Underwater propeller fault detection method, device and computer-readable storage medium

Abstract

The present application discloses an underwater propeller fault detection method, device, and computer-readable storage medium for improving the accuracy of fault detection. The method includes: establishing a functional similarity relationship and a control coupling relationship between the propellers based on the thrust capacity, thrust direction, and control allocation coefficient of each propeller in a multi-propeller system; constructing a joint state prediction model for the multi-propeller system based on the functional similarity relationship and the control coupling relationship; collecting current operating state data and control input data of each propeller; inputting the control input data into the joint state prediction model to calculate the predicted operating state data of each propeller; collecting actual operating state data of each propeller, and calculating a state residual based on the actual operating state data and the predicted operating state data; and comparing the state residual with a preset threshold to determine whether the multi-propeller system has a fault.

Description

Technical Field

[0001] The present application relates to the technical field of underwater thrusters, and in particular to a method, device and computer-readable storage medium for detecting underwater thruster faults. Background Art

[0002] As underwater operations become increasingly complex and diverse, reliable monitoring and fault detection of underwater thrusters, key power units, are becoming increasingly important. Accurately and promptly detecting abnormal thruster conditions not only impacts the normal operation of the equipment but also directly impacts the safety and efficiency of the entire underwater operation system.

[0003] Currently, the fault detection technologies commonly used in the market and within research primarily rely on the collection of thruster operating parameters and single threshold determination, or on fault identification based on fixed patterns. These methods typically focus on a single data dimension or indicator and lack the comprehensive utilization of multi-dimensional information. Their detection models struggle to fully capture the dynamic changes occurring under complex operating conditions during thruster operation, resulting in significant errors in detection results.

[0004] Therefore, in the existing technology, the thruster fault detection technology still has obvious shortcomings in terms of accuracy, and it is difficult to meet the requirements of modern underwater propulsion systems for high reliability and high-precision fault diagnosis. Summary of the Invention

[0005] The embodiments of the present application provide an underwater thruster fault detection method, device, and computer-readable storage medium, which can improve the accuracy of fault detection.

[0006] A first aspect of an embodiment of the present application provides a method for detecting underwater thruster faults, comprising:

[0007] According to the thrust capability, thrust direction and control distribution coefficient of each thruster in the multi-thruster system, the functional similarity relationship and control coupling relationship between the thrusters are established;

[0008] constructing a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship;

[0009] Collecting current operating status data and control input data of each thruster, wherein the operating status data includes speed, current, thrust feedback and attitude response;

[0010] Inputting the control input data into the joint state prediction model to calculate predicted operating state data of each thruster;

[0011] Collecting actual operating state data of each thruster, and calculating a state residual based on the actual operating state data and the predicted operating state data; wherein the actual operating state data is data at the next moment corresponding to the current operating state data;

[0012] Comparing the state residual with a preset threshold to determine whether the multi-thruster system has a fault;

[0013] If a fault exists, the faulty thruster and the fault type are determined based on the distribution characteristics of the state residual and the control coupling relationship.

[0014] Optionally, constructing a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship includes:

[0015] determining thruster groups according to the similarity relationship;

[0016] determining a joint state transfer matrix according to the thruster grouping and the control coupling relationship;

[0017] A joint state prediction model of the multi-thruster system is constructed based on the joint state transfer matrix.

[0018] Optionally, constructing the joint state prediction model of the multi-thruster system based on the joint state transfer matrix includes:

[0019] Determine the current joint state vector, input gain matrix and joint control input matrix according to the parameter information of each thruster;

[0020] A joint state prediction model of the multi-thruster system is constructed according to the current joint state vector, the input gain matrix, the joint control input matrix and the joint state transfer matrix.

[0021] Optionally, the joint state prediction model is specifically designed as follows:

[0022]

[0023]

[0024]

[0025]

[0026] in, A vector representing the predicted next state, represents the current joint state vector, represents the joint control input matrix, represents the input gain matrix, represents the joint state transition matrix.

[0027] Optionally, determining the faulty thruster and the fault type according to the distribution characteristics of the state residual and the control coupling relationship includes:

[0028] constructing a residual matrix according to the state residual;

[0029] Constructing a coupling coefficient matrix based on the control coupling relationship;

[0030] generating a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix;

[0031] The residual coupling influence matrix is ​​input into a pre-trained thruster fault identification model to determine the faulty thruster and the fault type.

[0032] Optionally, generating a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix includes:

[0033] Calculating a normalized coupling weight matrix using the residual vector amplitudes of each row of the residual matrix and corresponding elements of the coupling coefficient matrix;

[0034] Applying a double nonlinear transformation to each residual vector in the residual matrix to obtain a transformation vector;

[0035] The transformation vectors are weighted and aggregated by the normalized coupling weight matrix to obtain a residual coupling influence matrix.

[0036] Optionally, comparing the state residual with a preset threshold to determine whether the multi-thruster system has a fault includes:

[0037] Determine whether each item of the state residual is less than a corresponding preset threshold;

[0038] If not, it is determined that there is a fault in the multi-thruster system; if so, it is determined that there is no fault in the multi-thruster system.

[0039] A second aspect of an embodiment of the present application provides an underwater thruster fault detection device, comprising:

[0040] An establishment unit for establishing functional similarity relationships and control coupling relationships between thrusters in a multi-thruster system based on thrust capabilities, thrust directions, and control allocation coefficients of each thruster;

[0041] a construction unit, configured to construct a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship;

[0042] A first acquisition unit is used to collect current operating status data and control input data of each thruster, wherein the operating status data includes speed, current, thrust feedback and attitude response;

[0043] an input unit, configured to input the control input data into the joint state prediction model to calculate predicted operating state data of each thruster;

[0044] a second acquisition unit, configured to acquire actual operating state data of each thruster and calculate a state residual based on the actual operating state data and the predicted operating state data; wherein the actual operating state data is data at a next moment corresponding to the current operating state data;

[0045] a judgment unit, configured to compare the state residual with a preset threshold to determine whether the multi-thruster system has a fault;

[0046] A determination unit is configured to determine the faulty thruster and the fault type based on the distribution characteristics of the state residual and the control coupling relationship if a fault exists.

[0047] Optionally, the construction unit includes:

[0048] A first determining module, configured to determine a thruster grouping according to the similarity relationship;

[0049] a second determining module, configured to determine a joint state transfer matrix according to the thruster grouping and the control coupling relationship;

[0050] The first building module is used to build a joint state prediction model of the multi-thruster system based on the joint state transfer matrix.

[0051] Optionally, the first building module is specifically used to:

[0052] Determine the current joint state vector, input gain matrix and joint control input matrix according to the parameter information of each thruster;

[0053] A joint state prediction model of the multi-thruster system is constructed according to the current joint state vector, the input gain matrix, the joint control input matrix and the joint state transfer matrix.

[0054] Optionally, the joint state prediction model is specifically designed as follows:

[0055]

[0056]

[0057]

[0058]

[0059] in, A vector representing the predicted next state, represents the current joint state vector, represents the joint control input matrix, represents the input gain matrix, represents the joint state transition matrix.

[0060] Optionally, the determining unit includes:

[0061] A second construction module is used to construct a residual matrix according to the state residual;

[0062] A third building module is used to build a coupling coefficient matrix based on the control coupling relationship;

[0063] A generating module, configured to generate a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix;

[0064] An input module is used to input the residual coupling influence matrix into a pre-trained thruster fault identification model to determine the faulty thruster and the fault type.

[0065] Optionally, the generating module is specifically configured to:

[0066] Calculating a normalized coupling weight matrix using the residual vector amplitudes of each row of the residual matrix and corresponding elements of the coupling coefficient matrix;

[0067] Applying a double nonlinear transformation to each residual vector in the residual matrix to obtain a transformation vector;

[0068] The transformation vectors are weighted and aggregated by the normalized coupling weight matrix to obtain a residual coupling influence matrix.

[0069] Optionally, the judgment unit is specifically configured to:

[0070] Determine whether each item of the state residual is less than a corresponding preset threshold;

[0071] If not, it is determined that there is a fault in the multi-thruster system; if so, it is determined that there is no fault in the multi-thruster system.

[0072] A third aspect of the embodiments of the present application provides an underwater thruster fault detection device, comprising:

[0073] processor, memory, input and output units, and buses;

[0074] The processor is connected to the memory, the input and output unit, and the bus;

[0075] A program is stored in the memory, and the processor calls the program to execute the method in the first aspect and any possible implementation of the first aspect.

[0076] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, on which a program is stored. When the program is executed on a computer, the computer executes the method in the first aspect and any possible implementation of the first aspect.

[0077] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0078] In the embodiment of the present application, the joint state prediction model comprehensively considers the functional similarity and control coupling relationship between multiple thrusters, realizes the accurate prediction of the overall state of multiple thrusters, avoids the problems of information isolation and error accumulation when predicting a single thruster, and makes the state residual more truly reflect the abnormal situation, thereby greatly improving the accuracy of fault detection and positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 This is a flow chart of an embodiment of a method for detecting underwater propeller failure in an embodiment of the present application;

[0080] Figure 2 A flowchart of an embodiment of constructing a joint state prediction model for a multi-thruster system in an embodiment of the present application;

[0081] Figure 3 This is a flow chart of an embodiment of building a joint state prediction model for a multi-thruster system based on a joint state transfer matrix in an embodiment of the present application;

[0082] Figure 4 This is a flowchart of an embodiment of determining a faulty thruster and a fault type in an embodiment of the present application;

[0083] Figure 5 This is a flow chart of an embodiment of generating a residual coupling influence matrix in an embodiment of the present application;

[0084] Figure 6 This is a flow chart of an embodiment of determining whether a multi-thruster system has a fault in the embodiment of the present application;

[0085] Figure 7 This is a structural diagram of an embodiment of an underwater propeller fault detection device in an embodiment of the present application;

[0086] Figure 8 This is a structural diagram of another embodiment of the underwater propeller fault detection device in the embodiment of the present application. DETAILED DESCRIPTION

[0087] Embodiments of the present application provide an underwater thruster fault detection method, device, and computer-readable storage medium for improving the accuracy of fault detection.

[0088] The method of the present application can be applied to a server, a terminal or other device with logic processing capabilities, and the present application does not limit this. For the convenience of description, the following description is based on an example in which the execution subject is a server.

[0089] The embodiments of the present application will be described below with reference to the accompanying drawings.

[0090] See also Figure 1 , Figure 1 An embodiment of the underwater thruster fault detection method provided in this application includes:

[0091] 101. Based on the thrust capacity, thrust direction and control allocation coefficient of each thruster in the multi-thruster system, establish the functional similarity relationship and control coupling relationship between the thrusters;

[0092] The server analyzes the thrust capacity, thrust direction, and control allocation coefficients of each thruster in a multi-thruster system. It first extracts the upper limit of each thruster's thrust magnitude and direction vector. Combined with the system's control allocation strategy, it calculates the substitution capacity and response impact between thrusters when performing attitude adjustment or position control tasks. Based on the functional similarity of the thrusters in control tasks, the server establishes functional similarity relationships between thrusters. Furthermore, combining the correlation of each thruster's response to the same control objective, it forms a control coupling relationship, thereby forming a quantifiable inter-thruster correlation data structure for subsequent model construction.

[0093] 102. Based on the functional similarity relationship and control coupling relationship, a joint state prediction model of a multi-thruster system is constructed;

[0094] Based on the functional similarity and control coupling relationships established above, the server constructs a joint state prediction model. This model treats multiple thrusters as a dynamic, interconnected system and uses multivariable modeling methods, such as multi-input, multi-output (MIMO) system identification, neural network prediction models, or extended Kalman filter models, to model the mapping relationship between the control inputs and operating state changes of the entire thruster cluster. This joint model can predict the operating state of multiple thrusters after inputting control signals at any time, reflecting the collaborative response characteristics between systems. It should be noted that after constructing the joint state prediction model, the server can verify and adjust the joint state prediction model to make it more reliable.

[0095] 103. Collect the current operating status data and control input data of each thruster, the operating status data including speed, current, thrust feedback and attitude response;

[0096] During system operation, the server continuously collects the current operating status and control input data of each thruster. This operating status data includes the thruster's current speed, current, thrust feedback (e.g., obtained by a thrust sensor), and attitude response (such as angular velocity or attitude angle change). This data is collected in real time by sensors and uploaded to the server to reflect the current physical state of the thruster. Simultaneously, the server collects control input signals from the controller, such as voltage or PWM control signals, to drive each thruster.

[0097] 104. Input the control input data into the joint state prediction model to calculate the predicted operating state data of each thruster;

[0098] The server feeds the collected control input data into a joint state prediction model to predict the operating state of each thruster under the current control input. The predicted operating state data corresponds to parameters such as thrust, current, speed, and attitude response expected at the next moment or within a future time window. These predicted values ​​serve as a reference for subsequent comparison with the actual operating state to assess whether the thruster is behaving normally.

[0099] 105. Collect actual operating status data of each thruster, and calculate the state residual based on the actual operating status data and the predicted operating status data; wherein the actual operating status data is the data at the next moment corresponding to the current operating status data;

[0100] After completing the prediction, the server continues to collect data on the actual operating status of each thruster at the next moment—that is, the actual response of the thruster to the control input at the previous moment. The server compares this actual operating status data with the predicted operating status data item by item, calculating the difference between each state parameter to form a state residual. This is used to quantify the deviation between the predicted result and the actual response, thus providing a basis for fault diagnosis.

[0101] 106. Compare the state residual with a preset threshold to determine whether the multi-thruster system has a fault;

[0102] The server compares the state residual corresponding to each thruster with the preset threshold set by the system. The preset threshold represents the acceptable prediction error range of the system. If a state residual of a certain thruster exceeds the preset threshold, it is determined that the thruster has abnormal behavior, and then it is determined that there may be a fault in the multi-thruster system. This comparison process can adopt multi-dimensional residual judgment criteria, such as the Euclidean distance method or weighted error scoring. It should be noted that the state residual includes information in multiple dimensions, so the preset threshold needs to be set accordingly, and the server needs to compare the residual state with the corresponding preset threshold respectively.

[0103] 107. If there is a fault, the faulty propeller and fault type are determined based on the distribution characteristics of the state residual and the control coupling relationship.

[0104] After determining a fault, the server further analyzes the distribution of state residuals across the thrusters. For example, if multiple parameters of a particular thruster are simultaneously out of range or a certain type of state parameter exhibits consistency deviations across multiple coupled thrusters, the server analyzes the propagation path of the coupling signature based on previously established control coupling relationships and identifies the core thruster causing the fault. Combining residual characteristics (such as thrust anomalies, sudden current changes, and sluggish attitude response), the server determines the fault type, such as mechanical jamming, drive failure, or response deviation, thereby identifying the specific faulty thruster and its type.

[0105] In this embodiment, the joint state prediction model comprehensively considers the functional similarity and control coupling relationship between multiple thrusters, realizes the accurate prediction of the overall state of multiple thrusters, avoids the problems of information isolation and error accumulation when predicting a single thruster, and makes the state residual more realistically reflect the abnormal situation, thereby greatly improving the accuracy of fault detection and positioning.

[0106] See also Figure 2 In some embodiments of the present application, step 102 of the above embodiment constructs a joint state prediction model of a multi-thruster system based on the functional similarity relationship and the control coupling relationship, which may include the following steps:

[0107] 201. Determine thruster grouping based on similarity relationships;

[0108] The server determines how to group the thrusters based on their functional similarities. The server first reads each thruster's parameters, such as thrust capacity, response speed, control sensitivity, and thrust direction, and uses these parameters to calculate a similarity index between the thrusters. The server then uses a clustering algorithm (such as K-means, hierarchical clustering, or distance-based partitioning) to group thrusters with similar functional characteristics or similar roles in control tasks. For example, thrusters with consistent thrust directions and similar response characteristics can be grouped together. Grouping thrusters allows for a unified state description to be established for similar thrusters during subsequent modeling, reducing modeling complexity while retaining the representativeness of the system's internal structure.

[0109] 202. Determine a joint state transfer matrix based on thruster grouping and control coupling relationships;

[0110] The server constructs a joint state transfer matrix based on the thruster grouping results in step 201 and the control coupling relationship between the thrusters. The server first establishes an intra-group state transfer model within each thruster group to determine the mapping relationship between the current state of each thruster in the group and the state at the next moment. Subsequently, the server analyzes how the control input or state change in one group affects the thruster response in other groups based on the control coupling relationship, and converts this mutual influence into a coupling term in the matrix. Finally, the server integrates the local state transfer matrices and coupling relationship terms of all groups into a large-scale joint state transfer matrix, which comprehensively reflects the global state evolution characteristics of the multi-thruster system and retains the interaction and interference between the thrusters.

[0111] 203. Construct a joint state prediction model for a multi-thruster system based on the joint state transfer matrix.

[0112] The server uses the constructed joint state transition matrix to establish a joint state prediction model for the multi-thruster system. The server uses the current operating state and control inputs of the thrusters as inputs to the model. Based on the state changes and coupling relationships defined in the matrix, the server predicts the operating state of each thruster at the next moment. The prediction results include state variables such as each thruster's speed, current, thrust feedback, and attitude response. This model supports unified prediction of the states of all thrusters and reflects the coordinated response characteristics between thrusters, serving as the basis for subsequent residual analysis and fault detection.

[0113] In this embodiment, the server constructs a joint state transfer matrix based on the establishment of thruster grouping, and uses the matrix to establish a joint state prediction model, so that the prediction model can simultaneously consider the functional similarity and control coupling relationship between the thrusters, thereby improving the comprehensiveness and accuracy of the state prediction, providing a reliable basis for fault detection, and helping to improve the robustness and safety of the multi-thruster system.

[0114] See also Figure 3 In some embodiments of the present application, step 203 of the above embodiment constructs a joint state prediction model of the multi-thruster system based on the joint state transfer matrix, which may include the following steps:

[0115] 301. Determine the current joint state vector, input gain matrix, and joint control input matrix based on parameter information of each thruster;

[0116] Based on the parameter information of each thruster, the server determines the current joint state vector, input gain matrix, and joint control input matrix. The server first collects and integrates the real-time operating parameters of all thrusters, such as speed, thrust feedback, current, and attitude response, and combines these parameters into a joint state vector that describes the state of the entire multi-thruster system. At the same time, based on the characteristics and control configuration of the thrusters, the server calculates the input gain matrix to reflect the impact of the control input on the system state. In addition, based on the current control strategy and control allocation results, the server determines the joint control input matrix, which represents the distribution of the control inputs applied to each thruster in the overall system.

[0117] 302. Construct a joint state prediction model of the multi-thruster system based on the current joint state vector, input gain matrix, joint control input matrix and joint state transfer matrix.

[0118] The server constructs a joint state prediction model for the multi-thruster system using the joint state vector, input gain matrix, joint control input matrix, and the previously constructed joint state transition matrix determined in step 301. The server combines the current joint state with the control input and, by mapping the joint state transition matrix and input gain matrix, predicts the state change of the multi-thruster system at the next moment. This model simultaneously reflects the dynamic changes in each thruster's state and the effects of the control inputs, enabling accurate prediction of the system's overall operating state and providing precise data support for subsequent state residual analysis and fault detection.

[0119] Specifically, the joint state prediction model can be designed as follows:

[0120] Formula 1

[0121] Formula 2

[0122] Formula 3

[0123] Formula 4

[0124] in, A vector representing the predicted next state, represents the current joint state vector, represents the joint control input matrix, represents the input gain matrix, represents the joint state transition matrix.

[0125] Represents the state vector of a single thruster, which includes information in four dimensions: speed, current, thrust feedback, and attitude response. Represents the control input of a single thruster, which can include information in multiple dimensions, such as thrust, lateral force, vertical force, torque, speed, angular velocity, control signal strength, phase information, etc., which is not limited in this application.

[0126] It should be noted that for the joint state transfer matrix The diagonal submatrix of - , the thrusters in the same group use the same sub-matrix, for example, thruster 1 and thruster 2 belong to the same thruster group, then and are equal. For the joint state transfer matrix The coupling terms on the off-diagonal lines of , which is used to represent the mutual influence terms between different thrusters due to the control coupling relationship. These coupling terms are used to reflect the control coupling relationship, that is, the interaction and influence between different thrusters.

[0127] It should be noted that when each thruster is driven only by its own control input, It is expressed as a block diagonal matrix, as shown in the following formula:

[0128] Formula 5

[0129] However, if there are some coupled control channels (e.g., multiple thrusters share a control law), there may also be some non-zero off-diagonal terms. This application does not limit this.

[0130] In this embodiment, the server accurately determines the joint state vector and input correlation matrix, and constructs a joint state prediction model in combination with the joint state transfer matrix. This can comprehensively reflect the dynamic characteristics and control input influence of the multi-thruster system, improve the accuracy and real-time performance of state prediction, and thereby enhance the reliability and response speed of fault diagnosis, thereby promoting the safe and stable operation of the multi-thruster system.

[0131] See also Figure 4 In some embodiments of the present application, step 107 of the above embodiment determines the faulty thruster and the fault type based on the distribution characteristics of the state residual and the control coupling relationship, and may include the following steps:

[0132] 401. Construct a residual matrix based on the state residual;

[0133] The server collects the actual operating state data and the corresponding predicted operating state data for each thruster in the multi-thruster system. By comparing the differences between the two at the same moment, it calculates the state residuals of each thruster's operating parameters. The server categorizes and aggregates these residuals by thruster number and state type (such as speed, current, thrust feedback, and attitude response), forming a multi-dimensional residual matrix. This residual matrix provides a detailed picture of the deviation between each thruster's current state and the predicted state, facilitating subsequent analysis.

[0134] 402. Constructing a coupling coefficient matrix based on the control coupling relationship;

[0135] Based on the multi-thruster system design and control allocation scheme, the server analyzes the control coupling relationships between each thruster in detail and quantifies the mutual influence between different thrusters. By collecting and organizing information on the control input influence paths and coupling strengths between thrusters, the server constructs a coupling coefficient matrix, where each element represents the strength and direction of the control coupling between two corresponding thrusters. This matrix systematically characterizes the mutual coupling effects between thrusters and provides a foundation for understanding the residual propagation mechanism.

[0136] 403. Generate a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix;

[0137] Based on the residual matrix from step 401 and the coupling coefficient matrix from step 402, the server generates a residual coupling influence matrix using mathematical methods such as matrix multiplication or weighted superposition. This matrix reflects how the state residuals of a single thruster affect the states of other thrusters due to the control coupling relationship between the thrusters, thereby revealing the diffusion path of residuals in a multi-thruster system and the scope of their influence. Using this matrix, the server can identify areas where residuals are concentrated and thrusters that may have abnormalities, providing important clues for fault diagnosis.

[0138] 404. Input the residual coupling influence matrix into a pre-trained thruster fault identification model to determine the faulty thruster and the fault type.

[0139] The server inputs the residual coupling influence matrix into a pre-trained thruster fault identification model. This model uses machine learning algorithms (such as support vector machines, neural networks, or decision trees) to deeply analyze the input matrix and identify the corresponding relationships between residual patterns and known fault types. The server uses the model output to identify the specific thruster fault and the fault type, such as thrust anomaly, sensor failure, or control malfunction, providing clear diagnostic results for subsequent troubleshooting.

[0140] In this embodiment, the server constructs a detailed residual matrix and coupling coefficient matrix, and combines the residual coupling influence matrix as input, and uses the fault identification model to deeply mine the residual signal of the control coupling between thrusters, thereby significantly improving the accuracy of fault location and the precision of fault type discrimination, thereby enhancing the fault diagnosis capability and operational safety of the multi-thruster system, and helping to take maintenance measures in a timely manner to ensure stable operation of the system.

[0141] See also Figure 5 In some embodiments of the present application, step 403 of the above embodiment generates a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix, which may include the following steps:

[0142] 501. Calculate a normalized coupling weight matrix using the residual vector amplitude of each row of the residual matrix and the corresponding element of the coupling coefficient matrix;

[0143] The server uses the residual vector amplitudes corresponding to each row in the residual matrix, combined with the corresponding elements in the coupling coefficient matrix, to calculate a normalized coupling weight matrix. Specifically, the server first counts the residual amplitudes of each thruster in the residual matrix to assess their impact on the overall system state. Then, the server weights the residual amplitudes and coupling strengths, combined with the control coupling strengths between thrusters in the coupling coefficient matrix, to form a matrix that reflects the residual transfer weights between different thrusters. The server normalizes this matrix to ensure a reasonable weight distribution and facilitate subsequent calculations.

[0144] 502. Apply a double nonlinear transformation to each residual vector in the residual matrix to obtain a transformation vector;

[0145] The server applies a dual nonlinear transformation to each residual vector in the residual matrix to generate a transformed vector. This dual nonlinear transformation involves applying nonlinear functions to both the magnitude and direction of the residual vector. This enhances key information in the residual signal while suppressing the effects of noise and outliers. This transformation highlights features in the residual vector that may indicate a fault, making the fault information more visible and reliable in the subsequent weighted aggregation process.

[0146] 503. The transformation vectors are weighted and aggregated by the normalized coupling weight matrix to obtain the residual coupling influence matrix.

[0147] The server uses the normalized coupling weight matrix obtained in step 501 to perform weighted aggregation on the transformation vectors obtained in step 502 to generate a residual coupling influence matrix. Specifically, the server assigns different weights to the transformation vectors of different thrusters based on the elements in the coupling weight matrix and performs a linear combination or weighted average of all the transformation vectors according to the weights. The resulting residual coupling influence matrix comprehensively reflects the overall propagation and influence of each thruster's residual under the control coupling relationship, providing accurate input data for subsequent fault identification.

[0148] In a possible implementation, the server can absorb the state residual information of each thruster and construct it into a residual matrix:

[0149] Formula 6

[0150] in, is the state residual vector of the j-th thruster, which includes the prediction residuals in multiple state dimensions such as speed, current, thrust feedback, and attitude response.

[0151] Then, the server can establish a coupling coefficient matrix based on the control coupling information such as the structural layout, thrust direction, and control allocation matrix between the thrusters:

[0152] Formula 7

[0153] in, It represents the influence of the state of thruster j on thruster i, thereby reflecting the coupling weight between the thrusters in control.

[0154] Next, the server can combine the residual matrix R and the coupling coefficient matrix C to calculate the normalized coupling weight matrix:

[0155] Formula 8

[0156] Among them, formula 8 introduces the residual size As a regulating factor, other thrusters with larger residuals and strong coupling to the current thruster have higher weights in the influence matrix, which is helpful to highlight possible fault sources.

[0157] To improve the feature expression capability, the server can convert the residual vector of each thruster into Input the dual-channel nonlinear mapping function and get the transformation vector:

[0158] Formula 9

[0159] in, is the Sigmoid function; is the hyperbolic tangent function; is the weight matrix; is the bias term; Represents the multiplication of corresponding elements (i.e., Hadamard product).

[0160] The server introduces a nonlinear gating mechanism to enhance its response to complex changes in different residual signals, facilitating subsequent processing by the neural network.

[0161] Next, the server uses the normalized coupling weight matrix Perform weighted aggregation on the transformation vectors to calculate the residual coupling influence vector of each thruster . Calculated according to the following formula:

[0162] Formula 10

[0163] Finally, the server can The final residual coupling influence matrix is ​​formed by row combination:

[0164] Formula 11

[0165] The residual coupling influence matrix not only integrates the coupling relationship information and the amplitude change of the residual signal, but also undergoes nonlinear enhancement processing to obtain the final It can be used as input to the downstream thruster fault identification model to effectively improve fault identification accuracy and positioning precision.

[0166] In this embodiment, the server constructs a residual coupling influence matrix that reflects the residual propagation and coupling effects by calculating a normalized coupling weight matrix, performing a dual nonlinear transformation on the residual vector, and then weighting and aggregating the transformed vectors. This matrix effectively integrates the residual information and coupling relationships between thrusters, improving the ability to identify fault signals and the accuracy of overall system fault diagnosis, thereby enhancing the reliability and accuracy of multi-thruster system fault identification.

[0167] See also Figure 6In some embodiments of the present application, step 106 of the above embodiment compares the state residual with a preset threshold to determine whether the multi-thruster system has a fault, which may include the following steps:

[0168] 601. Determine whether each item of the state residual is less than a corresponding preset threshold;

[0169] The server evaluates the residual state of each thruster in the multi-thruster system, comparing each residual value to its corresponding preset threshold. The server then iterates through all residual data in the residual matrix, determining whether each residual is less than a pre-set safety threshold. This allows the server to assess whether the current operating state deviations are within normal limits, ensuring a comprehensive check of each thruster and indicator.

[0170] 602. Determine that there is a fault in the multi-thruster system.

[0171] When the server detects that at least one state residual reaches or exceeds the corresponding preset threshold, it determines that the multi-thruster system is faulty. The server marks this abnormal information as a fault signal, triggering subsequent fault diagnosis and location processes, providing a basis for system maintenance and repair, and ensuring the safe and stable operation of the multi-thruster system.

[0172] 603. Determine that there is no fault in the multi-thruster system.

[0173] When the server confirms that all state residuals are less than the corresponding preset thresholds, it determines that the multi-thruster system is not faulty. The server records that the current system status is within the normal operating range, without initiating the fault diagnosis program, and maintains normal system monitoring and control to ensure the stable operation of the multi-thruster system.

[0174] In this embodiment, the server achieves accurate monitoring of the operating status and fault identification of the multi-thruster system by strictly comparing the state residuals item by item with the preset thresholds. It can promptly detect anomalies and determine the occurrence of faults, or confirm that the system is safe and has no anomalies, thereby improving the reliability and safety of the system operation, reducing potential risks, and ensuring the stable and efficient operation of the multi-thruster system.

[0175] See also Figure 7 , Figure 7 An embodiment of the underwater propeller fault detection device provided in the present application includes:

[0176] An establishing unit 701 is configured to establish a functional similarity relationship and a control coupling relationship between the thrusters according to the thrust capability, thrust direction, and control allocation coefficient of each thruster in the multi-thruster system;

[0177] A construction unit 702 is configured to construct a joint state prediction model of a multi-thruster system based on the functional similarity relationship and the control coupling relationship;

[0178] The first acquisition unit 703 is used to collect the current operating status data and control input data of each thruster, the operating status data including speed, current, thrust feedback and attitude response;

[0179] An input unit 704 is used to input the control input data into the joint state prediction model to calculate the predicted operating state data of each thruster;

[0180] The second acquisition unit 705 is used to collect the actual operating state data of each thruster and calculate the state residual based on the actual operating state data and the predicted operating state data; wherein the actual operating state data is the data at the next moment corresponding to the current operating state data;

[0181] a judgment unit 706, configured to compare the state residual with a preset threshold to determine whether the multi-thruster system has a fault;

[0182] The determination unit 707 is configured to determine the faulty propeller and the fault type according to the distribution characteristics of the state residual and the control coupling relationship if a fault exists.

[0183] In this embodiment, the joint state prediction model comprehensively considers the functional similarity and control coupling relationship between multiple thrusters, realizes the accurate prediction of the overall state of multiple thrusters, avoids the problems of information isolation and error accumulation when predicting a single thruster, and makes the state residual more realistically reflect the abnormal situation, thereby greatly improving the accuracy of fault detection and positioning.

[0184] Optionally, the construction unit 702 includes:

[0185] A first determining module, configured to determine thruster groups based on similarity relationships;

[0186] A second determination module is used to determine a joint state transfer matrix according to the thruster grouping and control coupling relationship;

[0187] The first building module is used to build a joint state prediction model of the multi-thruster system based on the joint state transfer matrix.

[0188] Optionally, the first building block is specifically configured to:

[0189] Determine the current joint state vector, input gain matrix and joint control input matrix according to the parameter information of each thruster;

[0190] A joint state prediction model of the multi-thruster system is constructed based on the current joint state vector, input gain matrix, joint control input matrix and joint state transfer matrix.

[0191] Optionally, the joint state prediction model is specifically designed as follows:

[0192]

[0193]

[0194]

[0195]

[0196] in, A vector representing the predicted next state, represents the current joint state vector, represents the joint control input matrix, represents the input gain matrix, represents the joint state transition matrix.

[0197] Optionally, the determining unit 707 includes:

[0198] A second building module is used to build a residual matrix according to the state residual;

[0199] A third building module is used to build a coupling coefficient matrix based on the control coupling relationship;

[0200] A generation module, used for generating a residual coupling influence matrix according to a residual matrix and a coupling coefficient matrix;

[0201] The input module is used to input the residual coupling influence matrix into the pre-trained thruster fault identification model to determine the faulty thruster and the fault type.

[0202] Optionally, the generation module is specifically used to:

[0203] The normalized coupling weight matrix is ​​calculated using the residual vector amplitude of each row of the residual matrix and the corresponding element of the coupling coefficient matrix;

[0204] Apply a double nonlinear transformation to each residual vector in the residual matrix to obtain a transformation vector;

[0205] The transformation vectors are weighted and aggregated by the normalized coupling weight matrix to obtain the residual coupling influence matrix.

[0206] Optionally, the determining unit 706 is specifically configured to:

[0207] Determine whether each item of the state residual is less than the corresponding preset threshold;

[0208] If not, it is determined that there is a fault in the multi-thruster system; if so, it is determined that there is no fault in the multi-thruster system.

[0209] In this implementation, the functions of each unit and module are the same as those mentioned above. Figures 1 to 6 The steps in the illustrated embodiment correspond to each other and will not be repeated here.

[0210] See also Figure 8 , Figure 8 Another embodiment of the underwater propeller fault detection device provided by the present application includes:

[0211] Processor 801, memory 802, input and output unit 803 and bus 804;

[0212] The processor 801 is connected to the memory 802, the input and output unit 803 and the bus 804;

[0213] The memory 802 stores a program, and the processor 801 calls the program to execute Figures 1 to 6 Steps in the illustrated embodiment.

[0214] In this embodiment, the function of the processor 801 is the same as that of the aforementioned Figures 1 to 6 The steps in the illustrated embodiment correspond to each other and will not be repeated here.

[0215] The embodiment of the present application further provides a computer-readable storage medium having a program stored thereon, which, when executed on a computer, causes the computer to execute the aforementioned Figures 1 to 6 A method in any possible embodiment.

[0216] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0217] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0218] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0219] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0220] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk, and other media that can store program code.

Claims

1. A method for detecting underwater propeller failure, characterized in that: include: According to the thrust capability, thrust direction and control distribution coefficient of each thruster in the multi-thruster system, the functional similarity relationship and control coupling relationship between the thrusters are established; constructing a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship; Collecting current operating status data and control input data of each thruster, wherein the operating status data includes speed, current, thrust feedback and attitude response; Inputting the control input data into the joint state prediction model to calculate predicted operating state data of each thruster; Collecting actual operating state data of each thruster, and calculating a state residual based on the actual operating state data and the predicted operating state data; wherein the actual operating state data is data at the next moment corresponding to the current operating state data; Comparing the state residual with a preset threshold to determine whether the multi-thruster system has a fault; If a fault exists, the faulty thruster and the fault type are determined based on the distribution characteristics of the state residual and the control coupling relationship.

2. The underwater propeller fault detection method according to claim 1, characterized in that: The constructing of a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship includes: determining thruster groups according to the similarity relationship; determining a joint state transfer matrix according to the thruster grouping and the control coupling relationship; A joint state prediction model of the multi-thruster system is constructed based on the joint state transfer matrix.

3. The underwater propeller fault detection method according to claim 2, characterized in that: The constructing of the joint state prediction model of the multi-thruster system based on the joint state transfer matrix includes: Determine the current joint state vector, input gain matrix and joint control input matrix according to the parameter information of each thruster; A joint state prediction model of the multi-thruster system is constructed according to the current joint state vector, the input gain matrix, the joint control input matrix and the joint state transfer matrix.

4. The underwater propeller fault detection method according to claim 3, characterized in that: The joint state prediction model is specifically designed as follows: in, A vector representing the predicted next state, represents the current joint state vector, represents the joint control input matrix, represents the input gain matrix, represents the joint state transfer matrix, represents the state vector of a single thruster, represents the control input for a single thruster, - represents the joint state transition matrix The submatrix on the diagonal of It is used to represent the mutual influence terms between different thrusters due to the control coupling relationship.

5. The underwater propeller fault detection method according to claim 1, characterized in that: The determining of the faulty thruster and the fault type according to the distribution characteristics of the state residual and the control coupling relationship includes: constructing a residual matrix according to the state residual; Constructing a coupling coefficient matrix based on the control coupling relationship; generating a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix; The residual coupling influence matrix is ​​input into a pre-trained thruster fault identification model to determine the faulty thruster and the fault type.

6. The underwater propeller fault detection method according to claim 5, characterized in that: Generating a residual coupling influence matrix according to the residual matrix and the coupling coefficient matrix includes: Calculating a normalized coupling weight matrix using the residual vector amplitudes of each row of the residual matrix and corresponding elements of the coupling coefficient matrix; Applying a double nonlinear transformation to each residual vector in the residual matrix to obtain a transformation vector; The transformation vectors are weighted and aggregated by the normalized coupling weight matrix to obtain a residual coupling influence matrix.

7. The underwater propeller fault detection method according to any one of claims 1 to 6, characterized in that: Comparing the state residual with a preset threshold to determine whether the multi-thruster system has a fault includes: Determine whether each item of the state residual is less than a corresponding preset threshold; If not, it is determined that there is a fault in the multi-thruster system; if so, it is determined that there is no fault in the multi-thruster system.

8. Underwater propeller fault detection device, characterized in that: include: An establishment unit for establishing functional similarity relationships and control coupling relationships between thrusters in a multi-thruster system based on thrust capabilities, thrust directions, and control allocation coefficients of each thruster; a construction unit, configured to construct a joint state prediction model of the multi-thruster system based on the functional similarity relationship and the control coupling relationship; A first acquisition unit is used to collect current operating status data and control input data of each thruster, wherein the operating status data includes speed, current, thrust feedback and attitude response; an input unit, configured to input the control input data into the joint state prediction model to calculate predicted operating state data of each thruster; a second acquisition unit, configured to acquire actual operating state data of each thruster and calculate a state residual based on the actual operating state data and the predicted operating state data; wherein the actual operating state data is data at a next moment corresponding to the current operating state data; a judgment unit, configured to compare the state residual with a preset threshold to determine whether the multi-thruster system has a fault; A determination unit is configured to determine the faulty thruster and the fault type based on the distribution characteristics of the state residual and the control coupling relationship if a fault exists.

9. An underwater propeller fault detection device, characterized in that: include: processor, memory, input and output units, and buses; The processor is connected to the memory, the input and output unit, and the bus; A program is stored in the memory, and the processor calls the program to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, and when the program is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images