A fuel supply drive control method for low flash point fuel ship training

Abstract

The application discloses a fuel supply driving control method for low-flashpoint fuel ship training, relates to the field of fuel supply, and aims to improve the adaptability of fuel supply simulation and the practicality of training.

Description

Technical Field

[0001] This invention relates to the field of fuel supply technology, and more particularly, to a fuel supply drive control method for training low flash point fuel ships. Background Technology

[0002] The fuel supply drive control method for low flash point fuel ship training, as the core supporting module of the low flash point fuel ship operation training system, directly determines the degree of practical simulation, operational safety, and training objective achievement rate of the training by its control accuracy, safety adaptability, and dynamic adjustment capability. It has become a key technical tool to promote the transformation of ship fuel training towards intelligence and precision and to achieve safe training for high-risk fuel operations. However, the flammable and explosive characteristics of low flash point fuels, the dynamic coupling of multi-dimensional operating data, the performance degradation of long-cycle simulation equipment, and the complexity and variability of ship training scenarios are the core constraints affecting the performance of fuel supply simulation control. It is dynamically affected by multiple factors such as differences in fuel physicochemical properties, fluctuations in training operating conditions, different equipment wear rates, and frequent scenario switching.

[0003] Fuel supply-driven control technology based on full-cycle data acquisition and intelligent algorithm optimization is not only a key means to solve the problems of insufficient accuracy in supply simulation adaptation and imbalance in safety risk control caused by the coupling of multiple factors, but also directly ensures the safety and practicality of training for low flash point fuel ships. By dynamically adjusting supply parameters and control strategies, it reduces simulation operation risks and improves training effectiveness. Furthermore, by exploring the laws of fuel supply simulation, it optimizes control strategies and provides a basis for the model construction and data system improvement of the next generation of low flash point fuel ship operation training systems.

[0004] However, existing fuel supply control methods for low flash point fuel ship training often rely on empirical parameter settings, neglecting the coupling value of system operating status data, fuel characteristic data, and training scenario data. These methods fail to cover the temporal variation patterns of low flash point fuel supply, the long-term impact of equipment degradation, and the complex logic of scenario switching. This results in a one-sided approach to supply control, emphasizing parameter setting while neglecting dynamic adaptation. Furthermore, existing methods often employ static control models or single algorithms for parameter matching, failing to address the issues of standardized alignment and accurate feature extraction from multi-source heterogeneous data. This leads to loose data feature correlations and susceptibility to interference. Additionally, the lack of a closed-loop iterative mechanism based on simulated operating data prevents dynamic optimization of the control benchmark based on equipment degradation and scheme effectiveness. Long-term operation can lead to defects such as simulation accuracy decay and safety threshold failure, making it difficult to meet the high-precision, high-safety supply simulation requirements for low flash point fuel ship training.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] To address the problems in related technologies, this invention proposes a fuel supply drive control method for training low flash point fuel ships, thereby overcoming the aforementioned technical problems in existing related technologies.

[0007] To achieve the above objectives, the specific technical solution adopted by the present invention is as follows: A fuel supply drive control method for training on low flashpoint fuel ships includes the following steps: S1. Collect system simulation operation status data, fuel characteristic simulation adaptation data, ship training scenario data and safety condition simulation response data throughout the entire life cycle of low flash point fuel supply simulation in ship training equipment, and build a hierarchical system simulation operation archive and a multi-fuel characteristic feature library. S2. Based on the multi-fuel characteristic feature library, set the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark. Label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrate them to form a multi-dimensional fuel association dataset. As a preferred embodiment, the steps of setting system simulation operation safety thresholds, fuel supply simulation adaptation standards, and operating condition conflict judgment benchmarks based on a multi-fuel characteristic feature library, and then labeling and classifying system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrating them into a multi-dimensional fuel-related dataset, include the following: S21. Extract historical safe operation data, fault condition record data and typical condition adaptation data from the hierarchical system simulation operation archive to form a basic dataset; S22. Combine the basic dataset with the low flash ignition physicochemical characteristic parameters, safety characteristic parameters and supply adaptation characteristic parameters retrieved from the multi-fuel characteristic feature library, and set the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark. S23. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark, establish a three-level labeling rule, and label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive respectively. As a preferred embodiment, the establishment of a three-level labeling rule based on the system simulation operation safety threshold, fuel supply simulation adaptation standard, and operating condition conflict judgment benchmark, and the labeling and classification of system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, includes the following steps: S231. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark, establish a three-level labeling rule, which includes the first level safety compliance labeling rule, the second level fuel compatibility labeling rule and the third level working condition conflict risk labeling rule, and clarify the labeling content and labeling standard corresponding to each level of rule; S232. According to the established three-level annotation rules, the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data in the hierarchical system simulation operation archive are annotated in sequence, and the annotation results are summarized and classified according to the annotation attributes to form a dataset with annotation labels.

[0008] S24. Perform consistency verification on the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data that have completed the three-level labeling, and integrate them according to the time dimension and the operating condition dimension to form a multi-dimensional fuel association dataset.

[0009] S3. Preprocess the multi-dimensional fuel-related dataset, and use the working condition feature clustering algorithm and time series pattern mining algorithm according to the differences of the data groups to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. Then, combine the multi-fuel characteristic feature library to build a multi-group data weight dynamic allocation model and a fuel supply simulation safety control model. As a preferred embodiment, the preprocessing of the multi-dimensional fuel-related dataset, and the use of operating condition feature clustering algorithms and time series pattern mining algorithms based on the differences in data groups to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns, combined with a multi-fuel characteristic feature library to build a multi-set data weight dynamic allocation model and a fuel supply simulation safety control model, includes the following steps: S31. Preprocess the multi-dimensional fuel association dataset, remove outliers, redundant values ​​and missing values ​​from the data, and perform standardization transformation on different types of data to complete the data dimension alignment and obtain the standardized multi-dimensional fuel association dataset. S32. Based on the grouping differences of the multi-dimensional fuel association dataset after standardization, the working condition feature clustering algorithm and the time series operation related data in the multi-dimensional fuel association dataset are used to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. As a preferred embodiment, the step of extracting the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns from the fuel supply adaptation-related data and time-series operation-related data in the multi-dimensional fuel association dataset based on the grouping differences after standardization processing includes the following steps: S321. Group the standardized multi-dimensional fuel association dataset and clarify the data group related to fuel supply adaptation and the data group related to time-series operation. S322. Use the operating condition feature clustering algorithm to perform cluster analysis on the fuel supply adaptation related data group, obtain the supply characteristic parameters related to fuel supply adaptation and the correlation between parameters, and use the time series pattern mining algorithm to perform time series analysis on the time series operation related data group, extract the time series change pattern of system simulation operation and the operating condition switching correlation features. As a preferred embodiment, the step of using a working condition feature clustering algorithm to perform cluster analysis on the fuel supply adaptation-related data group, obtaining the supply characteristic parameters related to fuel supply adaptation and the correlation between parameters, and using a time series pattern mining algorithm to perform time series analysis on the time series operation-related data group, extracting the time series change patterns and working condition switching correlation features of the system simulation operation, includes the following steps: S3221. Identify the core parameter dimensions of the fuel supply adaptation related data group, remove redundant parameters that are not related to fuel supply adaptation from the data group, and obtain the purified fuel supply adaptation data group. S3222. The working condition feature clustering algorithm is used to perform cluster analysis on the purified fuel supply adaptation data group, and the fuel supply adaptation parameter clusters are divided to obtain the supply feature parameters and the correlation between parameters in each cluster. S3223. Define the time series parameter range of the time series operation related data group, and calibrate the time dimension of the time series operation related data group to align the time dimension of the data group. S3224. Use a time-series pattern mining algorithm to perform time-series analysis on the time-series operation-related data group after time-dimensional calibration, and extract the time-series change patterns of the system simulation operation and the parameter correlation characteristics during operating condition switching.

[0010] S323. Integrate supply characteristic parameters, relationships between parameters, time-series change patterns, and operating condition switching characteristics, and refine them to obtain fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns.

[0011] S33. Retrieve the low flash ignition physicochemical characteristics, safety characteristics, and supply adaptation characteristics from the multi-fuel characteristic feature library. Combine the fuel supply simulation adaptation rhythm, core safety risk simulation points, and working condition switching simulation adaptation rules to build a multi-set data weight dynamic allocation model. S34. The core inputs are the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation rules. Combined with a multi-fuel characteristic feature library, the correlation mapping relationship between fuel characteristics, operating conditions, and safety control is constructed to build a fuel supply simulation safety control model.

[0012] S4. Introduce the multi-set data weight dynamic allocation model into the multi-set data collaborative multi-objective optimization algorithm, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and perform control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range. As a preferred embodiment, the step of introducing a multi-set data weight dynamic allocation model into a multi-set data collaborative multi-objective optimization algorithm, defining the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and performing control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range includes the following steps: S41. Introduce the dynamic weight allocation model of multiple sets of data into the multi-objective optimization algorithm of multi-set data, clarify the constraints of the data weight allocation rules on the optimization algorithm, and perform collaborative adaptation between the dynamic weight allocation model of multiple sets of data and the multi-objective optimization algorithm of multi-set data. S42. Combining the system simulation operation safety threshold, fuel supply simulation adaptation standard, and low flash ignition physicochemical characteristic parameters and safety characteristic parameters in the multi-fuel characteristic feature library, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range. As a preferred embodiment, the step of defining the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range by combining the system simulation operation safety threshold, fuel supply simulation adaptation standard, and low flash ignition physicochemical characteristic parameters and safety characteristic parameters in the multi-fuel characteristic feature library includes the following steps: S421. Retrieve the system simulation operation safety threshold and fuel supply simulation adaptation standard, simultaneously extract the physicochemical and safety characteristic parameters of low flash point fuels from the multi-fuel characteristic feature library, conduct correlation analysis on various parameters, clarify the constraint logic and influence weight between different parameters, and establish a parameter correlation matrix. S422. Based on the parameter correlation matrix, combined with the low flash ignition culinary and chemical characteristics parameters and safety characteristics parameters, define the upper and lower limits and judgment criteria of the system simulation operation safety tolerance threshold, and delineate the range boundary and graded adjustment gradient of the supply parameter simulation adjustment range, thus forming the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range.

[0013] S43. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, and based on a multi-set data collaborative multi-objective optimization algorithm, perform multiple rounds of matching and optimization of fuel supply-related control parameters, and synchronously record the parameter adaptation effect of each round of optimization. As a preferred embodiment, the method of using the system's simulated operational safety tolerance threshold as a constraint, the simulated adjustment range of supply parameters as a range, and a multi-objective optimization algorithm based on multiple sets of data to perform multiple rounds of matching and optimization of fuel supply-related control parameters, while simultaneously recording the parameter adaptation effect of each round of optimization, includes the following steps: S431. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, determine the optimization basis conditions for the multi-set data collaborative multi-objective optimization algorithm. S432. Based on the optimization conditions, a multi-set data collaborative multi-objective optimization algorithm is used to perform multi-round matching optimization of fuel supply-related control parameters. In each round of optimization, the parameter combination and parameter adaptation effect corresponding to that round of optimization are recorded synchronously to form a multi-round optimization record.

[0014] S44. Screen and verify the results of multiple rounds of optimization, and remove parameter combinations that do not meet the fuel supply simulation adaptation standards and safety tolerance thresholds to obtain the multi-version fuel supply simulation benchmark parameters and the corresponding dynamic adaptation range of safety control.

[0015] S5. Substitute the multi-version fuel supply simulation benchmark parameters and the dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme. Simulate and implement the personalized fuel supply drive control scheme in the ship training equipment and collect real-time data of the simulation scheme and equipment simulation attenuation data. As a preferred embodiment, the step of substituting multi-version fuel supply simulation baseline parameters and dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme, and simulating the implementation of this personalized fuel supply drive control scheme in the ship training equipment, and collecting real-time data of the simulation scheme and equipment simulation attenuation data, includes the following steps: S51. Retrieve the baseline parameters of multiple versions of fuel supply simulation and the dynamic adaptation range of safety control, and substitute them one by one into the fuel supply simulation safety control model to perform parameter initialization and debugging. S52. After initializing and debugging the parameters, the calculation output of the fuel supply simulation safety control model is combined with the ship training scenario data to generate a personalized fuel supply simulation drive control scheme. S53. Based on the personalized fuel supply simulation drive control scheme, start the simulation implementation in the low flash point fuel supply simulation device of the ship training equipment, and simultaneously monitor the operation status of the entire process of scheme implementation. S54. During the simulation implementation, real-time operating parameters and safety response signals related to fuel supply simulation are collected as real-time data of the simulation scheme. At the same time, the wear and performance degradation parameters of each simulated component of the ship training equipment are collected as equipment simulation degradation data.

[0016] S6. Integrate the real-time data of the simulation scheme and the equipment simulation attenuation data to form a control scheme effectiveness simulation evaluation report and a long-term simulation operation impact report. Based on these two reports, iteratively update the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark.

[0017] The beneficial effects of this invention are as follows: 1. This invention achieves a panoramic and precise integration of system operation status, fuel characteristic adaptation, safety condition response, and training scenario data through multi-dimensional data acquisition throughout the entire lifecycle of low flashpoint fuel supply simulation and the construction of a dual-database system simulation operation archive and a multi-fuel characteristic feature library. Combined with a refined data governance mechanism involving three-level annotation rules and consistency verification, this overcomes the shortcomings of traditional ship training fuel supply simulation data acquisition, such as one-sidedness, disordered data classification, and vague feature characterization. Furthermore, through precise division of labor in data preprocessing and data group differentiation algorithms, it achieves targeted extraction of fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns, replacing the traditional single-algorithm extensive feature extraction method. The feature extraction mode enhances the relevance and effectiveness of the core feature set. It relies on the construction of a dual model, which combines a dynamic weight allocation model for multiple sets of data with a fuel supply simulation safety control model. Combined with the deep integration of a multi-objective optimization algorithm for multiple sets of data, it achieves the scientific definition of the system's simulated operation safety tolerance threshold and the simulated adjustment range of supply parameters, as well as the precise matching of control parameters. This solves the problems of insufficient supply adaptability and imbalance in safety risk control caused by traditional single-objective optimization and static parameter settings in simulation control. As a result, the generated personalized fuel supply simulation-driven control scheme can accurately match the differentiated needs of different low flash point fuel types and different ship training scenarios, thereby improving the adaptability of fuel supply simulation and the practicality of training.

[0018] 2. This invention achieves dynamic adaptive adjustment of the system's simulation operation safety threshold, fuel supply simulation adaptation standards, and operating condition conflict judgment benchmarks through a personalized fuel supply simulation-driven control scheme implementation and synchronous monitoring of real-time simulation data and equipment simulation attenuation data. It combines a dual-report collaborative evaluation system—a control scheme effectiveness simulation evaluation report and a long-term simulation operation impact report—and constructs a closed-loop iterative mechanism. This addresses the shortcomings of traditional ship training fuel supply simulations, which focus only on short-term parameter matching, neglecting long-term equipment attenuation and adaptation drift, leading to decreased simulation accuracy. Furthermore, through an iterative parameter archiving and update mechanism, it continuously enriches the hierarchical system simulation operation archive and multi-fuel characteristic feature library, forming a data-driven continuous optimization capability. This ensures the stable adaptation of the fuel supply simulation control scheme to the entire training process for low-flash-point fuel ships. Finally, through an end-to-end technical link closed loop, combined with a modular collaborative design of data acquisition, model building, scheme implementation, and iterative optimization, it ensures the operability and flexibility of the technical solution. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a fuel supply drive control method for training low flashpoint fuel ships according to an embodiment of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0022] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0023] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the fuel supply drive control method for low flashpoint fuel ships according to an embodiment of the present invention includes the following steps: S1. Collect system simulation operation status data, fuel characteristic simulation adaptation data, ship training scenario data and safety condition simulation response data throughout the entire life cycle of low flash point fuel supply simulation in ship training equipment, and build a hierarchical system simulation operation archive and a multi-fuel characteristic feature library. Specifically, based on equipment such as the simulation control console, electrical control system, and LPG refueling and supply simulation device, four types of core data are collected throughout the entire lifecycle of low flash point fuel supply simulation. Through industrial control computers, PLCs, and pressure and temperature sensors, system simulation operation status data such as tank pressure, vaporization temperature, and valve status are collected. Combined with the adaptation parameters of the LPG refueling and supply simulation device, fuel characteristic simulation adaptation data is collected. Operation and running data for different ship training scenarios, such as three filling methods and main and auxiliary engine gas supply, are recorded as training scenario data. Furthermore, through combustible gas detection sensors, alarm systems, and nitrogen purging devices, safety condition simulation response data such as audible and visual alarms, valve closure, and purging and inerting are collected. Based on the above collected data, a hierarchical system simulation operation archive is built, archiving the entire lifecycle operation data according to levels such as normal operation, fault alarm, and different training scenarios. Integrating low flash point ignition physicochemical characteristics, supply adaptation parameters, and equipment adaptation standards, a multi-fuel characteristic feature library is built to provide data support for subsequent system simulation operation and safety control.

[0024] S2. Based on the multi-fuel characteristic feature library, set the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark. Label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrate them to form a multi-dimensional fuel association dataset. In this embodiment of the invention, the steps of setting system simulation operation safety thresholds, fuel supply simulation adaptation standards, and operating condition conflict judgment benchmarks based on a multi-fuel characteristic feature library, and then labeling and classifying system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrating them to form a multi-dimensional fuel association dataset, include the following steps: S21. Extract historical safe operation data, fault condition record data and typical condition adaptation data from the hierarchical system simulation operation archive to form a basic dataset; Specifically, based on the established hierarchical system simulation operation archive, and combined with equipment such as the simulation control console, liquefied gas filling and supply simulation device, sensors, and alarm systems in the handover documents, three types of data are specifically extracted to form a basic dataset: First, historical safe operation data, extracting system operation parameters and equipment stable operation records that meet safety thresholds, such as tank pressure, vaporization temperature, and valve opening, throughout the simulation cycle under normal operating conditions in the archive; second, fault condition record data, extracting equipment response data, fault trigger parameters, and emergency response records for various fault scenarios such as fuel leakage and pressure exceeding limits under the fault alarm level; and third, typical operating condition adaptation data, extracting fuel supply adaptation parameters, operation process data, and system adaptation response records for typical training operating conditions such as the three filling methods and main and auxiliary machine gas supply in the handover documents under different training scenario levels. After preliminary sorting and deduplication of the three types of data, they are integrated to form the basic dataset.

[0025] S22. Combine the basic dataset with the low flash ignition physicochemical characteristic parameters, safety characteristic parameters and supply adaptation characteristic parameters retrieved from the multi-fuel characteristic feature library, and set the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark. Specifically, the core parameters of low flash point fuels, including physicochemical properties, safety characteristics, and supply compatibility, are retrieved from a multi-fuel characteristic feature library. These parameters are then deeply integrated with the existing basic dataset. Relying on the hardware support of the training equipment, such as sensors and simulation control systems, as outlined in the handover documents, data analysis is used to correlate the two types of data. Combined with historical safe operation data from the basic dataset, the system's simulated operation safety threshold is defined, clarifying the safe range of key parameters such as tank pressure and vaporization temperature. Furthermore, by combining typical operating condition compatibility data with fuel supply compatibility parameters, fuel supply simulation compatibility standards are set, standardizing the supply parameter compatibility requirements under different training conditions. Based on fault condition record data and fuel safety characteristics, a condition conflict judgment benchmark is established, clarifying the judgment criteria and levels for condition parameter conflicts under different training scenarios.

[0026] S23. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark, establish a three-level labeling rule, and label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive respectively. In this embodiment of the invention, the step of establishing a three-level labeling rule based on the system simulation operation safety threshold, fuel supply simulation adaptation standard, and operating condition conflict judgment benchmark, and labeling and classifying the system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, includes the following steps: S231. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark, establish a three-level labeling rule, which includes the first level safety compliance labeling rule, the second level fuel compatibility labeling rule and the third level working condition conflict risk labeling rule, and clarify the labeling content and labeling standard corresponding to each level of rule; Specifically, based on the established system simulation operation safety thresholds, fuel supply simulation adaptation standards, and operating condition conflict judgment benchmarks, and combined with the data annotation requirements for low flash point fuel ship training, a three-level annotation rule is established. This clarifies the annotation content and standards corresponding to each level of the rule, ensuring standardized and accurate annotation. The first level is the safety compliance annotation rule, focusing on system simulation operation status data and safety operating condition simulation response data. The annotation standard is to determine whether the operating parameters meet the safety thresholds and whether the safety response is timely and up-to-date. Compliant rules are marked with corresponding compliance indicators, while non-compliant rules are marked with the abnormal type and specific exceedance or inaccuracy. The second level is the fuel compatibility annotation rule, targeting fuel characteristic simulation adaptation data. The annotation standard is to determine the degree of matching between the fuel's physicochemical, safety, and supply adaptation parameters and the simulated supply requirements. If compatible, an adaptation indicator is marked; if incompatible, the deviation range and reason are indicated. The third level is the operating condition conflict risk annotation rule, focusing on ship training scenario data. The annotation standard is based on the conflict judgment benchmark, dividing the conflict into three levels: no conflict, minor conflict, and severe conflict, with corresponding low, medium, and high risk indicators, clearly defining the core conflict parameters.

[0027] S232. According to the established three-level annotation rules, the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data in the hierarchical system simulation operation archive are annotated in sequence, and the annotation results are summarized and classified according to the annotation attributes to form a dataset with annotation labels.

[0028] Specifically, following the established three-level annotation rules and the hierarchical division of the hierarchical system simulation operation archive, the system simulation operation status data, fuel characteristic simulation adaptation data, and safety condition simulation response data in the archive are annotated sequentially. During annotation, both data correlation and rule correspondence are considered. First, based on the first-level safety compliance rules, the system simulation operation status data and safety condition simulation response data are annotated for compliance. Then, based on the second-level fuel adaptability rules, the fuel characteristic simulation adaptation data is annotated for adaptability. Finally, based on the third-level operating condition conflict risk rules, the operating condition conflict risk levels of the three types of data are supplemented with annotations. After annotation is completed, the annotation results of all data are summarized, the annotation label types are sorted, and then classified and organized according to annotation attributes (safety compliance, fuel adaptability, and operating condition conflict risk). Duplicate and abnormal entries are removed, and the label format is standardized, ultimately forming a dataset with clear annotation labels.

[0029] S24. Perform consistency verification on the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data that have completed the three-level labeling, and integrate them according to the time dimension and the operating condition dimension to form a multi-dimensional fuel association dataset.

[0030] Specifically, consistency verification is conducted on the system simulation operation status data, fuel characteristic simulation adaptation data, and safety condition simulation response data that have completed three levels of labeling. The focus is on verifying the matching of the labeling tags to the data itself, identifying duplicate labels, contradictory labels, and anomalies where labels do not match the actual data. Problems identified are corrected and improved to ensure accurate, complete, and unambiguous labeling. After successful verification, the data is integrated along time and operating condition dimensions. In the time dimension, based on the time sequence of the entire low flash point fuel supply simulation cycle, the three types of data at the same time point are correlated to form a time-series coherent data chain. In the operating condition dimension, combined with various typical operating conditions in ship training, the three types of labeled data are categorized according to the corresponding operating conditions, clarifying the correspondence between different operating conditions and data characteristics. During the integration process, data dimension alignment is performed simultaneously, data association identifiers are supplemented, and redundant and invalid information is eliminated, ultimately forming a fuel-related dataset covering multiple dimensions such as time, operating condition, and labeling attributes. S3. Preprocess the multi-dimensional fuel-related dataset, and use the working condition feature clustering algorithm and time series pattern mining algorithm according to the differences of the data groups to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. Then, combine the multi-fuel characteristic feature library to build a multi-group data weight dynamic allocation model and a fuel supply simulation safety control model. In this embodiment of the invention, the preprocessing of the multi-dimensional fuel-related dataset, and the extraction of fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns based on the differences in data groups using operating condition feature clustering algorithms and time series pattern mining algorithms, combined with the construction of multiple data weight dynamic allocation models and fuel supply simulation safety control models using a multi-fuel characteristic feature library, includes the following steps: S31. Preprocess the multi-dimensional fuel association dataset, remove outliers, redundant values ​​and missing values ​​from the data, and perform standardization transformation on different types of data to complete the data dimension alignment and obtain the standardized multi-dimensional fuel association dataset. Specifically, the multi-dimensional fuel association dataset is first purified by using statistical analysis to identify and remove outliers that exceed reasonable ranges, and to delete duplicate records and irrelevant redundant values. For missing values, appropriate completion methods are used based on the characteristics of the dataset to ensure data integrity and unbiasedness. Subsequently, for different types of data in the dataset, such as system simulation operation status, fuel characteristic simulation adaptation, and safety condition simulation response, corresponding standardization transformation methods are used to eliminate the influence of dimensions and unify the data scale due to differences in their units of measurement and value ranges. After the transformation, data dimension alignment is further performed to adjust the dimensional specifications of different types of data to ensure that various types of data can be analyzed collaboratively, ultimately resulting in a standardized multi-dimensional fuel association dataset.

[0031] S32. Based on the grouping differences of the multi-dimensional fuel association dataset after standardization, the working condition feature clustering algorithm and the time series operation related data in the multi-dimensional fuel association dataset are used to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. In this embodiment of the invention, the step of extracting the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns from the fuel supply adaptation-related data and time-series operation-related data in the multi-dimensional fuel association dataset based on the grouping differences after standardization processing includes the following steps: S321. Group the standardized multi-dimensional fuel association dataset and clarify the data group related to fuel supply adaptation and the data group related to time-series operation. Specifically, the core principles of grouping were first clarified. Based on the actual needs of training on low-flashpoint fuel ships, and using the core content reflected in the data as the basis for division, the standardized dataset was systematically reviewed. Secondly, two data groups were precisely divided: a fuel supply adaptation-related data group, which focuses on collecting various data directly related to fuel supply adaptation, including fuel physicochemical properties, supply parameter adaptation standards, operating condition adaptation parameters, and related labeled data; and a time-series operation-related data group, which focuses on the time-series characteristics of system operation, collecting system simulation operation status data, safety condition simulation response time-series data, and corresponding time node labeling information recorded along the time dimension. During the division process, data correlation was verified, redundant cross-boundary data was eliminated, and unique identifiers for the two data groups were clearly defined to ensure clear, non-overlapping, and complete grouping. Finally, the fuel supply adaptation-related data group and the time-series operation-related data group were clearly defined. S322. Use the operating condition feature clustering algorithm to perform cluster analysis on the fuel supply adaptation related data group, obtain the supply characteristic parameters related to fuel supply adaptation and the correlation between parameters, and use the time series pattern mining algorithm to perform time series analysis on the time series operation related data group, extract the time series change pattern of system simulation operation and the operating condition switching correlation features. In this embodiment of the invention, the step of using a working condition feature clustering algorithm to perform cluster analysis on the fuel supply adaptation-related data group to obtain the supply characteristic parameters related to fuel supply adaptation and the correlation between parameters, and using a time series pattern mining algorithm to perform time series analysis on the time series operation-related data group to extract the time series change pattern and working condition switching correlation features of the system simulation operation includes the following steps: S3221. Identify the core parameter dimensions of the fuel supply adaptation related data group, remove redundant parameters that are not related to fuel supply adaptation from the data group, and obtain the purified fuel supply adaptation data group. Specifically, in conjunction with the training needs of low flash point fuel ships, focusing on the core objective of fuel supply adaptation, a comprehensive review of the standardized grouped fuel supply adaptation-related data sets was conducted to clarify their core parameter dimensions. These core dimensions mainly include three categories: fuel characteristic adaptation parameters (such as the physicochemical characteristics of low flash point fuels and supply adaptation parameters), operating condition adaptation-related parameters (such as supply adaptation standards under different training conditions and operating condition conflict adaptation parameters), and adaptation effect-related parameters (such as fuel adaptation-related annotation data). Subsequently, the data sets were checked one by one against the core parameter dimensions to identify and eliminate redundant parameters unrelated to fuel supply adaptation, including non-adaptation-related annotation information, irrelevant data unrelated to supply adaptation, and duplicate and redundant auxiliary parameters. After elimination, the remaining data was checked and verified to ensure that the data all focused on the core of fuel supply adaptation, with no missing key parameters and no residual redundant information, ultimately resulting in a purified fuel supply adaptation data set.

[0032] S3222. The working condition feature clustering algorithm is used to perform cluster analysis on the purified fuel supply adaptation data group, and the fuel supply adaptation parameter clusters are divided to obtain the supply feature parameters and the correlation between parameters in each cluster. Specifically, the first step is to confirm that the purified fuel supply matching data set is free of redundancy and anomalies, and that the parameter dimensions are consistent, laying the foundation for cluster analysis. Then, the working condition feature clustering algorithm is applied, and the clustering parameters (such as the number of clusters and the number of iterations) are reasonably set in combination with the fuel supply matching requirements of low flash point fuel ship training. The similarity of supply feature parameters is used as the core clustering basis to start cluster analysis on the data set.

[0033] During the analysis, the algorithm parameters are adjusted in real time to ensure that the clustering results fit the actual fuel supply adaptation scenarios. After the clustering is completed, different fuel supply adaptation parameter clusters are divided according to the parameter characteristics and adaptation condition correlation of each cluster. Each cluster corresponds to a typical fuel supply adaptation condition. Then, each cluster is analyzed one by one to extract the core supply characteristic parameters within the cluster. Correlation analysis is used to explore and sort out the inherent relationship between each parameter, clarify the parameter influence weight and interaction logic, and finally obtain the supply characteristic parameters within each cluster and the relationship between parameters.

[0034] S3223. Define the time series parameter range of the time series operation related data group, and calibrate the time dimension of the time series operation related data group to align the time dimension of the data group. Specifically, considering the full-cycle duration of the low flash point fuel supply simulation, data sampling frequency, and characteristics of ship training conditions, the core timing attributes of the relevant data sets were identified, the range of timing parameters was clarified, the start and end times of the data (simulation start time and simulation end time) were defined, a unified time sampling interval was determined, and abnormal timing data exceeding this range were eliminated to ensure that the timing parameters covered the entire simulation process. Subsequently, using the established unified time benchmark as the standard, the time dimension of the data set was aligned and calibrated, and the timestamps of all timing data were checked, time deviations were corrected, data with different sampling frequencies were adjusted to a unified time interval, corresponding data for missing time nodes were supplemented, and redundant data with duplicate timestamps were deleted to ensure that the timing data of the system simulation operation status and safety condition simulation response at the same time node are accurately matched and the timing is consistent. Finally, the time dimension calibration of the relevant data sets was completed.

[0035] S3224. Use a time-series pattern mining algorithm to perform time-series analysis on the time-series operation-related data group after time-dimensional calibration, and extract the time-series change patterns of the system simulation operation and the parameter correlation characteristics during operating condition switching.

[0036] Specifically, based on a time-coherent and unbiased dataset calibrated along the time dimension, and combined with the full-cycle characteristics of low flashpoint fuel supply simulation, the algorithm parameters for time-series pattern mining are reasonably set. The analysis focuses on core time-series data such as the system's simulated operating status and safety condition simulation response. The algorithm then mines the dynamic fluctuation trends, stable operating ranges, and inflection points of various parameters as the simulation progresses, extracting the time-series change patterns of the system's simulated operation and clarifying the operational characteristics of parameters at different time periods. Simultaneously, for operating condition switching scenarios, the algorithm accurately locates key switching time nodes, analyzes the abrupt changes and connection logic of various time-series parameters before and after the switch, mines the inherent correlations and influence weights between parameters during operating condition switching, clarifies the correspondence between different operating condition switches and parameter changes, and after analysis, summarizes and refines the results, eliminating invalid information to ensure that the patterns and characteristics align with actual training scenarios.

[0037] S323. Integrate supply characteristic parameters, relationships between parameters, time-series change patterns, and operating condition switching characteristics, and refine them to obtain fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns.

[0038] Specifically, based on the clustering and time-series analysis results above, four types of core information are first systematically integrated, including supply characteristic parameters, relationships between parameters, time-series change patterns, and operating condition switching characteristics. The inherent relationships among these types of information are then analyzed, information consistency is verified, and redundant or contradictory content is eliminated, forming a complete set of multi-dimensional characteristic information. Subsequently, based on the actual needs of training for low-flashpoint fuel ships, the integrated information is specifically refined. Combining supply characteristic parameters and time-series change patterns, the fuel supply simulation adaptation rhythm is extracted, clarifying the dynamic adaptation logic of fuel supply parameters under different time periods and operating conditions. Simultaneously, combining the relationships between parameters and time-series change patterns, core safety risk simulation points are identified, pinpointing key parameters and change nodes that are prone to safety hazards. Combining operating condition switching characteristics and relationships between parameters, the operating condition switching simulation adaptation pattern is extracted, clarifying the parameter connection standards and adaptation logic during different operating condition switches. After the extraction is completed, the results are analyzed and integrated to ensure that the patterns accurately match the actual supply simulation scenario, ultimately yielding the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns.

[0039] S33. Retrieve the low flash ignition physicochemical characteristics, safety characteristics, and supply adaptation characteristics from the multi-fuel characteristic feature library. Combine the fuel supply simulation adaptation rhythm, core safety risk simulation points, and working condition switching simulation adaptation rules to build a multi-set data weight dynamic allocation model. Specifically, firstly, three core parameters of low flash point fuels—physicochemical properties, safety properties, and supply compatibility—are precisely retrieved from the established multi-fuel characteristic database. These parameters are then verified for completeness and accuracy, ensuring a precise match between the parameters and the low flash point fuel type and the requirements of ship training equipment. Subsequently, these three parameters are deeply integrated with the previously extracted fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation rules. The inherent correlation between various parameters and the three core rules is analyzed to clarify the impact of different parameters on fuel supply adaptation, safety risk prevention and control, and operating condition switching adaptation. The core basis for weight allocation is defined, and based on the correlation analysis results and the safety-first and efficient adaptation requirements of low flash point fuel ship training, dynamic weight allocation rules are set. Multiple sets of data weight dynamic allocation models are built to ensure that the model can dynamically adjust the weight ratio of various data according to changes in fuel characteristics, safety risk levels, and operating condition switching scenarios, ultimately completing the model construction.

[0040] S34. The core inputs are the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation rules. Combined with a multi-fuel characteristic feature library, the correlation mapping relationship between fuel characteristics, operating conditions, and safety control is constructed to build a fuel supply simulation safety control model.

[0041] Specifically, the core input content is first standardized and organized. The extracted fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation rules are sorted out and verified to ensure that the input information is accurate, unbiased, and logically coherent, serving as the core foundation for model construction. Then, three core parameters of low flash point fuels—physicochemical properties, safety properties, and supply adaptation characteristics—are retrieved from a multi-fuel characteristic feature library. The correlation between the parameters and the core input is verified to ensure that the parameters meet the actual needs of supply simulation for low flash point fuel ship training. On this basis, the inherent logic of the interaction between fuel characteristics, operating conditions, and safety control is deeply integrated with the core input and feature library parameters. The safety control standards and response requirements corresponding to different fuel characteristics and different operating conditions are clarified, and the correlation mapping relationship between the three is constructed. Finally, based on this correlation mapping relationship and combined with the safety prevention and control requirements of the training scenario, a fuel supply simulation safety control model is built.

[0042] S4. Introduce the multi-set data weight dynamic allocation model into the multi-set data collaborative multi-objective optimization algorithm, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and perform control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range. In this embodiment of the invention, the step of introducing a multi-set data weight dynamic allocation model into a multi-set data collaborative multi-objective optimization algorithm, defining the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and performing control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range includes the following steps: S41. Introduce the dynamic weight allocation model of multiple sets of data into the multi-objective optimization algorithm of multi-set data, clarify the constraints of the data weight allocation rules on the optimization algorithm, and perform collaborative adaptation between the dynamic weight allocation model of multiple sets of data and the multi-objective optimization algorithm of multi-set data. Specifically, the existing multi-set data weight dynamic allocation model is first rationally introduced into the multi-set data collaborative multi-objective optimization algorithm. The combination logic of the model and the algorithm is clarified to ensure that the weight allocation rules output by the model effectively support the algorithm's optimization process. Then, the constraints are clearly defined. Combining the safety priority and efficient adaptation requirements of fuel supply simulation, the constraint range of the data weight allocation rules on the optimization algorithm is defined, including the dynamic adjustment boundary of weights and the priority of optimization objectives (such as safety threshold compliance and accurate supply adaptation). This ensures that the optimization process does not deviate from the core requirements set by the model. Finally, collaborative adaptation work is carried out to debug the correlation parameters of the model and the algorithm, calibrate the linkage logic of weight allocation and optimization process, correct adaptation deviations, and solve the connection problem in the collaborative process between the two. This ensures that the algorithm can accurately focus on the core objectives of fuel supply simulation under the constraints of the weight model, and realize the collaborative operation of multi-objective optimization and dynamic weight allocation.

[0043] S42. Combining the system simulation operation safety threshold, fuel supply simulation adaptation standard, and low flash ignition physicochemical characteristic parameters and safety characteristic parameters in the multi-fuel characteristic feature library, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range. In this embodiment of the invention, the step of defining the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range by combining the system simulation operation safety threshold, fuel supply simulation adaptation standard, and low flash ignition physicochemical characteristic parameters and safety characteristic parameters in the multi-fuel characteristic feature library includes the following steps: S421. Retrieve the system simulation operation safety threshold and fuel supply simulation adaptation standard, simultaneously extract the physicochemical and safety characteristic parameters of low flash point fuels from the multi-fuel characteristic feature library, conduct correlation analysis on various parameters, clarify the constraint logic and influence weight between different parameters, and establish a parameter correlation matrix. Specifically, the process begins by accurately retrieving the pre-set system simulation operation safety thresholds and fuel supply simulation adaptation standards. Simultaneously, the physicochemical and safety characteristics of low-flash-point fuels are extracted from a multi-fuel characteristic database. The completeness and accuracy of all parameters are verified to ensure they closely match the actual simulated operation of the training equipment, without deviation or omission. Subsequently, a systematic correlation analysis is conducted on all retrieved parameters to clarify the intrinsic relationships between thresholds, adaptation standards, and fuel characteristic parameters, and to define the interaction relationships between different types of parameters.

[0044] Furthermore, based on this, the constraint logic between different parameters is further defined, the value boundaries of each parameter and their mutual constraints are clarified, the influence weights of various parameters are divided, the priority of the role of core parameters and auxiliary parameters is distinguished, and finally, based on the correlation analysis results, constraint logic and influence weights, various parameters are used as rows and columns of a matrix, and the corresponding correlation relationships and weight values ​​are filled in to establish a parameter correlation matrix, which clearly presents the correlation of various parameters.

[0045] S422. Based on the parameter correlation matrix, combined with the low flash ignition culinary and chemical characteristics parameters and safety characteristics parameters, define the upper and lower limits and judgment criteria of the system simulation operation safety tolerance threshold, and delineate the range boundary and graded adjustment gradient of the supply parameter simulation adjustment range, thus forming the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range.

[0046] Specifically, firstly, based on the established parameter correlation matrix as the core support, the physicochemical and safety characteristic parameters of low flash point fuels in the multi-fuel characteristic feature library are simultaneously combined to verify the correlation and accuracy of various parameters, ensuring that the definition basis fits the actual simulation operation without logical deviation. Then, based on the parameter constraint logic and influence weight clearly defined in the parameter correlation matrix, combined with the safety control requirements of the flammable and explosive characteristics of fuels, the upper and lower limits of the safety tolerance threshold of the system simulation operation are reasonably defined, the specific standards for threshold compliance and exceeding the standard are clarified, and a feasible safety tolerance threshold judgment benchmark is formed.

[0047] Simultaneously, based on the fuel supply simulation adaptation requirements, the range boundaries of the supply parameter simulation adjustment range are defined to prevent adjustments from exceeding the safety and adaptation range. Then, according to the adaptation differences of different training working conditions, graded adjustment gradients are divided, and the adjustment range and adaptation scenarios of each gradient are clarified. Finally, the safety tolerance threshold (including upper and lower limits and judgment criteria) and the supply parameter simulation adjustment range (including range boundaries and graded gradients) are integrated to form a complete parameter control system.

[0048] S43. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, and based on a multi-set data collaborative multi-objective optimization algorithm, perform multiple rounds of matching and optimization of fuel supply-related control parameters, and synchronously record the parameter adaptation effect of each round of optimization. In this embodiment of the invention, the process of using the system's simulated operational safety tolerance threshold as a constraint, the simulated adjustment range of supply parameters as a range, and a multi-objective optimization algorithm based on multiple sets of data to perform multiple rounds of matching and optimization of fuel supply-related control parameters, while simultaneously recording the parameter adaptation effect of each round of optimization, includes the following steps: S431. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, determine the optimization basis conditions for the multi-set data collaborative multi-objective optimization algorithm. Specifically, the core criteria for defining the basic conditions for optimization are first clarified. The defined system simulation operation safety tolerance threshold is strictly used as the core constraint. The upper and lower limits of the threshold and the judgment criteria are fully incorporated into the constraint system of the algorithm optimization. It is clear that all parameter combinations in the algorithm optimization process must meet the safety tolerance threshold requirements, and optimization results that exceed the safety range are eliminated to adhere to the safety bottom line. At the same time, the simulated adjustment range of supply parameters is used as the optimization value range. The boundary restrictions and graded adjustment gradients of parameter optimization are clarified to ensure that the values ​​of supply-related control parameters in the optimization process are within the defined range and conform to the fuel supply simulation adaptation requirements. In addition, combined with the data weight allocation rules constraints determined in the early stage, auxiliary conditions such as the priority of optimization targets are added. By integrating the constraint requirements, value ranges and auxiliary conditions, the basic conditions for multi-group data collaborative multi-objective optimization algorithm are finally determined.

[0049] S432. Based on the optimization conditions, a multi-set data collaborative multi-objective optimization algorithm is used to perform multi-round matching optimization of fuel supply-related control parameters. In each round of optimization, the parameter combination and parameter adaptation effect corresponding to that round of optimization are recorded synchronously to form a multi-round optimization record.

[0050] Specifically, the optimization premise is first clarified, strictly based on the established optimization conditions. The system simulation operation safety tolerance threshold is used as a constraint, and the simulated adjustment range of supply parameters is used as the value range. A multi-set data collaborative multi-objective optimization algorithm is launched. During the optimization process, the goals are fuel supply simulation adaptation rhythm, core safety risk prevention and control, and operating condition switching adaptation. The fuel supply-related control parameters are iteratively matched and optimized in multiple rounds. In each round, the parameter combination is reasonably adjusted to ensure that the parameter values ​​meet the basic condition constraints. At the same time, the details of each round of optimization are recorded, and the corresponding control parameter combination and parameter adaptation effect (such as whether it meets the safety threshold, adaptation standard and operating condition requirements) are accurately retained. After the multi-round optimization is completed, the recorded information of all rounds is integrated, and the correspondence between parameter combination and adaptation effect is sorted out to form a complete multi-round optimization record.

[0051] S44. Screen and verify the results of multiple rounds of optimization, and remove parameter combinations that do not meet the fuel supply simulation adaptation standards and safety tolerance thresholds to obtain the multi-version fuel supply simulation benchmark parameters and the corresponding dynamic adaptation range of safety control.

[0052] Specifically, the core screening and verification criteria are first clarified, and the fuel supply simulation adaptation standards and system simulation operation safety tolerance thresholds are strictly compared with those set in the early stage to build a dual verification dimension. Then, the parameter combinations and adaptation effects in the multi-round optimization records are checked one by one. The focus is on verifying whether the parameter values ​​are within the upper and lower limits of the safety tolerance threshold, and whether the adaptation effect meets the supply adaptation requirements under different working conditions. Invalid parameter combinations that exceed the threshold range or do not meet the adaptation standards are eliminated.

[0053] After verification, the qualified parameter combinations are classified and organized according to operating conditions and applicable scenarios, and multi-version fuel supply simulation benchmark parameters are extracted. At the same time, the fluctuation range and applicable operating condition characteristics of each version of benchmark parameters are combined to define the corresponding dynamic adaptation range of safety control, clarify the adjustment boundary of parameters under different scenarios, and finally obtain multi-version benchmark parameters and corresponding dynamic adaptation range of safety control.

[0054] S5. Substitute the multi-version fuel supply simulation benchmark parameters and the dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme. Simulate and implement the personalized fuel supply drive control scheme in the ship training equipment and collect real-time data of the simulation scheme and equipment simulation attenuation data. In this embodiment of the invention, the step of substituting multiple versions of fuel supply simulation benchmark parameters and dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme, and simulating the implementation of this personalized fuel supply drive control scheme in the ship training equipment, and collecting real-time data of the simulation scheme and equipment simulation attenuation data, includes the following steps: S51. Retrieve the baseline parameters of multiple versions of fuel supply simulation and the dynamic adaptation range of safety control, and substitute them one by one into the fuel supply simulation safety control model to perform parameter initialization and debugging. Specifically, firstly, the baseline parameters for fuel supply simulation and their corresponding dynamic adaptation ranges for safety control obtained from the preliminary screening are accurately retrieved. The one-to-one correspondence between the parameter versions and the adaptation ranges is verified to ensure the completeness and accuracy of the parameters and to ensure that there are no version confusions or missing parameters. Then, the baseline parameters and their corresponding dynamic adaptation ranges for safety control are substituted into the fuel supply simulation safety control model one by one according to the version order to complete the parameter initialization configuration for the corresponding version. This ensures that the parameters and the model input interface are accurately matched and there are no adaptation conflicts. During the debugging process, the basic operating status of the model after parameter initialization is verified by combining the system simulation operation safety threshold and the fuel supply simulation adaptation standard. Problems such as parameter logic contradictions and value exceeding limits are investigated. At the same time, the model response data and debugging results after the parameter initialization of each version are recorded simultaneously.

[0055] S52. After initializing and debugging the parameters, the calculation output of the fuel supply simulation safety control model is combined with the ship training scenario data to generate a personalized fuel supply simulation drive control scheme. Specifically, firstly, retrieve the calculation output data of the fuel supply simulation safety control model after parameter initialization and debugging, including the supply adaptation effect corresponding to the benchmark parameters of each version, the safety risk prevention and control response logic, the working condition switching parameter connection scheme, etc. Simultaneously collect ship training scenario data, covering the working condition settings, equipment operating parameters, training target requirements, etc. of different training stages, and verify the correlation and matching degree of the two types of data.

[0056] Subsequently, the model's output is deeply integrated and analyzed with ship training scenario data. For the personalized needs of different training scenarios, corresponding fuel supply benchmark parameters and dynamic adaptation range of safety control are matched. The adjustment rhythm of supply parameters, safety risk warning thresholds and response strategies for switching operating conditions are clarified for different training stages. Finally, the analysis results are integrated to generate a personalized fuel supply simulation drive control scheme. The scheme is verified to meet the safety tolerance threshold and supply adaptation standards to ensure that the scheme can directly guide the fuel supply simulation operation in different training scenarios.

[0057] S53. Based on the personalized fuel supply simulation drive control scheme, start the simulation implementation in the low flash point fuel supply simulation device of the ship training equipment, and simultaneously monitor the operation status of the entire process of scheme implementation. Specifically, the process begins by retrieving the previously generated personalized fuel supply simulation drive control scheme and verifying its compatibility with the low flash point fuel supply simulation device in the ship training equipment. This ensures that the parameter settings and control strategies in the scheme meet the device's operating standards and safety requirements, and that there are no compatibility conflicts. Subsequently, the control scheme is imported into the simulation device, and the relevant parameters of the device are precisely configured. The simulation implementation program is then started, enabling the device to strictly follow the fuel supply rhythm, parameter adjustment gradient, and safety control strategy set in the scheme to carry out low flash point fuel supply simulation operations. Simultaneously, the device's supporting operation monitoring system is activated to monitor the operating status of the entire scheme implementation process in real time. The system focuses on tracking key indicators such as fuel supply parameters, device operating conditions, and safety risk warnings, and captures abnormal parameter fluctuations and operating condition deviations in real time, providing timely feedback.

[0058] S54. During the simulation implementation, real-time operating parameters and safety response signals related to fuel supply simulation are collected as real-time data of the simulation scheme. At the same time, the wear and performance degradation parameters of each simulated component of the ship training equipment are collected as equipment simulation degradation data.

[0059] Specifically, a multi-dimensional data acquisition system is first established, connecting the operation control module and safety monitoring module of the fuel supply simulation device. Real-time acquisition indicators are clearly defined, including core operating parameters such as fuel supply flow rate, pressure, and temperature, as well as safety response signals such as safety warning trigger signals, threshold over-limit feedback signals, and emergency response activation signals. A high-frequency acquisition frequency is set to ensure data real-time performance. Simultaneously, status monitoring sensors are installed on various simulated components of the ship training equipment (such as pumps, valves, and pipelines) to accurately collect loss and performance degradation parameters such as component operating time, wear, sealing performance degradation, and power output attenuation rate. Acquisition nodes are divided according to the simulation stage. Finally, timestamps are uniformly configured for both types of data to achieve synchronous linkage during the acquisition process, real-time verification of data integrity and accuracy, elimination of abnormal data, and synchronous storage to the data terminal.

[0060] S6. Integrate the real-time data of the simulation scheme and the equipment simulation attenuation data to form a control scheme effectiveness simulation evaluation report and a long-term simulation operation impact report. Based on these two reports, iteratively update the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark.

[0061] Specifically, the first step is to systematically integrate the real-time data of the simulation scheme with the equipment simulation attenuation data. This involves aligning operating parameters such as fuel supply flow, pressure, and temperature, as well as response signals such as safety warnings and threshold exceedances, with attenuation data such as wear of simulated components like pumps, valves, and pipelines, the rate of decrease in sealing performance, and the rate of power output attenuation. Timestamps and operating condition labels are then uniformly configured, the correlation and accuracy of the data are verified, and abnormal data is removed before constructing a standardized fusion dataset.

[0062] Based on this dataset, on the one hand, a simulation evaluation report on the effectiveness of the control scheme is generated, which focuses on quantitatively analyzing the supply adaptation accuracy, safety risk prevention and control response efficiency, and working condition switching effect of the scheme under different training scenarios, and clarifies the advantages and shortcomings to be optimized of the scheme. On the other hand, a long-term simulation operation impact report is compiled to deeply analyze the impact mechanism of equipment performance degradation on the stability of the scheme operation, and sort out the parameter deviation patterns and potential safety hazards caused by component wear during long-term operation.

[0063] Subsequently, based on the two reports, the iterative update of the core standards was initiated: for the system simulation operation safety threshold, combined with the parameter fluctuation data caused by long-term decay, the upper and lower limits of the threshold and the judgment sensitivity were adjusted to enhance the adaptability of the threshold to the equipment decay state; for the fuel supply simulation adaptation standard, referring to the scenario-based adaptation data in the results report, the supply parameter range for different training conditions was refined, and the adaptation parameter correction rules for the equipment decay stage were supplemented; for the working condition conflict judgment benchmark, based on the parameter conflict cases during working condition switching in the report, the conflict identification indicators and judgment logic were optimized, and the triggering conditions for conflict early warning and handling were improved.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fuel supply drive control method for training on low flashpoint fuel ships, characterized in that, Includes the following steps: S1. Collect system simulation operation status data, fuel characteristic simulation adaptation data, ship training scenario data and safety condition simulation response data throughout the entire life cycle of low flash point fuel supply simulation in ship training equipment, and build a hierarchical system simulation operation archive and a multi-fuel characteristic feature library. S2. Based on the multi-fuel characteristic feature library, set the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark. Label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrate them to form a multi-dimensional fuel association dataset. S3. Preprocess the multi-dimensional fuel-related dataset, and use the working condition feature clustering algorithm and time series pattern mining algorithm according to the differences of the data groups to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. Then, combine the multi-fuel characteristic feature library to build a multi-group data weight dynamic allocation model and a fuel supply simulation safety control model. S4. Introduce the multi-set data weight dynamic allocation model into the multi-set data collaborative multi-objective optimization algorithm, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and perform control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range. S5. Substitute the multi-version fuel supply simulation benchmark parameters and the dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme. Simulate and implement the personalized fuel supply drive control scheme in the ship training equipment and collect real-time data of the simulation scheme and equipment simulation attenuation data. S6. Integrate the real-time data of the simulation scheme and the equipment simulation attenuation data to form a control scheme effectiveness simulation evaluation report and a long-term simulation operation impact report. Based on these two reports, iteratively update the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark.

2. The fuel supply drive control method for training low flashpoint fuel ships according to claim 1, characterized in that, The process of setting system simulation operation safety thresholds, fuel supply simulation adaptation standards, and operating condition conflict judgment benchmarks based on a multi-fuel characteristic feature library, and then labeling and classifying system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, and integrating them into a multi-dimensional fuel association dataset includes the following steps: S21. Extract historical safe operation data, fault condition record data and typical condition adaptation data from the hierarchical system simulation operation archive to form a basic dataset; S22. Combine the basic dataset with the low flash ignition physicochemical characteristic parameters, safety characteristic parameters and supply adaptation characteristic parameters retrieved from the multi-fuel characteristic feature library, and set the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark. S23. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and operating condition conflict judgment benchmark, establish a three-level labeling rule, and label and classify the system simulation operation status data, fuel characteristic simulation adaptation data and safety operating condition simulation response data in the hierarchical system simulation operation archive respectively. S24. Perform consistency verification on the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data that have completed the three-level labeling, and integrate them according to the time dimension and the operating condition dimension to form a multi-dimensional fuel association dataset.

3. The fuel supply drive control method for training low flashpoint fuel ships according to claim 1, characterized in that, The process of preprocessing the multi-dimensional fuel-related dataset and using operating condition feature clustering and time-series pattern mining algorithms based on data group differences to extract fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns, combined with a multi-fuel characteristic feature library to build a multi-group data weight dynamic allocation model and a fuel supply simulation safety control model, includes the following steps: S31. Preprocess the multi-dimensional fuel association dataset, remove outliers, redundant values ​​and missing values ​​from the data, and perform standardization transformation on different types of data to complete the data dimension alignment and obtain the standardized multi-dimensional fuel association dataset. S32. Based on the grouping differences of the multi-dimensional fuel association dataset after standardization, the working condition feature clustering algorithm and the time series operation related data in the multi-dimensional fuel association dataset are used to extract the fuel supply simulation adaptation rhythm, core safety risk simulation points and working condition switching simulation adaptation rules. S33. Retrieve the low flash ignition physicochemical characteristics, safety characteristics, and supply adaptation characteristics from the multi-fuel characteristic feature library. Combine the fuel supply simulation adaptation rhythm, core safety risk simulation points, and working condition switching simulation adaptation rules to build a multi-set data weight dynamic allocation model. S34. The core inputs are the fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation rules. Combined with a multi-fuel characteristic feature library, the correlation mapping relationship between fuel characteristics, operating conditions, and safety control is constructed to build a fuel supply simulation safety control model.

4. The fuel supply drive control method for training low flashpoint fuel ships according to claim 1, characterized in that, The process of introducing a multi-set data weight dynamic allocation model into a multi-set data collaborative multi-objective optimization algorithm, defining the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range, and performing control parameter matching optimization to obtain multi-version fuel supply simulation benchmark parameters and safety control dynamic adaptation range includes the following steps: S41. Introduce the dynamic weight allocation model of multiple sets of data into the multi-objective optimization algorithm of multi-set data, clarify the constraints of the data weight allocation rules on the optimization algorithm, and perform collaborative adaptation between the dynamic weight allocation model of multiple sets of data and the multi-objective optimization algorithm of multi-set data. S42. Combining the system simulation operation safety threshold, fuel supply simulation adaptation standard, and low flash ignition physicochemical characteristic parameters and safety characteristic parameters in the multi-fuel characteristic feature library, define the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range. S43. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, and based on a multi-set data collaborative multi-objective optimization algorithm, perform multiple rounds of matching and optimization of fuel supply-related control parameters, and synchronously record the parameter adaptation effect of each round of optimization. S44. Screen and verify the results of multiple rounds of optimization, and remove parameter combinations that do not meet the fuel supply simulation adaptation standards and safety tolerance thresholds to obtain the multi-version fuel supply simulation benchmark parameters and the corresponding dynamic adaptation range of safety control.

5. A fuel supply drive control method for training on low flashpoint fuel ships according to claim 1, characterized in that, The process of substituting multi-version fuel supply simulation baseline parameters and dynamic adaptation range of safety control into the fuel supply simulation safety control model to generate a personalized fuel supply simulation drive control scheme, and then simulating and implementing this personalized fuel supply drive control scheme in the ship training equipment, and collecting real-time data of the simulation scheme and equipment simulation attenuation data, includes the following steps: S51. Retrieve the baseline parameters of multiple versions of fuel supply simulation and the dynamic adaptation range of safety control, and substitute them one by one into the fuel supply simulation safety control model to perform parameter initialization and debugging. S52. After initializing and debugging the parameters, the calculation output of the fuel supply simulation safety control model is combined with the ship training scenario data to generate a personalized fuel supply simulation drive control scheme. S53. Based on the personalized fuel supply simulation drive control scheme, start the simulation implementation in the low flash point fuel supply simulation device of the ship training equipment, and simultaneously monitor the operation status of the entire process of scheme implementation. S54. During the simulation implementation, real-time operating parameters and safety response signals related to fuel supply simulation are collected as real-time data of the simulation scheme. At the same time, the wear and performance degradation parameters of each simulated component of the ship training equipment are collected as equipment simulation degradation data.

6. A fuel supply drive control method for training on low flashpoint fuel ships according to claim 2, characterized in that, The establishment of a three-level annotation rule based on system simulation operation safety thresholds, fuel supply simulation adaptation standards, and operating condition conflict judgment criteria, and the subsequent annotation and classification of system simulation operation status data, fuel characteristic simulation adaptation data, and safety operating condition simulation response data in the hierarchical system simulation operation archive, includes the following steps: S231. Based on the system simulation operation safety threshold, fuel supply simulation adaptation standard and working condition conflict judgment benchmark, establish a three-level labeling rule, which includes the first level safety compliance labeling rule, the second level fuel compatibility labeling rule and the third level working condition conflict risk labeling rule, and clarify the labeling content and labeling standard corresponding to each level of rule; S232. According to the established three-level annotation rules, the system simulation operation status data, fuel characteristic simulation adaptation data and safety condition simulation response data in the hierarchical system simulation operation archive are annotated in sequence, and the annotation results are summarized and classified according to the annotation attributes to form a dataset with annotation labels.

7. A fuel supply drive control method for training on low flashpoint fuel ships according to claim 3, characterized in that, Based on the grouping differences of the standardized multi-dimensional fuel association dataset, the following steps are taken to extract the fuel supply adaptation-related data, core safety risk simulation points, and operating condition switching simulation adaptation patterns from the fuel supply adaptation-related data and time-series operation-related data in the multi-dimensional fuel association dataset using operating condition feature clustering algorithms and time-series pattern mining algorithms: S321. Group the standardized multi-dimensional fuel association dataset and clarify the data group related to fuel supply adaptation and the data group related to time-series operation. S322. Use the operating condition feature clustering algorithm to perform cluster analysis on the fuel supply adaptation related data group, obtain the supply characteristic parameters related to fuel supply adaptation and the correlation between parameters, and use the time series pattern mining algorithm to perform time series analysis on the time series operation related data group, extract the time series change pattern of system simulation operation and the operating condition switching correlation features. S323. Integrate supply characteristic parameters, relationships between parameters, time-series change patterns, and operating condition switching characteristics, and refine them to obtain fuel supply simulation adaptation rhythm, core safety risk simulation points, and operating condition switching simulation adaptation patterns.

8. A fuel supply drive control method for training low flashpoint fuel ships according to claim 4, characterized in that, The process of defining the system simulation operation safety threshold and the supply parameter simulation adjustment range by combining the system simulation operation safety threshold, fuel supply simulation adaptation standards, and low flash ignition physicochemical characteristics and safety parameters in the multi-fuel characteristic feature library includes the following steps: S421. Retrieve the system simulation operation safety threshold and fuel supply simulation adaptation standard, simultaneously extract the physicochemical and safety characteristic parameters of low flash point fuels from the multi-fuel characteristic feature library, conduct correlation analysis on various parameters, clarify the constraint logic and influence weight between different parameters, and establish a parameter correlation matrix. S422. Based on the parameter correlation matrix, combined with the low flash ignition culinary and chemical characteristics parameters and safety characteristics parameters, define the upper and lower limits and judgment criteria of the system simulation operation safety tolerance threshold, and delineate the range boundary and graded adjustment gradient of the supply parameter simulation adjustment range, thus forming the system simulation operation safety tolerance threshold and the supply parameter simulation adjustment range.

9. A fuel supply drive control method for training on low flashpoint fuel ships according to claim 8, characterized in that, The process of using the system's simulated operational safety threshold as a constraint, the simulated adjustment range of supply parameters as a range, and a multi-objective optimization algorithm based on multiple sets of data to perform multiple rounds of matching and optimization of fuel supply-related control parameters, while simultaneously recording the parameter adaptation effect of each round of optimization, includes the following steps: S431. Using the system simulation operation safety tolerance threshold as a constraint and the supply parameter simulation adjustment range as a range, determine the optimization basis conditions for the multi-set data collaborative multi-objective optimization algorithm. S432. Based on the optimization conditions, a multi-set data collaborative multi-objective optimization algorithm is used to perform multi-round matching optimization of fuel supply-related control parameters. In each round of optimization, the parameter combination and parameter adaptation effect corresponding to that round of optimization are recorded synchronously to form a multi-round optimization record.

10. A fuel supply drive control method for training low flashpoint fuel ships according to claim 7, characterized in that, The process of using a working condition feature clustering algorithm to perform cluster analysis on fuel supply adaptation-related data sets to obtain fuel supply adaptation-related supply feature parameters and the correlation between parameters, and using a time series pattern mining algorithm to perform time series analysis on time series operation-related data sets to extract the time series change patterns and working condition switching correlation features of the system simulation operation, includes the following steps: S3221. Identify the core parameter dimensions of the fuel supply adaptation related data group, remove redundant parameters that are not related to fuel supply adaptation from the data group, and obtain the purified fuel supply adaptation data group. S3222. The working condition feature clustering algorithm is used to perform cluster analysis on the purified fuel supply adaptation data group, and the fuel supply adaptation parameter clusters are divided to obtain the supply feature parameters and the correlation between parameters in each cluster. S3223. Define the time series parameter range of the time series operation related data group, and calibrate the time dimension of the time series operation related data group to align the time dimension of the data group. S3224. Use a time-series pattern mining algorithm to perform time-series analysis on the time-series operation-related data group after time-dimensional calibration, and extract the time-series change patterns of the system simulation operation and the parameter correlation characteristics during operating condition switching.

Images