Ultrasonic cleaning method for ship equipment based on universe and digital twinning

Abstract

The invention discloses an ultrasonic cleaning system and method for marine equipment based on meta universe and digital twinning, and belongs to the technical field of information and automatic control and meta universe. The system is composed of a virtual ultrasonic cleaning twin model based on the universe and an entity ultrasonic cleaning model, a three-layer network framework is adopted, and a user is connected with the ultrasonic cleaning twin system through the augmented reality technology. The virtual model and the entity model map an entity into reality in the form of virtual interactive fusion and virtual connection, a user can analyze and count cleaning data in the middle layer through a Web online browser, dynamic change information of ultrasonic cleaning is monitored in real time, and a decision is made. According to the method, the target twin model is updated in real time through various models and historical data, the cleaning of the ship mechanical equipment is tested and inspected by creating digital copies of actual production and operation, the actual cleaning process can be modeled with high efficiency and high precision, the cleaning cost is reduced, and the cleaning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of information and automatic control technology, and in particular to a system and method for ultrasonic cleaning of marine equipment based on metaverse and digital twins. Background Art

[0002] During ship operations, some mechanical equipment requires regular maintenance and cleaning. For example, marine high-pressure oil pumps and turbocharger components operate under high temperatures and shock loads, requiring each component to meet not only structural, material, shape, and geometric accuracy requirements, but also dynamic, aerodynamic, thermal, and strength performance specifications. Improper maintenance can lead to malfunctions such as surge, bearing burnout, and increased or decreased turbocharger pressure, impacting the ship's navigational performance and safety. Therefore, rigorous geometric analysis and quality control are essential during cleaning and maintenance. Effective cleaning monitoring is crucial to the performance and safety of ship operations.

[0003] In ship machinery and equipment, the main targets of cleaning are sludge and carbon deposits. Traditional cleaning methods include manual cleaning, grinders, small pneumatic or electric rust removal equipment such as wire brushes, shot (sand) cleaning, chemical cleaning and ultra-high pressure water cleaning. However, due to the low efficiency of manual rust removal, the cleaning effect of semi-mechanized equipment such as angle grinders, wire brushes, and pneumatic needle beam cleaners cannot achieve high-quality surface treatment quality, and the rust removal of shot (sand) rust removal equipment will cause serious environmental pollution. It is difficult to meet the cleaning standards. In the cleaning process, the mixed action of multiple cleaning methods may cause varying degrees of scratches and damage to the surface of parts, and the cleaning effect is poor and the cleaning efficiency is low. It can no longer meet the development of ships towards precision, automation and intelligence.

[0004] At present, ultrasonic cleaning technology has been widely used in production links such as precision parts assembly, molding and processing because of its high cleaning efficiency, fast speed and uniform cleaning. It can evenly clean away dust or particle attachments, oil, water spots and other contaminants without damaging the workpiece being cleaned. Although ultrasonic cleaning technology provides a foundation for the cleaning industry,

[0005] However, as the shipbuilding industry evolves towards digitalization and intelligence, the limitations of single physical cleaning methods are becoming increasingly apparent. During ultrasonic cleaning, the settings for cleaning machine parameters, such as the number of ultrasonic generators activated and the power level, are typically based on experience or fixed values. This results in inconsistent cleaning results for different types and models of components, varying targets and cleaning requirements. The lack of personalized cleaning parameter settings for individual equipment and parts results in high energy consumption and low cleaning efficiency, ultimately leading to wasted resources and increased costs.

[0006] With the increasing demand for cleanliness of ultrasonic cleaning products and the increasing complexity of cleaning components, the existing simulation models related to the ultrasonic cleaning process have problems in information transmission, such as low information utilization, poor integration, waste of resources, and lack of feature definition between processes, environments, and equipment. It is necessary to simulate and analyze the dynamic cleaning process in real time and predictably. The existing workshop cleaning system does not thoroughly analyze data such as ultrasonic generators, ultrasonic vibrators, cleaning temperatures, and changes in the status of workpieces to be cleaned. During the cleaning process, the complex changes in the real-time status of the workpieces and cleaning equipment will lead to increasing cleaning errors, which will ultimately affect the cleaning effect of the workpieces.

[0007] The traditional cleaning process involves empirical analysis of cleaning characteristics, manual determination of cleaning areas, manual definition of energy consumption control, and manual control of operating status, which is time-consuming and labor-intensive. Furthermore, existing cleaning model simulation technology cannot efficiently and accurately model marine turbochargers in complex production and design environments. Summary of the Invention

[0008] In order to solve the problems existing in the existing technology, the present invention will use the design scheme of the integration of virtual and reality in the metaverse and digital twin technology to optimize the maintenance and cleaning process of traditional ship equipment, and propose a modeling method based on the metaverse and digital twin.

[0009] Metaverse and digital twin technologies, as new approaches to connecting the physical and digital worlds, can test and verify ship turbocharger cleaning by simulating digital replicas of actual production and operations, thereby reducing costs and improving cleaning efficiency. Therefore, the power of metaverse and digital twin technologies, through their integrated consideration of engineering, real-world data, and historical performance, provides a real-time, efficient solution for cleaning ship machinery and equipment.

[0010] Based on the above problems, the present invention proposes a solution. The present invention proposes an ultrasonic cleaning system and method for marine equipment based on metaverse and digital twins.

[0011] In the context of the metaverse, this patent constructs an operating platform for the ultrasonic cleaning process through virtual combination with digital twin model technology. A marine equipment ultrasonic cleaning system based on the metaverse and digital twin mainly includes: a virtual ultrasonic cleaning twin model based on the metaverse and a physical ultrasonic cleaning model.

[0012] The virtual ultrasonic cleaning twin model of the metaverse is characterized by connecting users to the ultrasonic cleaning twin system through extended reality technology, wherein the ultrasonic cleaning twin model maps the physical ultrasonic cleaning model to the extended reality module through virtual interactive fusion and virtual connection and management technology, thereby improving the cleaning process system and optimizing the ultrasonic cleaning design process.

[0013] An ultrasonic cleaning system for ship equipment based on metaverse and digital twin, in which the system network mainly includes a three-layer architecture: ultrasonic cleaning equipment layer, middle layer and user layer.

[0014] In the three-layer network architecture, the ultrasonic cleaning equipment layer includes: ultrasonic cleaning servers, switches, and ultrasonic generators. The ultrasonic cleaning equipment layer is located in the third layer of the network framework, and the cleaning line information is transmitted to the middle layer through the cleaning data acquisition server. The cleaning data includes static information and dynamic information. The static information includes: the size and model of the parts to be cleaned and the ultrasonic cleaning machine, etc. The dynamic information includes: sensor data, cleaning status information, etc. The middle layer is located in the second layer of the network framework and includes: Web servers, real-time database servers, etc. The real-time database server contains historical data of the cleaning process.

[0015] , users can access the real-time database in the web server through the IE browser. The user layer is located at the top layer of the network framework and consists of multiple real-time browser clients. Users can remotely monitor ultrasonic cleaning through the digital twin model established based on the second-layer network framework data.

[0016] The user layer is at the highest level of the entire system. It is the part where users interact directly with the system and receives input from user commands. It can obtain production operation status data of the workshop layer through the middle layer or send control commands and decision-making commands to the workshop layer, and then directly exchange with the workshop layer data acquisition application server, thereby improving the real-time performance of system access.

[0017] In the system network framework, the principles include:

[0018] The status data of the ultrasonic cleaning line is collected by the cleaning line data collection server in the ultrasonic cleaning equipment layer. The collected data is then transmitted to the middle layer through the transmission device to transmit the cleaning status information. The user layer can analyze and count the cleaning data in the middle layer through the Web online browser, monitor the real-time dynamic changes of ultrasonic cleaning in real time, and make decisions.

[0019] A system and method for ultrasonic cleaning of marine equipment based on metaverse and digital twin, wherein the interaction process between the physical model and the digital twin model mainly includes: geometric data twin stage, physical parameter twin stage, drive data twin stage and update optimization stage.

[0020] The geometric data twin stage, as the first stage, has the following main features: using high-precision 3D scanning technology, point cloud data technology, CAD design, 3Dmax, PTC Creo tools or systems to convert data into three-dimensional parametric models, thereby achieving 1:1 geometric mapping of the solid model and ensuring the visual and structural consistency of the mapped object.

[0021] Among them, the characteristics of the physical digital twin stage include: modeling the physical properties and performance parameters of the entity through physical model software such as Simulink and Modelica, analyzing the coupling relationship between marine equipment parts and ultrasonic cleaning machine generators in stress, transmission, vibration, etc., and forming a multi-field physical test model, so that the model can ensure the accuracy of physical and performance parameters.

[0022] The driving data twin stage, in which the driving data includes actual static and dynamic data, wherein the actual static and dynamic data include: a large amount of data such as the material properties of the original parts to be washed, surface integrity, cleaning temperature, generator vibration parameters, etc., can be comprehensively analyzed, learned and predicted through the neural network model, thereby constructing a real-time model data set.

[0023] Among them, the update and optimization stage of the driving data twin stage can establish a model for the real-time data of the cleaning originals and cleaning machines as well as the historical cleaning information. It can not only predict the performance indicators of cleaning, but also dynamically update the geometric data twin stage, physical parameter twin stage and driving data twin stage, and continuously map the status and performance of the twin originals.

[0024] The present invention proposes a metaverse- and digital twin-based ultrasonic cleaning system and method for marine equipment, which is applied to the ultrasonic cleaning process of ship turbochargers. The ultrasonic cleaning modeling method is characterized by performing real-time statistical analysis and processing of data collected by sensors and other acquisition devices on the digital twin system. Finally, the obtained decision information is transmitted to the ultrasonic cleaning equipment in the physical space for timely adjustment and updating of the data, thereby effectively solving the problems existing in existing cleaning technologies in complex production and design environments.

[0025] The ultrasonic cleaning modeling analysis method is characterized by:

[0026] S1: Construct an initial judgment matrix C1 based on A monitoring time nodes and B cleaning cleanliness indicators of the predetermined cleaning object in the ultrasonic cleaning equipment;

[0027] S2: Based on the initial judgment matrix, perform standardization processing to construct a labeled decision matrix C2;

[0028] S3: Combine the standardized decision matrix with the weight matrix to obtain the weighted judgment matrix C3;

[0029] S4: Obtain the positive ideal value D1 and negative ideal value D2 of the evaluation target from the weighted judgment matrix;

[0030] S5: calculating the Euclidean distance between each of the cleaning quality indicators and the positive ideal value and the negative ideal value respectively;

[0031] S6: Calculating the relative fit of each of the cleaning cleanliness indicators based on the Euclidean distance between the positive ideal value and the negative ideal value, wherein the relative fit is negatively correlated with the deviation of the cleaning quality;

[0032] S7, normalizing the relative fit degree and mapping it to [0, 1] to classify the ultrasonic cleaning quality status;

[0033] S8. Based on the comparison between the normalized numerical value and the state classification value, the cleaning quality state is predicted and feedback is given.

[0034] The process of constructing the initial judgment matrix C1 in S1 includes:

[0035] According to the predetermined cleaning object, for the i-th (i= 1, 2, ..., A) monitoring time section and the j-th (j= 1, 2, ..., B) cleaning cleanliness index, the evaluation value is X ij , forming the initial judgment matrix C1:

[0036]

[0037] The process of constructing the labeled decision matrix C2 in S2 includes: standardizing the initial judgment matrix to obtain the standard value b of the jth cleaning quality index at the i-th monitoring time node ij , the standardized decision matrix is C1 ’ :

[0038]

[0039]

[0040] Where, is the evaluation value, is the mean;

[0041] The weighted judgment matrix process in S3 and S4 includes: combining the standardized decision matrix with the weight matrix to obtain the weighted judgment matrix T;

[0042]

[0043] The positive ideal value D1 and negative ideal value D2 of the evaluation target are obtained by the weighted judgment matrix;

[0044]

[0045]

[0046] In the formula, the profitability index set D1 is the optimal solution of the i-th index value; the loss index set D2 is the worst solution of the i-th index value; n is the total number of processing quality indicators; d ij represents the element of the jth processing quality index of the i-th monitoring node in the weighted judgment matrix;

[0047] The steps S5 and S6 are to calculate the Euclidean distance Q between each cleaning quality index and the positive ideal value and the negative ideal value. i1 , Q i2 ;

[0048]

[0049] Where D1 represents the positive ideal value; D2 represents the negative ideal value;

[0050] Based on the Euclidean distance between the positive ideal value and the negative ideal value, the relative fit P of each cleaning cleanliness index is calculated. i1 ,wherein the relative fit is negatively correlated with the deviation of cleaning quality;

[0051] The S7 fit normalization process includes: dividing the ultrasonic cleaning quality status into N1 and N2, wherein N1 is a qualified cleaning status and N2 is an unqualified cleaning status.

[0052] A system and method for ultrasonic cleaning of marine equipment based on Metaverse and digital twins, wherein the traditional cleaning process includes:

[0053] S1: Manually obtain basic information of the components to be cleaned and control parameters of the ultrasonic cleaning machine;

[0054] S2: Manually perform subjective analysis on the original part to be cleaned to determine the characteristic parameters of oil sludge and carbon deposits on the workpiece;

[0055] S3: Manually determine the relationship between the oil sludge and carbon deposit content of the workpiece to be cleaned and the control of the ultrasonic cleaning machine;

[0056] S4: Manually control component cleaning and machine operation status.

[0057] The cleaning method based on the metaverse and digital twins is characterized by: in the physical and digital twin stages, the digital twin system performs real-time statistical analysis and processing of data collected by sensors and other acquisition devices, and finally transmits the obtained decision information to the ultrasonic cleaning equipment in the physical space for timely adjustment and update of the data.

[0058] The modeling process includes:

[0059] S1: Based on the information collected by the sensor, obtain geometric data and build a geometric model based on the turbocharger and ultrasonic cleaning machine;

[0060] S2: The system constructs a physical model of the turbocharger's physical properties and performance parameters based on the collected data;

[0061] S3: The system builds a data model based on the real-time cleaning data of the turbocharger components and the historical storage data;

[0062] S4: Use sensors and historical data models to update geometric models, physical models, and data models in real time;

[0063] S5: Create a digital twin of the turbocharger model based on the updated geometric model, physical model, and data model.

[0064] Among them, the cleaning twin principle includes:

[0065] The twin model of the metaverse twin layer is determined by the physical layer cleaning target and the ultrasonic cleaning machine. The consistency verification module is used to determine the twin rule model information and the twin behavior model information. The system calculates the twin control relationship based on the determined twin behavior model information and determines the optimal solution for the ultrasonic generator energy consumption relationship. The system controls the cleaning machine generator based on this optimal solution information, and then controls the number of ultrasonic vibrators turned on and the power level based on this optimal solution information. During the twinning process, the system updates the model, consistency verification module, and twin control relationship in real time based on the twin database, cloud database, and existing historical database.

[0066] Among them, the ultrasonic cleaning energy-saving control process includes:

[0067] S1: Obtain basic information of the cleaning target and identify the cleaning area of the cleaning target to obtain a cleaning feature set;

[0068] S2: Obtain basic information, cleaning components, and control parameters of the ultrasonic cleaning machine, and perform cleaning characteristic analysis on the cleaning components and control parameters based on the basic information to determine the cleaning characteristic parameters;

[0069] S3: Based on the cleaning characteristic parameters, the cleaning parameters of the control energy consumption analysis, fitting the energy-saving relationship between the cleaning characteristic parameters;

[0070] S4: Matching the cleaning feature set with the cleaning feature parameters to determine a cleaning-component control relationship;

[0071] S5: Based on the cleaning - element control relationship, energy-saving relationship, energy-saving optimization, determine the control parameters of each element;

[0072] S6: Based on the control parameters of each component, a control instruction is generated and sent to the console to control the operation of the generator.

[0073] The beneficial effects of the present invention are:

[0074] The system adopts a three-layer network architecture consisting of ultrasonic cleaning equipment layer, middle layer and user layer to optimize the cleaning and maintenance process of marine equipment, and proposes a new modeling method for ultrasonic cleaning digital twin based on metaverse and digital twin; the system can monitor the dynamic change information of ultrasonic cleaning in real time, and provide cleaning equipment decision-making and feedback based on the status data of the data acquisition server; the system proposes a consistency judgment method for ultrasonic cleaning of marine equipment, and judges the status of equipment cleaning in real time by constructing a cleaning judgment matrix. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a schematic diagram of an ultrasonic cleaning system for marine equipment based on the metaverse and digital twin.

[0076] Figure 2 It is a network architecture diagram of virtual-real interactive integration.

[0077] Figure 3 It is the overall structural principle diagram.

[0078] Figure 4 It is a flow chart of the ultrasonic cleaning twin process.

[0079] Figure 5 It is a digital twin virtual-reality interaction diagram.

[0080] Figure 6 This is a flow chart of the ultrasonic cleaning process for traditional marine equipment.

[0081] Figure 7 It is a schematic diagram of the ultrasonic cleaning modeling process.

[0082] Figure 8 It is the relationship diagram of the sludge cleaning rate of the data update layer.

[0083] Figure 9 It is a flow chart of the system energy-saving control method.

[0084] Figure 10 It is a schematic diagram of the geometric modeling of the ultrasonic cleaning machine.

[0085] Figure 11 This is a schematic diagram of the generator distribution on the side of the ultrasonic cleaning machine.

[0086] In the figure: 1, the first ultrasonic vibrator on the side; 2, the second ultrasonic vibrator on the side; 3, the third ultrasonic vibrator on the side; 4, the fourth ultrasonic vibrator on the side; 5, the fifth ultrasonic vibrator on the side. DETAILED DESCRIPTION

[0087] The present invention will be further described below with reference to specific examples. It should be understood that this example is intended to illustrate the present invention only and is not intended to limit the scope of the present invention. It should also be understood that after reading the contents of the present invention, those skilled in the art may make various changes or modifications to the present invention, and that these equivalent forms also fall within the scope defined by the appended claims.

[0088] The specific structure of the present invention is further described below with reference to the accompanying drawings.

[0089] pass Figure 1 The Metaverse and Digital Twin marine equipment ultrasonic cleaning system displays the virtual state of ultrasonic cleaning in real time. Figure 2 The virtual-real interactive fusion network architecture connects all systems and information. Figure 3 The overall structural principle diagram illustrates the real-time twin principle of digital twin technology. Figure 4 Ultrasonic cleaning twin process flow chart real-time update twin process. Figure 5 The digital twin virtual-real interaction diagram transforms physical parts into virtual parts. Figure 6 The ultrasonic cleaning flow chart of traditional marine equipment completes the comparative analysis of traditional technology and digital twin cleaning technology. Figure 4 Schematic diagram of the ultrasonic cleaning modeling process, converting the physical entity assembly model into a virtual assembly model.

[0090] like Figure 1 Figure 2 shows a schematic diagram of an ultrasonic cleaning system for marine equipment based on the Metaverse and digital twin. The physical ultrasonic cleaning model is integrated into the Metaverse system through virtual-reality interaction. Extended reality technology within the Metaverse system is used to display the status of components within the ultrasonic cleaning system. The virtual status displayed by the Metaverse system is fed back into the physical ultrasonic cleaning model through virtual connections and management technologies for real-time optimization and management.

[0091] like Figure 2 As shown in the diagram, the virtual-real interactive fusion network architecture reads data from physical devices at the ultrasonic cleaning equipment layer. This information is then processed by the workshop cleaning station servers, switches, and control machines at the middle layer. The virtual information of the cleaning equipment is then connected to the Metaverse system via a web client browser at the user layer.

[0092] The ultrasonic cleaning equipment layer, located in the third layer of the network framework, includes an ultrasonic cleaning server, switches, and ultrasonic generators. It transmits cleaning line information to the middle layer via a cleaning data acquisition server. Cleaning data includes both static and dynamic information. Static information includes the size and model of the parts being cleaned and the ultrasonic cleaning machine. Dynamic information includes sensor data and the status of the parts being cleaned.

[0093] The middle layer is located in the second layer of the network framework, including: Web server, real-time database server, etc. The real-time database server contains historical data of the cleaning process, and users can access the real-time database in the Web server through the IE browser.

[0094] The user layer is located at the top layer of the network framework and consists of multiple real-time browser clients. The user layer is at the highest layer of the entire system. It is the part where users interact directly with the system and receives input from user commands, so that the production operation status data of the workshop layer can be obtained through the middle layer.

[0095] like Figure 3 As shown in the figure, the framework diagram of the ultrasonic cleaning twin module includes: geometric data twin stage, physical parameter twin stage, drive data twin stage and update optimization stage.

[0096] Geometric Data Twin Phase: This phase establishes a digital twin geometric model, using modeling software to geometrically twin the ultrasonic cleaning machine's transducer and generator. After the ultrasonic cleaning machine's physical data is determined, the twin geometric model information and the twin physical model information are verified for consistency, forming a pending target.

[0097] Specific steps:

[0098] S1: Predetermine the monitoring time nodes and cleaning cleanliness indicators of the cleaning object in the ultrasonic cleaning equipment and construct the initial judgment matrix C1;

[0099] According to the predetermined cleaning object, for the i-th (i= 1, 2, ..., A) monitoring time section and the j-th (j= 1, 2, ..., B) cleaning cleanliness index, the evaluation value is X ij , forming the initial judgment matrix C1:

[0100]

[0101] S2: Standardize the initial judgment matrix to obtain the standard value b of the jth cleaning quality index at the i-th monitoring time node ij , the standardized decision matrix is C1 ’ :

[0102]

[0103]

[0104] Where, is the evaluation value, is the mean;

[0105] S3: Combine the standardized decision matrix with the weight matrix to obtain the weighted judgment matrix T;

[0106]

[0107] S4: Obtain the positive ideal value D1 and negative ideal value D2 of the evaluation target from the weighted judgment matrix;

[0108]

[0109]

[0110] In the formula, the profitability index set D1 is the optimal solution of the i-th index value; the loss index set D2 is the worst solution of the i-th index value; n is the total number of processing quality indicators; d ij represents the element of the jth processing quality index of the i-th monitoring node in the weighted judgment matrix;

[0111] S5: Calculating the Euclidean distance Q between each cleaning quality index and the positive ideal value and the negative ideal value i1 , Q i2 ;

[0112]

[0113] Where D1 represents the positive ideal value; D2 represents the negative ideal value;

[0114] S6: Based on the Euclidean distance between the positive ideal value and the negative ideal value, the relative fit P of each cleaning cleanliness index is calculated. i1 ,wherein the relative fit is negatively correlated with the deviation of cleaning quality;

[0115] S7, normalizing the relative fit and mapping it to [0, 1] to classify the ultrasonic cleaning quality status;

[0116] S8. Based on the comparison between the normalized numerical value and the state classification value, the ultrasonic cleaning quality state is divided into N1 and N2, where N1 is a qualified cleaning state and N2 is an unqualified cleaning state; the cleaning quality state is predicted and fed back.

[0117] Finally, the undetermined target is transferred to the data twin model, and the cleaned information of the twin is verified for consistency. This process is repeated to find the optimal solution. The entity layer information and the twin layer information are continuously optimized and updated through the Metaverse database.

[0118] like Figure 4As shown in the figure, the schematic diagram of the ultrasonic cleaning structure based on the metaverse and digital twin is shown. The modeling process includes:

[0119] S1: Based on the information collected by the sensor, obtain geometric data and build a geometric model based on the turbocharger and ultrasonic cleaning machine;

[0120] S2: The system constructs a physical model of the turbocharger's physical properties and performance parameters based on the collected data;

[0121] S3: The system builds a data model based on the real-time cleaning data of the turbocharger components and the historical storage data;

[0122] S4: Use sensors and historical data models to update geometric models, physical models, and data models in real time;

[0123] S5: Create a digital twin of the turbocharger model based on the updated geometric model, physical model, and data model.

[0124] like Figure 5 As shown in the figure, the digital twin virtual-reality interaction diagram, the specific implementation steps are as follows:

[0125] Step 1: Use 3D scanning, point cloud data and modeling tools to convert the data into a three-dimensional parametric model;

[0126] Step 2: Use Simulink and Modelica physical model software to model the physical properties and performance parameters of the entity, and analyze the coupling relationship between the stress, transmission, and vibration of the marine equipment parts and the ultrasonic cleaning machine generator;

[0127] Step 3: Analyze, learn, and predict data through a neural network model to build a real-time model dataset;

[0128] Step 4: Build a model based on the real-time data of the cleaning parts and cleaning machines as well as historical cleaning information to predict the cleaning performance indicators, and dynamically update the geometric data twin stage, physical parameter twin stage, and drive data twin stage to continuously map the status and performance of the twin parts.

[0129] like Figure 6 As shown in the figure, the ultrasonic cleaning process flow chart of traditional marine equipment mainly includes: empirical analysis of cleaning characteristics, manual determination of cleaning areas, manual definition of energy consumption control and manual control of operating status. The characteristics of the cleaning process include:

[0130] S1: Manually obtain basic information of the components to be cleaned and control parameters of the ultrasonic cleaning machine;

[0131] S2: Manually perform subjective analysis on the original part to be cleaned to determine the characteristic parameters of oil sludge and carbon deposits on the workpiece;

[0132] S3: Manually determine the relationship between the oil sludge and carbon deposit content of the workpiece to be cleaned and the control of the ultrasonic cleaning machine;

[0133] S4: Manually control component cleaning and machine operation status.

[0134] Traditional cleaning requires manual control of start and stop states, and energy consumption settings based on experience. Cleaned parts must be manually judged for cleanliness, with no fixed standard. Manual cleaning requires visual recognition of parameter information and the cleaning machine's control system parameters, making automated cleaning impossible.

[0135] like Figure 7 As shown in the figure, the ultrasonic cleaning twin model structure schematic diagram uses sensor technology to read the dimensions, weight, and fluorescence of the physical model and save them to the historical data model. Based on this historical information, the virtual model is optimized and rebuilt, thus enabling the Metaverse database to update and iterate the physical and twin layers, and finally display them within the overall Metaverse system.

[0136] The characteristics of the process of determining the optimal solution include: Figure 8 As shown, the system uses data to update the sludge cleaning rate in the layer The optimal solution for energy consumption is determined by the relationship with temperature, power, frequency and cleaning time.

[0137] like Figure 9 As shown in FIG, a flow chart of the system energy-saving control method is shown, and the process includes:

[0138] S1: Obtain basic information of the cleaning target and identify the cleaning area of the cleaning target to obtain a cleaning feature set;

[0139] S2: Obtain basic information, cleaning components, and control parameters of the ultrasonic cleaning machine, and perform cleaning characteristic analysis on the cleaning components and control parameters based on the basic information to determine the cleaning characteristic parameters;

[0140] S3: Based on the cleaning characteristic parameters, the cleaning parameters of the control energy consumption analysis, fitting the energy-saving relationship between the cleaning characteristic parameters;

[0141] S4: Matching the cleaning feature set with the cleaning feature parameters to determine a cleaning-component control relationship;

[0142] S5: Based on the cleaning - element control relationship, energy-saving relationship, energy-saving optimization, determine the control parameters of each element;

[0143] S6: Based on the control parameters of each component, a control instruction is generated and sent to the console to control the operation of the generator.

[0144] like Figure 10 、11 As shown in the schematic diagram of the geometric modeling of the ultrasonic generator's side distribution, the system controls the number and power of the first, second, third, fourth, and fifth ultrasonic vibrators on the side, 5 based on the optimal solution information. During the twinning process, the system updates the model, consistency verification module, and twin control relationship in real time based on the twin database, cloud database, and existing historical database.

Claims

1. The ultrasonic cleaning method for ship equipment based on metaverse and digital twin is characterized by: The following steps are involved: Step 1: The data collection server in the ultrasonic cleaning equipment layer collects the status data of the ultrasonic cleaning line, and the collected cleaning data is then transmitted to the middle layer through the transmission device to transmit the cleaning status information; Step 2: Access the real-time database in the web server through a browser; Step 3: The user remotely monitors ultrasonic cleaning through the digital twin model established based on the second-layer network framework data; Step 4: Use VR, AR, and MR extended reality technologies to integrate users with the ultrasonic cleaning twin system, and use virtual interaction fusion, virtual connection, and management technologies to map the physical ultrasonic cleaning model to the extended reality module. Step 5: Use 3D scanning, point cloud data and modeling tools to convert the part cleaning data into a three-dimensional parametric model; Step 6: Use Simulink and Modelica physical model software to model the physical properties and performance parameters of the entity, and analyze the coupling relationship between the stress, transmission, and vibration of the marine equipment parts and the ultrasonic cleaning machine generator; Step 7: Analyze, learn, and predict the parts cleaning data through a neural network model to build a real-time model data set; Step 8: Build a model based on the real-time data of the cleaning component and the cleaning machine, as well as historical cleaning information, to predict cleaning performance indicators. Dynamically update the geometric data twin stage, physical parameter twin stage, and drive data twin stage to continuously map the status and performance of the twin component. Step 9: Conduct digital twin 3D modeling of the turbocharger and ultrasonic cleaning machine; Step 10: Determine the consistency of the parts oil cleaning data; Step 11, feeding the above results back to the generator of the ultrasonic cleaning equipment; Step 12: Based on the feedback results, the number of generators turned on; Step 13: Store the cleaned parts data in the cloud database and update iteratively.

2. The ultrasonic cleaning method for ship equipment based on metaverse and digital twin according to claim 1 is characterized in that: In step 9, modeling includes the following sub-steps: S9.1: Based on the information collected by the sensors, obtain geometric data and construct a geometric model based on the turbocharger and ultrasonic cleaning machine; S9.2: The system constructs a physical model of the turbocharger's physical properties and performance parameters based on the collected data; S9.3: The system builds a data model based on the real-time cleaning data of the turbocharger components and the historical storage data; S9.4: Use sensors and historical data models to update geometric models, physical models, and data models in real time; S9.5: Create digital twins of the turbocharger and ultrasonic cleaning machine based on the updated geometric model, physical model, and data model.

3. The ultrasonic cleaning method for ship equipment based on metaverse and digital twin according to claim 1 is characterized in that: In step 10, determining consistency specifically includes the following sub-steps: S10.1: Predetermine the monitoring time nodes and cleaning cleanliness indicators of the cleaning object in the ultrasonic cleaning equipment and construct the initial judgment matrix C1; According to the predetermined cleaning object, for the i-th (i= 1, 2, ..., A) monitoring time section and the j-th (j= 1, 2, ..., B) cleaning cleanliness index, the evaluation value is X ij , forming the initial judgment matrix C1: ； S10.2: Standardize the initial judgment matrix to obtain the standard value b of the jth cleaning quality index at the i-th monitoring time node ij , the standardized decision matrix is C1 ’ : ； ； Where, is the evaluation value, m ij represents the minimum value of the jth processing quality index of the i-th monitoring node; M j represents the maximum value of the jth processing quality index; m j Indicates the maximum value of the j-th processing quality index; [*] sxt Indicates that the * matrix is standardized; S10.3: Combine the standardized decision matrix with the weight matrix to obtain the weighted judgment matrix T; ； Where W represents the weight matrix, C1 ’ is a standardized decision matrix; S10.4: Obtain the positive ideal value D1 and negative ideal value D2 of the evaluation target from the weighted judgment matrix; ； ； In the formula, the profitability index set D1 is the optimal solution of the i-th index value; the loss index set D2 is the worst solution of the i-th index value; n is the total number of processing quality indicators; d ij represents the element of the jth processing quality index of the i-th monitoring node in the weighted judgment matrix; S10.5: Calculate the Euclidean distance Q between each cleaning quality index and the positive ideal value and the negative ideal value respectively i1 , Q i2 ; ； Where D1 represents the positive ideal value; D2 represents the negative ideal value; S10.6: Based on the Euclidean distance between the positive ideal value and the negative ideal value, calculate the relative fit P of each cleaning cleanliness index i1 ,wherein the relative fit is negatively correlated with the deviation of cleaning quality; S10.

7. Normalize the relative fit and map it to [0, 1] to classify the ultrasonic cleaning quality status; S10.

8. Based on the comparison between the normalized numerical value and the state classification value, the ultrasonic cleaning quality state is divided into N1 and N2, where N1 is a qualified cleaning state and N2 is an unqualified cleaning state; the cleaning quality state is predicted and feedback is given.

Images