Real-time evaluation and autonomous monitoring method and system for service state of underwater equipment

Abstract

The invention relates to a real-time evaluation and autonomous monitoring method and system for the service state of underwater equipment. The method comprises the steps that multiple operation parameters and multiple environment parameters of the underwater equipment are collected; inputting the plurality of operation parameters and the plurality of environment parameters into a state evaluation model as network nodes of a Bayesian network model to obtain a state probability of an equipment state corresponding to each network node; acquiring a real-time equipment state of the underwater equipment according to the state probability; based on a semi-Markov decision, obtaining an optimal detection strategy of the underwater equipment according to the real-time equipment state of the underwater equipment; and according to the influence of the plurality of environmental parameters on the health index of the underwater equipment and a preset current influence threshold value, correcting the detection frequency of the detection strategy to obtain a detection scheme of the underwater equipment, and continuing to collect the plurality of operation parameters and the plurality of environmental parameters of the underwater equipment according to the detection scheme. According to the method, the incidence relation between the detection scheme and the operation parameters and the environment parameters is improved, and safe and stable operation of the underwater equipment is facilitated.

Description

Technical Field

[0001] The present application relates to the technical field of status monitoring of underwater equipment, and in particular to a method and system for real-time evaluation and autonomous monitoring of the service status of underwater equipment. Background Art

[0002] With the rapid development of marine resource development, underwater scientific research and exploration, and marine engineering construction, the application of underwater equipment has become increasingly extensive and in-depth. Underwater equipment undertakes important tasks such as underwater detection, data collection, and operation execution in complex marine environments. The stability and reliability of its operating status are directly related to the success of the mission and the smooth progress of related work.

[0003] However, the marine environment is characterized by high pressure, low temperatures, strong corrosion, complex and variable currents, and dim lighting. These factors can have a number of adverse effects on the normal operation of underwater equipment. For example, high pressure can compromise the seal of the equipment's casing, potentially damaging internal electronic components through moisture. High corrosion can gradually erode the equipment's metal components, reducing its structural strength. The impact and fluctuations of water currents can affect the equipment's positioning and attitude, reducing its operating accuracy.

[0004] Currently, monitoring methods for underwater equipment status face numerous limitations. Traditional manual periodic inspections are not only inefficient but also fail to promptly detect potential failures and abnormal changes in equipment during operation. Simple sensor data collection methods only capture information on a subset of equipment parameters, making it difficult to comprehensively and in-depth assess the overall service status of the equipment. Furthermore, most existing monitoring systems lack the ability to dynamically assess equipment status in real time, making it impossible to adjust monitoring strategies and implement effective countermeasures based on the equipment's actual operating conditions and environmental changes.

[0005] Furthermore, underwater equipment experiences varying state changes and failure modes in different operating phases and mission scenarios. Existing monitoring technologies often fail to fully account for these factors, resulting in inaccurate and unreliable equipment status assessments. Summary of the Invention

[0006] Based on this, the purpose of this application is to provide a method and system for real-time evaluation and autonomous monitoring of the service status of underwater equipment, which can overcome the shortcomings of the existing technology.

[0007] In order to achieve the above objectives, the technical solutions adopted in this application are:

[0008] A method for real-time evaluation and autonomous monitoring of the service status of underwater equipment, comprising:

[0009] Collect multiple operating parameters and environmental parameters of underwater equipment;

[0010] Input the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability;

[0011] Based on semi-Markov decision making, an optimal detection strategy for the underwater equipment is obtained according to the real-time equipment status of the underwater equipment;

[0012] According to the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, the detection frequency of the detection strategy is corrected to obtain the detection plan of the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection plan.

[0013] Compared with traditional technologies, the beneficial effects of this application are:

[0014] The real-time evaluation and autonomous monitoring method of the service status of underwater equipment of the present application obtains the real-time equipment status of the underwater equipment based on multiple operating parameters and multiple environmental parameters of the underwater equipment, and then obtains the optimal detection strategy corresponding to the real-time equipment status through semi-Markov decision making. Then, combined with the influence of multiple environmental parameters on the health indicators of the underwater equipment and the preset current influence threshold, the detection frequency of the detection strategy is corrected to obtain a detection scheme for the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme. The detection scheme can be determined based on the multiple operating parameters and multiple environmental parameters of the underwater equipment, thereby adjusting the detection strategy more comprehensively and timely, improving the correlation between the detection scheme and the multiple operating parameters and multiple environmental parameters, and facilitating the safe and stable operation of the underwater equipment.

[0015] As an embodiment, the step of inputting the multiple operating parameters and the multiple environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of each network node corresponding to a device state includes:

[0016] The state probability is obtained by using the Bayesian formula:

[0017] ;

[0018] in, is the state probability; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state.

[0019] As an embodiment, the step of obtaining the real-time device status of the underwater device according to the status probability includes:

[0020] Obtaining a real-time risk level of the underwater equipment according to the state probability;

[0021] The real-time device status of the underwater device is obtained according to the real-time risk degree and a plurality of preset risk range thresholds; wherein the plurality of risk range thresholds correspond to a plurality of fault states of the underwater device.

[0022] As an embodiment, the step of obtaining the real-time risk of the underwater equipment according to the state probability includes:

[0023] The real-time risk is obtained by the following formula:

[0024] ;

[0025] in, is the real-time risk; M is the total number of network nodes; is the state probability; For the The association weight of each network node to the operation of the underwater equipment.

[0026] As an embodiment, the step of obtaining the optimal detection strategy for the underwater device based on the semi-Markov decision and according to the real-time device status of the underwater device includes:

[0027] Inputting the real-time device state into a trained strategy return expectation prediction model to obtain expected returns of several detection strategies;

[0028] The detection strategy with the highest expected return is determined as the optimal detection strategy for the underwater equipment.

[0029] As an embodiment, the step of inputting the real-time device state into a trained strategy reward expectation prediction model to obtain the expected rewards of several detection strategies includes:

[0030] The expected return of each detection strategy is obtained through the following formula:

[0031] ;

[0032] in, For the expected return; The real-time device status; For time; For detection strategy expectations; for The discount factor at the moment, where ; Device status Adopt a detection strategy Actions included immediate returns.

[0033] As an embodiment, the step of modifying the detection frequency of the detection strategy based on the impact of the multiple environmental parameters on the health index of the underwater equipment and the preset current impact threshold to obtain the detection plan for the underwater equipment includes:

[0034] Optimizing the previous impact threshold using a gradient descent method based on the underwater equipment's historical operating data, real-time equipment status changes, equipment performance indicators, and environmental condition parameters to obtain the current impact threshold;

[0035] Obtaining impact prediction values ​​corresponding to the plurality of environmental parameters based on parameter changes of the plurality of environmental parameters and a trained underwater equipment health indicator impact prediction model;

[0036] Obtaining a frequency correction value according to the impact prediction value and the current impact threshold;

[0037] The detection frequency of the detection strategy is corrected according to the frequency correction value to obtain a detection scheme for the underwater equipment.

[0038] As an implementation manner, the step of optimizing the previous impact threshold using a gradient descent method to obtain the current impact threshold includes:

[0039] The current impact threshold is obtained by the following formula:

[0040] ;

[0041] in, is the current impact threshold; is the previous impact threshold; is the learning rate; The objective function is The gradient at .

[0042] As an implementation method, it further includes:

[0043] The state assessment model is reversely optimized using the multiple operating parameters, multiple environmental parameters and the real-time device state.

[0044] The second embodiment of the present application provides a system for real-time evaluation and autonomous monitoring of the service status of underwater equipment, including:

[0045] A data acquisition module, used to collect multiple operating parameters and multiple environmental parameters of underwater equipment;

[0046] A state assessment module is configured to input the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability;

[0047] A self-service monitoring module, configured to obtain an optimal detection strategy for the underwater equipment based on a semi-Markov decision and according to the real-time equipment status of the underwater equipment;

[0048] A detection scheme acquisition module is used to correct the detection frequency of the detection strategy based on the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, and obtain the detection scheme of the underwater equipment to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme.

[0049] Compared with traditional technologies, the beneficial effects of this application are:

[0050] The real-time evaluation and autonomous monitoring system for the service status of underwater equipment of the present application obtains the real-time equipment status of the underwater equipment based on multiple operating parameters and multiple environmental parameters of the underwater equipment, and then obtains the optimal detection strategy corresponding to the real-time equipment status through semi-Markov decision making. Then, combined with the influence of multiple environmental parameters on the health indicators of the underwater equipment and the preset current influence threshold, the detection frequency of the detection strategy is corrected to obtain a detection scheme for the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme. The detection scheme can be determined based on the multiple operating parameters and multiple environmental parameters of the underwater equipment, thereby adjusting the detection strategy more comprehensively and timely, improving the correlation between the detection scheme and multiple operating parameters and multiple environmental parameters, and facilitating the safe and stable operation of the underwater equipment.

[0051] For better understanding and implementation, the present application is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to one embodiment of the present application;

[0053] Figure 2 This is a flow chart of data collection for a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to one embodiment of the present application;

[0054] Figure 3 A flowchart of a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to an embodiment of the present application for obtaining real-time equipment status;

[0055] Figure 4 This is a flow chart of obtaining real-time status indicators of a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to one embodiment of the present application;

[0056] Figure 5 This is a flow chart of a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to an embodiment of the present application;

[0057] Figure 6 This is a schematic diagram of module connections of a system for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to one embodiment of the present application;

[0058] 101. Data collection module; 102. Status assessment module; 103. Self-service monitoring module; 104. Detection plan acquisition module. DETAILED DESCRIPTION

[0059] In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

[0060] It should be clear that the embodiments described are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the embodiments of the present application.

[0061] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of this application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to the specific circumstances. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates other meanings. The words "if" / "if" used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0062] In this application, unless otherwise specified, "plurality" refers to two or more. "And / or" describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0063] See also Figure 1 , which is a flow chart of a method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to a first embodiment of the present application, the method comprising:

[0064] S1: Collect multiple operating parameters and environmental parameters of underwater equipment.

[0065] See also Figure 2 Among them, multiple operating parameters and multiple environmental parameters of underwater equipment can be collected in real time and comprehensively with the help of multiple sensors installed in key parts of underwater equipment (such as power system, control system, casing, etc.) and surrounding environment (such as different depths and different directions of the water area where the equipment is located), including pressure sensors, temperature sensors, current sensors, acceleration sensors, water quality sensors, etc., for physical parameters (such as pressure, temperature, vibration, displacement, etc.), electrical parameters (such as current, voltage, power, etc.) and environmental parameters (such as water flow rate, water temperature, salinity, pH, etc.) during the operation of the equipment. The collected data will be immediately transmitted to the pre-processing unit for pre-processing operations such as denoising (using filtering algorithms to remove noise interference, such as median filtering, Gaussian filtering, etc.) and normalization (uniformly mapping parameter values ​​in different ranges to specific intervals, such as [0, 1]) to ensure the accuracy and consistency of the data and provide a high-quality data foundation for the analysis and processing of subsequent modules.

[0066] S2: Input the multiple operating parameters and multiple environmental parameters as network nodes of the Bayesian network model into the state assessment model to obtain the state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability.

[0067] The plurality of operating parameters and the plurality of environmental parameters are used as network nodes, and a Bayesian network model can be constructed based on the parameter relationship between the plurality of operating parameters and the plurality of environmental parameters.

[0068] S3: Based on a semi-Markov decision, an optimal detection strategy for the underwater device is obtained according to the real-time device status of the underwater device;

[0069] S4: Based on the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, the detection frequency of the detection strategy is modified to obtain a detection plan for the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection plan.

[0070] Compared with traditional technologies, the beneficial effects of this application are:

[0071] The real-time evaluation and autonomous monitoring method of the service status of underwater equipment of the present application obtains the real-time equipment status of the underwater equipment based on multiple operating parameters and multiple environmental parameters of the underwater equipment, and then obtains the optimal detection strategy corresponding to the real-time equipment status through semi-Markov decision making. Then, combined with the influence of multiple environmental parameters on the health indicators of the underwater equipment and the preset current influence threshold, the detection frequency of the detection strategy is corrected to obtain a detection scheme for the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme. The detection scheme can be determined based on the multiple operating parameters and multiple environmental parameters of the underwater equipment, thereby adjusting the detection strategy more comprehensively and timely, improving the correlation between the detection scheme and the multiple operating parameters and multiple environmental parameters, and facilitating the safe and stable operation of the underwater equipment.

[0072] In a feasible embodiment, the step S2: inputting the multiple operating parameters and multiple environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of each network node corresponding to the device state includes:

[0073] The state probability is obtained by using the Bayesian formula:

[0074] ;

[0075] in, is the state probability; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state.

[0076] See also Figure 3In a feasible embodiment, the step S2: obtaining the real-time device status of the underwater device according to the state probability includes:

[0077] S21: Obtaining a real-time risk level of the underwater equipment according to the state probability;

[0078] S22: Acquire the real-time device status of the underwater device according to the real-time risk degree and a plurality of preset risk range thresholds; wherein the plurality of risk range thresholds correspond to a plurality of fault states of the underwater device.

[0079] After obtaining the real-time status of underwater equipment, state transition trends can be visualized. For example, advanced data visualization tools and algorithms can be used to present the changing trends of equipment status over time in intuitive and easy-to-understand charts (such as line charts showing the changes in key parameters over time, and state transition diagrams showing the transition relationships between different equipment states). This visual interface allows operators to clearly understand the evolution of equipment status, identify potential state change trends in advance, and take timely measures to prevent failures.

[0080] In a feasible embodiment, the step of S21: obtaining the real-time risk of the underwater equipment according to the state probability includes:

[0081] The real-time risk is obtained by the following formula:

[0082] ;

[0083] in, is the real-time risk; M is the total number of network nodes; is the state probability; For the The association weight of each network node to the operation of the underwater equipment.

[0084] It can be obtained through the parameters and states in historical data (including normal operation data and fault case data). It is a weight value used to objectively and scientifically quantify the importance of each parameter to state prediction. It can be obtained by the following formula:

[0085] ;

[0086] in, For the The impact factor of each network node; is the total number of network nodes; For the In one embodiment, Determined based on the information gain method, the calculation formula is:

[0087] ;

[0088] in, For parameters Target device status Information gain; Target device status The entropy of For the given parameters Conditional entropy when . Parameters Refers to the state variable of network node j , which can also be understood as the parameter of network node j , which is represented by data variables collected by the equipment, such as temperature, because the above formula is to quantify the importance of each parameter to state prediction.

[0089] See also Figure 4 In a feasible embodiment, other real-time status indicators other than the real-time risk can also be obtained based on the status probability, wherein the real-time status indicator is an indicator parameter corresponding to the real-time device status, and other real-time status indicators include but are not limited to transfer rate and reliability.

[0090] The transfer rate is used to measure the speed of device state changes and can be obtained by the following formula:

[0091] ;

[0092] in, is the transfer rate; for Probability distribution of device status at each moment; for Probability distribution of device status at each moment; is the time interval.

[0093] Reliability is used to express the probability that the equipment is in normal operation, which can be obtained by the following formula:

[0094] ;

[0095] in, For reliability.

[0096] In a feasible embodiment, the step of obtaining the optimal detection strategy for the underwater equipment based on the semi-Markov decision and the real-time equipment status of the underwater equipment includes:

[0097] S31: Inputting the real-time device status into the trained strategy return expectation prediction model to obtain the expected returns of several detection strategies;

[0098] S32: Determine the detection strategy with the highest expected return as the optimal detection strategy for the underwater equipment.

[0099] By calculating the expected returns of different inspection strategies and comparing them, the optimal inspection strategy is selected, which in turn determines the inspection time node, specific inspection content (such as which key parts to inspect and which parameters to inspect), and inspection methods (such as non-destructive testing, performance testing, etc.). At the same time, the inspection process can be displayed visually on the operation interface, including but not limited to real-time display of inspection progress (such as the number of completed inspection steps and the proportion of the total inspection tasks), inspection results (such as the status assessment of each equipment component, the measured values ​​of the inspection parameters and whether they meet the standards, etc.), and other information, allowing operators to fully control the inspection process.

[0100] In a feasible embodiment, the step S31 of inputting the real-time device state into a trained strategy reward expectation prediction model to obtain expected rewards of several detection strategies includes:

[0101] The expected return of each detection strategy is obtained through the following formula:

[0102] ;

[0103] in, For the expected return; The real-time device status; For time; For detection strategy expectations; for The discount factor at the moment, where ; Device status Adopt a detection strategy Actions included immediate returns.

[0104] Among them, action For detection strategy Related actions and operations, such as determining the detection time node, detecting key parts, selecting detection parameters, and adopting non-destructive testing or performance testing.

[0105] In a feasible embodiment, the step S4: modifying the detection frequency of the detection strategy based on the impact of the multiple environmental parameters on the health index of the underwater equipment and a preset current impact threshold to obtain a detection plan for the underwater equipment includes:

[0106] S41: Optimizing the previous impact threshold using a gradient descent method based on the underwater equipment's historical operating data, real-time equipment status changes, equipment performance indicators, and environmental condition parameters to obtain the current impact threshold;

[0107] S42: Obtaining impact prediction values ​​corresponding to the plurality of environmental parameters based on parameter changes of the plurality of environmental parameters and the trained underwater equipment health index impact prediction model;

[0108] See also Figure 5 , where if the parameter change of the environmental parameter exceeds the corresponding preset normal fluctuation range, and the amplitude and rate of change reach a certain standard, the environmental parameter is determined to be a dynamic shock factor. For example, if the water flow velocity changes by more than 15% in a short period of time, or the water pressure fluctuation exceeds ±500 Pa, etc., the corresponding environmental parameters are all dynamic shock factors. Once the dynamic shock factor is detected, it is immediately analyzed and evaluated through the trained underwater equipment health index impact prediction model. Using a mathematical model constructed based on equipment design parameters, historical operating data and actual working conditions, the type (such as water flow velocity change, water pressure anomaly, etc.) and intensity (such as water flow velocity change, water pressure fluctuation value) of the dynamic shock factor are input, and the underwater equipment health index impact prediction model is used to calculate its potential impact on the equipment health index (such as power system stability, shell sealing, etc.), that is, the impact prediction value. .

[0109] S43: Obtaining a frequency correction value according to the impact prediction value and the current impact threshold;

[0110] Current impact threshold Judgment and response is to compare the impact of dynamic shock factors on equipment health indicators with the perception threshold determined based on multiple factors of the equipment. , initiating countermeasures. Building on the existing detection strategy, the detection frequency is increased proportionally based on the severity of the dynamic shock factor. For example, for mild impacts, the detection frequency is increased by 20%; for moderate impacts, the detection frequency is doubled; and for severe impacts, real-time monitoring is implemented. Simultaneously, equipment operating parameters are adjusted (such as reducing operating power and switching operating modes), and early warning signals are issued to alert operators to potential equipment risks.

[0111] S44: Correcting the detection frequency of the detection strategy according to the frequency correction value to obtain a detection plan for the underwater equipment.

[0112] In a feasible embodiment, the step of S41: optimizing the previous impact threshold using a gradient descent method to obtain the current impact threshold, includes:

[0113] The current impact threshold is obtained by the following formula:

[0114] ;

[0115] in, is the current impact threshold; is the previous impact threshold; is the learning rate; The objective function is The gradient at .

[0116] Threshold optimization dynamically optimizes the perception threshold using adaptive algorithms (such as gradient descent) based on the device's long-term operating history, actual operating conditions, device performance indicators, and environmental parameters. Device performance indicators are captured through built-in sensors and monitoring modules, covering areas such as power system output power stability, sensor measurement accuracy, and communication module signal strength. Environmental parameters are collected by external environmental monitoring equipment, including temperature, humidity, water pressure, and water flow rate.

[0117] In a feasible embodiment, it also includes:

[0118] S5: Reversely optimize the state assessment model using the multiple operating parameters, multiple environmental parameters and the real-time device state.

[0119] Compared with the prior art, the present invention has the following beneficial effects:

[0120] (1) Through multi-source data collection and in-depth analysis using a dynamic Bayesian network, this invention comprehensively covers the operating parameters and environmental factors of underwater equipment and accurately assesses the service status of the equipment. Compared with traditional monitoring methods, this method can not only accurately identify the current fault status of the equipment, but also predict the occurrence of potential faults in advance. The accuracy and reliability of the assessment are greatly improved, providing a scientific and reliable basis for equipment maintenance and management.

[0121] (2) Based on a semi-Markov decision process and a dynamic threshold optimization mechanism, the system can intelligently and autonomously adjust its monitoring strategy based on the real-time status of the equipment and environmental changes. Without frequent human intervention, it can respond to changes in equipment status in a timely manner, achieving intelligent monitoring. This greatly improves the efficiency and timeliness of monitoring, enabling immediate response measures when equipment anomalies occur, effectively preventing the escalation of faults and ensuring stable operation of the equipment.

[0122] (3) The back-propagation association module of the present invention enables the two core modules of state assessment and autonomous monitoring to form a close closed-loop collaborative relationship. Through continuous data interaction and model optimization, the system can better adapt to the complex and changing underwater environment and continuously improve the accuracy and effectiveness of assessment and monitoring. This collaborative optimization mechanism not only improves system performance, but also reduces equipment maintenance costs, reduces downtime and economic losses caused by equipment failures, has significant economic and social benefits, and provides a strong guarantee for the safe operation of underwater equipment.

[0123] See also Figure 6 The second embodiment of the present application provides a system for real-time evaluation and autonomous monitoring of the service status of underwater equipment, including:

[0124] The data acquisition module 101 is used to collect multiple operating parameters and multiple environmental parameters of the underwater equipment;

[0125] The state assessment module 102 is configured to input the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability;

[0126] The self-service monitoring module 103 is configured to obtain an optimal detection strategy for the underwater equipment based on a semi-Markov decision and according to the real-time equipment status of the underwater equipment;

[0127] The detection scheme acquisition module 104 is used to correct the detection frequency of the detection strategy according to the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, and obtain the detection scheme of the underwater equipment to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme.

[0128] In a feasible embodiment, it also includes:

[0129] A reverse optimization module is used to reversely optimize the state evaluation model using the multiple operating parameters, multiple environmental parameters and the real-time device state.

[0130] The real-time status indicators of the present application have an impact on the self-service monitoring module 103, the detection solution acquisition module 104, and the reverse optimization module. For example:

[0131] 1. For the self-service monitoring module 103, the higher the real-time risk, the greater the probability of equipment failure. The self-service monitoring module will give priority to high-frequency and comprehensive detection strategies. For example, when the real-time risk exceeds the preset threshold, the system may trigger real-time monitoring or emergency maintenance instead of regular periodic detection. Reflects the speed of state change. If the device status deteriorates rapidly within a short period of time, the system will recalculate the expected return of the detection strategy using a semi-Markov decision model and dynamically adjust the detection interval. For example, the original weekly detection schedule can be changed to daily detection to capture sudden changes in status.

[0132] When reliability is low, the inspection strategy will focus on in-depth inspections of key components (such as powertrains and sealing modules) rather than routine parameter collection. For example, the frequency of nondestructive testing or performance testing will be increased to ensure that high-risk components are closely monitored.

[0133] 2. For the detection scheme acquisition module 104, the detection scheme acquisition module 104 will optimize the current impact threshold by gradient descent method based on the risk R and the impact of environmental parameters on the predicted value (such as the impact of sudden changes in water flow speed on the equipment). If the risk level corresponding to R is superimposed on the predicted value of the impact of environmental parameters, it exceeds , the detection frequency will be increased proportionally (for example, the original detection cycle was 24 hours, which will be shortened to 8 hours after correction).

[0134] Transfer rate triggers an emergency detection mechanism: When the transfer rate V exceeds a preset threshold (e.g., a state probability distribution change of more than 20% within an hour), the system immediately generates an emergency detection plan without waiting for the regular detection cycle. For example, a sudden increase in equipment vibration parameters triggers real-time vibration monitoring of components such as bearings and motors.

[0135] Reliability-assisted adjustments to inspection methods: When reliability falls below a safety threshold, the inspection plan will incorporate more sophisticated detection methods (such as sonar imaging and ultrasonic flaw detection) rather than relying on conventional sensor data. For example, when the reliability of the housing seal decreases, the frequency of water pressure testing and leak detection will be increased.

[0136] 3. For the reverse optimization module, risk R, transfer rate V and reliability It will be reversely input into the state assessment model (such as Bayesian network) together with the original operating parameters and environmental parameters to adjust the weights of network nodes and the prior distribution of state probabilities. For example, if the pre-fault transfer rate V is not effectively captured, the model will increase the weights of parameters such as vibration and displacement to increase the sensitivity to sudden changes in state in the future.

[0137] Closed-loop optimization improves assessment accuracy: Reverse optimization enables the model to adapt to factors such as equipment aging and environmental changes. For example, after long-term operation in a high-salinity environment, the reliability calculation automatically adjusts the weighting of corrosion-related parameters (such as shell potential and coating thickness), making the condition assessment more accurate for actual operating conditions.

[0138] Real-time status indicators bridge the gap between "condition assessment" and "autonomous monitoring": risk metric quantifies overall failure risk, driving the priority and urgency of detection strategies; transfer rate captures the speed of state changes, optimizing detection timelines and emergency response; and reliability measures equipment availability, guiding the depth of detection content and method selection. These three metrics continuously iterate the model through a reverse optimization mechanism, forming a closed loop of "data collection - condition assessment - strategy formulation - solution revision - model optimization." Ultimately, this achieves dynamic matching of detection plans with equipment status and environmental parameters, ensuring the safe operation of underwater equipment.

[0139] The device embodiments described above are merely illustrative, wherein the components described as separate parts 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 the modules may be selected according to actual needs to achieve the purpose of the present application. Those of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0140] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0141] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the function selected in the process. Figure 1 a process or multiple processes and / or boxes Figure 1function selected in a box or multiple boxes.

[0142] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 steps for the function selected in a box or multiple boxes.

[0143] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0144] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0145] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0146] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0147] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for real-time evaluation and autonomous monitoring of the service status of underwater equipment, characterized in that: include: Collect multiple operating parameters and environmental parameters of underwater equipment; Input the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability; Based on semi-Markov decision making, an optimal detection strategy for the underwater equipment is obtained according to the real-time equipment status of the underwater equipment; According to the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, the detection frequency of the detection strategy is corrected to obtain the detection plan of the underwater equipment, so as to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection plan.

2. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 1, characterized in that: The step of inputting the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of a device state corresponding to each network node includes: The state probability is obtained by using the Bayesian formula: ； in, is the state probability; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state; The underwater equipment evaluated by the state evaluation model is in equipment state Corresponding network node probability; The underwater equipment evaluated by the state assessment model is in the The probability of a device state.

3. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 1, characterized in that: The step of obtaining the real-time device status of the underwater device according to the status probability includes: Obtaining a real-time risk level of the underwater equipment according to the state probability; The real-time device status of the underwater device is obtained according to the real-time risk degree and a plurality of preset risk range thresholds; wherein the plurality of risk range thresholds correspond to a plurality of fault states of the underwater device.

4. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 3, characterized in that: The step of obtaining the real-time risk of the underwater equipment according to the state probability includes: The real-time risk is obtained by the following formula: ； in, is the real-time risk; M is the total number of network nodes; is the state probability; For the The association weight of each network node to the operation of the underwater equipment.

5. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 1, characterized in that: The step of obtaining the optimal detection strategy for the underwater equipment based on the semi-Markov decision and according to the real-time equipment status of the underwater equipment includes: Inputting the real-time device state into a trained strategy return expectation prediction model to obtain expected returns of several detection strategies; The detection strategy with the highest expected return is determined as the optimal detection strategy for the underwater equipment.

6. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 5, characterized in that: The step of inputting the real-time device state into the trained strategy return expectation prediction model to obtain the expected returns of several detection strategies includes: The expected return of each detection strategy is obtained through the following formula: ； in, For the expected return; The real-time device status; For time; For detection strategy expectations; for The discount factor at the moment, where ; Device status Adopt a detection strategy Actions included immediate returns.

7. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 1, characterized in that: The step of modifying the detection frequency of the detection strategy according to the impact of the multiple environmental parameters on the health index of the underwater equipment and the preset current impact threshold to obtain the detection plan for the underwater equipment includes: Optimizing the previous impact threshold using a gradient descent method based on the underwater equipment's historical operating data, real-time equipment status changes, equipment performance indicators, and environmental condition parameters to obtain the current impact threshold; Obtaining impact prediction values ​​corresponding to the plurality of environmental parameters based on parameter changes of the plurality of environmental parameters and a trained underwater equipment health indicator impact prediction model; Obtaining a frequency correction value according to the impact prediction value and the current impact threshold; The detection frequency of the detection strategy is corrected according to the frequency correction value to obtain a detection scheme for the underwater equipment.

8. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 7, characterized in that: The step of optimizing the previous impact threshold by using the gradient descent method to obtain the current impact threshold includes: The current impact threshold is obtained by the following formula: ； in, is the current impact threshold; is the previous impact threshold; is the learning rate; The objective function is The gradient at .

9. The method for real-time evaluation and autonomous monitoring of the service status of underwater equipment according to claim 1, characterized in that: Also includes: The state assessment model is reversely optimized using the multiple operating parameters, multiple environmental parameters and the real-time device state.

10. A real-time evaluation and autonomous monitoring system for the service status of underwater equipment, characterized in that: include: A data acquisition module, used to collect multiple operating parameters and multiple environmental parameters of underwater equipment; A state assessment module is configured to input the plurality of operating parameters and the plurality of environmental parameters as network nodes of a Bayesian network model into a state assessment model to obtain a state probability of the device state corresponding to each network node; and obtain the real-time device state of the underwater device based on the state probability; A self-service monitoring module, configured to obtain an optimal detection strategy for the underwater equipment based on a semi-Markov decision and according to the real-time equipment status of the underwater equipment; A detection scheme acquisition module is used to correct the detection frequency of the detection strategy based on the impact of the multiple environmental parameters on the health indicators of the underwater equipment and the preset current impact threshold, and obtain the detection scheme of the underwater equipment to continue to collect multiple operating parameters and multiple environmental parameters of the underwater equipment according to the detection scheme.

Images