Marine engine gas system health assessment method and system based on digital twinning

Abstract

The invention provides a ship engine gas system health assessment method and system based on digital twinning. The method comprises the steps that a gas system hierarchical structure tree is acquired; through a subjective and objective weight fusion method, carrying out importance degree evaluation on each component of the gas system; obtaining a health degree data set of components in the gas system; according to the health degree data set of the components in the gas system, combining a crown porcupine optimization algorithm to optimize a variational mode decomposition method and a sparrow algorithm to optimize a long short-term memory neural network, and predicting and obtaining the health degree of the gas system; the gas system is monitored in real time based on digital twinning, and the current health degree is displayed in real time and subsequently predicted. According to the method, the importance degree of the components of the gas system is accurately evaluated through subjective and objective weight fusion, real-time monitoring and health degree visualization are realized based on digital twinning, the health degree prediction precision is improved by adopting a double-optimization mechanism of optimization of a crown porcupine optimization algorithm and optimization of a sparrow algorithm, and efficient prediction of single input and single output is supported.

Description

Technical Field

[0001] The present application relates to the technical field of marine engines, and particularly to a method and system for health assessment of a marine engine gas system based on digital twin. Background Art

[0002] As the heart of marine ship transportation, the performance and health status of marine engines directly affect the operation efficiency, safety and economy of ships. With the acceleration of global trade and the continuous growth of marine transportation volume, the fault situations faced by marine engines have increased significantly, thus posing higher requirements for the health assessment of marine engines. The health assessment technology of marine engines mainly relies on the collection of mechanical equipment data and analysis and judgment through the collected data. These technologies include methods such as vibration analysis, acoustic emission monitoring, oil analysis, and thermal imaging. Although there are currently a variety of detection means and health assessment methods widely used, there are still some problems with these technologies. First, the working environment of marine engines is complex and affected by various factors such as temperature, humidity, and salt spray, which often interferes with the equipment monitoring data and affects the accuracy of the data. Second, the structure of marine engines is complex, and maintenance and repair are difficult, and traditional monitoring means are difficult to comprehensively cover all key components.

[0003] To ensure the safe navigation of ships and the normal operation of ship engine equipment, it is necessary to realize the health assessment of marine engines. With the improvement of shipping demand and the rapid development of ship engines, the structure of ship engines has become more complex, and the corresponding health assessment methods for ship equipment are more and more complex. In order to ensure the operating state of the engine and obtain more operating state information, it is necessary to collect data on vibration, temperature, pressure, speed, etc. through sensors. The health status of the engine system and components is determined by the abnormal signal data collected by the sensors.

[0004] In response to these problems, to ensure the normal and stable operation of marine engines. In recent years, there have also been many studies on health assessment methods for engines or ship equipment.

[0005] Naval University of Engineering, Chinese People's Liberation Army (Hu Shifeng; Yu Xiang; Feng Shaowei; Zhang Jing. Ship Status Assessment System and Its Construction Method and Storage Medium [P]: CN 118723013 A) designed a ship status assessment system and its construction method. By constructing the logical relationships between ship equipment, the logical relationships between ship equipment and hierarchical systems, and the logical relationships between hierarchical systems, a functional topology network of the whole ship is formed. The system includes an equipment construction module, an equipment association module, and a status assessment module, which are used to construct multiple ship equipment and their hierarchical systems, set the association relationships and weights between equipment and hierarchical systems, and calculate the overall status score of the ship according to these relationships and weights, so as to more accurately assess the operating state of the ship.

[0006] This method can detect system anomalies in a timely manner and trigger alarms, improving the reliability and maintenance efficiency of the system. However, it does not evaluate the importance of the system, determine the processing priorities for system anomaly components, or accurately predict the system health, increasing the complexity and time cost of system maintenance and diagnosis. Summary of the Invention

[0007] The present application provides a method and system for health assessment of a marine engine gas system based on digital twin, which can solve the key technical bottlenecks existing in the application of existing marine engine health assessment technologies in dual-fuel ship power equipment, such as poor adaptability to multi-fuel modes, real-time response delay, high misjudgment rate of cross-faults, low system integration, and insufficient prediction ability, resulting in the technical problem that timely and accurate health status assessment and fault warning cannot be achieved.

[0008] In a first aspect, the present application provides a method for health assessment of a marine engine gas system based on digital twin, including the following steps:

[0009] Perform functional hierarchical decomposition on the entire marine engine to obtain a hierarchical structure tree of the gas system;

[0010] Based on the hierarchical structure tree of the gas system, through a subjective weight and objective weight fusion allocation method, evaluate the importance of each component of the gas system to obtain the importance ranking of the components in the gas system;

[0011] Collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain a health dataset of the components in the gas system based on the evaluation indicator values collected for each component;

[0012] According to the health dataset of the components in the gas system, combine the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm to predict and obtain the health of the gas system;

[0013] Based on digital twin technology, monitor the gas system in real time, display the current health of the gas system in real time, and predict the subsequent health.

[0014] Optionally, the step of evaluating the importance of each component of the gas system based on the hierarchical structure tree of the gas system through a subjective weight and objective weight fusion allocation method to obtain the importance ranking of the components in the gas system specifically includes the following steps:

[0015] Construct a health assessment index system for the components in the marine engine gas system to obtain the health assessment criteria for the components in the gas system;

[0016] Collect data on the evaluation indicators in the health assessment criteria to obtain the evaluation indicator value matrix of each component;

[0017] Standardize the evaluation indicator values of each component collected to generate a standardized matrix of the evaluation indicator values of the components;

[0018] According to the standardized matrix of the evaluation indicator values of all components in the gas system, use the analytic hierarchy process and the entropy weight method to calculate in parallel to obtain the comprehensive weight of each evaluation indicator in the gas system;

[0019] According to the comprehensive weight of each evaluation indicator, obtain the importance ranking of each component in the gas system.

[0020] Optionally, the standardized evaluation indicator values of each component collected to generate a standardized matrix of the evaluation indicator values of the components in the gas system specifically include the following steps:

[0021] Classify the evaluation indicator data of the important components collected into cost type and benefit type to obtain the standardized component evaluation indicator values;

[0022] According to the obtained standardized component evaluation indicator values, generate a standardized matrix of the evaluation indicator values of the components in the gas system.

[0023] Optionally, the step of using the analytic hierarchy process and the entropy weight method to calculate in parallel according to the standardized matrix of the evaluation indicator values of all components in the gas system to obtain the comprehensive weight of each evaluation indicator in the gas system specifically includes the following steps:

[0024] Use the analytic hierarchy process to calculate the standardized matrix of the evaluation indicator values constructed to obtain the subjective weight of each evaluation indicator in the dual-fuel gas system;

[0025] Use the entropy weight method to calculate and obtain the objective weights of the engine sensor data and the historical operation records, and obtain the objective weights of each evaluation indicator in the dual-fuel gas system;

[0026] Linearly combine the subjective weight and the objective weight, and comprehensively evaluate to obtain the comprehensive weight of each evaluation indicator in the gas system.

[0027] Optionally, the step of obtaining the importance ranking of each component in the gas system according to the comprehensive weight of each evaluation indicator specifically includes the following steps:

[0028] According to the comprehensive weight of each evaluation indicator, use the method of distance between superior and inferior to calculate through the weighted formula to obtain the weighted standardized matrix of each evaluation indicator of each component in the gas system;

[0029] Obtain the positive and negative ideal solutions according to the weighted standardized matrix;

[0030] Calculate and obtain the Euclidean distance according to the positive and negative ideal solutions;

[0031] Calculate the closeness of each component in the gas system according to the Euclidean distance;

[0032] Obtain the importance ranking of each component in the gas system according to the closeness.

[0033] Optionally, collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain the health dataset of the components in the gas system according to the evaluation indicator values of each component obtained by the collection, specifically including the following steps:

[0034] Determine the priority processing level of the gas system components through the importance ranking in the marine engine gas system;

[0035] Collect the real-time operation data of the components at each priority processing level;

[0036] Obtain the health of each component through the historical operation data and real-time operation data of the component;

[0037] Obtain the health dataset of the components in the gas system according to the health of each component.

[0038] Optionally, according to the health dataset of the components in the gas system, combined with the Crown Porcupine Optimization Algorithm (C res t e dPorcupine Optimizer, CPO) to optimize the variational mode decomposition method and the Sparrow Algorithm to optimize the long short-term memory neural network, predict and obtain the health of the gas system, specifically including the following steps:

[0039] Determine the optimal decomposition mode number and penalty factor of the variational mode decomposition through the Crown Porcupine Optimization Algorithm;

[0040] Simulate the foraging and anti-predation behaviors of the sparrow group through the Sparrow Search Algorithm to optimize the hyperparameters of the long short-term memory neural network;

[0041] Taking the health dataset as the input and using the root mean square error, mean absolute error, and median of the absolute percentage error as the evaluation error evaluation indicators, predict the health of the dual-fuel gas system.

[0042] Optionally, after predicting and obtaining the health of the gas system by combining the Crown Porcupine Optimization Algorithm to optimize the variational mode decomposition method and the Sparrow Algorithm to optimize the long short-term memory neural network according to the health dataset of the components in the gas system, the following steps are further included:

[0043] Build and implement real-time monitoring of the gas system based on digital twin technology, display the current health of the gas system in real time, and predict the subsequent health.

[0044] Optionally, based on digital twin technology, the gas system is monitored in real time, the current health status of the gas system is displayed in real time, and the subsequent health status is predicted. The specific steps are as follows:

[0045] Construct a three-dimensional model of a marine engine with geometric properties;

[0046] Construct a digital twin model of the marine engine through the three-dimensional model of the marine engine and sensor data;

[0047] Through the UI interface design, the operating status of the digital twin model of the marine engine and the operating data of the marine engine are displayed in real time, and the subsequent health status is predicted.

[0048] In a second aspect, the present application provides a health assessment system for a marine engine gas system based on digital twins, including:

[0049] A functional hierarchical decomposition module for functionally hierarchically decomposing the entire marine engine to obtain a hierarchical structure tree of the gas system;

[0050] An importance evaluation module, communicatively connected to the functional hierarchical decomposition module, for evaluating the importance of each component of the gas system based on the hierarchical structure tree of the gas system through a method of fusing subjective weights and objective weights, and obtaining the importance ranking of the components in the gas system;

[0051] A component health status dataset acquisition module, communicatively connected to the importance evaluation module, for collecting data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtaining a health status dataset of the components in the gas system based on the obtained evaluation indicator values of each component;

[0052] A system health status evaluation module, communicatively connected to the component health status dataset acquisition module, for predicting and obtaining the health status of the gas system based on the health status dataset of the components in the gas system, in combination with the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm;

[0053] A real-time monitoring and warning module, communicatively connected to the system health status evaluation module, for real-time monitoring of the gas system based on digital twin technology, for real-time displaying the current health status of the gas system and predicting the subsequent health status.

[0054] The beneficial effects brought by the technical solutions provided in the embodiments of the present application at least include:

[0055] By fusing subjective expert weights and objective weights, the present application can more accurately judge the importance of the components of the gas system;

[0056] The real-time monitoring of the marine engine gas system can be carried out by means of digital twin, and the health status and health prediction of the gas system are displayed in real time to ensure the long-term reliable operation of the marine engine;

[0057] This application uses the variational mode decomposition method optimized by the crown porcupine optimization algorithm combined with the sparrow algorithm to optimize the long short-term memory neural network for health prediction, realizing the prediction of single input and single output of health data. By adopting a double optimization mechanism, the prediction accuracy is improved. Brief Description of the Drawings

[0058] Figure 1 It is a schematic flow chart of the health assessment method for the marine engine gas system based on digital twin provided by this application;

[0059] Figure 2 It is a structural diagram of the health assessment system for the engine gas system based on digital twin provided by this application;

[0060] Figure 3 It is a structural diagram of the engine gas system provided by this application;

[0061] Figure 4 It is a framework diagram of the evaluation standard model for the engine gas system components provided by this application;

[0062] Figure 5 It is a flow chart for predicting the health of the engine gas system components provided by this application;

[0063] Figure 6 It is a schematic diagram of the digital twin model framework of the engine gas system provided by this application;

[0064] Figure 7 It is a block diagram of the functional modules of the health assessment method for the marine engine gas system based on digital twin provided by this application. Detailed Description of the Embodiments

[0065] In order to enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0066] In the description of the specification and claims of this application and the above-mentioned drawings, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products, or devices. Descriptions such as "first", "second", and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit that "first", "second", and "third" are different types.

[0067] In the description of the embodiments of this application, "exemplary", "for example", or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary", "for example", or "for instance" in the embodiments of this application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary", "for example", or "for instance" is intended to present relevant concepts in a specific manner.

[0068] In the description of the embodiments of this application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B; "and / or" in the text is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "a plurality of" means two or more than two.

[0069] In some processes described in the embodiments of this application, a plurality of operations or steps appear in a specific order. However, it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of this application or may be executed in parallel. The serial numbers of the operations are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in order or in parallel, and these operations or steps may be combined.

[0070] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below in conjunction with the drawings.

[0071] In a first aspect, as Figure 1 and Figure 2 shown, this application provides a method for health assessment of a marine engine gas system based on digital twin, including the following steps:

[0072] Step S1: Perform a functional hierarchical decomposition on the entire marine engine to obtain a hierarchical structure tree of the gas system;

[0073] Step S2: Based on the hierarchical structure tree of the gas system, through the fusion and distribution method of subjective weight and objective weight, evaluate the importance of each component of the gas system, and obtain the importance ranking of the components in the gas system;

[0074] Step S3: Collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain the health dataset of the components in the gas system based on the collected evaluation indicator values of each component;

[0075] Step S4: According to the health dataset of the components in the gas system, combine the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network (LSTM) optimized by the sparrow algorithm to predict and obtain the health of the gas system;

[0076] Step S5: Based on digital twin technology, monitor the gas system in real time, display the current health of the gas system in real time, and predict the subsequent health.

[0077] In this application, by fusing subjective expert weights and objective weights, the importance of each component of the gas system can be judged more accurately;

[0078] Through the digital twin method, the gas system of the marine engine can be monitored in real time, the health of the gas system and the health prediction can be displayed in real time, ensuring the long-term reliable operation of the marine engine;

[0079] By using the variational mode decomposition (VMD) method optimized by the crown porcupine optimization algorithm combined with the sparrow algorithm to optimize the long short-term memory neural network to predict the health of the gas system, the prediction of single input and single output of health data is realized. By adopting a double optimization mechanism, the prediction accuracy is improved.

[0080] In this application, the marine engine is more specifically a marine dual-fuel engine, where the dual fuel is specifically fuel oil and gas.

[0081] In one embodiment, the decomposition levels of the functional hierarchical decomposition in step S1 include the system level, the component level, and the part level; specifically, the whole marine engine is functionally hierarchically decomposed into 13 components or subsystems through the WBS work breakdown structure method, such as Figure 3As shown, specifically including structural components, moving components, fuel systems, gas systems, lubricating oil systems, and cooling systems; further, the components of the gas system are gas storage tanks, gas transmission pipelines, gas filters, gas pressure regulating valves, gas booster pumps, gas injectors, gas inlet valves, gas fuel control modules, gas-air mixers, gas leakage detection systems, gas flow meters, gas control units, and ventilation and scavenging systems; furthermore, the main parts of each component are as follows: the main parts of the gas storage tank are the gas input interface and the safety relief valve; the main parts of the gas transmission pipeline are the high-pressure gas pipeline, flexible connection hose, and sealing gasket; the main parts of the gas filter are the filter element and the drain valve; the main parts of the gas pressure regulating valve are the regulating spring, diaphragm, safety valve, and pressure feedback pipeline; the main parts of the gas booster pump are the pump body, drive motor, and coupling; the main parts of the gas injector are the nozzle head, electromagnetic coil, needle valve, and sealing assembly; the main parts of the gas inlet valve are the valve body, valve core, and spring assembly; the main parts of the gas fuel control module are the main control board, power supply module, and communication interface; the main parts of the gas-air mixer are the mixing chamber, Venturi tube, and throttle valve plate; the main parts of the gas leakage detection system are the multi-point detector, alarm controller, and audible and visual alarm; the main parts of the gas flow meter are the signal processor and calibration module; the main parts of the gas control unit are the microprocessor, fault diagnosis module, and redundant power supply; the main parts of the ventilation and scavenging unit are the scavenging pump, ventilation pipeline, and check valve.

[0082] In one embodiment, the step S2: Based on the hierarchical structure tree of the gas system, through the subjective weight and objective weight fusion allocation method, the importance of each component of the gas system is evaluated, and the importance ranking of the components in the gas system is obtained, specifically including the following steps:

[0083] Step S21: Construct a health assessment index system for the components in the marine engine gas system, and obtain the health assessment criteria for the components in the gas system; specifically, as Figure 4 shown, the health assessment criteria for the components in the gas system of the present application are divided into 5 dimensions, namely failure frequency O, maintenance cost C, criticality H, life cycle P, and supply chain stability S; further, the dimensions of the health assessment criteria for the components in the gas system are historical failure rate O1, failure impact range O2, single repair cost C1, average repair time C2, impact on system performance H1, safety risk level H2, design life P1, remaining life prediction P2, number of suppliers S1, and average delivery cycle S2;

[0084] Step S22: Collect data for the evaluation indicators in the health assessment criteria to obtain the evaluation index value matrix of each component; specifically, the present application collects 10 standard data of 13 components of the gas system and constructs the evaluation index value matrix of the components in the gas system: where xij Denote the j-th evaluation index value of the i-th component;

[0085] Step S23: Standardize the evaluation index values of each component collected to generate a standardized matrix of the evaluation index values of the components;

[0086] Step S24: According to the standardized matrix of the evaluation index values of all components in the gas system, use the analytic hierarchy process and the entropy weight method to calculate in parallel to obtain the comprehensive weights of each evaluation index in the gas system;

[0087] Step S25: According to the comprehensive weights of each evaluation index, obtain the importance ranking of each component in the gas system.

[0088] In one embodiment, the step S23: Standardize the evaluation index values of each component collected to generate a standardized matrix of the evaluation index values of the components in the gas system, specifically including the following steps:

[0089] Step S231: Classify the evaluation index data of the important components collected into cost-type and benefit-type evaluation indexes to obtain the standardized component evaluation index value r ij , where the calculation formula of r ij is as follows:

[0090]

[0091] In the formula, r ij is the standardized j-th evaluation index value of the i-th component, min(x j ) and max(x j ) are respectively the maximum value and the minimum value of the j-th evaluation index;

[0092] Step S232: According to the obtained standardized component evaluation index values, generate a standardized matrix of the evaluation index values of the components in the gas system, as shown in the following formula:

[0093]

[0094] In one embodiment, the step S24: According to the standardized matrix of the evaluation index values of all components in the gas system, use the analytic hierarchy process and the entropy weight method to calculate in parallel to obtain the comprehensive weights of each evaluation index in the gas system, specifically including the following steps:

[0095] Step S241: Use the analytic hierarchy process to calculate the standardized matrix of the constructed evaluation index values to obtain the subjective weights of each evaluation index in the dual-fuel gas system;

[0096] Step S242: Use the entropy weight method to calculate and obtain the objective weights objectively assigned to the engine sensor data and the historical operation records, and obtain the objective weights of each evaluation index in the dual-fuel gas system;

[0097] Step S243: Linearly combine the subjective weight and the objective weight, and comprehensively evaluate to obtain the comprehensive weight of each evaluation index in the gas system.

[0098] By fusing the subjective weight and the objective weight, this application can more accurately evaluate the importance of each component in the gas system during the health assessment of the gas system.

[0099] In one embodiment, Step S241: Use the analytic hierarchy process to calculate the standardized matrix of the constructed evaluation index values to obtain the subjective weights of each evaluation index of the dual-fuel gas, specifically including the following steps:

[0100] Step S2411: Compare the 10 evaluation indexes pairwise through the method of expert evaluation, and construct the judgment matrix A as follows:

[0101]

[0102] where a 12 is the relative importance comparison result of pairwise comparison between the first evaluation index and the second evaluation index.

[0103] Step S2412: Calculate the maximum eigenvalue λ max of the judgment matrix A as shown in the following formula:

[0104]

[0105] where w AHP is the preliminary weight vector.

[0106] Step S2413: Calculate the degree of deviation from consistency of the judgment matrix A through the test coefficient C R . If the value of the test coefficient C R is less than 0.1, it is determined that the consistency test is passed. The calculation formula of the test coefficient C R is:

[0107]

[0108] In the formula, C1 is the consistency evaluation index, R1 is the average consistency evaluation index, λ max is the maximum eigenvalue of the judgment matrix; n = 10 is the order of the judgment matrix;

[0109] Step S2413: If the judgment matrix passes the consistency test, extract the characteristic weight, calculate the eigenvector corresponding to the maximum eigenvalue of the judgment matrix A, and obtain the characteristic weight w AHP after normalization as:

[0110] AwAHP = λ max w AHP 。

[0111] Specifically, solve the equation A·w = λmax·W to obtain the eigenvector w, and divide each component of w by the sum to obtain the normalized eigenweight w AHP 。For example: if the eigenvector is [1, 2, 3] T , then the eigenweight is [1 / 6, 2 / 6, 3 / 6];

[0112] Obtain the j-th evaluation index value w of the gas system AHP,j

[0113] In one embodiment, the step S242: Calculate and obtain the objective weights of the engine sensor data and the historical operation records by using the entropy weight method, and obtain the objective weights of the evaluation indicators in the dual-fuel gas system, which specifically includes the following steps:

[0114] Step S2421: Transform the standardized evaluation index values r of each component ij into the relative proportion p of each component on each evaluation indicator ij , and the relative proportion p of the importance of 13 components in the gas system on each evaluation indicator ij :

[0115]

[0116] where p ij is the relative proportion of the importance of the i-th component on the j-th evaluation indicator; ∈ is a minimum constant.

[0117] Step S2422: Calculate the entropy value of the j-th evaluation indicator to measure the evaluation indicator e according to the relative proportion of the importance of each component in the gas system on each evaluation evaluation indicator j , and determine the degree of chaos of each evaluation indicator measured by the entropy value. The larger the entropy value, the more uniform the data; j In the formula, n = 13 is the 13 components in the engine gas system.

[0118]

[0119]

[0120] Step S2423: Calculate and obtain the information utility value of each evaluation indicator according to the entropy value of each evaluation indicator to measure the evaluation indicator, which is used to reflect the effective information volume of each evaluation indicator:

[0121] d j = 1 - e j

[0122] In the formula, d j ​is the information utility value of the j-th evaluation index, d j The larger it is, the more it indicates that the evaluation index e j has a higher degree of variation, and relatively higher weights are assigned to the components of the gas system;

[0123] Step S2422: Determine the objective weight of the evaluation index according to the information utility value of each evaluation index:

[0124] In the formula, n is 10 evaluation indexes.

[0125] In one embodiment, the step S243: linearly combine the subjective weight and the objective weight as shown in the following formula, and comprehensively evaluate to obtain the comprehensive weight w of the j-th evaluation index in the dual-fuel gas system j :

[0126] w j = α·w AHP,j + β·w 熵权,j .

[0127] In the formula, α + β = 1.

[0128] In one embodiment, the step S25: Obtain the importance ranking of each component in the gas system according to the comprehensive weight of each evaluation index, specifically including the following steps:

[0129] Step S251: According to the comprehensive weight of each evaluation index, use the method of distance between superiority and inferiority to calculate and obtain the weighted standardized matrix of each evaluation index of each component in the gas system through the weighted formula:

[0130] The weighted formula is:

[0131] v ij = w j ·r ij ;

[0132] In the formula, v ij is the weighted value of the j-th evaluation index of the i-th component;

[0133] The weighted standardized matrix V is:

[0134]

[0135] Step S252: Obtain the positive and negative ideal solutions of the j-th evaluation index according to the weighted standardized matrix as shown in the following formula:

[0136]

[0137] In the formula, is the positive ideal solution of the j-th evaluation index; is the negative ideal solution of the j-th evaluation index.

[0138] Step S253: Calculate the Euclidean distance of each component in the gas system according to the positive and negative ideal solutions of each evaluation index, as shown in the following formula:

[0139]

[0140] Step S254: Calculate the closeness degree of each component in the gas system according to the Euclidean distance, as shown in the following formula:

[0141] Formula

[0142] Obtain the importance ranking of each component in the gas system according to the closeness degree. Arrange the closeness degree C i in descending order. The larger the C i , the higher the priority index of the component;

[0143] The ranking of the closeness degree is the importance ranking of the components in the gas system. The larger the C i , the higher the priority index of the component. Determine the importance of the component through the importance ranking. Arrange from large to small, and the 4 important components in the marine engine gas system are: gas pressure regulating valve, gas control unit, gas injector, and gas leakage detector.

[0144] In one embodiment, step S3: Collect data on the evaluation indexes of each component of the engine gas system according to the obtained importance ranking, and obtain the health data set of the components in the gas system according to the evaluation index values of each component collected. The specific steps are as follows:

[0145] Step S31: Determine the priority processing level of the components in the gas system through the importance ranking in the marine engine gas system, and divide the priority processing level of the components into Class A, Class B, Class C, and Class D, as shown in Table 1.

[0146] Table 1

[0147] Label Fault Handling Priority Description 1 Class A Very important, need to be processed with priority 2 Class B Important, need to be processed with secondary priority 3 Class C Generally important, general processing 4 Class D Design redundancy, processed at the end

[0148] Step S32: Determine the importance of each component in the marine engine gas system, select sensors with appropriate volume for deployment, collect the real-time operation data of the components at each priority processing level, and transmit the real-time data collected by the sensors through the TCP communication protocol;

[0149] Step S33: Obtain the health degree of each component through the historical operation data and real-time operation data of the component, and determine the operation and maintenance method of the engine through the health degree range;

[0150] Step S34: Obtain the health data set of the components in the gas system according to the health degree of each component.

[0151] In a specific embodiment, in step S3: data collection is performed on the evaluation indexes of the components of the engine gas system according to the obtained importance ranking, and a health dataset of the components in the gas system is obtained according to the obtained evaluation index values of each component. The specific implementation is as follows:

[0152] In this application, health judgment is performed through the exhaust temperature data collected by the sensor, and the exhaust temperatures of six combustion chambers in the marine engine are respectively recorded, that is, T r represents the currently real-time collected exhaust temperature.

[0153] A dynamically adjusted method is adopted to determine the threshold of the components of the gas system, and the temperature numerical threshold of the engine is dynamically adjusted in real time, that is, D max (t) = D max (t - 1)+Δt, where D max (t) is the threshold at the current moment, and Δt is the amplitude of the dynamic adjustment of the threshold. The thresholds at different moments are determined through the historical operation data of the marine engine gas system.

[0154] By using the Euclidean distance D formula to determine the currently real-time collected exhaust temperature T r and the historical exhaust temperature T n That is:

[0155]

[0156] where, T (r，i) is the real-time temperature value at the i-th time point, T (n，i) is the historical temperature value at the historical i-th time point, and n is the length of the sensor data.

[0157] The calculation formula for the health degree H is:

[0158]

[0159] The value range of the health degree is (0 - 1), 0 represents a dangerous and unhealthy state, and 1 represents a completely healthy state.

[0160] The health status of each component of the marine engine gas system is divided into a healthy state, a good state, a critical state, a warning state, and a dangerous state, as shown in Table 3 below:

[0161] Table 3

[0162] Label Status Health Degree Range Description Maintenance Strategy 1 Health Status [0.8,1] Excellent Status Normal Operation 2 Good Status [0.6,0.8) The equipment has certain deterioration Included in the maintenance plan 3 Critical Status [0.4,0.6) Close to the damaged state Arrange maintenance 4 Warning Status [0.2,0.4) Exceeding the normal value of the equipment, engine performance drops Repair as soon as possible 5 Danger Status [0,0.2) Danger Status, the engine cannot work properly Repair immediately

[0163] The health degree data of each component of the gas system is calculated and put into an Excel table to provide a dataset for subsequent health degree prediction of the gas system.

[0164] In one embodiment, in order to realize the health prediction of the marine engine gas system, the variational mode decomposition method optimized by the crown porcupine optimization algorithm is combined with the sparrow algorithm to optimize the long short-term memory neural network for health prediction, so as to realize the single-input single-output prediction of the health data, as Figure 5 shown. The step S4: According to the health dataset of the components in the gas system, combine the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm to predict and obtain the health of the gas system, which specifically includes the following steps:

[0165] Step S41: Decompose the health of the gas system into intrinsic mode functions with different center frequencies and bandwidths. Usually, the number of decomposition modes k and the penalty factor α in the variational mode decomposition are determined by the method of expert experience;

[0166] In this application, the crown porcupine optimization algorithm is used to determine the number of variational mode decomposition modes k and the penalty factor α, automatically search for the optimal k and α in the variational mode decomposition, and then improve the prediction accuracy; the mathematical model adopted by the crown porcupine optimization algorithm is as follows:

[0167]

[0168] In the formula, W is the number of crown porcupine populations; T is a variable that determines the number of cycles; T max is the maximum number of cycles of function evaluation; W min is the minimum number of iterations to generate the crown porcupine population.

[0169] When this application uses the crown porcupine optimization algorithm to optimize the parameters of the variational mode decomposition, the minimum envelope entropy is used as the fitness function. The envelope entropy formula is defined as:

[0170]

[0171] In the formula, N(i) is the envelope entropy; n is the number of sampling points; a(i) is the envelope signal.

[0172] For the above steps, first initialize a group of candidate solutions k1 and α1, and then perform variational mode decomposition on each candidate solution, calculate the envelope entropy of each intrinsic mode function after decomposition as the fitness value, and the smaller the envelope entropy, the better the decomposition effect. The crown porcupine optimization algorithm optimizes by continuously iterating and updating the number and position of the crown porcupine population, dynamically adjusting the weights of the defense and foraging behaviors, and gradually searching for the combination of k and α that minimizes the envelope entropy, and finally determines the optimal parameters of the variational mode decomposition.

[0173] Step S42: Obtain the optimal number of decomposition modes k and the penalty factor α of the multi-modal decomposition through Step S41. In this application, the sparrow search algorithm (SSA) is used to simulate the foraging and anti-predation behaviors of the sparrow population to optimize the hyperparameters of the long short-term memory neural network to minimize the prediction error. The Crested Porcupine Optimizer (CPO) is used to dynamically optimize the number of modes K and the penalty factor α of the variational mode decomposition to minimize the envelope entropy of each mode after decomposition. The parameters are updated by simulating the collision and repulsive force behaviors of the porcupine population: By continuously updating the porcupine population, the over-decomposition or under-decomposition problems caused by fixed parameters in traditional VMD are avoided.

[0174] Step S43: Use the intrinsic mode functions (IMFs) optimized by the crested porcupine optimization algorithm as the input of the LSTM. The sparrow algorithm is used to optimize the long short-term memory neural network. First, initialize the population and randomly generate combinations of long short-term memory neural network hyperparameters. Divide the sparrow population into discoverers, followers, and vigilantists. The discoverers explore the global optimum through an exponential decay strategy: where α is the decay coefficient, T is the maximum number of iterations, and R2 and the vigilance threshold ST determine the search mode. The followers develop the local space around the current optimal solution x best Develop the local space:

[0175] The output gate formula of the long short-term memory neural network is: o t = σ(W o ·o t-1 ⊙tanh(C t-1 ), x t ) + b o ) The final output layer formula of the hidden state h of the LSTM t is mapped to the predicted value y through the fully connected layer t = W y ·h t + b y , where, W y is the weight matrix of the output layer, and b y is the bias vector of the output layer.

[0176] Finally, use the root mean square error, mean absolute error, and median absolute percentage error as the evaluation error evaluation metrics:

[0177] The root mean square error E RSME is:

[0178]

[0179] The mean absolute error is:

[0180]

[0181] Median of absolute percentage error is:

[0182]

[0183] where x i is the true value of the health of the gas system, is the predicted value of the health of the gas system, and n is the number of test sample sets.

[0184] By predicting the health of the gas system of the marine engine, the health of the marine engine can be judged in advance, thereby reducing unnecessary economic losses caused by untimely health detection of the engine gas system.

[0185] In a more specific embodiment, as Figure 6 shown, step S5: Based on digital twin technology, the gas system is monitored in real time, the current health of the gas system is displayed in real time, and the subsequent health is predicted, which specifically includes the following steps:

[0186] Step S51: Construct a three-dimensional model of the marine engine with geometric properties;

[0187] Step S52: Construct a digital twin model of the marine engine through the three-dimensional model of the marine engine and sensor data;

[0188] Step S53: Through the UI interface design, the operating state of the digital twin model of the marine engine and the operating data of the marine engine are displayed in real time, and the subsequent health is predicted.

[0189] In one embodiment, step S51: Construct a three-dimensional model of the marine engine with geometric properties, which specifically includes the following steps:

[0190] Step S511: According to the physical entity of the marine engine of this model and the instruction manual of the relevant marine engine, determine the components and part models of the marine engine, and model the parts of the engine in the three-dimensional modeling software SolidWorks.

[0191] Step S512: Obtain various parts of the marine engine through step S1, classify the parts according to different subsystems, and model each subsystem separately. After the three-dimensional modeling of the subsystems is completed, use the assembly function in SolidWorks to assemble the subsystems into a complete marine engine model.

[0192] Step S513: When using the assembly function in SolidWorks, pay attention to the assembly between systems. Use the mating tools to define the relative positions and constraint relationships between the subsystems of the marine engine, and complete the three-dimensional construction of the whole marine engine model through fits such as concentric, distance, and coincidence.

[0193] Step S514: Verify the three-dimensional model of the marine engine to ensure the stability of the operation of the marine engine model. First, use the Move Component tool in SolidWorks to ensure that the movement of the engine components meets the expectations; second, through the built-in evaluation system in SolidWorks, perform interference checks on each subsystem of the marine engine to check whether there are interference problems between the subsystems and parts of the marine engine; finally, save the three-dimensional model of the whole marine engine in STEP format.

[0194] In one embodiment, Step S52: Construct a digital twin model of the marine engine through the three-dimensional model and sensor data of the marine engine, which specifically includes the following steps:

[0195] Step S521: According to Step S1, the whole marine engine model saved in STEP format is first processed in format through intermediate software. The intermediate software can be selected as 3ds Max or Blender, and in this invention, 3ds Max is selected. Import the marine engine model in STEP format into 3ds Max, and use the ProOptimizer module in 3ds Max to optimize the engine model, reduce the number of faces of the engine model and simplify the mesh to improve the smoothness of the operation of the digital twin model. Finally, select the triangular mesh to ensure the normal display of the marine engine model in Unity and export it in FBX format as a Unity-compatible format;

[0196] Step S522: Use models of different shapes in Unity. In this application, a small cube model is constructed to simulate the temperature sensor, a flat cylinder model is constructed to simulate the vibration sensor, and the materials of the sensor models at the key and important positions of the marine engine are set to red, select the RGB value as 255, and mark the important sensor models in the marine engine model in red;

[0197] Step S523: Select the moving parts of the marine engine model, set rigid bodies for the parts in the moving parts in Unity, add the Rigidbody component, and select discrete collision detection in the component. Add the MeshCollider collision component to the moving parts to ensure the stable operation of the particle system. Create a C# script in the C# programming software, drag the script to the moving parts of the marine engine model to simulate the motion state of the engine moving parts;

[0198] Step S524: Set up a particle system on the marine engine model, including a fuel injection particle system, an intake and exhaust particle system, a fuel-gas liquid particle system, and an in-cylinder combustion particle system. The marine engine model models diesel injection and natural gas injection with two different types of particles. In this application, the diesel injection simulation is set with coarser and slower-injecting particles, while the natural gas injection simulation is set with finer and faster-injecting particles. The start and stop of the diesel and natural gas injection particle injection systems are controlled through C# scripts. The intake and exhaust particle systems use blue and red particle colors respectively to simulate the intake temperature of the gas passing through the air cooler and the exhaust temperature of the gas discharged after combustion in the combustion chamber. Through particle systems in different states, the motion states of liquids and gases in the marine engine model are simulated, and the gas temperatures under different operating states are visually displayed.

[0199] In one embodiment, the said Step S53: Through UI interface design, the operating state of the digital twin model of the marine engine and the operating data of the marine engine are displayed in real time, and the subsequent health condition is predicted; specifically, after the digital twin model of the marine engine is completed, the engine model is visually displayed in Unity. The operating mode and operating state of the digital twin model of the marine engine, as well as the gas motion, oil flow, and in-cylinder combustion conditions during the operation of the engine, are visually displayed. The operating data and sensor data of the marine engine are displayed on the UI design interface, and at the same time, the sensor data is recorded and the abnormal data and alarm records are saved to a log file; more specifically, it includes the following steps:

[0200] Step S531: Design the UI layout. First, determine the parameters to be displayed. In this application, a status panel, a control panel, a data chart, and an alarm area are set. The status panel displays the rotational speed of the marine engine model and the temperature of the gas, etc. The control panel controls the start and stop of the marine engine model and the switching of the operating mode. The data chart displays the sensor data and creates a line chart. The alarm area displays fault or abnormal information;

[0201] Step S532: Split the marine engine model into subsystems, create a transparent material and adjust the color and transparency of the material. Set up a transparency switching logic and control the material switching of the subsystems through scripts. Control the camera in Unity to achieve the function of zooming in and out of the field of view;

[0202] Step S533: In this application, the Unity Charts plugin is selected to obtain real-time data from the digital twin model of the marine engine or sensors, and multiple line chart and bar chart objects are created in the scene to display data such as rotational speed, temperature, and pressure respectively;

[0203] Step S534: Display the fault data sensors of the marine engine model, highlight the sensors with fault data, and display the fault location and alarm information;

[0204] Step S535: Archive and organize the fault data and alarm information of the marine engine model, and record the time of fault or alarm occurrence, fault type, fault details, and alarm level.

[0205] In a second aspect, as Figure 7 shown, the present application provides a health assessment system for a marine engine gas system based on digital twin, including a functional hierarchical decomposition module 100, an importance assessment module 200, a component health data set acquisition module 300, a system health assessment module 400, and a real-time monitoring and warning module 500; the functional hierarchical decomposition module 100 is used to perform functional hierarchical decomposition on the entire marine engine to obtain a hierarchical structure tree of the gas system; the importance assessment module 200 is communicatively connected to the functional hierarchical decomposition module 100, and is used to perform importance assessment on each component of the gas system based on the hierarchical structure tree of the gas system through a subjective weight and objective weight fusion allocation method to obtain the importance ranking of the components in the gas system; the component health data set acquisition module 300 is communicatively connected to the importance assessment module 200, and is used to collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain a health data set of the components in the gas system based on the evaluation indicator values collected for each component; the system health assessment module 400 is communicatively connected to the component health data set acquisition module 300, and is used to predict and obtain the health of the gas system based on the health data set of the components in the gas system, in combination with the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm; the real-time monitoring and warning module 500 is communicatively connected to the system health assessment module 400, and is used to perform real-time monitoring on the gas system based on digital twin technology, and is used to display the current health of the gas system in real time and predict the subsequent health.

[0206] In a third aspect, an embodiment of the present application provides a health assessment device for a marine engine gas system based on digital twin. The health assessment device for a marine engine gas system based on digital twin can be a device with data processing functions such as a personal computer (PC), a laptop computer, a server, etc.

[0207] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces, etc., which are used to implement the interconnection of components inside the digital-twin-based marine engine gas system health assessment device, as well as interfaces for implementing the interconnection between the digital-twin-based marine engine gas system health assessment device and other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber optic interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.

[0208] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical memory, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0209] The processor can be a general-purpose processor, which can call the digital-twin-based marine engine gas system health assessment program stored in the memory and execute the digital-twin-based marine engine gas system health assessment method provided in the embodiments of the present application. For example, the general-purpose processor can be a central processing unit (CPU). Among them, the method executed when the digital-twin-based marine engine gas system health assessment program is called can refer to the various embodiments of the digital-twin-based marine engine gas system health assessment method of the present application, which will not be elaborated here.

[0210] Fourthly, the embodiments of the present application also provide a readable storage medium.

[0211] The digital-twin-based marine engine gas system health assessment program is stored on the readable storage medium of the present application. When the digital-twin-based marine engine gas system health assessment program is executed by a processor, it realizes the steps of the digital-twin-based marine engine gas system health assessment method as described above.

[0212] Among them, the method realized when the digital-twin-based marine engine gas system health assessment program is executed can refer to the various embodiments of the digital-twin-based marine engine gas system health assessment method of the present application, which will not be elaborated here.

[0213] It should be noted that the serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0214] Through the description of the above embodiments, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above and includes several instructions to enable a terminal device to execute the methods described in the various embodiments of the present application.

[0215] The above are only the preferred embodiments of the present application and do not limit the patent scope of the present application accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A health assessment method for the gas system of a marine engine based on digital twin, characterized in that It includes the following steps: Conduct a functional level decomposition of the marine engine as a whole to obtain the hierarchical structure tree of the gas system; Based on the hierarchical structure tree of the gas system, through the fusion assignment method of subjective weight and objective weight, evaluate the importance of each component of the gas system, and obtain the importance ranking of the components in the gas system; Collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain the health dataset of the components in the gas system based on the collected evaluation indicator values of each component; According to the health dataset of the components in the gas system, combine the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm to predict and obtain the health of the gas system; Based on digital twin technology, conduct real-time monitoring of the gas system, display the current health of the gas system in real-time, and predict the subsequent health; 2. The method for health assessment of the gas system of a marine engine based on digital twin according to claim 1, wherein The step of evaluating the importance of each component of the gas system based on the hierarchical structure tree of the gas system through the fusion assignment method of subjective weight and objective weight to obtain the importance ranking of the components in the gas system specifically includes the following steps: Construct a health evaluation index system for the components in the marine engine gas system to obtain the health evaluation criteria for the components in the gas system; Collect data on the evaluation indicators in the health evaluation criteria to obtain the evaluation indicator value matrix of each component; Standardize the evaluation indicator values of each component collected to generate the standardized matrix of the evaluation indicator values of the components; According to the standardized matrix of the evaluation indicator values of all components in the gas system, use the analytic hierarchy process and the entropy weight method to calculate in parallel to obtain the comprehensive weight of each evaluation indicator in the gas system; According to the comprehensive weight of each evaluation indicator, obtain the importance ranking of each component in the gas system; 3. The method for health assessment of the marine engine gas system based on digital twin according to claim 2, wherein, The step of standardizing the evaluation indicator values of each component collected to generate the standardized matrix of the evaluation indicator values of the components in the gas system specifically includes the following steps: Classify the evaluation indicator data of the important components collected according to cost type and benefit type to obtain the standardized component evaluation indicator values; According to the obtained standardized component evaluation indicator values, generate the standardized matrix of the evaluation indicator values of the components in the gas system; 4. The method for health assessment of a marine engine gas system based on digital twin according to claim 2, wherein, The step of using the analytic hierarchy process and the entropy weight method to calculate in parallel according to the standardized matrix of the evaluation indicator values of all components in the gas system to obtain the comprehensive weight of each evaluation indicator in the gas system specifically includes the following steps: Use the analytic hierarchy process to calculate the standardized matrix of the constructed evaluation indicator values to obtain the subjective weight of each evaluation indicator in the dual-fuel gas system; Use the entropy weight method to calculate and obtain the objective weights of the engine sensor data and the historical operation records, and obtain the objective weights of each evaluation indicator in the dual-fuel gas system; Linearly combine the subjective weight and the objective weight, and comprehensively evaluate to obtain the comprehensive weight of each evaluation indicator in the gas system; 5. The method for health assessment of a marine engine gas system based on digital twin according to claim 2, characterized in that, The step of obtaining the importance ranking of each component in the gas system according to the comprehensive weight of each evaluation indicator specifically includes the following steps: According to the comprehensive weight of each evaluation indicator, use the method of distance between superiority and inferiority to calculate through the weighted formula to obtain the weighted standardized matrix of each evaluation indicator of each component in the gas system; Obtain the positive and negative ideal solutions according to the weighted standardized matrix; Calculate the Euclidean distance according to the positive and negative ideal solutions; Calculate the closeness degree of each component in the gas system according to the Euclidean distance, and obtain the importance ranking of each component in the gas system based on the closeness degree.

6. The method for health assessment of the marine engine gas system based on digital twin according to claim 1, characterized in that, Collect data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtain the health dataset of the components in the gas system based on the evaluation indicator values of each component collected. The specific steps are as follows: Determine the priority processing level of the components in the gas system through the importance ranking in the marine engine gas system; Collect the real-time operation data of the components at each priority processing level; Obtain the health of each component through the historical operation data and real-time operation data of the component; Obtain the health dataset of the components in the gas system according to the health of each component.

7. The method for health assessment of a marine engine gas system based on digital twin according to claim 1, wherein, After predicting the health of the gas system by combining the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm based on the health dataset of the components in the gas system, the following steps are also included: Determine the optimal decomposition mode number and penalty factor of the variational mode decomposition through the crown porcupine optimization algorithm; Optimize the hyperparameters of the long short-term memory neural network by simulating the foraging and anti-predation behaviors of the sparrow population through the sparrow search algorithm; Take the health dataset as the input, and use the root mean square error, mean absolute error, and median of the absolute percentage error as the evaluation error evaluation indicators to predict the health of the dual-fuel gas system.

8. The method for health assessment of the marine engine gas system based on digital twin according to claim 1, wherein After predicting the health of the gas system by combining the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm based on the health dataset of the components in the gas system, the following steps are also included: Construct and implement real-time monitoring of the gas system based on digital twin technology, and display the current health of the gas system in real time and predict the subsequent health.

9. The method for health assessment of a marine engine gas system based on digital twin according to claim 1, characterized in that The implementation of real-time monitoring of the gas system based on digital twin technology, and display the current health of the gas system in real time and predict the subsequent health, specifically includes the following steps: Construct a three-dimensional model of the marine engine with geometric attributes; Construct a digital twin model of the marine engine through the three-dimensional model of the marine engine and sensor data; Through the UI interface design, display the running state of the digital twin model of the marine engine and the running data of the marine engine in real time, and predict the subsequent health.

10. A marine engine gas system health assessment system based on digital twin, characterized in that, Include: A functional hierarchical decomposition module for functionally hierarchically decomposing the entire marine engine to obtain a hierarchical structure tree of the gas system; An importance evaluation module, communicatively connected to the functional hierarchical decomposition module, for evaluating the importance of each component of the gas system based on the hierarchical structure tree of the gas system through a subjective weight and objective weight fusion allocation method, and obtaining the importance ranking of the components in the gas system; A component health dataset acquisition module, communicatively connected to the importance evaluation module, for collecting data on the evaluation indicators of each component of the engine gas system according to the obtained importance ranking, and obtaining the health dataset of the components in the gas system based on the evaluation indicator values of each component collected. The system health assessment module, which is communicatively connected to the component health data set acquisition module, is used to predict and obtain the health of the gas system according to the health data set of components in the gas system, by combining the variational mode decomposition method optimized by the crown porcupine optimization algorithm and the long short-term memory neural network optimized by the sparrow algorithm; The real-time monitoring and warning module, which is communicatively connected to the system health assessment module, is used to perform real-time monitoring on the gas system based on digital twin technology, and is used to display the current health of the gas system in real time and predict the subsequent health.

Images