A method and system for cooperative control of electrical devices

Abstract

The application discloses an electric equipment cooperative control method and system, and relates to the technical field of unmanned ship control; the technical points are as follows: through space-time-electromagnetic double-dimension synchronous preprocessing, taking the geomagnetic detection sampling time sequence as a core reference, the phase reference alignment of the power supply time sequence, the equipment operation time sequence and the detection sampling time sequence is carried out, so that the electric interference peak value is avoided from falling into the detection sampling window; in addition, on the basis, through matching the cooperative control decision value, three cooperative operations of power supply ripple reverse offset, equipment time sequence and detection window avoidance, and ship body posture and detection sampling phase lock are executed, active offset and accurate avoidance of electromagnetic interference are carried out from the power supply source, equipment operation and detection sampling, the problem that the prior art can only realize passive interference attenuation through physical isolation and hardware shielding, and cannot eliminate the interference source through electric cooperative control is solved, and thus the electromagnetic interference suppression efficiency of the geomagnetic detection channel and the signal-to-noise ratio of the geomagnetic detection data are improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned surface vessel control technology, and in particular to a method and system for coordinated control of electrical equipment. Background Technology

[0002] In marine engineering and national defense fields such as marine resource exploration, underwater target detection, and seabed geological mapping, marine geomagnetic measurement is a core geophysical exploration method. Unmanned surface vessels (USVs) have become the mainstream platform for marine geomagnetic exploration due to their advantages of no personnel operation risk, maneuverability, ability to operate continuously for long periods, and adaptability to complex waters such as shallow seas and nearshore areas. Due to the inherent design limitations of general-purpose USVs, the industry generally adopts an application model of modifying a general-purpose USV body with geomagnetic detection equipment. This model achieves geomagnetic detection functions by adding geomagnetic detection sensors, supporting electrical systems, and data processing units to existing USVs. This model does not require redesigning the hull platform, has low modification costs, and a short implementation cycle, and has become a common implementation method in the industry.

[0003] Currently, the general implementation path for the electrical system design and equipment control scheme of unmanned surface vessel (USV) geomagnetic detection retrofit is as follows: The USV's original main power battery or a reserved redundant battery is used as the power supply. This power is converted to the rated operating voltage of the geomagnetic detection equipment via a DC-DC step-down module. Then, independent power supply circuits are distributed to multiple devices such as the magnetometer main unit, data processing unit, and attitude sensors via a splitter. The magnetometer probe is mounted on the front of the hull using a non-magnetic support rod, isolating the hull from electromagnetic interference through physical distance. Simultaneously, the electrical circuits are equipped with magnetic rings and filtering circuits for hardware shielding. The equipment operates using a preset fixed logic. Upon entering the detection area, the detection equipment and its supporting electrical system are activated, maintaining fixed operating parameters and power supply mode throughout the process. After detection, the equipment is shut down, and finally, the collected data is corrected for errors using a post-detection magnetic compensation algorithm. However, in actual engineering implementation, the following technical defects exist:

[0004] Firstly, relying solely on passive methods such as physical distance isolation and hardware shielding filtering to reduce interference does not involve coordinated design for the power supply timing, equipment switching timing, and geomagnetic detection sampling timing of the electrical system. The peak values ​​of power supply ripple, switching noise, and electromagnetic radiation generated by the operation of electrical equipment inside the submarine are very likely to fall within the sampling window of geomagnetic detection and directly enter the detection channel through spatial coupling and line conduction. Especially in high-precision marine geomagnetic measurement scenarios, weak geomagnetic signals at the nT level are easily overwhelmed by electrical interference. Even with post-event algorithm compensation, it is impossible to recover the distorted original detection data, resulting in low signal-to-noise ratio and accuracy deviations exceeding requirements.

[0005] Secondly, the use of preset fixed control cycles and operating parameters cannot dynamically adjust according to real-time accuracy deviations in geomagnetic detection, changes in electromagnetic interference intensity, and fluctuations in the hull's navigation conditions. Under high-dynamic wind and wave conditions, data acquisition and control response are prone to lag, while under low-interference and stable conditions, it results in a waste of computing power and power supply resources. At the same time, the existing solution can only perform post-event data correction for interference that has already occurred, and cannot predict the timing characteristics and peak patterns of electrical interference in advance, let alone actively cancel interference or avoid interference peaks through collaborative control. It is a passive mode of post-event remediation. In complex marine environments and variable navigation scenarios, fixed control logic is very easy to become disconnected from actual detection needs, and the system has extremely poor robustness and environmental adaptability, which cannot meet the requirements of long-term, large-scale, and high-precision unmanned surface vessel geomagnetic detection operations. Therefore, we propose a collaborative control method and system for electrical equipment. Summary of the Invention

[0006] (a) Technical problems to be solved

[0007] To address the shortcomings of existing technologies, this invention provides a method and system for coordinated control of electrical equipment, solving the technical problems of disconnect between the electrical system and geomagnetic detection requirements, inability to actively prevent and control electromagnetic interference, and inability of control strategies to dynamically adapt to operating conditions in the scenario of unmanned surface vessel geomagnetic detection modification.

[0008] (II) Technical Solution

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A method for coordinated control of electrical equipment is applied to the onboard electrical cluster in a scenario of retrofitting an unmanned surface vessel for geomagnetic detection. The electrical cluster adopts an architecture with one input and multiple independent isolated power supplies adapted to onboard redundant batteries. The method includes the following steps:

[0011] S1. Acquire multiple sets of operational source data from the magnetic detection core equipment in various locations within the electrical cluster. The operational source data includes magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data.

[0012] S2. Perform spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on all operational source data to eliminate acquisition time deviation, align the phase reference of power supply timing, equipment operation timing and detection sampling timing, and generate a standardized operational dataset;

[0013] S3. Based on the standardized operation dataset, construct a multi-physics field coupling interference prediction model of electrical system-geomagnetic detection, extract multi-dimensional coupling interference features, quantify the interference contribution, and generate an electromagnetic interference prediction sequence with time-series phase labels;

[0014] S4. Based on the standardized operation dataset, electromagnetic interference prediction sequence and detection accuracy requirements, obtain differentiated collaborative control cycle information for each device and power supply circuit. The cycle information is dynamically linked with the detection sampling window, interference intensity, and hull operating conditions.

[0015] S5. Generate a multi-batch equipment operation sample set with multi-dimensional unified labels based on the differentiated collaborative control cycle information;

[0016] S6. By integrating and analyzing multiple batches of equipment operation sample sets, multi-dimensional evaluation values ​​of equipment operation status, power supply system stability, geomagnetic detection interference degree and navigation condition adaptability are obtained;

[0017] S7. By combining multi-dimensional evaluation values, preset detection accuracy thresholds, and electromagnetic interference prediction sequences, dynamically adaptive and updated collaborative control decision values ​​are generated;

[0018] S8. Based on the coordinated control decision value, match the coordinated control strategy and perform coordinated operations on the electrical cluster, including power supply ripple phase cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking.

[0019] S9. Continuously track the detection accuracy deviation, perform online iterative correction of the collaborative control decision value and control strategy, and simultaneously generate pre-adjustment plans for operating conditions.

[0020] Furthermore, step S1, which involves acquiring multiple sets of runtime source data, includes:

[0021] S101. Obtain data access nodes with hardware timestamps from various types of electrical clusters, and synchronously collect the original operating data of each device based on the unified hardware clock of the shipborne main control system.

[0022] S102. Perform noise reduction, outlier removal and phase calibration on the raw operating data, and determine the data characteristics as magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data and environmental condition data respectively.

[0023] Furthermore, step S3, which involves constructing a multi-physics coupling interference prediction model, includes:

[0024] S301. Based on the standardized operation dataset, extract multi-dimensional features of the electric field, magnetic field, hull mechanical disturbance, and environmental flow field disturbance of the electrical system to construct a multi-physics field feature dataset;

[0025] S302. By training with coupled simulation and historical measured data, a nonlinear mapping relationship between multi-physics field characteristics and geomagnetic detection noise is established to generate an initial prediction model.

[0026] S303. Input the real-time standardized running dataset into the initial prediction model and output the interference contribution of a single device and a single loop, the cluster coupling interference intensity, and the electromagnetic interference prediction sequence with time-series phase labels.

[0027] S304. Based on the real-time detection accuracy deviation, the model is iteratively corrected online to generate the final coupling interference prediction model.

[0028] Furthermore, step S4, which involves obtaining differentiated collaborative control cycle information, includes:

[0029] S401. Determine the timing and phase reference of the detection sampling window based on the magnetometer detection data, and set the initial first control cycle;

[0030] S402. Based on the timing of the interference peak of the electromagnetic interference prediction sequence and the sampling window avoidance requirements, the first control cycle is modified.

[0031] S403. The second, third and fourth control cycles are set according to the hull attitude disturbance amplitude, the power supply system voltage regulation requirements and the equipment switching interference sensitivity, respectively.

[0032] S404. Perform time-phase alignment and integration on the four types of control cycles to generate differentiated collaborative control cycle information that is dynamically bound to the detection sampling window and the interference prediction time.

[0033] Furthermore, step S6, which involves obtaining the multi-dimensional evaluation value, includes:

[0034] S601. Based on multiple batches of equipment operation sample sets, calculate the evaluation value of data time sequence continuity and phase matching degree, and combine the evaluation results of multi-dimensional content information to generate the equipment operation status evaluation value.

[0035] S602. Extract the voltage fluctuation, current stability, ripple phase, and load balance parameters of the power supply system, and calculate the stability evaluation value of the power supply system using the analytic hierarchy process.

[0036] S603. Combining the interference prediction results with the real-time detection accuracy deviation, calculate the interference ratio of a single device and a single loop, and generate a geomagnetic detection interference assessment value.

[0037] S604. Combining environmental conditions, hull attitude, and detection mission requirements, the navigation condition adaptability assessment value is calculated.

[0038] Furthermore, step S7, which involves generating collaborative control decision values, includes:

[0039] S701. Based on the core requirements of the current exploration mission, determine the dynamic weight ratio of various operational source data;

[0040] S702. Combine multi-dimensional evaluation values, dynamic weighting of data, effective data ratio and phase matching degree to calculate single-dimensional operational quality evaluation value;

[0041] S703, combining the power supply system baseline parameters, single-dimensional operation quality assessment value, multi-dimensional comprehensive assessment results, and the difference between real-time detection accuracy deviation and preset threshold, calculates the collaborative control decision value of the electrical cluster.

[0042] An electrical equipment collaborative control system is applied to the onboard electrical cluster in the scenario of unmanned surface vessel (USV) geomagnetic detection modification. The electrical cluster adopts a single-input, multi-channel independent power supply architecture adapted to onboard redundant batteries. The system includes:

[0043] The first acquisition module is used to acquire multiple sets of operational source data from the magnetic detection core equipment in various locations within the electrical cluster. The operational source data includes magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data.

[0044] The dual-synchronization preprocessing module is used to perform spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on all operational source data, eliminate acquisition time deviation, align the phase reference of power supply timing, equipment operation timing and detection sampling timing, and generate a standardized operational dataset.

[0045] The coupling interference prediction module is used to build a multi-physics coupling interference prediction model of electrical system-geomagnetic detection based on standardized operating datasets, extract multi-dimensional coupling interference features, quantify interference contribution, and generate electromagnetic interference prediction sequences with time-series phase labels.

[0046] The second acquisition module is used to acquire differentiated collaborative control cycle information of each device and power supply circuit based on the standardized operation dataset, electromagnetic interference prediction sequence and detection accuracy requirements. The cycle information is dynamically linked with the detection sampling window, interference intensity and hull operating conditions.

[0047] The third acquisition module is used to generate a multi-batch equipment operation sample set with multi-dimensional unified labels based on the differentiated collaborative control cycle information;

[0048] The fourth acquisition module is used to integrate and analyze multiple batches of equipment operation sample sets to obtain multi-dimensional evaluation values ​​of equipment operation status, power supply system stability, geomagnetic detection interference degree and navigation condition adaptability.

[0049] The decision generation module is used to combine multi-dimensional evaluation values, preset detection accuracy thresholds and electromagnetic interference prediction sequences to generate dynamically adaptive and updated collaborative control decision values.

[0050] The collaborative control execution module is used to match the collaborative control strategy according to the collaborative control decision value, and to perform collaborative operations on the electrical cluster, such as power supply ripple phase reversal cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking.

[0051] The online iteration module is used to continuously track the detection accuracy deviation, perform online iterative correction of the collaborative control decision value and control strategy, and simultaneously generate pre-adjustment plans for operating conditions.

[0052] Furthermore, the first acquisition module specifically includes:

[0053] The access node acquisition unit is used to acquire various data access nodes with hardware timestamps within the electrical cluster.

[0054] The synchronous acquisition unit is used to synchronously acquire the raw operating data of each device based on the unified hardware clock of the shipborne main control system;

[0055] The preprocessing unit is used to perform noise reduction, outlier removal, and phase calibration on the raw running data;

[0056] The classification and determination unit is used to classify the raw operating data into magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data based on data characteristics.

[0057] Furthermore, the second acquisition module specifically includes:

[0058] The initial cycle setting unit is used to determine the timing and phase reference of the detection sampling window based on the magnetometer detection data, and to set the initial first control cycle.

[0059] The period correction unit is used to correct the first control period by combining the timing of the interference peak of the electromagnetic interference prediction sequence with the sampling window avoidance requirements.

[0060] The cycle setting unit is used to set the second, third and fourth control cycles respectively according to the hull attitude disturbance amplitude, the power supply system voltage regulation requirements and the equipment switching interference sensitivity.

[0061] The cycle integration unit is used to perform time-phase alignment and integration of four types of control cycles to generate differentiated collaborative control cycle information that is dynamically bound to the detection sampling window and the interference prediction timing.

[0062] Furthermore, the third acquisition module specifically includes:

[0063] The data acquisition task generation unit is used to generate data acquisition tasks with a unified spatiotemporal and phase reference based on differentiated collaborative control cycle information.

[0064] The parameter configuration unit is used to determine the acquisition configuration parameters corresponding to the acquisition task.

[0065] The data source location unit is used to locate the original data traceability source corresponding to various types of operational source data based on the collection configuration parameters;

[0066] The acquisition sequence determination unit is used to determine the acquisition sequence corresponding to each data source;

[0067] The data acquisition unit is used to synchronously collect data from various data sources according to the acquisition sequence and generate collected information with multi-dimensional unified labels.

[0068] The sample generation unit is used to align and integrate various types of collected information according to spatiotemporal and phase references to form multiple batches of equipment operation samples.

[0069] (III) Beneficial Effects

[0070] 1. By employing spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing, and using the geomagnetic detection sampling time sequence as the core benchmark, phase benchmark alignment is performed for the power supply time sequence, equipment operation time sequence, and detection sampling time sequence. This avoids electrical interference peaks falling into the detection sampling window from the root cause of timing issues. Furthermore, based on this, by matching and coordinating control decision values ​​to execute three major coordinated operations—power supply ripple phase cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking—active cancellation and precise avoidance of electromagnetic interference are achieved from the three core links of power supply head, equipment operation, and detection sampling. This solves the technical problem that existing technologies can only achieve passive interference attenuation through physical isolation and hardware shielding, and cannot eliminate the root cause of interference through electrical coordinated control. This improves the electromagnetic interference suppression efficiency of the geomagnetic detection channel, and also enhances the signal-to-noise ratio of the geomagnetic detection data.

[0071] 2. Based on the standardized operational dataset after spatiotemporal synchronization, multi-dimensional features of the electric field, magnetic field, hull mechanical disturbance, and environmental flow field disturbance of the electrical system are extracted. A multi-physics field coupling interference prediction model for the electrical system and geomagnetic detection is constructed through coupled simulation and training with historical measured data. This model can quantify the interference contribution of a single device or a single power supply circuit to geomagnetic detection and output electromagnetic interference prediction sequences with high-precision time-phase labels in advance. Simultaneously, the model is iteratively corrected online based on the real-time accuracy deviation of geomagnetic detection to ensure prediction accuracy. This solves the technical problem of existing technologies that can only correct errors through post-event compensation algorithms after geomagnetic data is contaminated by interference, and cannot avoid distortion of the original data. It achieves early prediction and proactive prevention of interference, fundamentally avoiding the contamination of the original weak geomagnetic signal by electrical interference, increasing the proportion of effective geomagnetic detection data under complex operating conditions, significantly reducing the workload of subsequent data processing, and improving the reliability of geomagnetic detection results.

[0072] 3. Based on the geomagnetic detection sampling window, an initial control cycle is set. Combining the interference peak timing of the electromagnetic interference prediction sequence, the amplitude of the hull attitude disturbance, the voltage stabilization requirements of the power supply system, and the sensitivity of equipment switching interference, four types of differentiated control cycles are modified and set respectively. Finally, a collaborative control cycle is generated that is dynamically bound to the detection sampling window, the interference prediction timing, and the hull operating conditions. This abandons the inherent mode of fixed-cycle control in existing technologies. Under high interference and high dynamic conditions, the control cycle is automatically shortened, increasing the system response speed by more than 2 times and ensuring the real-time performance of interference prevention and control. Under low interference and stable navigation conditions, the control cycle is automatically extended, reducing the system's computing power and power supply resource consumption by more than 60%. This solves the dual drawbacks of existing fixed-cycle control, namely data lag in high-demand scenarios and resource waste in low-demand scenarios. While ensuring control accuracy, it significantly extends the continuous operation time of the unmanned surface vessel.

[0073] 4. Through the integrated analysis of multiple batches of operational sample sets, a four-dimensional comprehensive evaluation system was constructed, encompassing equipment operating status, power supply system stability, geomagnetic detection interference, and navigation condition adaptability. This system provides a comprehensive quantitative evaluation of the system's operating status and operational environment. Furthermore, by combining real-time detection accuracy deviation, electromagnetic interference prediction results, and changes in navigation conditions, the system dynamically updates the hierarchical matching benchmark and control strategy for collaborative control decision values. It also continuously iterates and corrects the strategy online, simultaneously generating pre-adjustment plans for future operating conditions. This achieves real-time matching between the control strategy and actual operational needs. Even in complex marine environments such as sea state 4, strong background magnetic field interference, and navigation under varying operating conditions, the system maintains stable control performance and detection accuracy. It can complete the entire process of adaptive adjustment without manual intervention, significantly improving the system's robustness and adaptability to complex environments.

[0074] 5. It can interface with the natively reserved redundant batteries, multiple independent power supply circuits, and various geomagnetic detection and sensing equipment of unmanned surface vessels (USVs). It does not require large-scale modifications to the original structure and electrical system of the USV, strictly adhering to the core principle of minimizing structural damage and disturbance to the original system when modifying USVs. This significantly reduces the difficulty and cost of implementing the solution. It can be directly applied to geomagnetic detection modification projects of various large, medium, and small USVs without redesigning the electrical architecture of the USV. At the same time, it has reserved sufficient functional expansion interfaces and can be compatible with the collaborative control of other marine detection equipment. It has extremely high engineering practical value and industry promotion prospects. Attached Figure Description

[0075] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0076] Figure 1 This is an overall flowchart of an embodiment of the present invention;

[0077] Figure 2 This is a block diagram of the power supply connection of the electrical system in an embodiment of the present invention;

[0078] Figure 3 This is a power supply circuit diagram of the Huawi No. 6 unmanned surface vessel in an embodiment of the present invention. Detailed Implementation

[0079] This application provides an electrical equipment collaborative control method and system to solve the technical problems of disconnect between the electrical system and geomagnetic detection requirements, inability to actively prevent and control electromagnetic interference, and inability of control strategies to dynamically adapt to working conditions in the scenario of unmanned surface vessel geomagnetic detection modification.

[0080] The general-purpose application carrier is the Huawi 6 unmanned surface vessel (USV) geomagnetic detection modification platform. The onboard electrical cluster uses the USV's original reserved third 32V / 100Ah redundant lithium battery as the power supply core. The 32V DC input is converted to a stable 28V DC output through a DC-DC step-down module. The power supply circuit is divided into four electrically isolated independent power supply circuits through a one-to-four quick-release splitter, which power the helium optical pump magnetometer host, the USV data acquisition and magnetic compensation instrument, the GPS receiving system, and the display and control system.

[0081] Example 1: Electrical Equipment Cooperative Control Method

[0082] Primarily designed for high-precision marine geomagnetic measurement operations in nearshore shallow waters, this system is adapted to the entire operational process of unmanned surface vessels (USVs) modified for geomagnetic detection. It achieves full-cycle coverage from pre-operation data acquisition to in-operation cyclical control and online optimization. The specific process is as follows:

[0083] The first step is to acquire operational source data, which involves acquiring multiple sets of operational source data from the core magnetic detection equipment in various locations within the electrical cluster. Specifically, this includes six types of core data: magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data. This completes the collection and classification of basic data sources.

[0084] The second step is dual-dimensional synchronous preprocessing, which performs spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on all running source data to eliminate the acquisition time deviation of different data sources, align the phase reference of power supply timing, equipment running timing and detection sampling timing, and generate a standardized running dataset that is unified in spatiotemporal and electromagnetic phase dimensions.

[0085] The third step is to build an interference prediction model. Based on the standardized operation dataset, an electrical system-geomagnetic detection multi-physics field coupling interference prediction model is built. Multi-dimensional coupling interference features are extracted, the interference contribution of a single device and power supply circuit is quantified, and an electromagnetic interference prediction sequence with time-series phase labels is generated.

[0086] The fourth step is to obtain the differentiated control cycle. Based on the standardized operation dataset, electromagnetic interference prediction sequence and current detection accuracy requirements, the differentiated collaborative control cycle information of each device and power supply circuit is obtained. This cycle information is dynamically linked with the detection sampling window, interference intensity and hull operating conditions.

[0087] The fifth step is to generate a running sample set. Based on the differentiated collaborative control cycle information, a multi-batch equipment running sample set with unified labels of multiple dimensions is generated.

[0088] The sixth step is a multi-dimensional status assessment, which integrates and analyzes multiple batches of equipment operation sample sets to obtain multi-dimensional assessment values ​​of equipment operation status, power supply system stability, geomagnetic detection interference degree and navigation condition adaptability.

[0089] The seventh step is to generate collaborative control decision values. By combining multi-dimensional evaluation values, preset detection accuracy thresholds and electromagnetic interference prediction sequences, dynamic adaptive updated collaborative control decision values ​​are generated. The hierarchical matching benchmark of the collaborative control decision values ​​is based on four dimensions: accuracy requirements of marine magnetic measurement specifications, real-time navigation conditions, electromagnetic interference prediction results, and power supply circuit load status, which are dynamically and adaptively updated.

[0090] The eighth step is to perform coordinated operations. Based on the coordinated control decision value, the coordinated control strategy is matched to perform full coordinated operations on the electrical cluster, including power supply ripple phase reversal cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking.

[0091] The ninth step is online iterative optimization, which continuously tracks the detection accuracy deviation, iteratively corrects the collaborative control decision value and control strategy online, and generates pre-adjustment plans for operating conditions simultaneously. This constructs a collaborative control system from data acquisition, preprocessing, interference prediction, cycle adaptation, state assessment, decision generation, control execution to iterative optimization, breaking through the core defects of existing open-loop control and passive interference prevention and control technologies, and realizing deep collaboration between the electrical system and geomagnetic detection requirements.

[0092] Example 2: Acquisition of multi-source runtime data and dual-dimensional synchronous preprocessing:

[0093] This embodiment focuses on multi-source data acquisition, preprocessing, classification and judgment, and spatiotemporal-electromagnetic dual-dimensional synchronization. It is applicable to the synchronous acquisition and standardized processing of multi-source heterogeneous data in unmanned surface vessel geomagnetic exploration operations, and solves the technical problems of disordered timing of multi-source data, large time synchronization error and poor data quality in existing technologies. It is mainly divided into two execution stages.

[0094] The first step involves acquiring and classifying multiple sets of source data:

[0095] First, acquire data access nodes with nanosecond-level hardware timestamps within the electrical cluster, including the synchronous acquisition interface of the helium optical pump magnetometer, the RS485 synchronous communication interface of the electrical equipment controller, the status monitoring interface of the DC-DC step-down module, the load monitoring node of the 1-to-4 fast splitter, the fiber optic interface of the three-component attitude meter, the synchronous serial port of the GPS receiver system, and the interface of the multi-parameter sensor of the aquatic environment.

[0096] Then, using the unified hardware clock of the thermostatic crystal oscillator of the unmanned surface vessel main control system as a reference, the original operating data of each device is collected synchronously, and the time synchronization reference error is controlled within 1μs. Subsequently, adaptive wavelet denoising is performed on the collected original operating data to eliminate random noise, and jump outliers are eliminated by the 3σ criterion. At the same time, initial phase calibration is completed to improve the effectiveness and consistency of the original data.

[0097] Finally, based on the inherent characteristics of the data, the preprocessed raw data is classified. First, magnetometer detection data is obtained by extracting geomagnetic detection gradient parameters and noise characteristics from the raw data. This includes parameters such as total geomagnetic field strength, geomagnetic gradient value, sampling sequence, and data signal-to-noise ratio, serving as the core basis for judging detection accuracy and interference levels. Second, equipment operating parameters are obtained by extracting equipment operating parameters, switching sequence, and electromagnetic radiation spectrum characteristics from the raw data. This includes parameters such as the start / stop status, operating current, switching frequency, and operating mode of each electrical device, serving as the core execution object for coordinated control. Third, power supply system status data is obtained by extracting input / output voltage of the step-down module, power ripple, line voltage drop, overcurrent protection status, and power supply sequence characteristics from the raw data. This includes 32V input voltage, 28V output voltage, current of each power supply circuit, peak-to-peak ripple, and stable voltage. Parameters such as pressure state are the core control targets for active interference cancellation; fourth, hull attitude data is obtained by extracting the hull roll, pitch, bow angle and three-axis acceleration change characteristics from the original data, including parameters such as hull three-axis attitude angle, angular velocity, acceleration, and attitude change rate, providing data support for detection sampling phase-locking and attitude compensation; fifth, positioning data is obtained by extracting positioning coordinates, timestamps and positioning accuracy characteristics from the original data, including parameters such as hull latitude and longitude coordinates, UTC timestamp, positioning accuracy factor, speed and heading, providing spatial location marking for geomagnetic detection data; sixth, environmental condition data is obtained by extracting the water background magnetic field, seawater salinity, water flow velocity, wind and wave level and hull vibration characteristics from the original data, including parameters such as water background magnetic field strength, salinity, current velocity, wave height, and hull vibration frequency amplitude, providing environmental basis for condition adaptability assessment.

[0098] The second step is spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing:

[0099] First, perform spatiotemporal synchronization. Using the unified hardware clock of the main control system's thermostatic crystal oscillator as the absolute reference, linearly align the hardware timestamps of the six types of running source data to eliminate the acquisition time deviation. Perform cubic spline interpolation resampling on data with different sampling frequencies to unify them to a time resolution of 100Hz and generate a dataset with a unified time dimension.

[0100] Then perform electromagnetic dimension synchronization, using the detection and sampling timing of the helium optical pump magnetometer as the absolute phase reference, align the phase of the power supply switch timing of the four independent power supply circuits, the operation timing of each electrical device, and the attitude data acquisition timing, and establish a fixed 90° phase offset relationship to avoid the interference peak of electrical actions falling into the detection and sampling window.

[0101] Ultimately, a standardized operational dataset with unified spatiotemporal and electromagnetic phase dimensions is generated, providing a unified and accurate data foundation for subsequent stages. This enables nanosecond-level synchronous acquisition and standardized processing of multi-source data, controlling the time synchronization error of multi-source data to within 1μs. It fundamentally avoids electrical interference peaks falling into the detection sampling window, solving the core defect of existing technologies that can only be aligned through post-event software timestamps.

[0102] Example 3, Prediction model of multi-physics coupling interference between electrical system and geomagnetic detection:

[0103] This embodiment clarifies the entire implementation process of constructing, training, predicting, and iterating a multiphysics coupling interference prediction model. It addresses the long-standing industry pain point of existing technologies, which can only provide post-hoc compensation after geomagnetic data contamination and cannot prevent distortion of the original data, specifically for the pre-emptive prediction and prevention of electromagnetic interference in unmanned surface vessel (USV) geomagnetic detection scenarios. The specific implementation process is as follows:

[0104] The first step is to construct a multiphysics feature dataset. Based on the standardized running dataset, four major categories of coupling interference features are extracted to construct the multiphysics feature dataset, which specifically includes electric field features, magnetic field features, mechanical disturbance features, and flow field disturbance features. Among them, electric field features cover the ripple spectrum of the power supply system, voltage fluctuation gradient, and electric field radiation intensity of the line; magnetic field features cover the electromagnetic radiation spectrum of electrical equipment, the induced magnetic field intensity generated by current changes, and the magnetic field radiation features of the power supply circuit; mechanical disturbance features cover the influence of hull attitude changes and mechanical disturbances generated by hull vibration on the magnetometer probe; and flow field disturbance features cover the hull attitude disturbances and background magnetic field changes generated by water flow velocity and wave level changes.

[0105] The second step is to train the initial prediction model. Through COMSOL multiphysics coupling simulation, combined with 120 sets of valid operational data from the historical modification and actual measurement of this unmanned surface vessel, a 3-layer BP neural network model is constructed to establish a nonlinear mapping relationship between multiphysics features and geomagnetic detection noise, thereby generating an initial coupling interference prediction model. The model input is the multiphysics feature dataset, and the output is the amplitude, occurrence time, and duration of geomagnetic detection noise. The initial prediction accuracy of the model is not less than 85%.

[0106] The third step is to perform interference prediction and contribution quantification. The standardized operational dataset collected in real time is input into the initial prediction model, and three core results are output: the interference contribution of a single device and a single power supply circuit to the geomagnetic detection channel, the coupling interference intensity of the electrical cluster as a whole, and the electromagnetic interference prediction sequence with 1ms precision time and phase label within the next 5 seconds. The occurrence time, amplitude and duration of the interference peak are clearly defined, providing a preliminary decision basis for subsequent coordinated control.

[0107] The fourth step involves online iterative correction of the model. Based on the real-time detection accuracy deviation of the helium-pumped magnetometer, incremental learning is used to iteratively correct the model online. During the iteration process, effective data within the last 10 seconds are selected as incremental samples. The iteration stops when the model prediction error is less than 5% for three consecutive times. The model weights are updated every 10 seconds to ensure that the model prediction accuracy remains above 90%. The final electrical system-geomagnetic detection multi-physics field coupling interference prediction model is generated, realizing the early prediction and accurate quantification of electrical interference. It also represents a technological leap from post-event remediation to pre-event prevention, avoiding the contamination of the original weak geomagnetic signal by electrical interference.

[0108] Example 4: Differentiated collaborative control cycle and generation of multi-batch running sample sets:

[0109] Regarding the dynamic adaptation of control cycle and operating conditions, and the construction of standardized sample sets during unmanned surface vessel (USV) geomagnetic exploration, this approach addresses the dual drawbacks of existing technologies: data lag in high-demand scenarios and resource waste in low-demand scenarios. It is divided into two stages, as detailed below:

[0110] The first step is to obtain differentiated collaborative control cycle information:

[0111] First, an initial first control cycle is set. Based on the helium-pumped magnetometer detection data, the accuracy requirements of the current detection mission, the timing of the geomagnetic sampling window, and the phase reference are determined. The initial first control cycle is set to 10ms, which is completely synchronized with the magnetometer detection sampling window. Then, the first control cycle is dynamically corrected. Combining the interference peak timing of the electromagnetic interference prediction sequence and the sampling window avoidance requirements, the initial first control cycle is corrected. If an interference peak is predicted to occur within the next 1 second, the first control cycle is shortened to 5ms to improve the control response speed. If there is no interference peak prediction and the detection accuracy is stable, the first control cycle is extended to 20ms to reduce resource consumption. Subsequently, the other three types of control cycles are set independently. The second control cycle is set based on the hull attitude disturbance amplitude and the timing requirements of geomagnetic detection attitude compensation. The control cycle is 10ms when the roll and pitch angles are greater than 5° and 50ms when the roll and pitch angles are less than 5°. The third control cycle is set according to the voltage regulation requirements of the power supply system and the phase matching requirements for ripple interference cancellation. The control cycle is 5ms when the output voltage fluctuation is greater than 0.5V and 20ms when the voltage fluctuation is less than 0.5V. The fourth control cycle is set according to the sensitivity of the equipment to switching interference. The control cycle is 2ms when the equipment starts up, stops, or switches modes, and 100ms when the equipment is running in steady state. Finally, the cycle is integrated and aligned. The four types of control cycles are integrated and aligned in time and phase to ensure that the phase reference of all control cycles is synchronized with the geomagnetic detection sampling window. Finally, differentiated collaborative control cycle information that is dynamically bound to the detection sampling window and interference prediction timing is generated.

[0112] The second step is to generate a sample set of multiple batches of equipment operation data:

[0113] First, the data acquisition task and configuration parameters are generated. Based on the differentiated collaborative control cycle information, a data acquisition task with a unified spatiotemporal and phase reference is generated. The acquisition time window, sampling frequency, data bit width, phase calibration rules, power supply circuit identification rules, and electromagnetic interference feature marking rules are configured. Then, the data source is located and the acquisition sequence is determined. The original data traceability source corresponding to the six types of operational data sources is located, and the acquisition sequence corresponding to each data source is determined, including the acquisition execution order, acquisition execution count, acquisition execution interval, timestamp calibration rules, and phase matching rules. Subsequently, synchronous acquisition and tagging are performed. Synchronous acquisition is performed on each data source according to the acquisition sequence, and each piece of acquired information is tagged with the corresponding spatiotemporal reference tag and electromagnetic phase tag. The system uses labels for power supply circuit identification and operating condition matching. The label coding adopts a hexadecimal combination coding rule of "timestamp-phase value-circuit number-operating condition level" to ensure that all collected information is traceable and aligned. Finally, the sample set is integrated and generated. The collected information corresponding to the six types of data is aligned and integrated in two dimensions according to the spatiotemporal reference label and the electromagnetic phase label to form a multi-batch equipment operation sample set with continuous time, unified phase and complete labels. This realizes full-dimensional dynamic adaptation of control cycle and detection requirements and operating condition changes, improves the system response speed in high interference scenarios, reduces resource consumption in stable scenarios, and generates a standardized and traceable operation sample set, providing a high-quality data foundation for subsequent status assessment.

[0114] Example 5: Multi-dimensional state assessment and collaborative control decision value generation:

[0115] This embodiment clarifies the construction method of the four-dimensional comprehensive evaluation system and the dynamic generation logic of collaborative control decision values. It addresses the core aspects of quantitative evaluation of system state and generation of control decisions during unmanned surface vessel (USV) geomagnetic exploration, overcoming the shortcomings of existing technologies that can only evaluate using a single indicator and cannot comprehensively reflect the system state. The main components are as follows:

[0116] The first step is the calculation of multi-dimensional evaluation values:

[0117] First, calculate the equipment operating status assessment value. Based on multiple batches of equipment operating sample sets, calculate the data time series continuity and phase matching degree assessment values. Simultaneously, extract six core indicators: data completeness, accuracy, timeliness, clarity, relevance, and continuity. Use the Analytic Hierarchy Process (AHP) to determine the weight proportions of these six indicators. When implementing AHP, first construct a judgment matrix, assign values ​​using a 1-9 scale, and determine the weights after a consistency test (CR < 0.1). Calculate the content information assessment value, and then... The formula R = α × A + β × B is used to calculate the equipment operating status evaluation value, where R is the equipment operating status evaluation value (interval [0,1]), A is the data timing continuity and phase matching evaluation value (interval [0,1]), B is the content information evaluation value (interval [0,1]), α = 0.4, β = 0.6 are weighting coefficients, and α + β = 1. Example: If A = 0.92 and B = 0.88, then R = 0.4 × 0.92 + 0.6 × 0.88 = 0.896. Then, the power supply... The system stability assessment value is obtained by extracting five core parameters of the power supply system: voltage fluctuation, current stability, ripple phase, load balance, and protection device triggering status. The weight ratio of each parameter is determined by the analytic hierarchy process (AHP), and the weighted calculation yields the power supply system stability assessment value in the 0-1 interval. The closer the value is to 1, the more stable the power supply system. Subsequently, the geomagnetic detection interference assessment value is calculated. Combining the interference contribution output by the multi-physics coupling interference prediction model, the deviation between the electromagnetic interference prediction sequence and the real-time detection accuracy, the interference ratio of a single device and a single loop on the detection channel is calculated. Combined with the overall electromagnetic environment characteristics of the cluster, the geomagnetic detection interference assessment value in the 0-1 interval is generated. The closer the value is to 1, the more severe the interference. Finally, the navigation condition adaptability assessment value is calculated. Combining the wind and wave levels, water flow speed, and hull attitude change amplitude of the environmental conditions with the current detection mission's condition adaptability requirements, the navigation condition adaptability assessment value in the 0-1 interval is calculated. The closer the value is to 1, the higher the adaptability of the control strategy to the operating conditions.

[0118] The second step is the generation of collaborative control decision values:

[0119] First, determine the dynamic weighting percentages. Based on the core accuracy requirements of the current detection mission, determine the dynamic weighting percentages for the six types of operational data sources. In high-precision detection mode, the magnetometer detection data accounts for 35% of the weight, the power supply system status data accounts for 30%, and the remaining four types of data account for a combined 35%. In conventional detection mode, the magnetometer detection data accounts for 25%, the power supply system status data accounts for 25%, and the remaining four types of data account for a combined 50%. Then, calculate the single-dimensional operational quality assessment value. Combining the multi-dimensional comprehensive assessment value, the dynamic weighting percentages of the data, the effective data percentage, and the phase matching degree, calculate the single-dimensional operational quality assessment values ​​for each of the six types of data. Finally, use the formula D=ω1×R. s +ω2×R p +ω3×Rd +ω4×R w +γ×ΔE, complete the fusion calculation of collaborative control decision values, where D is the collaborative control decision value, in the interval [0,1], R s R is the equipment operating status assessment value. p R is the stability assessment value for the power supply system. d R is the evaluation value for geomagnetic detection interference. w The values ​​for the navigation condition adaptability assessment are all in the range [0,1]. ω1, ω2, ω3, and ω4 are dynamic weighting coefficients. ∑ω i =1, in high-precision detection mode, ω1=0.15, ω2=0.3, ω3=0.35, ω4=0.2; in normal mode, ω1=0.25, ω2=0.25, ω3=0.25, ω4=0.25; ΔE is the normalized difference between the real-time detection accuracy deviation and the preset threshold, in the interval [0,1]; γ=0.3 is the deviation correction coefficient. Numerical example: in high-precision detection mode, if R... s =0.896, R p =0.92, R d =0.25, R w =0.85, ΔE=0.15, then D=0.15×0.896+0.3×0.92+0.35×0.25+0.2×0.85+0.3×0.15=0.7129. Simultaneously, the hierarchical interval rules for collaborative control decision values ​​are clearly defined: D≥0.7 corresponds to a high-precision interference cancellation strategy, 0.3<D<0.7 corresponds to a conventional voltage stabilization strategy, and D≤0.3 corresponds to maintaining the original state. This hierarchical matching benchmark for decision values ​​is dynamically and adaptively updated based on four dimensions: the accuracy requirements of marine magnetic measurement specifications, real-time navigation conditions, electromagnetic interference prediction results, and the load status of the power supply circuit. A comprehensive and accurate four-dimensional integrated evaluation system is constructed, achieving full-dimensional quantitative perception of the system's operating status, interference level, and condition adaptability. Simultaneously, collaborative control decision values ​​dynamically match operational requirements and changes in operating conditions, providing precise core instructions for subsequent collaborative control execution.

[0120] Example 6: Cooperative control execution and online iterative optimization:

[0121] This embodiment clarifies the three major collaborative control operations and the entire process of online iterative optimization and pre-adjustment plan generation. Specifically, it addresses the execution phases of active electromagnetic interference cancellation and adaptive control strategy optimization during unmanned surface vessel (USV) geomagnetic detection. Breaking away from the industry's common technical bias that existing technologies can only achieve passive interference attenuation through physical isolation, this embodiment is mainly divided into the following two execution phases:

[0122] The first step is the execution of collaborative control operations:

[0123] Based on the control strategy corresponding to the range of the collaborative control decision value, three core collaborative operations are executed. The first is active interference cancellation, where the power supply ripple phase and the detection sampling timing are inversely matched. When the collaborative control decision value is ≥0.7, a high-precision ripple cancellation strategy is activated. Based on the ripple phase and amplitude of the electromagnetic interference prediction sequence, the DC-DC step-down module outputs a compensation voltage that is inversely phase and equal in amplitude to the ripple noise, keeping the peak-to-peak value of the output ripple below 10mV. When the collaborative control decision value is between 0.3 and 0.7, a conventional voltage regulation strategy is activated to maintain the output voltage stable within the range of 28V±0.2V. When the decision value is ≤0.3, the original power supply state is maintained. The second is the active interference cancellation, where the power supply ripple phase and the detection sampling timing are inversely matched. When the collaborative control decision value is ≥0.7, a high-precision ripple cancellation strategy is activated. Based on the ripple phase and amplitude of the electromagnetic interference prediction sequence, the DC-DC step-down module outputs a compensation voltage that is inversely phase and equal in amplitude to the ripple noise, keeping the peak-to-peak value of the output ripple below 10mV. When the collaborative control decision value is between 0.3 and 0.7, a conventional voltage regulation strategy is activated to maintain the output voltage stable within the range of 28V±0.2V. When the decision value is ≤0.3, the original power supply state is maintained. The first step is to achieve spatiotemporal avoidance of the detection window. Based on the interference peak timing of the electromagnetic interference prediction sequence, the switching actions and mode switching timing of unnecessary electrical equipment are controlled to completely avoid the core detection sampling window of the helium optical pump magnetometer. For equipment that must run continuously, its operating frequency is adjusted so that its electromagnetic radiation spectrum avoids the detection passband of the magnetometer, thereby reducing interference at the source. The second step is to coordinate phase-locking between the hull attitude and the detection sampling. Based on the change pattern of the hull attitude data, the time window of the hull's stable attitude is predicted. The core sampling timing of the magnetometer is locked and matched with the stable attitude window. At the same time, the pitch angle of the magnetometer probe is adjusted to ensure that the detection surface is always perpendicular to the water surface, eliminating the impact of hull attitude disturbances on the detection accuracy.

[0124] The second step is online iterative correction and pre-adjustment plan generation:

[0125] The system continuously tracks the real-time detection accuracy deviation of the helium optical pump magnetometer. When the deviation exceeds 20% of the preset threshold, it iteratively corrects the weight allocation of the collaborative control decision value and the parameters of the cyclic control strategy online to optimize the control effect. At the same time, based on the electromagnetic interference prediction sequence and the trend of navigation conditions, it generates a pre-adjustment plan for the next 30 seconds. When the conditions change abruptly, the plan can be executed directly, improving the system's response speed and anti-interference capability. This breaks through the industry bias of passive interference prevention and control in existing technologies, improves the electromagnetic interference suppression efficiency of the geomagnetic detection channel, and increases the signal-to-noise ratio of geomagnetic detection data by more than one order of magnitude. It also achieves adaptive optimization of the control strategy. Even in the complex environment of sea state 4, the system can still maintain stable control effect and detection accuracy.

[0126] Example 7, Electrical Equipment Collaborative Control System:

[0127] This system is used for the deployment and implementation of an embedded control system for an unmanned surface vessel (USV) geomagnetic detection modification platform. It is deployed on the IMX8 core board of the USV's main control system, running an embedded Linux operating system. The system adopts a modular design, with each module connected via an internal high-speed CAN bus. Data interaction uses the Modbus-RTU protocol, as detailed below:

[0128] The first module is the first acquisition module, which acquires six types of operational source data from the core magnetic detection equipment in various locations within the electrical cluster. This module contains four functional units: an access node acquisition unit, a synchronous acquisition unit, a preprocessing unit, and a classification and determination unit. The access node acquisition unit scans and acquires data access nodes with nanosecond-level hardware timestamps to maintain communication links. The synchronous acquisition unit synchronously acquires raw operational data based on a unified hardware clock. The preprocessing unit performs noise reduction, outlier removal, and phase calibration on the raw data. The classification and determination unit classifies the raw data into the six types of operational source data.

[0129] The second module is the dual-synchronization preprocessing module, which performs spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on all running source data to generate a standardized running dataset. It is the core unit for the system to achieve time synchronization.

[0130] The third module is the coupling interference prediction module, which constructs a multi-physics field coupling interference prediction model of electrical system-geomagnetic detection and generates an electromagnetic interference prediction sequence with time and phase labels. It is the core unit for the system to realize pre-interference prevention and control.

[0131] The fourth module is the second acquisition module, which acquires the differentiated collaborative control cycle information of each device and power supply circuit. The module is equipped with four functional units: initial cycle setting unit, cycle correction unit, sub-cycle setting unit, and cycle integration unit. The initial cycle setting unit is used to set the initial first control cycle, the cycle correction unit is used to correct the first control cycle in combination with the interference prediction sequence, the sub-cycle setting unit is used to set the second, third, and fourth control cycles respectively, and the cycle integration unit is used to perform timing and phase alignment integration of the four types of control cycles to generate the final differentiated collaborative control cycle information.

[0132] The fifth module is the third acquisition module, which generates a multi-batch equipment operation sample set with multi-dimensional unified labels. This module contains six functional units: an acquisition task generation unit, a parameter configuration unit, a data source location unit, an acquisition sequence determination unit, a data acquisition unit, and a sample generation unit. Specifically, the acquisition task generation unit generates data acquisition tasks with unified spatiotemporal and phase references; the parameter configuration unit determines the configuration parameters for the acquisition tasks; the data source location unit locates the original data traceability source for various types of operational data; the acquisition sequence determination unit determines the acquisition sequence corresponding to each data source; the data acquisition unit synchronously acquires data and labels it with multi-dimensional unified labels; and the sample generation unit aligns and integrates the acquired information to form a multi-batch equipment operation sample set.

[0133] The sixth module is the fourth acquisition module, which integrates and analyzes multiple batches of equipment operation sample sets to generate evaluation values ​​in four dimensions. It is the core intermediate link connecting data collection and decision generation.

[0134] The seventh module is the decision generation module, which combines multi-dimensional evaluation values, preset detection accuracy thresholds and electromagnetic interference prediction sequences to generate dynamically adaptive and updated collaborative control decision values. It is the decision brain of the entire system.

[0135] The eighth module is the collaborative control execution module, which matches the collaborative control strategy according to the collaborative control decision value and executes three core collaborative operations. It is the implementation link of the system's core innovation.

[0136] The ninth module is the online iteration module, which continuously tracks the detection accuracy deviation, performs online iterative correction of the collaborative control decision value and control strategy, and synchronously generates pre-adjustment plans for operating conditions to achieve adaptive optimization of the system. It strictly follows the core principle of minimizing structural damage and minimizing disturbance to the original system when modifying unmanned surface vessels, and can be directly adapted to geomagnetic detection modification projects of various unmanned surface vessels, with extremely high engineering practical value and industry promotion prospects.

[0137] Finally, it should be noted that the above embodiments are merely examples for clearly illustrating the present invention and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for coordinated control of electrical equipment, characterized in that, The method includes the following steps: Acquire multiple sets of operational source data from the core magnetic detection equipment in various locations within the electrical cluster. These operational source data include magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data. Perform spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on the source data to eliminate acquisition time deviation, align the phase reference of power supply timing, equipment operation timing and detection sampling timing, and generate a standardized operation dataset; Based on a standardized operational dataset, a multi-physics field coupling interference prediction model for electrical systems and geomagnetic detection is constructed. Multi-dimensional coupling interference features are extracted, interference contribution is quantified, and an electromagnetic interference prediction sequence with time-series phase labels is generated. Based on the standardized operation dataset, electromagnetic interference prediction sequence and detection accuracy requirements, the differentiated collaborative control cycle information of each device and power supply circuit is obtained. The cycle information is dynamically linked with the detection sampling window, interference intensity and hull operating conditions. Based on the differentiated collaborative control cycle information, generate a multi-batch equipment operation sample set with multi-dimensional unified labels; By integrating and analyzing multiple batches of equipment operation sample sets, multi-dimensional evaluation values ​​of equipment operation status, power supply system stability, geomagnetic detection interference degree and navigation condition adaptability were obtained. By combining multi-dimensional evaluation values, preset detection accuracy thresholds, and electromagnetic interference prediction sequences, dynamic adaptive updated collaborative control decision values ​​are generated. Based on the coordinated control decision value matching the coordinated control strategy, the electrical cluster performs coordinated operations such as power supply ripple phase cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking. Continuously track detection accuracy deviations, perform online iterative corrections to collaborative control decision values ​​and control strategies, and simultaneously generate pre-adjustment plans for operating conditions.

2. The method for coordinated control of electrical equipment according to claim 1, characterized in that, The data obtained from multiple sets of runtime source data includes: The system acquires data from various data access nodes with hardware timestamps within the electrical cluster, and synchronously collects the raw operating data of each device based on the unified hardware clock of the onboard main control system. The raw operating data is then processed for noise reduction, outlier removal, and phase calibration. Based on the data characteristics, it is identified as magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data.

3. The method for coordinated control of electrical equipment according to claim 1, characterized in that, The steps for constructing a multiphysics coupled interference prediction model include: Based on the standardized operational dataset, multi-dimensional features of the electric field, magnetic field, hull mechanical disturbance, and environmental flow field disturbance of the electrical system are extracted to construct a multi-physics feature dataset; By training with coupled simulation and historical measured data, a nonlinear mapping relationship between multi-physics field characteristics and geomagnetic detection noise is established, and an initial prediction model is generated. Input the real-time standardized running dataset into the initial prediction model, and output the interference contribution of a single device and a single loop, the cluster coupling interference intensity, and the electromagnetic interference prediction sequence with time-series phase labels; The model is iteratively corrected online based on the real-time detection accuracy deviation to generate the final coupling interference prediction model.

4. The method for coordinated control of electrical equipment according to claim 1, characterized in that, The steps to obtain differentiated collaborative control cycle information include: Determine the timing and phase reference of the detection sampling window based on the magnetometer detection data, and set the initial first control cycle; The first control cycle is modified by combining the timing of the interference peak in the electromagnetic interference prediction sequence with the sampling window avoidance requirements; The second, third, and fourth control cycles are set according to the amplitude of the hull attitude disturbance, the voltage regulation requirements of the power supply system, and the sensitivity of equipment switching interference, respectively. The four types of control cycles are integrated with time-phase alignment to generate differentiated collaborative control cycle information that is dynamically bound to the detection sampling window and the interference prediction time.

5. The method for coordinated control of electrical equipment according to claim 1, characterized in that, The steps to obtain multidimensional evaluation values ​​include: Based on multiple batches of equipment operation sample sets, the evaluation values ​​of data temporal continuity and phase matching degree are calculated, and combined with the evaluation results of multi-dimensional content information, the evaluation value of equipment operation status is generated. Extract parameters such as voltage fluctuation, current stability, ripple phase, and load balance of the power supply system, and calculate the stability assessment value of the power supply system using the analytic hierarchy process. By combining the interference prediction results with the real-time detection accuracy deviation, the interference ratio of a single device and a single loop is calculated, and an evaluation value of geomagnetic detection interference is generated. Based on the environmental conditions, hull attitude, and detection mission requirements, the navigation condition adaptability assessment value was calculated.

6. The method for coordinated control of electrical equipment according to claim 1, characterized in that, The steps for generating collaborative control decision values ​​include: Based on the core requirements of the current exploration mission, determine the dynamic weight ratio of various operational source data; By combining multi-dimensional evaluation values, dynamic weighting of data, proportion of effective data, and phase matching degree, a single-dimensional operational quality evaluation value is calculated. By combining the power supply system's baseline parameters, single-dimensional operational quality assessment values, multi-dimensional comprehensive assessment results, and the difference between real-time detection accuracy deviation and preset thresholds, the collaborative control decision value of the electrical cluster is calculated.

7. A collaborative control system for electrical equipment, characterized in that, An onboard electrical cluster for use in the modification of unmanned surface vessels for geomagnetic detection, the electrical cluster adopts a single-input, multi-channel independent power supply architecture adapted to onboard redundant batteries. The system includes: The first acquisition module is used to acquire multiple sets of operational source data from the magnetic detection core equipment in various locations within the electrical cluster. The operational source data includes magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data. The dual-synchronization preprocessing module is used to perform spatiotemporal-electromagnetic dual-dimensional synchronous preprocessing on all operational source data, eliminate acquisition time deviation, align the phase reference of power supply timing, equipment operation timing and detection sampling timing, and generate a standardized operational dataset. The coupling interference prediction module is used to build a multi-physics coupling interference prediction model of electrical system-geomagnetic detection based on standardized operating datasets, extract multi-dimensional coupling interference features, quantify interference contribution, and generate electromagnetic interference prediction sequences with time-series phase labels. The second acquisition module is used to acquire differentiated collaborative control cycle information of each device and power supply circuit based on the standardized operation dataset, electromagnetic interference prediction sequence and detection accuracy requirements. The cycle information is dynamically linked with the detection sampling window, interference intensity and hull operating conditions. The third acquisition module is used to generate a multi-batch equipment operation sample set with multi-dimensional unified labels based on the differentiated collaborative control cycle information; The fourth acquisition module is used to integrate and analyze multiple batches of equipment operation sample sets to obtain multi-dimensional evaluation values ​​of equipment operation status, power supply system stability, geomagnetic detection interference degree and navigation condition adaptability. The decision generation module is used to combine multi-dimensional evaluation values, preset detection accuracy thresholds and electromagnetic interference prediction sequences to generate dynamically adaptive and updated collaborative control decision values. The collaborative control execution module is used to match the collaborative control strategy according to the collaborative control decision value, and to perform collaborative operations on the electrical cluster, such as power supply ripple phase reversal cancellation, equipment timing and detection window avoidance, and hull attitude and detection sampling phase locking. The online iteration module is used to continuously track the detection accuracy deviation, perform online iterative correction of the collaborative control decision value and control strategy, and simultaneously generate pre-adjustment plans for operating conditions.

8. The electrical equipment collaborative control system according to claim 7, characterized in that, The first acquisition module includes: The access node acquisition unit is used to acquire various data access nodes with hardware timestamps within the electrical cluster. The synchronous acquisition unit is used to synchronously acquire the raw operating data of each device based on the unified hardware clock of the shipborne main control system; The preprocessing unit is used to perform noise reduction, outlier removal, and phase calibration on the raw running data; The classification and determination unit is used to classify the raw operating data into magnetometer detection data, equipment operating parameters, power supply system status data, hull attitude data, positioning data, and environmental condition data based on data characteristics.

9. The electrical equipment collaborative control system according to claim 7, characterized in that, The second acquisition module includes: The initial cycle setting unit is used to determine the timing and phase reference of the detection sampling window based on the magnetometer detection data, and to set the initial first control cycle. The period correction unit is used to correct the first control period by combining the timing of the interference peak of the electromagnetic interference prediction sequence with the sampling window avoidance requirements. The cycle setting unit is used to set the second, third and fourth control cycles respectively according to the hull attitude disturbance amplitude, the power supply system voltage regulation requirements and the equipment switching interference sensitivity. The cycle integration unit is used to perform time-phase alignment and integration of four types of control cycles to generate differentiated collaborative control cycle information that is dynamically bound to the detection sampling window and the interference prediction timing.

10. The electrical equipment collaborative control system according to claim 7, characterized in that, The third acquisition module includes: The data acquisition task generation unit is used to generate data acquisition tasks with a unified spatiotemporal and phase reference based on differentiated collaborative control cycle information. The parameter configuration unit is used to determine the acquisition configuration parameters corresponding to the acquisition task. The data source location unit is used to locate the original data traceability source corresponding to various types of operational source data based on the collection configuration parameters; The acquisition sequence determination unit is used to determine the acquisition sequence corresponding to each data source; The data acquisition unit is used to synchronously collect data from various data sources according to the acquisition sequence and generate collected information with multi-dimensional unified labels. The sample generation unit is used to align and integrate various types of collected information according to spatiotemporal and phase references to form multiple batches of equipment operation sample sets.

Images