Method and system for monitoring digital twin state of marine dual-fuel engine

Abstract

The invention discloses a marine dual-fuel engine digital twin state monitoring method, and relates to the technical field of ship power equipment, and the method comprises the steps: carrying out the importance grade evaluation of parts of a dual-fuel engine, and sequentially deploying sensors and obtaining the data of the sensors based on the importance grade of the parts; based on the sensor data, the health state of the dual-fuel engine is determined, and a maintenance mode is determined; constructing a three-dimensional model of the dual-fuel engine, driving the three-dimensional model of the dual-fuel engine by using sensor data, and simulating operation of the dual-fuel engine; and visually displaying the three-dimensional model of the dual-fuel engine and the simulated running state of the dual-fuel engine. According to the method, state monitoring and digital twinning are combined, omnibearing and high-fidelity modeling is carried out on the marine dual-fuel engine, displaying is carried out on a UI interface, state monitoring is upgraded to dynamic, visual and visual displaying of the operation state and the fault phenomenon of the dual-fuel engine.

Description

Technical Field

[0001] The present invention relates to the technical field of ship power equipment, and in particular to a digital twin state monitoring method and system for a ship dual-fuel engine. Background Art

[0002] At present, energy and environmental pollution issues are becoming more and more prominent. In the shipping industry, fuel consumption has become a huge cost for ship operation, and the exhaust gas generated by the engine has become an increasingly important factor in the earth's air pollution. In order to reduce ship operating costs and save energy, the use of dual-fuel engines to replace traditional diesel engines has a bright future. Dual-fuel engines can use both diesel and natural gas. As a clean energy source, natural gas has the advantages of high calorific value, full combustion, low impurity content, and very few pollutants after combustion, which meets the requirements of the International Maritime Organization's emission regulations.

[0003] Compared with diesel engines, dual-fuel engines have more parts and a more complex system. Therefore, it is of great significance to monitor the status of dual-fuel engines, improve the reliability of dual-fuel engines, and reduce maintenance costs, which is of great significance to the promotion and application of dual-fuel engines. The traditional marine diesel engine status monitoring method has shortcomings when dealing with dual-fuel engines. It is difficult to intuitively display fault phenomena and realize two-way interaction with dual-fuel engines, and the efficiency and accuracy of monitoring alarms need to be improved. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a digital twin state monitoring method and system for a marine dual-fuel engine.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A digital twin state monitoring method for a marine dual-fuel engine, comprising:

[0007] Evaluate the importance level of the parts of the dual-fuel engine, and deploy sensors based on the importance level of the parts, monitor the data of the dual-fuel engine, and obtain sensor data;

[0008] Determining a health status of the dual-fuel engine based on the sensor data and determining a maintenance method for the dual-fuel engine;

[0009] Build a 3D model of a dual-fuel engine and use sensor data to drive the 3D model of the dual-fuel engine to simulate the operation of the dual-fuel engine;

[0010] The three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating status are visualized.

[0011] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the importance level assessment of the parts of the dual-fuel engine includes:

[0012] Decompose the dual-fuel engine structure and identify the subsystems of the dual-fuel engine and the subordinate parts of the subsystems;

[0013] A fuzzy comprehensive evaluation factor set is established. The indicators of the factor set include impact degree, failure frequency, maintenance cost, and monitorability. The factor set is recorded as:

[0014] U = {impact level U1, failure frequency U2, maintenance cost U3, monitorability U4};

[0015] The four levels of very important, relatively important, important and general are used to evaluate the various factors of the importance level of the dual-fuel engine parts, and an evaluation set is established. The evaluation set is recorded as:

[0016] V = {very important V1, relatively important V2, important V3, general V4}, the corresponding score of the evaluation set S = (100, 80, 60, 40);

[0017] The experts' evaluation on the importance level of dual-fuel engine parts is summarized, and the data is normalized to obtain a fuzzy comprehensive evaluation matrix, which is:

[0018]

[0019] The weights between the factor sets are determined to be ω=(ω1, ω2, ω3, ω4), and the weights of the factor sets are multiplied by the fuzzy comprehensive evaluation components to obtain the total evaluation membership matrix of the dual-fuel engine parts. The data are normalized and multiplied by the scores corresponding to the evaluation sets to obtain the final scores of the dual-fuel engine parts. The calculation formula is:

[0020] score = ω × R × S T .

[0021] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the weights between the determination factor sets are:

[0022] ω = (ω1, ω2, ω3, ω4) includes:

[0023] Summarize the experts’ evaluations of the factor set weights and normalize the data using positive indicators;

[0024] Sum the weight evaluations of different samples and divide by the number of samples to obtain the subjective weights K = (k1, k2, k3, k4) between different factors;

[0025] According to the normalized data, calculate the proportion p of the i-th sample value under the j-th indicator to the sum of all sample values ​​of the indicator ij , and calculate the information entropy of the data under each indicator;

[0026] The final weight value of each indicator is calculated by the subjective weight and information entropy between different factors. The calculation formula is: Among them, m is the number of factor sets.

[0027] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the method of determining the health state of the dual-fuel engine based on the sensor data and determining the maintenance method of the dual-fuel engine includes:

[0028] The data of each sensor in the normal operation state of the dual-fuel engine are obtained for several times, and a fault threshold is set for each sensor of the dual-fuel engine. The calculation formula of the fault threshold is:

[0029] N(t)=S′(t)-S(t), where K is the fluctuation coefficient of the sensor parameter threshold;

[0030] is the average value of the sensor data measured when the dual-fuel engine is running smoothly after the noise is removed; S(t) is the standard deviation of the sensor data measured when the dual-fuel engine is running smoothly after the noise is removed; S′(t) is the standard deviation of the sensor data measured when the dual-fuel engine is running smoothly; N(t) is the noise parameter;

[0031] The sensor data collected in real time is compared with the corresponding fault threshold. If the real-time data of the sensor is within the fault threshold range, it means that the dual-fuel engine is operating normally; if the real-time data of the sensor exceeds the set fault threshold range for three consecutive times, an alarm is processed;

[0032] The health of the dual-fuel engine is calculated based on the real-time data of the sensor, and the calculation method of the health is:

[0033] The Mahalanobis distance calculation formula is:

[0034] Where N is the number of sensors; x i is the real-time data of the sensor; i is the average value of the normal operation data of the dual-fuel engine; i +ε i and i -ε i are the upper and lower limits of the sensor threshold range;

[0035] The calculation formula of energy entropy is:

[0036] Where x[n] is the sensor signal; ε is a value used to avoid mathematical errors in taking the logarithm of zero; N is the length of the intercepted signal; the sensor signal of length N is intercepted, the energy entropy of the signal during this period is calculated, and then divided by log(N) for normalization;

[0037] The calculation formula of health is: H = 1-α·D-(1-a)·I, where α is a coefficient with a value range of (0-1), and the value range of health H is (0-1), and the larger the value, the healthier the dual-fuel engine;

[0038] The health status of the dual-fuel engine is classified based on the health degree, and the maintenance method of the dual-fuel engine is determined based on the health status of the dual-fuel engine.

[0039] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the health state of the dual-fuel engine is divided based on the health degree, and the maintenance method of the dual-fuel engine is determined based on the health state of the dual-fuel engine, including:

[0040] The health level of [0.75, 1] ​​is classified as healthy, the health level of [0.5, 0.75) is classified as good, the health level of [0.25, 0.5) is classified as degraded, and the health level of [0, 0.25) is classified as faulty.

[0041] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the step of constructing a three-dimensional model of the dual-fuel engine includes:

[0042] Modeling various parts of the dual-fuel engine based on the physical entity of the dual-fuel engine;

[0043] Assemble the parts according to the subsystem structure of the dual-fuel engine to obtain a three-dimensional model of each subsystem;

[0044] Assemble the subsystems and build a 3D model of the dual-fuel engine;

[0045] Analyze the moving parts of the dual-fuel engine and simulate the movement relationship and movement structure between the moving parts of the dual-fuel engine;

[0046] Use a particle system to model the flow of gas and liquid inside a dual-fuel engine.

[0047] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the method of using sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine includes:

[0048] The real-time data collected by sensors is used to simulate the motion relationship and motion structure in the three-dimensional model of the dual-fuel engine. According to the real-time status of the dual-fuel engine, the particle motion is controlled to simulate the internal operating status of the dual-fuel engine in real time.

[0049] As a preferred solution of the digital twin state monitoring method for a marine dual-fuel engine of the present invention, the visual display of the three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating state includes:

[0050] Import the 3D model of the dual-fuel engine into the UI interface, and use charts to display the real-time operating data, sensor data, health and fault information of the dual-fuel engine.

[0051] The present invention also provides a digital twin state monitoring system for a marine dual-fuel engine, comprising:

[0052] The condition monitoring module is used to evaluate the importance level of the parts of the dual-fuel engine, and deploy sensors in sequence based on the importance level of the parts, monitor the data of the dual-fuel engine, and obtain sensor data;

[0053] a state diagnosis module, for determining a health state of the dual-fuel engine based on the sensor data, and determining a maintenance method for the dual-fuel engine;

[0054] The digital twin module is used to build a three-dimensional model of the dual-fuel engine and use sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine;

[0055] The human-computer interaction module is used to visualize the three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating status.

[0056] As a preferred solution of the digital twin state monitoring system for a marine dual-fuel engine of the present invention, it also includes:

[0057] The data management module is used to store and manage the characteristic status data of the dual-fuel engine and all data generated during the operation process.

[0058] The beneficial effects of the present invention are:

[0059] (1) The present invention combines condition monitoring and digital twins to perform all-round, high-fidelity modeling of marine dual-fuel engines, establishes two-way interaction between physical entities and virtual entities, and displays them on the UI interface. The invisible state monitoring is upgraded to dynamic and visual, and the operating status and fault phenomena of the dual-fuel engine are intuitively displayed.

[0060] (2) The present invention uses fuzzy comprehensive evaluation method and improved entropy weight method to determine the importance level of dual-fuel engine parts. The improved entropy weight method combines subjective weight and information entropy, and the calculated weight is more real and effective, and the importance level of dual-fuel engine parts is more credible. The type and layout of sensors are determined according to the importance level and actual conditions of dual-fuel engine parts, making the layout of sensors more reasonable.

[0061] (3) In the state monitoring module of the present invention, the dual-fuel engine is divided into six subsystems according to the working principle and function, and sensors are deployed according to the subsystems. Each subsystem uses the CAN bus for communication, and the MODBUSTCP communication protocol is used to transmit the real-time data of the sensor to the monitoring alarm module and the digital twin module. This makes the transmission of sensor data more efficient, flexible and reliable.

[0062] (4) The status diagnosis module of the present invention uses the fault threshold method to monitor the operating status of each component of the dual-fuel engine and performs alarm processing on abnormal data; uses health assessment to evaluate the overall health status of the dual-fuel engine and adopts different maintenance strategies according to the health status. Ensure the monitoring coverage of each component, timely discover the signs of single component failure, and reduce the risk of overall system failure of the engine, ensuring the long-term reliable operation of the dual-fuel engine. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0064] Figure 1 A schematic diagram of the process of the digital twin state monitoring method for a marine dual-fuel engine provided by the present invention;

[0065] Figure 2 A schematic diagram of a digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0066] Figure 3 A schematic diagram of a state monitoring module in a digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0067] Figure 4 A schematic diagram of the transmission of sensor data in the digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0068] Figure 5A schematic diagram of a digital twin module in a digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0069] Figure 6 A schematic diagram of a state diagnosis module in a digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0070] Figure 7 A schematic diagram of a human-computer interaction module in the digital twin state monitoring system for a marine dual-fuel engine provided by the present invention;

[0071] Figure 8 A schematic diagram of a data management module in the digital twin state monitoring system for a marine dual-fuel engine provided by the present invention. DETAILED DESCRIPTION

[0072] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific implementation modes and in combination with the accompanying drawings.

[0073] Embodiment 1:

[0074] Figure 1 A schematic flow chart of a method for monitoring the state of a digital twin of a marine dual-fuel engine provided in this embodiment. The method specifically comprises the following steps:

[0075] Step S101: Evaluate the importance level of the parts of the dual-fuel engine, and deploy sensors in sequence based on the importance level of the parts, monitor the data of the dual-fuel engine, and obtain sensor data.

[0076] Specifically, before evaluating the importance level of the parts of the dual-fuel engine, it is necessary to decompose the working structure of the dual-fuel engine, decompose it into three levels: the whole machine, subsystems, and parts, and establish a decomposition structure tree. The decomposed subsystems are the structural component system, the moving component system, the fuel system, the gas system, the lubricating oil system, and the cooling system. After that, the importance level of the parts under each subsystem is evaluated, and the parts of the dual-fuel engine are divided into different importance levels. The details are as follows:

[0077] A fuzzy comprehensive evaluation analysis method is introduced to determine the importance level of dual-fuel engine parts from multiple aspects, and a fuzzy comprehensive evaluation factor set is established. The indicators of the factor set include impact degree, failure frequency, maintenance cost, and monitorability. The factor set is recorded as: U = {impact degree U1, failure frequency U2, maintenance cost U3, monitorability U4}.

[0078] Use four levels: very important, relatively important, important, and general to evaluate the various factors of the importance level of dual-fuel engine parts and establish an evaluation set. The evaluation set is recorded as:

[0079] V = {very important V1, relatively important V2, important V3, general V4}, and the corresponding score of the evaluation set is S = (100, 80, 60, 40).

[0080] In the form of a table, relevant experts and researchers were asked to evaluate the importance of dual-fuel engine parts. The results of the table were summarized, and the number of different evaluation levels in each factor in the evaluation table of relevant experts and researchers was counted. The data was normalized to obtain a fuzzy comprehensive evaluation matrix, which is:

[0081]

[0082] The weights between the factor sets are determined to be ω=(ω1, ω2, ω3, ω4), and the weights of the factor sets are multiplied by the fuzzy comprehensive evaluation components to obtain the total evaluation membership matrix of the dual-fuel engine parts. The data are normalized and multiplied by the scores corresponding to the evaluation sets to obtain the final scores of the dual-fuel engine parts. The calculation formula is:

[0083] score = ω × R × S T .

[0084] The importance level of dual-fuel engine parts is determined by the scores. Parts with scores greater than 80 are classified as A, parts with scores between 80 and 60 are classified as B, and parts with scores less than 60 are classified as C.

[0085] Preferably, in the above step of determining the weights between the factor sets, the improved entropy weight method is used to determine the weights of the factor sets in the fuzzy comprehensive evaluation. The steps of the improved entropy weight method are as follows:

[0086] In the form of a questionnaire, relevant experts and researchers were asked to evaluate the weights of the factor set. The data was normalized using positive indicators. The normalization formula is as follows:

[0087] Among them, X ij is the unnormalized data, Y ij is the normalized data, i represents the number of samples, and j represents the number of factor sets.

[0088] After normalization, the weight evaluations of different samples are summed and divided by the number of samples to obtain the subjective weights K = (k1, k2, k3, k4) between different factors.

[0089] According to the normalized data, calculate the proportion p of the i-th sample value under the j-th indicator to the sum of all sample values ​​of the indicator ij , and calculate the information entropy of the data under each indicator. The calculation formula of information entropy is as follows:

[0090] The final weight value of each indicator is calculated by the subjective weight and information entropy between different factors. The calculation formula is: Among them, m is the number of factor sets.

[0091] After completing the importance level assessment of dual-fuel engine parts, the appropriate sensors are selected to monitor the dual-fuel engine based on the applicable scope, volume and other factors of the sensors. Specifically, the deployment priority of the corresponding sensors of the parts is sorted based on the importance level of the parts, and the corresponding sensors are deployed in sequence according to the sorting.

[0092] In this embodiment, the sensors include speed sensors, pressure sensors, temperature sensors, vibration sensors, position sensors, etc. The parameters monitored by the sensors include the speed of the dual-fuel engine, the speed and temperature of each cylinder bearing, the pressure of fuel gas inlet, etc.

[0093] It should be noted that sensors are deployed according to subsystems, and each subsystem communicates using the CAN bus, while the MODBUS TCP communication protocol is used to transmit the real-time data of the sensors.

[0094] After the sensors are deployed, the operation and shutdown of the dual-fuel engine, the switching of the dual-fuel engine fuel, and the adjustment of the dual-fuel engine speed are controlled through hardware design and control algorithms.

[0095] Step S102: determining the health status of the dual-fuel engine based on the sensor data, and determining the maintenance method of the dual-fuel engine.

[0096] Specifically, first, obtain data of each sensor in the normal operation state of the dual-fuel engine for several times, and set a fault threshold for each sensor of the dual-fuel engine. The calculation formula of the fault threshold is:

[0097]

[0098] Wherein, N(t)=S′(t)-S(t), K is the fluctuation coefficient of the sensor parameter threshold; is the average value of the sensor data measured when the dual-fuel engine is running smoothly after noise is removed; S(t) is the standard deviation of the sensor data measured when the dual-fuel engine is running smoothly after noise is removed; S′(t) is the standard deviation of the sensor data measured when the dual-fuel engine is running smoothly; N(t) is the noise parameter.

[0099] After setting the fault threshold of each sensor, the real-time data collected by the sensor can be compared with the set fault threshold. If the real-time data of the sensor is within the fault threshold range, it means that the dual-fuel engine is operating normally; if the real-time data of the sensor exceeds the set fault threshold range for three consecutive times, an alarm will be processed.

[0100] Afterwards, the health of the dual-fuel engine is calculated using the real-time data from the sensor. The health calculation method is as follows:

[0101] ①The calculation formula of Mahalanobis distance is:

[0102] Where N is the number of sensors; x i is the real-time data of the sensor; i is the average value of the normal operation data of the dual-fuel engine; i +ε i and i -ε i are the upper and lower limits of the sensor threshold range;

[0103] ②The calculation formula of energy entropy is:

[0104] Where x[n] is the sensor signal; ε is a value used to avoid mathematical errors in taking the logarithm of zero; N is the length of the intercepted signal; the sensor signal of length N is intercepted, the energy entropy of the signal during this period is calculated, and then divided by log(N) for normalization;

[0105] ③ The calculation formula for health is: H = 1-α·D-(1-a)·I, where α is a coefficient with a value range of (0~1), and the value range of health H is (0~1), and the larger the value, the healthier the dual-fuel engine.

[0106] The above health calculation method is used to calculate the health of each subsystem and evaluate the health status of each subsystem. The health of each subsystem is summed and normalized to obtain the overall health of the dual-fuel engine and evaluate the overall health status of the dual-fuel engine.

[0107] Finally, the health status of the dual-fuel engine is divided based on the health degree, and the maintenance method of the dual-fuel engine is determined based on the health status of the dual-fuel engine.

[0108] Specifically, based on the health level, the health status of the dual-fuel engine is divided into four levels, namely healthy, good, degraded and faulty. The specific status description and corresponding maintenance methods are shown in Table 1:

[0109]

[0110] Table 1

[0111] Step S103: construct a three-dimensional model of the dual-fuel engine, and use the sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine.

[0112] Specifically, constructing a three-dimensional model of a dual-fuel engine includes the following steps:

[0113] Step S103a: Modeling various parts of the dual-fuel engine using 3D modeling software based on the physical entity of the dual-fuel engine and materials such as the instruction manual.

[0114] Step S103b: assemble the parts according to the structures of the six subsystems of the dual-fuel engine, namely, the structural component system, the moving component system, the fuel system, the gas system, the lubricating oil system and the cooling system, to obtain a three-dimensional model of each subsystem.

[0115] Step S103c: assemble the six subsystems, construct a three-dimensional model of the dual-fuel engine, and render and map the three-dimensional model. At the same time, the surrounding environment of the dual-fuel engine can also be modeled to make the three-dimensional model of the dual-fuel engine closer to reality.

[0116] Step S103d: simulate the movement inside the dual-fuel engine, analyze the moving parts of the dual-fuel engine, and write a script to simulate the movement relationship and movement structure between the parts.

[0117] Step S103e: Use Unity's particle system to model the flow conditions of gas and liquid inside the dual-fuel engine. Use particles of different colors and particle motion conditions to simulate the state of gas and liquid inside the dual-fuel engine, including the flow conditions of fuel and gas, the combustion conditions of the fuel chamber, the flow and temperature conditions of cooling water, etc.

[0118] After the construction of the dual-fuel engine 3D model is completed, the digital twin model is driven by sensor data. The motion relationship and motion structure in step S103d are driven by the real-time data collected by the sensor to simulate the motion of the dual-fuel engine in real time. At the same time, according to the real-time state of the dual-fuel engine, the movement of the particles in step S103e is controlled to simulate the internal condition of the dual-fuel engine in real time.

[0119] Step S104: Visually displaying the three-dimensional model of the dual-fuel engine and the simulated operating state of the dual-fuel engine.

[0120] Specifically, through the visualization of the dual-fuel engine, users can more conveniently observe the operating status of the dual-fuel engine. At the same time, the data is presented in the form of tables, graphs, etc., so that users can more intuitively understand the meaning of the data and observe the operating trend of the sensor data, so as to better analyze and make decisions. The specific steps are as follows:

[0121] Step S104a: Import the digital twin model of the dual-fuel engine into the Unity software, and present the digital twin model in the Unity software.

[0122] Step S104b: The movement of the 3D model of the dual-fuel engine is driven by various sensor information collected by the sensor and the movement structure of the dual-fuel engine to simulate the movement of the mechanical structure of the dual-fuel engine entity. The gas and liquid inside the dual-fuel engine are rendered using a particle system to characterize the flow of the fuel in the dual-fuel engine, the combustion of the combustion chamber, and the flow and temperature of the cooling water.

[0123] Step S104c: Use models of different shapes to replace different types of sensors and deploy them to corresponding positions of the dual-fuel engine three-dimensional model. Add a highlighting function to facilitate observation of sensor positions.

[0124] Step S104d: For sensor data such as temperature and pressure, different colors are used to indicate whether the component monitored by the sensor is in a normal state.

[0125] Step S104e: realizing the functions of viewing angle rotation, zooming in, zooming out and resetting of the dual-fuel engine, and realizing transparent processing of each subsystem according to the subsystem of the dual-fuel engine, so that the user can observe the dual-fuel engine from various viewing angles.

[0126] Step S104f: Obtain the fault information and real-time health of the dual-fuel engine through the TCP / IP communication protocol and display them on the UI interface. When the dual-fuel engine fails, the fault location, fault information and solution are displayed on the UI interface.

[0127] Step S104g: Various data of the dual-fuel engine are displayed in the form of charts, etc., so that the user can more intuitively understand the meaning of the data and observe the operating trend of the sensor data, so as to better analyze and make decisions.

[0128] Embodiment 2:

[0129] Figure 2 A schematic diagram of a digital twin state monitoring system for a marine dual-fuel engine provided in this embodiment. The system includes a dual-fuel engine entity, a state monitoring module, a state diagnosis module, a digital twin module, and a human-computer interaction module.

[0130] Specifically, the dual-fuel engine entity not only provides various information of the dual-fuel engine, but is also connected to a state monitoring module of the dual-fuel engine.

[0131] See also Figure 3 The condition monitoring module is used to evaluate the importance level of the parts of the dual-fuel engine, and deploy sensors in sequence based on the importance level of the parts, monitor the data of the dual-fuel engine, obtain sensor data, and transmit the monitored data to the condition diagnosis module and the digital twin module. Figure 3 , the implementation process of this module is as follows:

[0132] Step a: Decompose the dual-fuel engine's working structure into three levels: whole machine, subsystem, and parts, and establish a decomposition structure tree. The decomposed subsystems are structural parts, moving parts, fuel system, gas system, lubricating oil system, and cooling system.

[0133] Step b: Evaluate the importance level of the parts under each subsystem and classify the parts of the dual-fuel engine into different importance levels.

[0134] Step c: Based on the importance levels classified in step 2, combined with factors such as the applicable scope and volume of the sensor, select a suitable sensor to monitor the dual-fuel engine.

[0135] Step d: Deploy sensors according to subsystems. Each subsystem uses the CAN bus to communicate with each other. At the same time, the MODBUS TCP communication protocol is used to transmit the real-time data of the sensors to the monitoring and alarm module and the digital twin module. Figure 4 .

[0136] Step e: Control the operation and shutdown of the dual-fuel engine, the switching of the fuel of the dual-fuel engine and the adjustment of the speed of the dual-fuel engine through hardware design and control algorithm.

[0137] It should be noted that the sensors used in the condition monitoring module include speed sensors, pressure sensors, temperature sensors, vibration sensors, position sensors, etc. The parameters monitored by the sensors include the speed of the dual-fuel engine, the speed and temperature of each cylinder bearing, the pressure of fuel gas entering, etc.

[0138] See also Figure 5 The digital twin module is used to build a three-dimensional model of the dual-fuel engine and use sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine. The digital twin module consists of two parts: the digital model and the data driver. The implementation process of this module is as follows:

[0139] Step a: Using the physical entity of the dual-fuel engine and materials such as the instruction manual, use 3D modeling software to model each part of the dual-fuel engine.

[0140] Step b: Assemble the parts according to the structures of the six subsystems of the dual-fuel engine structural component system, moving component system, fuel system, gas system, lubricating oil system and cooling system to obtain the three-dimensional models of each subsystem.

[0141] Step c: Assemble the six subsystems, construct the 3D model of the dual-fuel engine, render and map the 3D model, and model the surrounding environment of the dual-fuel engine to make the 3D model of the dual-fuel engine closer to reality.

[0142] Step d: simulate the movement inside the dual-fuel engine, analyze the moving parts of the dual-fuel engine, and write scripts to simulate the movement relationship and movement structure between the parts.

[0143] Step e: Use Unity's particle system to model the flow of gas and liquid inside the dual-fuel engine. Use particles of different colors and particle motion to simulate the state of gas and liquid inside the dual-fuel engine, including the flow of fuel and gas, the combustion of the fuel chamber, the flow and temperature of cooling water, etc.

[0144] Step f: Use sensor data to drive the digital twin model. The motion relationship and motion structure in step 4 are driven by the real-time data collected by the sensor to simulate the motion of the dual-fuel engine in real time. At the same time, according to the real-time state of the dual-fuel engine, the motion of the particles in step 5 is controlled to simulate the internal operating state of the dual-fuel engine in real time.

[0145] The status diagnosis module is used to determine the health status of the dual-fuel engine based on the sensor data and determine the maintenance method of the dual-fuel engine.

[0146] For details, see Figure 6 The status diagnosis module diagnoses the data transmitted by the status monitoring module, determines the location and type of the fault through the abnormal data of the sensor, and obtains the maintenance decision of the corresponding fault in combination with the maintenance manual and expert experience. The implementation process of this module is as follows:

[0147] Step a: The data of each sensor in the normal operation state of the dual-fuel engine is obtained multiple times through the state monitoring module as a basis for setting the threshold value.

[0148] Step b: according to the data of the dual-fuel engine under normal operating conditions obtained in step a, respectively set a fault threshold for each sensor of the dual-fuel engine.

[0149] Step c: The data monitored by the status monitoring module is transmitted to the monitoring alarm module in real time and compared with the fault threshold. If the sensor data is within the set fault threshold range, the engine is running normally; if the sensor data exceeds the set fault threshold range for three consecutive times, an alarm is processed.

[0150] Step d: Classify the health status level. Calculate the health of the dual-fuel engine through the real-time data of the sensor, judge the health status of the dual-fuel engine according to the divided health status level, and take different maintenance measures.

[0151] In step d, the health status is divided into four levels, namely healthy, good, degraded, and faulty. The specific status description is the same as Table 1 in Example 1.

[0152] The human-computer interaction module is used to visualize the three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating status.

[0153] For details, see Figure 7 The human-computer interaction module is used to provide an intuitive and friendly interface, allowing users to more conveniently observe the operating status of the dual-fuel engine. At the same time, the data is presented in the form of tables, graphs, etc., so that users can more intuitively understand the meaning of the data and observe the operating trend of the sensor data, so as to better analyze and make decisions. The human-computer interaction module is implemented in Unity software, and the implementation process is as follows:

[0154] Step a: Import the digital twin model of the dual-fuel engine into Unity software and model the surrounding environment.

[0155] Step b: The motion of the 3D model of the dual-fuel engine is driven by various sensor information collected by sensors and the motion structure of the dual-fuel engine to simulate the motion of the mechanical structure of the dual-fuel engine entity. The gas and liquid inside the dual-fuel engine are rendered using a particle system to characterize the flow of fuel in the dual-fuel engine, the combustion of the combustion chamber, and the flow and temperature of the cooling water.

[0156] Step c: Use different shapes to replace different types of sensors and deploy them to the corresponding positions of the dual-fuel engine 3D model. Add a highlight function to facilitate observation of sensor positions.

[0157] Step d: For sensor data such as temperature and pressure, different colors are used to indicate whether the component monitored by the sensor is in a normal state.

[0158] Step e: realizing the functions of viewing angle rotation, zooming in, zooming out and resetting of the dual-fuel engine, and realizing transparent processing of each subsystem according to the subsystem of the dual-fuel engine, so that the user can observe the dual-fuel engine from various angles.

[0159] Step f: Connect the monitoring and alarm module through the TCP / IP communication protocol to obtain the fault information and real-time health of the dual-fuel engine and display it on the UI interface. When the dual-fuel engine fails, the fault location, fault information and solution are displayed on the UI interface.

[0160] Step g: Use charts and other forms to display various types of information, so that users can more intuitively understand the meaning of the data and observe the operating trends of sensor data, so as to better analyze and make decisions.

[0161] In addition, the marine dual-fuel engine digital twin condition monitoring system also includes a data management module. Figure 8 The data management module is used to store and manage the characteristic state data of the dual-fuel engine and all data generated during the operation process. The data is stored and managed using a database, and the stored and managed content includes the engineering technical data, sensor data, operating state data and fault data of the dual-fuel engine.

[0162] Therefore, the technical solution of this application combines condition monitoring and digital twins to conduct all-round, high-fidelity modeling of marine dual-fuel engines, establish two-way interaction between physical entities and virtual entities, and display them on the UI interface. The invisible state monitoring is upgraded to dynamic and visual, and the operating status and fault phenomena of the dual-fuel engine are intuitively displayed.

[0163] In addition to the above embodiments, the present invention may also have other implementation modes; any technical solutions formed by equivalent replacement or equivalent transformation shall fall within the protection scope required by the present invention.

Claims

1. A digital twin state monitoring method for a marine dual-fuel engine, characterized in that: include: Evaluate the importance level of the parts of the dual-fuel engine, deploy sensors based on the importance level of the parts, monitor the data of the dual-fuel engine, and obtain sensor data; Determining a health status of the dual-fuel engine based on the sensor data and determining a maintenance method for the dual-fuel engine; Build a 3D model of a dual-fuel engine and use sensor data to drive the 3D model of the dual-fuel engine to simulate the operation of the dual-fuel engine; The three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating status are visualized.

2. The method for monitoring the state of a digital twin of a marine dual-fuel engine according to claim 1, characterized in that: The importance level assessment of dual fuel engine parts includes: Decompose the dual-fuel engine structure and identify the subsystems of the dual-fuel engine and the subordinate parts of the subsystems; A fuzzy comprehensive evaluation factor set is established. The indicators of the factor set include impact degree, failure frequency, maintenance cost, and monitorability. The factor set is recorded as: ; The four levels of very important, relatively important, important and general are used to evaluate the various factors of the importance level of the dual-fuel engine parts, and an evaluation set is established. The evaluation set is recorded as: , the evaluation set corresponds to the score ; The experts' evaluation on the importance level of dual-fuel engine parts is summarized, and the data is normalized to obtain a fuzzy comprehensive evaluation matrix, which is: ; Determine the weights between factor sets as , multiply the weight of the factor set by the fuzzy comprehensive evaluation component to obtain the total evaluation membership matrix of the dual-fuel engine parts, and normalize the data and multiply it by the score corresponding to the evaluation set to obtain the final score of the dual-fuel engine parts. The calculation formula is: .

3. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 2, characterized in that: The weights between the determination factor sets are include: Summarize the experts’ evaluations of the factor set weights and normalize the data using positive indicators; Sum the weight evaluations of different samples and divide them by the number of samples to get the subjective weights between different factors. ; Based on the normalized data, calculate the proportion of the i-th sample value under the j-th indicator to the sum of all sample values ​​of the indicator , and calculate the information entropy of the data under each indicator; The final weight value of each indicator is calculated by the subjective weight and information entropy between different factors. The calculation formula is: ,in, is the number of factor sets.

4. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 1, characterized in that: The determining the health status of the dual-fuel engine based on the sensor data and determining the maintenance method of the dual-fuel engine includes: The data of each sensor in the normal operation state of the dual-fuel engine are obtained for several times, and a fault threshold is set for each sensor of the dual-fuel engine. The calculation formula of the fault threshold is: , ,in, is the fluctuation coefficient of the sensor parameter threshold; It is the average value of the sensor data measured when the dual-fuel engine is running smoothly after removing noise; The standard deviation of the measured sensor data after removing noise when the dual-fuel engine is running smoothly; is the standard deviation of the measured sensor data when the dual-fuel engine is running smoothly; is the noise parameter; The sensor data collected in real time is compared with the corresponding fault threshold. If the real-time data of the sensor is within the fault threshold range, it means that the dual-fuel engine is operating normally; if the real-time data of the sensor exceeds the set fault threshold range for three consecutive times, an alarm is processed; The health of the dual-fuel engine is calculated based on the real-time data of the sensor, and the health calculation method is: The Mahalanobis distance calculation formula is: ,in, is the number of sensors; It is the real-time data from the sensor; is the average value of the normal operating data of the dual-fuel engine; and are the upper and lower limits of the sensor threshold range; The calculation formula of energy entropy is: , ,in, is the sensor signal; It is a numerical value used to avoid the mathematical error of taking the logarithm of zero; is the length of the intercepted signal; the intercepted length is The sensor signal is calculated, and the energy entropy of the signal during this period of time is divided by Normalization; The calculation formula for health is: ,in, is a coefficient with a value range of (0~1), health The value range of is (0~1), and the larger the value, the healthier the dual-fuel engine; The health status of the dual-fuel engine is classified based on the health degree, and the maintenance method of the dual-fuel engine is determined based on the health status of the dual-fuel engine.

5. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 4, characterized in that: The method of classifying the health status of the dual-fuel engine based on the health degree and determining the maintenance method of the dual-fuel engine based on the health status of the dual-fuel engine includes: The health level of [0.75, 1] ​​is classified as healthy, the health level of [0.5, 0.75) is classified as good, the health level of [0.25, 0.5) is classified as degraded, and the health level of [0, 0.25) is classified as faulty.

6. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 1, characterized in that: The three-dimensional model of the dual-fuel engine is constructed as follows: Modeling various parts of the dual-fuel engine based on the physical entity of the dual-fuel engine; Assemble the parts according to the subsystem structure of the dual-fuel engine to obtain a three-dimensional model of each subsystem; Assemble the subsystems and build a 3D model of the dual-fuel engine; Analyze the moving parts of the dual-fuel engine and simulate the movement relationship and movement structure between the moving parts of the dual-fuel engine; Use a particle system to model the flow of gas and liquid inside a dual-fuel engine.

7. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 6, characterized in that: The method of using the sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine includes: The real-time data collected by sensors is used to simulate the motion relationship and motion structure in the three-dimensional model of the dual-fuel engine. According to the real-time status of the dual-fuel engine, the particle motion is controlled to simulate the internal operating status of the dual-fuel engine in real time.

8. The method for digital twin state monitoring of a marine dual-fuel engine according to claim 1, characterized in that: The visual display of the three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating state includes: Import the 3D model of the dual-fuel engine into the UI interface, and use charts to display the real-time operating data, sensor data, health and fault information of the dual-fuel engine.

9. A digital twin state monitoring system for a marine dual-fuel engine, characterized in that: include: The condition monitoring module is used to evaluate the importance level of the parts of the dual-fuel engine, and deploy sensors in sequence based on the importance level of the parts, monitor the data of the dual-fuel engine, and obtain sensor data; a state diagnosis module, for determining a health state of the dual-fuel engine based on the sensor data, and determining a maintenance method for the dual-fuel engine; The digital twin module is used to build a three-dimensional model of the dual-fuel engine and use sensor data to drive the three-dimensional model of the dual-fuel engine to simulate the operation of the dual-fuel engine; The human-computer interaction module is used to visualize the three-dimensional model of the dual-fuel engine and the simulated dual-fuel engine operating status.

10. The marine dual-fuel engine digital twin state monitoring system according to claim 9, characterized in that: Also includes: The data management module is used to store and manage the characteristic status data of the dual-fuel engine and all data generated during the operation process.

Images