Progressive diagnosis method and system for electromagnetic interference of underwater platform

Abstract

The invention provides an underwater platform electromagnetic interference progressive diagnosis method and system, and the method comprises the steps: carrying out the real-time monitoring of a loop electromagnetic state through electromagnetic interference distributed diagnosis, and fully obtaining the electromagnetic characteristic data of an underwater platform electromagnetic interference loop; and the checking period of the electromagnetic interference reason and the coupling path is improved from month to day, so that the function of quickly diagnosing the electromagnetic interference of the loop is realized. Progressive electromagnetic interference diagnosis is realized by applying a method of fusing the expert system model based on the fault tree and the BP neural network, manual inspection and engineer experience are not depended on, and the diagnosis efficiency and accuracy are improved. The method is convenient to popularize in an actual underwater platform electromagnetic environment, and has high engineering practical value.

Description

Technical Field

[0001] This invention belongs to the field of overall electromagnetic compatibility design technology, specifically relating to a progressive diagnostic method and system for electromagnetic interference on underwater platforms. Background Technology

[0002] Underwater platforms are characterized by confined spaces, intricate cables, and a wide range of equipment frequencies. Low-frequency interference and harmonics generated by high-power power electronic converters in the power supply network propagate throughout the entire vessel, easily interfering with sensitive equipment through cable coupling and spatial radiation. With the large-scale application of high-power power electronic equipment and highly sensitive devices, electromagnetic compatibility is increasingly approaching a critical state.

[0003] The electromagnetic characteristics of electronic and electrical equipment components gradually decrease over time. Combined with changes in electromagnetic compatibility (EMC) control processes, overall layout, and cable laying, the electromagnetic environment of underwater platforms gradually deviates from its factory specifications, frequently leading to various electromagnetic interference (EMI) problems. Currently, the troubleshooting and analysis of EMC problems in China mainly relies on the engineering experience of EMC designers. Faced with the complex electromagnetic environment of actual vessels, resolving EMC issues often takes one to several months, or even more than a year. Therefore, there is an urgent need to research EMC diagnostic methods to quickly and accurately locate the faulty part of the interference problem when it occurs, significantly improving the efficiency of EMC control throughout the entire lifespan. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a progressive diagnostic method and system for electromagnetic interference of underwater platforms, which is used to quickly diagnose electromagnetic interference in circuits.

[0005] The technical solution adopted by this invention to solve the above-mentioned technical problems is as follows: a progressive diagnostic method for electromagnetic interference on underwater platforms, comprising the following steps: S1: Distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; S2: Use a fault tree-based expert system model to make a preliminary diagnosis of the coupling path and interference source type of electromagnetic interference; S3: Accurate diagnosis of the specific causes of electromagnetic interference based on BP neural network.

[0006] According to the above scheme, the specific steps in step S1 are as follows: Sensors are placed at the main current line of the power grid, the signal line of the sensitive equipment, the ground line of the sensitive equipment, the radio frequency port of the sensitive equipment, and the aperture port of the sensitive equipment in the potential interference loop to collect the electromagnetic characteristics of the potential interference loop in a distributed real-time manner. The sensor includes a current monitoring probe, an electric field antenna, and a magnetic field antenna.

[0007] According to the above scheme, in step S2, the preliminary diagnosis targets include the electromagnetic interference coupling path and interference source type of the underwater platform. The coupling paths of electromagnetic interference include conducted electromagnetic interference and radiated electromagnetic interference; the coupling paths of conducted electromagnetic interference include power lines, signal lines, and ground lines; radiated electromagnetic interference includes electric field interference and magnetic field interference. Interference sources include bio-simulated interference sources with single-frequency characteristics and digital interference sources with harmonic characteristics.

[0008] According to the above scheme, the specific steps in step S2 are as follows: S21: Construct a fault tree for electromagnetic interference diagnosis; S22: Based on the established electromagnetic interference fault tree and historical electromagnetic interference fault information, construct the reasoning rule table of the expert knowledge system. S23: Diagnose electromagnetic interference simply based on inference rules.

[0009] Furthermore, in step S21, the specific steps are as follows: S211: For the research objective, electromagnetic interference faults are taken as the top event of the fault tree; S212: Decompose step by step to obtain the sub-fault tree corresponding to the abnormal situation of each monitoring point.

[0010] Furthermore, in step S23, the specific steps are as follows: S231: Based on the current monitoring results, select the matching fault tree nodes; S232: Combine the expert reasoning rule table with monitoring results to match reasoning rules; S233: Output all possible inference results based on the inference rule table.

[0011] According to the above scheme, the specific steps in step S3 are as follows: S31: Construct the training dataset; S32: Construct and train a BP neural network model to establish a high-dimensional mapping relationship between spectrum monitoring data with different characteristics and different types of interference causes; S33: Accurately diagnose the causes of complex electromagnetic interference.

[0012] Furthermore, in step S32, the specific steps are as follows: S321: Determine the topology of the BP neural network based on the input data and output characteristics of the interference branch; S322: Initialize network weights and bias parameters, target error, and activation function; S323: Based on the characteristics of the interference branch, monitor the spectrum data of different features under different states; S324: Input training data, calculate network error, iterate repeatedly to train until the target error is reached.

[0013] A progressive diagnostic system for electromagnetic interference on underwater platforms. The distributed acquisition submodule is used for distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; The preliminary diagnosis submodule is used to make preliminary diagnoses of the coupling path and interference source type of electromagnetic interference using a fault tree-based expert system model; The Precision Diagnosis submodule is used to accurately diagnose the specific causes of electromagnetic interference based on a BP neural network.

[0014] A computer memory storing a computer program executable by a computer processor, the computer program performing a progressive diagnostic method for electromagnetic interference on an underwater platform.

[0015] The beneficial effects of this invention are as follows: 1. The present invention provides a progressive diagnostic method and system for electromagnetic interference of underwater platforms. By conducting distributed electromagnetic interference diagnosis, the electromagnetic state of the circuit is monitored in real time, and the electromagnetic characteristic data of the electromagnetic interference circuit of the underwater platform is fully obtained. The investigation cycle of electromagnetic interference causes and coupling paths is reduced from a month to a day, thereby realizing the function of rapid diagnosis of electromagnetic interference of the circuit.

[0016] 2. This invention applies a method that combines a fault tree-based expert system model with a backpropagation neural network to achieve progressive electromagnetic interference diagnosis. It does not rely on manual inspections or engineers' experience, thus improving the efficiency and accuracy of diagnosis.

[0017] 3. This invention is easy to promote in the electromagnetic environment of actual underwater platforms and has strong engineering practical value.

[0018] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of a typical interference branch in an embodiment of the present invention.

[0022] Figure 3 This is a structural diagram of the distributed electromagnetic interference diagnostic device according to an embodiment of the present invention.

[0023] Figure 4 This is a basic fault tree diagram of an embodiment of the present invention.

[0024] Figure 5 This is a schematic diagram of the fault tree of the electric field monitoring point according to an embodiment of the present invention.

[0025] Figure 6 This is a schematic diagram of the fault tree of the magnetic field monitoring point in an embodiment of the present invention.

[0026] Figure 7 This is a schematic diagram of the fault tree of the power line of the interference loop according to an embodiment of the present invention.

[0027] Figure 8 This is a schematic diagram of the fault tree of the coupled circuit signal line in an embodiment of the present invention.

[0028] Figure 9 This is a schematic diagram of the fault tree of the ground wire of the coupled loop in an embodiment of the present invention.

[0029] Figure 10 This is a schematic diagram of an electromagnetic interference fault tree according to an embodiment of the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0031] Example 1 See Figure 1 The specific steps of a progressive diagnostic method for electromagnetic interference on underwater platforms are as follows: S1: Distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; Distributed diagnostic devices for electromagnetic interference of potential interference loops of underwater platforms are deployed at locations such as the main current line of the power grid, the signal line of sensitive equipment, the ground wire of sensitive equipment, the radio frequency port of sensitive equipment, and the port of sensitive equipment. Electromagnetic field sensors such as current monitoring probes, electric field antennas, and magnetic field antennas are used to collect the electromagnetic characteristics of potential interference loops in a distributed and real-time manner.

[0032] S2: Use a fault tree-based expert system model to make a preliminary diagnosis of electromagnetic interference; The initial diagnostic objective is to identify the electromagnetic interference (EMI) coupling paths and interference source types on the underwater platform. EMI coupling paths include conducted EMI and radiated EMI. Conducted EMI coupling paths include power lines, signal lines, and ground lines, while radiated EMI includes electric field interference and magnetic field interference. Based on the characteristics of the interference, a preliminary judgment of the interference source is needed, classifying it into analog and digital interference sources based on single-frequency and harmonic characteristics. This step provides engineers with a simple and effective way to quickly understand the system's reliability and identify potential failure modes. The specific steps are as follows: S21: Constructing an electromagnetic interference diagnostic fault tree S211: For the research objective, electromagnetic interference faults are taken as the top event of the fault tree; S212: Decompose step by step to obtain sub-fault trees corresponding to the abnormal situations of each monitoring point; S22: Based on the established system electromagnetic interference fault tree and historical electromagnetic interference fault information, construct the reasoning rule table of the expert knowledge system. S23: Simple diagnosis of electromagnetic interference based on inference rules; perform inference diagnosis according to the following steps: S231: Node Filtering: Based on the current monitoring results, filter out matching fault tree nodes; S232: Rule matching: Combine the expert reasoning rule table and monitoring results to find matching reasoning rules; S233: Comprehensive Reasoning: Output all possible reasoning results based on the reasoning rule table; S3: Accurate diagnosis of electromagnetic interference based on BP neural network; Considering that the subject of this invention faces real-time acquisition of electromagnetic monitoring data and fault reasoning, and that the monitoring data is often unknown and variable, the electromagnetic interference diagnosis method based solely on expert reasoning systems with fixed rules has certain limitations. Therefore, it is necessary to further combine machine learning-based intelligent diagnosis methods, construct a suitable machine learning network model, and fully leverage its advantages in nonlinear mapping capability, adaptive learning capability, flexible network features, and scalability to achieve accurate diagnosis of real-time electromagnetic interference monitoring data. Furthermore, it is necessary to diagnose the factors generating electromagnetic interference, such as the operating status of the load on the underwater platform's electrical grid and changes in load parameters, to achieve precise location and efficient diagnosis of electromagnetic interference. The specific steps are as follows: S31: Construct the training dataset; S32: Construct and train the BP neural network model; the specific steps are as follows: S321: Determine the topology of the BP neural network based on the input data and output characteristics of the interference branch; S322: Initialize network weights and bias parameters, target error, and activation function; S323: Based on the characteristics of the interference branch, monitor the spectrum data of different features under different states; S324: Input training data, calculate network error, iterate training repeatedly until the target error is achieved; In this way, a high-dimensional mapping relationship is established between spectrum monitoring data with different characteristics and different types of interference causes through a BP neural network.

[0033] S33: Accurate diagnosis of complex interference; In practical applications, a well-trained BP neural network can be regarded as a two-port network with a known transfer function. After the initial diagnosis is completed by the expert system model based on the fault tree, if it is indicated that further accurate diagnosis is needed, the spectrum data monitored in the current actual interference loop needs to be used as input. The BP network can then identify and classify the features of this data, thereby determining which type of electromagnetic interference cause the data is associated with, and achieving accurate diagnosis.

[0034] Example 2 The steps in this embodiment are the same as in Embodiment 1, the difference being that each step is applied to a typical interference branch in a specific instance. See [link / reference needed]. Figure 2 Specifically, it includes the following steps: S1: Deploy a distributed monitoring device for electromagnetic interference loops, including one electric field probe, one magnetic field probe, three current probes, and one set of 5-channel electromagnetic measurement and receiving equipment; the performance specifications of the electric field probe, magnetic field probe, and current probe are as follows: 1) Electric field monitoring: 10kHz~30MH; 2) Magnetic field monitoring: 25Hz~100kHz; 3) Current monitoring channels 1 to 3: 25Hz to 100kHz, spectral resolution 100Hz, dynamic range 55dBuA to 140dBuA.

[0035] Three current monitoring probes are respectively positioned on the main power supply line of the power grid, the signal line of the sensitive equipment, and the ground line of the sensitive equipment. The electric field probe is positioned at the radio frequency port of the sensitive equipment, and the magnetic field probe is positioned at the aperture of the sensitive equipment. The 5-channel electromagnetic measurement receiving equipment has a test bandwidth covering 25Hz to 30MHz, a spectral resolution better than 10Hz, and an amplitude measurement error better than 3dB.

[0036] S2: Use an expert system to make a preliminary diagnosis of simple electromagnetic interference; In order to make full use of the interconnection relationships, historical fault conditions and other data and experience in the typical electromagnetic interference circuits of underwater platforms, and to transform them into knowledge that can be used by expert systems, this embodiment first uses the expert system method to achieve the preliminary diagnosis of simple electromagnetic interference.

[0037] S21: Constructing an electromagnetic interference diagnostic fault tree S211: The occurrence of electromagnetic interference (EMI) faults in typical interference branches of an underwater platform is designated as the top event in the fault tree. When an EMI fault occurs in a typical interference circuit, abnormal electromagnetic signals may be detected at different locations within the circuit. Common types and locations for EMI monitoring include: magnetic fields, electric fields, power lines of the interference circuit, signal lines of the coupled circuit, and ground lines of the coupled circuit. Based on this, a basic EMI fault tree for the typical interference branches of an underwater platform can be established, such as... Figure 4 As shown.

[0038] S212: Based on the above analysis of the electromagnetic interference source fault tree, and further decomposing the abnormal conditions at each monitoring point, sub-fault trees corresponding to the abnormal conditions at each monitoring point are obtained. For different interference coupling paths, electromagnetic interference sources can be divided into two main categories: electromagnetic interference with relatively obvious single-frequency abnormalities and electromagnetic interference with relatively obvious multi-frequency abnormalities and significant harmonic characteristics. The interference sources generating these two types of electromagnetic interference are respectively inferred as analog electromagnetic interference signals and digital electromagnetic interference signals; simultaneously, the specific interference causes are inferred from the frequency characteristics of the current interference signal and the electromagnetic emission characteristics of the working equipment in the system, such as... Figures 5-9 As shown.

[0039] Based on the established basic fault tree and sub-fault tree models, the electromagnetic interference fault trees of typical interference branches are summarized as follows: Figure 10 As shown.

[0040] S22: Constructing reasoning rules for the expert system; Based on the typical interference branch electromagnetic interference fault tree established in step S21 and the historical electromagnetic interference faults of the underwater platform, the reasoning rule table of the expert system is compiled, as shown in Table 1.

[0041] Table 1 Reasoning Rules

[0042] According to the expert system reasoning rule table mentioned above, once the electromagnetic signal monitoring hardware detects a typical phenomenon in the "IF" column, it can directly locate the corresponding electromagnetic interference source using the aforementioned reasoning rules, thus enabling rapid diagnosis of simple electromagnetic interference faults. If multiple IFs are used to infer a single THEN (AND gate structure), all IFs of the preconditions must be true before a result can be derived.

[0043] S23: Simple diagnosis of electromagnetic interference based on reasoning rules; S231: Based on the current monitoring results, select the matching fault tree nodes; S232: Combine the expert reasoning rule table and monitoring results to find matching reasoning rules; S233: Output all possible results based on the inference rule table.

[0044] S3: Accurate diagnosis of complex interference based on BP neural network; After electromagnetic interference pre-diagnosis based on an expert system, interference paths such as magnetic field interference, electric field interference, and line-to-line coupling interference can be preliminarily located. However, the specific causes of line-to-line coupling interference are difficult to pinpoint directly. Therefore, for this part of the precise diagnostic requirement, this embodiment proposes to conduct analysis using a BP neural network model as the basic technical approach.

[0045] This step involves classifying the spectrum monitoring data in the interference loop to be diagnosed, and establishing a high-dimensional mapping relationship between spectrum monitoring data with different characteristics and different types of interference causes through a BP neural network, thereby achieving accurate diagnosis of specific electromagnetic interference causes.

[0046] S31: Construct the training dataset; The precise diagnosis of complex interference based on BP neural networks mainly uses electromagnetic field monitoring data at typical locations in typical interference branches as input, and the causes and coupling paths of interference in typical interference branches as output. The electromagnetic field monitoring data at typical locations includes electric and magnetic field radiation emissions near sensitive equipment, conducted emissions from the main power lines in the power supply circuit of sensitive equipment, conducted emissions from the signal lines of sensitive equipment, and conducted emissions from the ground wire of sensitive equipment.

[0047] By simulating different interference causes and coupling paths in typical interference branch loops, electromagnetic field monitoring data at typical locations are obtained. 500 sets of data pairs consisting of "monitoring data → interference cause and coupling path" are established to construct a training dataset for a BP neural network.

[0048] S32: Construct and train the BP neural network model; the specific steps are as follows: S321: Determine the network topology; In this use case, the input sample data has three features: signal amplitude at three frequency points (2kHz, 4kHz, and 6kHz). Therefore, the number of input layer neurons in the neural network is set to 3. There are eight possible causes of line-to-line coupling interference in this use case, therefore, the number of output layer neurons in the BP neural network model is set to 8. By convention, the hidden layers are set to 3, and the number of neurons is set to 18. S322: Initialize network weights, bias parameters, and activation function; randomly determine the weights and bias parameters of the BP neural network within the range [-3,3]; select the Sigmoid function as the activation function; S323: Based on the characteristics of the interference branch, monitor the spectrum data of different features under different states; S324: Input training data, calculate network error, and iterate training repeatedly until the target error is achieved; the specific steps are as follows: S3241: Calculate the network error function; input the training data from the network input layer into the BP neural network, and calculate the mean square error between the predicted output and the actual target output; S3242: Error backpropagation; optimizing and adjusting the weights and biases in the network through the backpropagation algorithm; S3243: Iterative training; Steps S3241 and S3242 constitute one training cycle of the BP neural network, with the maximum number of iterations set to 2000; Stopping condition: When the network iteration process reaches the maximum number of iterations (2000) or the set target error is 10. -5 When the iteration stops, the BP neural network training is complete; S33: Accurate diagnosis of complex interference; After the initial diagnosis is completed by the fault tree-based expert system, if further precise diagnosis is required, the spectrum data monitored in the actual interference loop should be used as input. The BP network can then identify and classify the features of this data to determine which type of electromagnetic interference is associated with the data, thus achieving precise diagnosis.

[0049] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0050] Example 3 This embodiment is used to implement the principle of the above method embodiment to construct an underwater platform electromagnetic interference progressive diagnostic system, including a distributed acquisition submodule, a preliminary diagnosis submodule, and a precise diagnosis submodule; The distributed acquisition submodule is used for distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; The preliminary diagnosis submodule is used to make preliminary diagnoses of the coupling path and interference source type of electromagnetic interference using a fault tree-based expert system model; The Precision Diagnosis submodule is used to accurately diagnose the specific causes of electromagnetic interference based on a BP neural network.

[0051] Each submodule is mainly used to implement the various steps of the method implementation, which will not be elaborated here.

[0052] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0053] This embodiment also includes a processor, a communication interface, a memory, and a communication bus; wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory stores a computer program, and when the program is executed by the processor, the processor performs the steps of a progressive diagnostic method for electromagnetic interference on an underwater platform.

[0054] This embodiment also provides a computer-readable storage medium storing executable instructions that, when executed by a processor, enable the processor to implement a progressive diagnostic method for electromagnetic interference on an underwater platform.

[0055] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0056] Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0057] This application is described with reference to the flowchart of the method and computer program product according to Embodiment 1 and the block diagram of the device (system) according to Embodiment 3. It should be understood that each step or block in the flowchart or block diagram, as well as combinations of steps or blocks in the flowchart or block diagram, can be implemented by computer program instructions.

[0058] These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which are executable by the processor of the computer or other programmable data processing device, produce instructions for implementing the process. Figure 1 One or more processes or boxes Figure 1 An underwater platform electromagnetic interference progressive diagnostic system that specifies functions within one or more boxes.

[0059] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes or boxes Figure 1The function specified in one or more boxes.

[0060] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes or boxes Figure 1 The steps of a progressive diagnostic method for electromagnetic interference on an underwater platform are specified in one or more boxes.

[0061] The above embodiments are only used to illustrate the design concept and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes or modifications made based on the principles and design ideas disclosed in the present invention are within the protection scope of the present invention.

Claims

1. A progressive diagnostic method for electromagnetic interference on underwater platforms, characterized in that: Includes the following steps: S1: Distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; S2: Use a fault tree-based expert system model to make a preliminary diagnosis of the coupling path and interference source type of electromagnetic interference; S3: Accurate diagnosis of the specific causes of electromagnetic interference based on BP neural network.

2. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 1, characterized in that: The specific steps in step S1 are as follows: Sensors are placed at the main current line of the power grid, the signal line of the sensitive equipment, the ground line of the sensitive equipment, the radio frequency port of the sensitive equipment, and the aperture port of the sensitive equipment in the potential interference loop to collect the electromagnetic characteristics of the potential interference loop in a distributed real-time manner. The sensor includes a current monitoring probe, an electric field antenna, and a magnetic field antenna.

3. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 1, characterized in that: In step S2, the preliminary diagnosis targets include the electromagnetic interference coupling path and interference source type of the underwater platform; The coupling paths of electromagnetic interference include conducted electromagnetic interference and radiated electromagnetic interference; the coupling paths of conducted electromagnetic interference include power lines, signal lines, and ground lines; radiated electromagnetic interference includes electric field interference and magnetic field interference. Interference sources include bio-simulated interference sources with single-frequency characteristics and digital interference sources with harmonic characteristics.

4. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 1, characterized in that: The specific steps in step S2 are as follows: S21: Construct a fault tree for electromagnetic interference diagnosis; S22: Based on the established electromagnetic interference fault tree and historical electromagnetic interference fault information, construct the reasoning rule table of the expert knowledge system. S23: Diagnose electromagnetic interference simply based on inference rules.

5. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 4, characterized in that: The specific steps in step S21 are as follows: S211: For the research objective, electromagnetic interference faults are taken as the top event of the fault tree; S212: Decompose step by step to obtain the sub-fault tree corresponding to the abnormal situation of each monitoring point.

6. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 4, characterized in that: The specific steps in step S23 are as follows: S231: Based on the current monitoring results, select the matching fault tree nodes; S232: Combine the expert reasoning rule table with monitoring results to match reasoning rules; S233: Output all possible inference results based on the inference rule table.

7. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 1, characterized in that: The specific steps in step S3 are as follows: S31: Construct the training dataset; S32: Construct and train a BP neural network model to establish a high-dimensional mapping relationship between spectrum monitoring data with different characteristics and different types of interference causes; S33: Accurately diagnose the causes of complex electromagnetic interference.

8. The progressive diagnostic method for electromagnetic interference on an underwater platform according to claim 7, characterized in that: The specific steps in step S32 are as follows: S321: Determine the topology of the BP neural network based on the input data and output characteristics of the interference branch; S322: Initialize network weights and bias parameters, target error, and activation function; S323: Based on the characteristics of the interference branch, monitor the spectrum data of different features under different states; S324: Input training data, calculate network error, iterate repeatedly to train until the target error is reached.

9. A progressive diagnostic system for electromagnetic interference on underwater platforms, characterized in that: The distributed acquisition submodule is used for distributed real-time acquisition of the electromagnetic characteristics of potential interference loops; The preliminary diagnosis submodule is used to make preliminary diagnoses of the coupling path and interference source type of electromagnetic interference using a fault tree-based expert system model; The Precision Diagnosis submodule is used to accurately diagnose the specific causes of electromagnetic interference based on a BP neural network.

10. A computer memory, characterized in that: It contains a computer program that can be executed by a computer processor, which performs a progressive diagnostic method for electromagnetic interference on an underwater platform as described in any one of claims 1 to 8.

Images