection in these changing environments. With the increasing complexity of underwater operations and the increasing requirements for detection accuracy, it is urgent to develop an efficient, accurate, non-contact current detection and control method that can adapt to complex underwater environments. The underwater magnetic induction non-contact current detection and control method came into being. It uses underwater electromagnetic characteristics, builds electromagnetic models and intelligent detection strategies to achieve accurate detection and effective control of underwater currents, overcomes the shortcomings of traditional methods, and provides more reliable technical support for underwater operations. Summary of the invention

[0003] In order to overcome the shortcomings and deficiencies of the above-mentioned prior art, the first purpose of the present invention is to provide an underwater magnetic induction type non-contact current detection and control system; the second purpose of the present invention is to provide an underwater magnetic induction type non-contact current detection and control system.

[0004] The first object of the present invention adopts the following technical solution: Underwater magnetic induction non-contact current detection and control system, including: Data preprocessing and model building module: used to collect and preprocess underwater environmental parameters and detection requirement parameters; build underwater electromagnetic models; use deep learning combined with genetic algorithms to build intelligent detection strategy models; Virtual detection area division module: divides the underwater electromagnetic three-dimensional virtual model into virtual detection areas according to the trained detection strategy model; generates control signals and transmits them to the detection equipment; Detection area priority calibration module: used to obtain the magnetic field attribute parameter characteristics of each detection area; automatically calibrate the detection priority value for each detection area according to the magnetic field attribute parameters; Data processing model construction and optimization module: used to construct data processing models, including basic algorithm sub-models, optimization algorithm sub-models, hybrid algorithm sub-models and algorithm working switching conditions; optimize and simulate the data processing model using regional detection priorities, underwater environmental parameters and detection requirement parameters; Control model building and detection execution module: build a control model based on reinforcement learning algorithm and control parameters; use data processing model and control model to control detection equipment to detect underwater current; Real-time monitoring and self-correction module: Use sensors to monitor magnetic field changes, interference and equipment status during the detection process in real time; analyze monitoring parameters and feed back analysis results to the data processing model and control model; self-correct the detection strategy and adjust the control parameters in real time.

[0005] The second object of the present invention adopts the following technical solution: The underwater magnetic induction type non-contact current detection control method is used to realize the underwater magnetic induction type non-contact current detection control system. The method flow is as follows: Step 1: Collect and preprocess underwater environmental parameters and detection requirement parameters, build an underwater electromagnetic model, and use deep learning combined with genetic algorithms to build an intelligent detection strategy to adaptively divide the underwater electromagnetic model into detection areas; Step 2: Based on the detection area division results, obtain the magnetic field attribute parameter characteristics of each detection area, and use the local priority strategy to automatically calibrate the detection priority value for each detection area; Step 3: Use the hybrid algorithm to build a data processing model, and use the regional detection priority, underwater environment parameters and detection requirement parameters to optimize the data processing model for simulation training; Step 4: Build and train a control model based on the reinforcement learning algorithm and control parameters, and use the data processing model and the control model to control the detection equipment to detect underwater currents; Step 5: Use sensors to monitor the detection process in real time, analyze the monitoring parameters, and feed the analysis results back to the data processing model and control model for self-correction and adjustment.

[0006] Preferably, underwater environment parameters and detection requirement parameters are collected and preprocessed, specifically including: collecting electromagnetic parameters of the underwater environment, detection requirement parameters and historical detection strategies, and preprocessing the collected parameters through normalization and standardization to make them conform to a unified input format.

[0007] Preferably, constructing an underwater electromagnetic model specifically includes: constructing o-xyz three-dimensional coordinates with the center of the detection area as the origin, inputting the preprocessed underwater environmental parameters into the three-dimensional modeling software, constructing an underwater electromagnetic three-dimensional virtual model within the three-dimensional coordinates, and marking the attribute parameters including magnetic field intensity and current density on the constructed three-dimensional virtual model.

[0008] Preferably, constructing an intelligent detection strategy specifically includes: constructing a detection strategy model using a BP neural network combined with a genetic algorithm, constructing a model input based on the acquired underwater environmental parameters, detection requirement parameters and historical detection strategies, and constructing a model output as a detection strategy, wherein a computer software is used to extract the underwater environmental parameter vectors marked on the three-dimensional virtual model. ;in, represents the nth type of underwater environmental parameter, where n represents the number of categories of underwater environmental parameters; represents the detection requirement attribute parameter of the mth category, and m represents the number of categories of the detection requirement attribute parameter; The initial detection strategy and the historical detection strategy are used to construct the initial population of the genetic algorithm. The obtained initial population is used to construct the qth chromosome of the genetic algorithm input, and the following constraint condition set S is set for each variable in the chromosome: ;in, , , , They represent the number of detection areas, detection accuracy, detection angle, and speed of the detection equipment moving process in the qth chromosome respectively; , , , are the minimum values ​​of the corresponding parameters respectively; , , , are the maximum values ​​of the corresponding parameters respectively; The fitness function corresponding to the genetic algorithm is constructed using the initial population. The specific formula is: ;in, Indicates the detection time corresponding to the qth chromosome; is an indicator function, indicating that when the qth chromosome corresponds to the detection time If the preset maximum time is exceeded, the value is 1, otherwise it is 0; Indicates the penalty coefficient when the detection time of the qth chromosome exceeds the preset maximum time; Indicates the detection quality score corresponding to the qth chromosome, It represents the penalty term for violating the sth constraint in the constraint set S corresponding to the qth chromosome. If the sth constraint in the constraint set S is not satisfied, the term is 1, otherwise it is 0, which represent the weighted coefficients of the fitness function in turn; , , , All are weight factors; The iteration cycle and stop condition are set, the constructed qth chromosome, constraint set S, and fitness function are input into the genetic algorithm, the above process is iterated, and when the iteration cycle or stop condition is met, the iteration is stopped, and the obtained optimal detection strategy is output; the trained detection strategy model is integrated into the three-dimensional modeling software, the underwater electromagnetic three-dimensional virtual model is divided into virtual detection areas, and the virtual division process uses the PID algorithm to generate corresponding control signals, and the control signals are transmitted to the corresponding detection equipment using sensors, and the detection equipment is controlled to divide the underwater area into detection areas to ensure that the detection area division is consistent with the underwater magnetic induction current detection.

[0009] Preferably, after the underwater electromagnetic model is adaptively divided into detection areas, each area after the detection area is divided is gridded in a fixed-size square grid form in the three-dimensional modeling software and the divided grids are numerically labeled, and the magnetic field properties related to the underwater magnetic induction current in each grid after the labeling are obtained, including magnetic field strength and direction parameter characteristics, which are input into the clustering algorithm with the corresponding labels for clustering operation, and the grid labels belonging to the same category after clustering are merged to obtain each corresponding detection area; historical attribute parameters and detection requirement parameters of the same underwater environment detection area are collected, and the detection priority of each detection area is marked according to the collected parameters by expert experience to obtain priority label data, and the priority is divided into Class A, Class B, and Class C, where Class C is a low priority, Class B is a medium priority, and Class A is a high priority; a local priority model is constructed by using support vector regression SVR, and the obtained priority label data set is input into the local priority model for training to obtain the trained local priority model, and the trained local priority model is integrated into the underwater electromagnetic three-dimensional virtual model to automatically mark the detection area with priority.

[0010] Preferably, in the process of constructing and training the control model, control constraints are constructed, specifically including: constructing control constraints based on underwater environment parameters, detection requirement parameters and control parameters, including: ; in, It indicates the magnetic field induction intensity of the detection device at time t. It indicates the optimal threshold corresponding to the magnetic field induction strength of the detection device at time t. Indicates the change of magnetic field in the detection area at time t, Indicates the target magnetic field change in the detection area at time t, represents the data processing sub-model executed by the detection device at time t, Represents the data processing sub-model executed by the detection device target at time t, Indicates the anti-interference capability of the detection device at time t, Indicates the target anti-interference capability of the detection equipment. Indicates the speed of the device moving at time t. Indicates the target speed at which the detection device moves; , , , , is the preset error threshold.

[0011] Preferably, after constructing the control constraint conditions, constructing the control trigger conditions, specifically including: when When , the first control condition is triggered; when When , the second control condition is triggered; when When , the third control condition is triggered; when When , the fourth control condition is triggered; when When , the fifth control condition is triggered; When two or more of the above control constraints are not met, the sixth control condition is triggered.

[0012] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: 1. The present invention constructs an intelligent detection strategy model through deep learning combined with genetic algorithm, and adaptively divides the underwater electromagnetic model into detection areas. This method can flexibly adjust key parameters such as the number, accuracy, angle and moving speed of the detection area according to specific underwater environmental parameters and detection requirement parameters, thereby ensuring that each detection area can obtain the most optimized resource allocation and improving the overall accuracy and efficiency of the detection.

[0013] 2. The present invention introduces a local priority strategy to automatically calibrate the detection priority value according to the magnetic field attribute parameters of each detection area, so that the system can dynamically adjust resource allocation and give priority to those problem areas with higher potential value or more urgent to be solved, thereby enhancing the flexibility and adaptability of the system; by introducing a data processing model constructed by a hybrid algorithm, different data processing strategies are selected according to the priority of the detection area, which improves the effectiveness and pertinence of data processing, and allows the system to make the best choice when facing tasks of different complexity, thereby improving overall work efficiency.

[0014] 3. The present invention monitors the magnetic field changes, interference conditions and equipment status in real time during the detection process, and analyzes the monitoring parameters through the configured monitoring algorithm. These analysis results are fed back to the data processing model and the control model for self-correction of the detection strategy and real-time adjustment of the control parameters. This mechanism ensures that even in a constantly changing underwater environment, it can maintain high detection accuracy and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 The module diagram of the underwater magnetic induction non-contact current detection control system of the present invention is shown; Figure 2 A flow chart showing the underwater magnetic induction non-contact current detection control method of the present invention is shown; Figure 3 The specific flow chart of step 2 of the present invention is shown. DETAILED DESCRIPTION

[0017] 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.

[0018] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0019] Embodiment 1: See also Figure 1 As shown, the underwater magnetic induction non-contact current detection control system of this embodiment includes Data preprocessing and model building module: used to collect and preprocess underwater environmental parameters and detection requirement parameters; build underwater electromagnetic models; use deep learning combined with genetic algorithms to build an intelligent detection strategy model.

[0020] Virtual detection area division module: divides the underwater electromagnetic three-dimensional virtual model into virtual detection areas according to the trained detection strategy model; generates control signals and transmits them to the detection equipment.

[0021] Detection area priority calibration module: used to obtain the magnetic field attribute parameter characteristics of each detection area; automatically calibrate the detection priority value for each detection area according to the magnetic field attribute parameters.

[0022] Data processing model construction and optimization module: used to construct data processing models, including basic algorithm sub-models, optimization algorithm sub-models, hybrid algorithm sub-models and algorithm working switching conditions; use regional detection priorities and underwater environmental parameters and detection requirement parameters to optimize the data processing model for simulation training.

[0023] Control model construction and detection execution module: Build a control model based on reinforcement learning algorithm and control parameters; use data processing model and control model to control the detection equipment to detect underwater current.

[0024] Real-time monitoring and self-correction module: Use sensors to monitor magnetic field changes, interference and equipment status during the detection process in real time; analyze monitoring parameters and feed back analysis results to the data processing model and control model; self-correct the detection strategy and adjust the control parameters in real time.

[0025] The beneficial effects of this embodiment are: realizing intelligent division and priority calibration of detection areas, thereby improving detection efficiency; ensuring detection accuracy by optimizing data processing models and control models; and enhancing the stability and adaptability of the system with real-time monitoring and self-correction functions, thereby improving overall detection performance.

[0026] Embodiment 2: See also Figure 2 As shown, the underwater magnetic induction non-contact current detection control method of this embodiment has the following process: Step 1: For underwater magnetic induction non-contact current detection control, collect and preprocess underwater environmental parameters and detection requirement parameters, use the preprocessed underwater environmental parameters to build an underwater electromagnetic model, and based on the constructed electromagnetic model, use deep learning combined with the genetic algorithm to build an intelligent detection strategy to adaptively divide the underwater electromagnetic model into detection areas, and use the preprocessed detection requirement parameters to adaptively adjust the division process.

[0027] The details are as follows: S101, collecting electromagnetic parameters, detection requirement parameters and historical detection strategies of the underwater environment, pre-processing the collected parameters to make them conform to a unified input format, and especially focusing on processing parameters related to underwater magnetic induction characteristics.

[0028] S102, constructing o-xyz three-dimensional coordinates with the center of the detection area as the origin, inputting the preprocessed underwater environmental parameters into the three-dimensional modeling software, and constructing an underwater electromagnetic three-dimensional virtual model within the three-dimensional coordinates. The model focuses on reflecting the characteristics such as the magnetic field distribution generated by the underwater current, and annotating the attribute parameters on the constructed three-dimensional virtual model.

[0029] S103. Utilize the BP neural network combined with the genetic algorithm to construct a detection strategy model. The model input is constructed according to the acquired underwater environmental parameters, detection requirement parameters and historical detection strategies. The model output is the detection strategy. The detection strategy covers the number of detection areas N, detection accuracy P, detection angle A and real-time speed V of the detection equipment during movement. Each parameter is closely set around the underwater magnetic induction current detection requirements.

[0030] S104: Input the constructed input and output into the detection strategy model for training, and obtain the trained detection strategy model; specifically including: S1041, using computer software to extract the underwater environment parameter vector marked on the three-dimensional virtual model, and at the same time taking the detection requirement parameters as the target control variables in the training process and performing numerical preprocessing, using the preprocessed detection requirement parameters to construct the detection requirement attribute parameter vector, and cascading the underwater environment parameter vector and the detection requirement attribute parameter vector to form a joint input vector ; in, represents the nth type of underwater environmental parameter, where n represents the number of categories of underwater environmental parameters; It represents the detection requirement attribute parameter of the mth category, and m represents the number of categories of the detection requirement attribute parameter.

[0031] S1042. Construct a BP neural network using three convolutional layers and two fully connected layers connected in series, input the constructed joint input vector into the BP neural network, and use the collected detection requirement attribute parameters and the set target detection requirement attribute parameters to obtain the mean square error as the training error, train the constructed BP neural network, and output the initial detection strategy.

[0032] S1043, constructing an initial population of the genetic algorithm using the initial detection strategy and the historical detection strategy; constructing the qth chromosome of the genetic algorithm input using the obtained initial population, and setting the following constraint condition set S for each variable in the chromosome: ; in, , , , They represent the number of detection areas, detection accuracy, detection angle, and speed of the detection equipment moving process in the qth chromosome respectively; , , , are the minimum values ​​of the corresponding parameters respectively; , , , are the maximum values ​​of the corresponding parameters respectively.

[0033] S1044. Use the initial population to construct a fitness function corresponding to the genetic algorithm. The specific formula is: ; in, Indicates the detection time corresponding to the qth chromosome; is an indicator function, indicating that when the qth chromosome corresponds to the detection time If the preset maximum time is exceeded, the value is 1, otherwise it is 0; Indicates the penalty coefficient when the detection time of the qth chromosome exceeds the preset maximum time; Indicates the detection quality score corresponding to the qth chromosome, It represents the penalty term for violating the sth constraint in the constraint set S corresponding to the qth chromosome. If the sth constraint in the constraint set S is not satisfied, the term is 1, otherwise it is 0, which represent the weighted coefficients of the fitness function in turn; , , , are all weight factors.

[0034] S1045, set the iteration cycle and stop condition, input the constructed qth chromosome, constraint condition set S, and fitness function into the genetic algorithm, iterate the above S1041 - S1044 process, and when the iteration cycle or stop condition is met, stop the iteration and output the obtained optimal detection strategy.

[0035] S105. Integrate the trained detection strategy model into the three-dimensional modeling software, divide the underwater electromagnetic three-dimensional virtual model into virtual detection areas, and use the PID algorithm to generate corresponding control signals for the virtual division process. Use sensors to transmit the control signals to corresponding detection equipment, control the detection equipment to divide the underwater area into detection areas, and ensure that the detection area division is consistent with the underwater magnetic induction current detection.

[0036] Step 2: Based on the detection area division results, obtain the magnetic field attribute parameter characteristics of each detection area, and use the configured local priority strategy to automatically calibrate the detection priority value for each detection area according to the magnetic field attribute parameters of each detection area, including: See also Figure 3 As shown, the specific steps are as follows: S201. In the three-dimensional modeling software, each area after the detection area is divided into grids using a fixed-size square grid, and the divided grids are numerically labeled to obtain the attribute parameter characteristics in each labeled grid, with a focus on the magnetic field properties related to the underwater magnetic induction current.

[0037] S202, input the acquired attribute parameter features and corresponding labels in each grid into the clustering algorithm, perform clustering operation, and merge the grid labels belonging to the same category after clustering to obtain each corresponding detection area.

[0038] S203, construct a local priority model, and input the acquired attribute parameter features of each detection area into the local priority model for training, and integrate the trained local priority model into the underwater electromagnetic three-dimensional virtual model in S105 to automatically mark the corresponding detection area priority of the three-dimensional virtual model detection area after the detection area is divided. The specific steps of local priority model training include: S2031. Collect historical attribute parameters and detection requirement parameters of the same underwater environment detection area, and use expert experience to mark the detection priority of each detection area according to the collected parameters to obtain priority label data; the priority is divided into Class A, Class B, and Class C, of ​​which Class C is a low priority, Class B is a medium priority, and Class A is a high priority.

[0039] S2032. Use support vector regression SVR to build a local priority model, and input the obtained priority label data set into the local priority model for training to obtain a trained local priority model.

[0040] Step 3: Introduce a hybrid algorithm, construct a data processing model, and use the acquired regional detection priority, underwater environmental parameters, and detection requirement parameters to optimize and simulate the constructed data processing model in a simulation environment to obtain a trained data processing model; wherein the data processing model includes a basic algorithm sub-model, an optimization algorithm sub-model, a hybrid algorithm sub-model, and an algorithm work switching condition, and the algorithm work switching condition specifically includes: S3031. When the priority of the detection area is low, the basic algorithm sub-model in the data processing model is used to process the data.

[0041] S3032. When the priority of the detection area is medium priority, the hybrid algorithm sub-model in the data processing model is used to process the data.

[0042] S3033. When the priority of the detection area is high priority, the optimization algorithm sub-model in the data processing model is used to process the data. Step 4: construct and train a control model based on the reinforcement learning algorithm and control parameters, and use the acquired data processing model and control model to control the detection equipment to detect underwater currents; wherein the specific steps of constructing the control model include: S401, construct control constraints based on underwater environmental parameters, detection requirement parameters and control parameters, including magnetic induction intensity and optimal magnetic induction intensity Comparison of magnetic field changes and optimal magnetic field variation Comparison of data processing sub-models and Best Execution Comparison and anti-interference ability and optimal anti-interference capability Compare and check the speed and optimal speed of mobile devices of comparison; among them, represents the optimal value; in, It indicates the magnetic field induction intensity of the detection device at time t. It indicates the optimal threshold corresponding to the magnetic field induction strength of the detection device at time t. Indicates the change of magnetic field in the detection area at time t, Indicates the target magnetic field change in the detection area at time t, represents the data processing sub-model executed by the detection device at time t, Represents the data processing sub-model executed by the detection device target at time t, Indicates the anti-interference capability of the detection device at time t, Indicates the target anti-interference capability of the detection equipment. Indicates the speed of the device moving at time t. Indicates the target speed of the detection device.

[0043] S402. According to the constraint conditions in S401, construct control trigger conditions, specifically including: when and do not satisfy the control constraint conditions, trigger the first control condition. When and do not satisfy the control constraint conditions, trigger the second control condition. When and do not satisfy the control constraint conditions, trigger the third control condition. When and do not satisfy the control constraint conditions, trigger the fourth control condition. When and do not satisfy the control constraint conditions, trigger the fifth control condition. When two or more control constraints in S402 are not satisfied, trigger the sixth control condition. The first control condition: It is triggered when the magnetic field induction intensity of the detection device at time t is not equal to the corresponding optimal threshold. This means that the current magnetic field induction intensity has not reached the expected optimal level, and the position or setting of the detection device may need to be adjusted.

[0044] The second control condition: The trigger is when the magnetic field change in the detection area does not match the target magnetic field change at time t. This means that the actual detected magnetic field change is different from the expected target, which may indicate a change in the environment or a problem with the device's operating status.

[0045] The third control condition: This condition indicates that the currently used data processing method may not be suitable for the current detection task or environmental conditions, and needs to be switched to a more suitable data processing strategy.

[0046] The fourth control condition: It is triggered when the anti-interference capability of the detection device is inconsistent with the target anti-interference capability at time t. This usually means that the current anti-interference performance of the device does not meet the requirements, which may affect the accuracy of the detection results, and measures need to be taken to enhance its stability.

[0047] The fifth control condition: When the speed of the detection device at time t is different from the target speed, the condition is triggered. This condition indicates that the actual moving speed of the detection device deviates from the planned speed, which may lead to low detection efficiency or miss the key area.

[0048] The sixth control condition is triggered when two or more of the above control constraints are not met at the same time. This situation is more serious, meaning that the system has deviated from the optimal working state in many aspects and needs to be fully checked and adjusted.

[0049] Step 5. Use the sensors integrated on the detection equipment to monitor the magnetic field changes, interference and equipment status during the detection process in real time, and use the monitoring algorithm configured in the sensor to analyze the monitoring parameters. The analysis results are fed back to the data processing model and the control model to self-correct the detection strategy and adjust the control parameters in real time to ensure the accuracy and stability of the underwater magnetic induction non-contact current detection control.

[0050] The beneficial effects of this embodiment are: dividing the detection area through intelligent strategies, automatically calibrating priorities, using hybrid algorithms to optimize data processing, building an adaptive control model, real-time monitoring and self-correction, improving detection efficiency and accuracy, enhancing system stability, and being suitable for complex underwater environments.

[0051] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0052] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. Underwater magnetic induction non-contact current detection and control system, characterized in that: The system comprises: Data preprocessing and model building module: used to collect and preprocess underwater environmental parameters and detection requirement parameters; build underwater electromagnetic models; use deep learning combined with genetic algorithms to build intelligent detection strategy models; Virtual detection area division module: divides the underwater electromagnetic three-dimensional virtual model into virtual detection areas according to the trained detection strategy model; generates control signals and transmits them to the detection equipment; Detection area priority calibration module: used to obtain the magnetic field attribute parameter characteristics of each detection area; automatically calibrate the detection priority value for each detection area according to the magnetic field attribute parameters; Data processing model construction and optimization module: used to construct data processing models, including basic algorithm sub-models, optimization algorithm sub-models, hybrid algorithm sub-models and algorithm working switching conditions; optimize and simulate the data processing model using regional detection priorities, underwater environmental parameters and detection requirement parameters; Control model building and detection execution module: build a control model based on reinforcement learning algorithm and control parameters; use data processing model and control model to control detection equipment to detect underwater current; Real-time monitoring and self-correction module: Use sensors to monitor magnetic field changes, interference and equipment status during the detection process in real time; analyze monitoring parameters and feed back analysis results to the data processing model and control model; self-correct the detection strategy and adjust the control parameters in real time.

2. An underwater magnetic induction type non-contact current detection control method, used to implement the underwater magnetic induction type non-contact current detection control system as claimed in claim 1, characterized in that: The process of the underwater magnetic induction non-contact current detection control method is as follows: Step 1: Collect and preprocess underwater environmental parameters and detection requirement parameters, build an underwater electromagnetic model, and use deep learning combined with genetic algorithms to build an intelligent detection strategy to adaptively divide the underwater electromagnetic model into detection areas; Step 2: Based on the detection area division results, obtain the magnetic field attribute parameter characteristics of each detection area, and use the local priority strategy to automatically calibrate the detection priority value for each detection area; Step 3: Use the hybrid algorithm to build a data processing model, and use the regional detection priority, underwater environment parameters and detection requirement parameters to optimize the data processing model for simulation training; Step 4: Build and train a control model based on the reinforcement learning algorithm and control parameters, and use the data processing model and the control model to control the detection equipment to detect underwater currents; Step 5: Use sensors to monitor the detection process in real time, analyze the monitoring parameters, and feed the analysis results back to the data processing model and control model for self-correction and adjustment.

3. The underwater magnetic induction non-contact current detection control method according to claim 2, characterized in that: The collecting and preprocessing of underwater environment parameters and detection requirement parameters specifically includes: collecting electromagnetic parameters of the underwater environment, detection requirement parameters and historical detection strategies, and preprocessing the collected parameters through normalization and standardization to make them conform to a unified input format.

4. The underwater magnetic induction non-contact current detection control method according to claim 2, characterized in that: The construction of the underwater electromagnetic model specifically includes: constructing o-xyz three-dimensional coordinates with the center of the detection area as the origin, inputting the pre-processed underwater environmental parameters into the three-dimensional modeling software, constructing an underwater electromagnetic three-dimensional virtual model within the three-dimensional coordinates, and marking the attribute parameters including magnetic field intensity and current density on the constructed three-dimensional virtual model.

5. The underwater magnetic induction non-contact current detection control method according to claim 2, characterized in that: The construction of intelligent detection strategy specifically includes: using BP neural network combined with genetic algorithm to construct detection strategy model, constructing model input according to the acquired underwater environment parameters, detection requirement parameters and historical detection strategy, constructing model output as detection strategy, wherein computer software is used to extract underwater environment parameter vectors marked on the three-dimensional virtual model ;in, represents the nth type of underwater environmental parameter, where n represents the number of categories of underwater environmental parameters; represents the detection requirement attribute parameter of the mth category, and m represents the number of categories of the detection requirement attribute parameter; The initial detection strategy and the historical detection strategy are used to construct the initial population of the genetic algorithm. The obtained initial population is used to construct the qth chromosome of the genetic algorithm input, and the following constraint condition set S is set for each variable in the chromosome: ;in, , , , They represent the number of detection areas, detection accuracy, detection angle, and speed of the detection equipment moving process in the qth chromosome respectively; , , , are the minimum values ​​of the corresponding parameters respectively; , , , are the maximum values ​​of the corresponding parameters respectively; The fitness function corresponding to the genetic algorithm is constructed using the initial population. The specific formula is: ;in, Indicates the detection time corresponding to the qth chromosome; is an indicator function, indicating that when the qth chromosome corresponds to the detection time If the preset maximum time is exceeded, the value is 1, otherwise it is 0; Indicates the penalty coefficient when the detection time of the qth chromosome exceeds the preset maximum time; Indicates the detection quality score corresponding to the qth chromosome, It represents the penalty term for violating the sth constraint in the constraint set S corresponding to the qth chromosome. If the sth constraint in the constraint set S is not satisfied, the term is 1, otherwise it is 0, which represent the weighted coefficients of the fitness function in turn; , , , All are weight factors; The iteration cycle and stop condition are set, the constructed qth chromosome, constraint set S, and fitness function are input into the genetic algorithm, the above process is iterated, and when the iteration cycle or stop condition is met, the iteration is stopped, and the obtained optimal detection strategy is output; the trained detection strategy model is integrated into the three-dimensional modeling software, the underwater electromagnetic three-dimensional virtual model is divided into virtual detection areas, and the virtual division process uses the PID algorithm to generate corresponding control signals, and the control signals are transmitted to the corresponding detection equipment using sensors, and the detection equipment is controlled to divide the underwater area into detection areas to ensure that the detection area division is consistent with the underwater magnetic induction current detection.

6. The underwater magnetic induction non-contact current detection control method according to claim 2, characterized in that: After the underwater electromagnetic model is adaptively divided into detection areas, each area after the detection area is divided is gridded in a fixed-size square grid form in the three-dimensional modeling software and the divided grids are numerically labeled. The magnetic field properties related to the underwater magnetic induction current in each grid after the labeling, including magnetic field intensity and directional parameter characteristics, are obtained, and the corresponding labels are input into the clustering algorithm for clustering operation, and the grid labels belonging to the same category after clustering are merged to obtain each corresponding detection area; The historical attribute parameters and detection requirement parameters of the same underwater environment detection area are collected, and the detection priority of each detection area is marked according to the collected parameters using expert experience to obtain priority label data. The priority is divided into Class A, Class B, and Class C, where Class C is a low priority, Class B is a medium priority, and Class A is a high priority. Support vector regression (SVR) is used to build a local priority model, and the obtained priority label data set is input into the local priority model for training to obtain the trained local priority model. The trained local priority model is integrated into the underwater electromagnetic three-dimensional virtual model to automatically mark the priority of the detection area.

7. The underwater magnetic induction non-contact current detection and control method according to claim 2, characterized in that: In the process of building and training the control model, control constraints are constructed, specifically including: building control constraints based on underwater environment parameters, detection requirement parameters and control parameters, including: ； in, It indicates the magnetic field induction intensity of the detection device at time t. It indicates the optimal threshold corresponding to the magnetic field induction strength of the detection device at time t. It represents the change of magnetic field in the detection area at time t. Indicates the target magnetic field change in the detection area at time t, represents the data processing sub-model executed by the detection device at time t, Represents the data processing sub-model executed by the detection device target at time t, Indicates the anti-interference capability of the detection device at time t, Indicates the target anti-interference capability of the detection equipment. Indicates the speed of the device moving at time t. Indicates the target speed at which the detection device moves; , , , , is the preset error threshold.

8. The underwater magnetic induction non-contact current detection control method according to claim 7, characterized in that: After constructing the control constraints, construct the control trigger conditions, including: when When , the first control condition is triggered; when When , the second control condition is triggered; when When , the third control condition is triggered; when When , the fourth control condition is triggered; when When , the fifth control condition is triggered; When two or more of the above control constraints are not met, the sixth control condition is triggered.

Images