Ship propulsion system operation state diagnosis method and system based on intelligent algorithm

Abstract

The invention relates to the field of ship propulsion systems, in particular to a ship propulsion system operation state diagnosis method and system based on an intelligent algorithm. The method comprises the steps that current and historical shafting data and operation plans are acquired, whether a current operation route is a specific operation route or not is determined based on the historical operation plans, the operation routes are classified and distinguished, and the specific operation route is subjected to data processing; the corresponding target historical shafting data are obtained, the ship operation state is determined in combination with the current shafting data, and the current ship state under the route can be more accurately judged by using the reference value of the historical data under the specific route; for a non-specific operation route, historical shafting data, current shafting data and a current operation plan are comprehensively considered to determine the operation state of the ship, the accuracy and reliability of diagnosis are improved through multi-dimensional analysis, abnormity of a ship propulsion system can be found in time, and safe operation of the ship is guaranteed.

Description

Technical Field

[0001] This application relates to the field of ship propulsion systems, and in particular to a method and system for diagnosing the operational status of ship propulsion systems based on intelligent algorithms. Background Technology

[0002] The ship propulsion system is the core power unit of a ship, and its operational status diagnosis is a key technology to ensure safe navigation and reduce maintenance costs. The ship shafting is a critical subsystem responsible for power transmission in the ship propulsion system. It is the core bridge connecting the "power source" (main engine) and the "actuator" (thruster) of the propulsion system, and its functionality directly determines the power transmission efficiency and operational safety of the propulsion system.

[0003] The ship's shafting system mainly includes the thrust shaft, intermediate shaft, stern shaft, propeller shaft, and their corresponding bearings. Its primary function is to transmit the torsional torque output from the main engine to the propeller, causing it to rotate and thus generating longitudinal thrust. During shafting operation, torsional vibration, longitudinal vibration, and gyroscopic vibration can damage the shafting system itself. In severe cases, this can lead to serious accidents such as main engine crankshaft failure, shafting fatigue damage, stern bearing deformation and seal failure, and even extreme situations like shaft breakage. In recent years, with the increasing length of ship shafting spans, shafting failures have become more frequent. Therefore, understanding and monitoring the operating status of the ship's shafting system during operation, and thus enabling timely detection of potential faults, is a crucial safety issue for ships. Summary of the Invention

[0004] In order to detect abnormalities in the ship propulsion system in a timely manner, this application provides a method and system for diagnosing the operating status of the ship propulsion system based on intelligent algorithms.

[0005] Firstly, this application provides a method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms, employing the following technical solution: A method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms, comprising: Obtain the current shafting data and historical shafting data corresponding to the current vessel, and obtain the current operation plan and historical operation plan corresponding to the current vessel. The current shafting data is the shafting data corresponding to the current moment, and the historical shafting data is the shafting data corresponding to each historical operation plan. The shafting data includes vibration sub-data, operating condition sub-data, and fault sub-data. The current operation plan is the operation plan corresponding to the current vessel, and the operation plan includes the operation route. Based on the historical operation plan, determine whether the operation route corresponding to the current operation plan is a specific operation route, wherein the specific operation route is a route whose historical operation count exceeds a preset threshold; If the current operating plan corresponds to a specific operating route, then the target historical shafting data corresponding to the specific operating route is obtained, and the ship operating status corresponding to the current ship in the current period is determined based on the current shafting data and the target historical shafting data. The ship operating status is either an abnormal state or a normal state. If the route corresponding to the current operation plan is a non-specific route, then the ship operation status corresponding to the current ship in the current period is determined based on the historical shafting data, the current shafting data, and the current operation plan.

[0006] By adopting the above technical solution, and by acquiring current and historical shafting data and operational plans, comprehensive information related to ship operation is collected. Based on historical operational plans, it is determined whether the current route is a specific route. The routes are then categorized and processed. For specific routes, corresponding historical shafting data is acquired and combined with current shafting data to determine the ship's operational status. This utilizes the reference value of historical data for specific routes, enabling a more accurate assessment of the current ship status for that route. For non-specific routes, historical shafting data, current shafting data, and the current operational plan are comprehensively considered to determine the ship's operational status. This multi-dimensional analysis improves the accuracy and reliability of the diagnosis, helps to promptly detect anomalies in the ship's propulsion system, and ensures the safe operation of the ship.

[0007] In one possible implementation, the fault sub-data includes the number of faults, fault location, fault cause, and fault area. When the current operating plan corresponds to a specific operating route, determining the ship's operating status within the current period based on the current shafting data and the target historical shafting data includes: First historical axle data is filtered from the target historical axle data, and the number of first historical axle data is determined. The historical target working condition sub-data of the first historical axle data is consistent with the current working condition sub-data. The working condition sub-data includes load. If the number exceeds a threshold, the number of fault combinations in different first historical operation plans and the time series length corresponding to each fault combination are determined. Based on the time series length and the number of fault combinations, the first historical operation plans are processed using unsupervised time series data to obtain fault event matrix data. Future fault matrix data is generated based on the fault event matrix data. The future fault matrix data is processed to determine the fault sub-data of the current ship in the current period, so as to determine the ship operation status corresponding to the current ship. If the number does not exceed the number threshold, the operating condition sub-data and vibration sub-data of each target historical shafting data are analyzed to determine the influence curve of the operating condition sub-data on the vibration sub-data, and the ship operation status corresponding to the current ship is determined based on the current shafting data and the influence curve.

[0008] In one possible implementation, the step of generating future fault matrix data based on the fault event matrix data, processing the future fault matrix data, and determining the fault sub-data of the current vessel in the current period includes: Time series pattern mining is performed on the fault event matrix data to identify and extract typical fault evolution sequence patterns; The current shaft system data is matched with the initial stage of the identified typical fault evolution sequence pattern; Based on the successfully matched typical fault evolution sequence patterns, sequence pattern extrapolation is performed to generate future fault development prediction sequences, which constitute the future fault matrix data. Data processing is performed on the data contained in the future fault matrix data to obtain fault event data. Based on the fault event data, the fault event occurrence volume of the current ship in the current period is generated to obtain the fault sub-data of the current ship in the current period.

[0009] In one possible implementation, the operating condition sub-data further includes speed, torque, and power; the vibration sub-data includes torsional vibration, lateral vibration, and longitudinal vibration; and the analysis of the operating condition sub-data and vibration sub-data for each target historical shaft system data to determine the influence curve of the operating condition sub-data on the vibration sub-data includes: Feature extraction is performed on the vibration sub-data of each target historical shaft system data to form a vibration feature dataset; For any vibration feature in the vibration feature dataset, a regression correlation model is established between the vibration feature and at least one working condition parameter in the normalized working condition data, and the quantitative influence coefficient of the vibration feature for each working condition parameter is calculated. Based on the quantified influence coefficient, a single-parameter influence curve of each working condition parameter on the vibration characteristic is generated, with the change value of the single working condition parameter on the x-axis and the relative change of the vibration characteristic caused by the single working condition parameter on the y-axis. The single-parameter influence curves of each working condition parameter are fused to obtain the influence degree curve characterizing the comprehensive influence of the working condition sub-data on the vibration sub-data.

[0010] In one possible implementation, determining the ship's operational status within the current period based on the historical shafting data, the current shafting data, and the current operational plan includes: Based on the historical shaft system data and the current operating plan, determine the future operating condition sub-data within the current cycle; Determine the vibration reference range corresponding to the future operating condition sub-data; Based on the current shafting data and the vibration reference range, the current shafting status of the ship is determined.

[0011] In one possible implementation, determining the future operating condition sub-data within the current period includes: The current operating plan's route is compared with the operating routes of all historical operating plans. Based on the similarity of origin and destination ports, major turning points, and route length, one or more historical voyages with route similarity exceeding a preset similarity threshold are selected from the historical axis data as reference historical voyages. Extract historical operating condition sub-data of all reference historical voyages within the expected flight time period corresponding to the current operation plan to form a reference operating condition sequence set; Based on the reference operating condition sequence set, a time-series prediction model is used, combined with the current operating plan, to determine the future operating condition sub-data sequence of the ship shafting in the future voyage within the current period, so as to obtain the future operating condition sub-data within the current period.

[0012] In one possible implementation, determining the vibration reference range corresponding to the future operating condition sub-data includes: Historical vibration sub-data corresponding to the reference historical voyage is extracted from the historical shaft system data. For each voyage condition included in the future operating condition sub-data, historical vibration characteristic data in which the differences in speed, torque, and load parameters with the voyage condition are all within a preset difference threshold are selected. The historical vibration characteristic data obtained from the screening were statistically analyzed to obtain the sample mean and sample standard deviation of the vibration characteristics under the operating conditions of each flight segment; Based on the changes in operating parameters of the future operating condition sub-data, operating condition correction coefficients are calculated. These correction coefficients include a speed correction coefficient, a load correction coefficient, and a sea state correction coefficient. The speed correction coefficient is the ratio of the speed of the future segment to the speed of the corresponding segment of the reference historical voyage. The load correction coefficient is the ratio of the load of the future segment to the load of the corresponding segment of the reference historical voyage. The sea state correction coefficient is a correction ratio determined based on the difference between the future sea state level and the sea state level of the reference historical voyage. Based on the sample mean, the sample standard deviation, and the operating condition correction coefficient, the vibration reference lower limit and reference upper limit corresponding to each operating condition of the flight segment are calculated to obtain the vibration reference sub-range. The vibration reference lower limit is the sample mean multiplied by the minimum operating condition correction coefficient minus the correction value of the sample standard deviation, and the vibration reference upper limit is the sample mean multiplied by the maximum operating condition correction coefficient plus the correction value of the sample standard deviation. By integrating the vibration reference sub-ranges corresponding to each flight segment operating condition within the current period, the complete vibration reference range corresponding to the future operating condition sub-data within the current period is obtained.

[0013] Secondly, this application provides a ship propulsion system operation status diagnosis system based on intelligent algorithms, which adopts the following technical solution: A ship propulsion system operation status diagnosis system based on intelligent algorithms, including...

[0014] By adopting the above technical solution, a ship propulsion system operation status diagnostic device based on intelligent algorithms is developed. An electronic device, the electronic device being used to execute the intelligent algorithm-based ship propulsion system operation status diagnosis method according to any one of claims 1 to 7.

[0015] In one possible implementation, the intelligent algorithm-based ship propulsion system operation status diagnostic device includes: The acquisition module is used to acquire the current shafting data and historical shafting data corresponding to the current vessel, and to acquire the current operation plan and historical operation plan corresponding to the current vessel. The current shafting data is the shafting data corresponding to the current moment, and the historical shafting data is the shafting data corresponding to each historical operation plan. The shafting data includes vibration sub-data, operating condition sub-data, and fault sub-data. The current operation plan is the operation plan corresponding to the current vessel, and the operation plan includes the operation route. The route determination module is used to determine, based on the historical operation plan, whether the operation route corresponding to the current operation plan is a specific operation route, wherein the specific operation route is a route whose historical operation count exceeds a preset threshold. The first state determination module is used to obtain the target historical shafting data corresponding to the specific operating route if the operating route corresponding to the current operating plan is a specific operating route, and determine the ship operating status corresponding to the current ship in the current period based on the current shafting data and the target historical shafting data, wherein the ship operating status is an abnormal state or a normal state. The second state determination module is used to determine the ship operation status of the current ship in the current period based on the historical shafting data, the current shafting data, and the current operation plan if the operation route corresponding to the current operation plan is a non-specific operation route.

[0016] In one possible implementation, the electronic device includes: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: perform the intelligent algorithm-based ship propulsion system operation status diagnosis method as described in any of the first aspects above.

[0017] Thirdly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium includes: a computer program stored thereon that can be loaded by a processor and execute the method described in any of the first aspects above.

[0018] In summary, this application includes the following beneficial technical effects: By acquiring current and historical shafting data and operational plans, and determining whether the current route is a specific route based on historical operational plans, the operational routes are categorized and processed. For specific routes, the corresponding target historical shafting data is acquired and combined with current shafting data to determine the ship's operational status. This utilizes the reference value of historical data for specific routes, enabling a more accurate assessment of the ship's current status for that route. For non-specific routes, the ship's operational status is determined by comprehensively considering historical shafting data, current shafting data, and the current operational plan. This multi-dimensional analysis improves the accuracy and reliability of the diagnosis, helps to promptly detect anomalies in the ship's propulsion system, and ensures the safe operation of the ship. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a method for diagnosing the operational status of a ship propulsion system provided in an embodiment of this application; Figure 2 This is a block diagram of a ship propulsion system operation status diagnosis system according to an embodiment of this application; Figure 3 This is a block diagram of a ship propulsion system operation status diagnostic device provided in an embodiment of this application; Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0020] The following is in conjunction with the appendix Figure 1-4This application will be described in further detail.

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] To facilitate understanding of the technical solutions proposed in this application, several elements that will be introduced in the description of this application are first presented here. It should be understood that the following description is only for the purpose of understanding these elements and the content of the embodiments of this application, and does not necessarily cover all possible situations.

[0023] Ship shafting typically includes the thrust shaft, intermediate shaft, stern shaft, propeller shaft, and their corresponding bearings. Its primary function is to transmit the torsional torque output from the main engine to the propeller, causing it to rotate and generating longitudinal thrust. During shafting operation, torsional vibration, longitudinal vibration, and gyroscopic vibration can damage the shafting itself. In severe cases, this can lead to serious accidents such as main engine crankshaft failure, shafting fatigue damage, stern bearing deformation and seal failure, and even extreme situations like shaft breakage. In recent years, with the increasing length of ship shafting spans, shafting failures have become more frequent. Therefore, understanding and monitoring the operating status of a ship's shafting during operation, and promptly detecting and repairing potential faults, is a crucial safety issue for ships.

[0024] In view of this, embodiments of this application provide a method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms. This method is executed by an electronic device, which can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication. This application embodiment does not impose any limitations on this connection. See [link to relevant documentation]. Figure 1 The method includes steps S101-S104, wherein: Step S101: Obtain the current shafting data and historical shafting data corresponding to the current vessel, and obtain the current operation plan and historical operation plan corresponding to the current vessel.

[0025] The current shaft system data refers to the shaft system data corresponding to the current moment, while the historical shaft system data refers to the shaft system data corresponding to each historical operating plan. The shaft system data includes vibration sub-data, operating condition sub-data, and fault sub-data. Specifically, vibration sub-data refers to the vibration signals of the shaft system in the torsional, longitudinal, and lateral directions, which can be time-domain waveforms, spectra, etc.; operating condition sub-data refers to the external conditions and load parameters of the shaft system during operation, such as speed, torque, power, and lubricating oil temperature; and fault sub-data refers to fault event information recorded or marked in history, including the fault occurrence time, type, and maintenance measures.

[0026] The current operational plan is the operational plan corresponding to the current vessel, and it includes the operational route. Specifically, the current operational plan refers to the instruction document for a vessel's voyage mission. In electronic systems, it is usually represented as structured data, the core of which is the operational route, and it also includes the operational segment tasks corresponding to each cycle. More specifically, the current shafting data is the shafting data corresponding to the current cycle. The current cycle is the cycle corresponding to the current moment in the current operational plan. The current operational plan is divided into multiple sub-plans, with one sub-plan corresponding to each cycle.

[0027] The electronic equipment collects data such as vibration, rotational speed, and torque of the current shafting in real time through the ship's sensor network, packaging it into current shafting data. Simultaneously, the electronic equipment can retrieve all past voyage records from the local database and link them to the corresponding historical shafting data. Furthermore, the electronic equipment can obtain the current operational plan (usually an electronic chart instruction including waypoints and speed) issued for this mission from the navigation management system and query all historical operational plans. The ship's sensors can include torque meters, vibration accelerometers, and temperature sensors, which are pre-installed on the ship to detect the relevant data.

[0028] Step S102: Based on historical operation plans, determine whether the operation route corresponding to the current operation plan is a specific operation route.

[0029] Among them, the specific operating routes are those whose historical number of runs exceeds a preset threshold.

[0030] Because some vessels may perform the same tasks repeatedly or cyclically, resulting in multiple runs on the same route, a large amount of historical shafting data matching that route is accumulated during repeated runs. This data may contain unique sea conditions, channel conditions, operating condition changes, and fault evolution patterns specific to that route. For example, if a specific section of a repeated route is prone to increased shafting vibration due to currents, or if a certain type of fault occurs frequently on that route, the target historical shafting data specific to that route can be directly accessed. Accurate predictions can be achieved through comparison with data on the same operating conditions and fault pattern mining. Specifically, electronic equipment can call a pre-built data analysis module to parse route information from all historical operating plans. More specifically, it can statistically analyze the number of times each historical operating route has been executed in history, defined by consecutive origin and destination ports and major turning points. Upon receiving the current operating plan, its route characteristics are matched against the historical route database. If the number of executions of a matched historical route exceeds a preset threshold, the route of the current operating plan is marked as a specific operating route; otherwise, it is marked as a non-specific operating route. The preset threshold can be set according to the actual situation, and this application embodiment does not limit it.

[0031] Step S103: If the current operating plan corresponds to a specific operating route, then obtain the target historical shafting data corresponding to the specific operating route, and determine the ship operating status corresponding to the current ship in the current cycle based on the current shafting data and the target historical shafting data.

[0032] The ship's operating status is either abnormal or normal.

[0033] Specifically, if a route is determined to be a specific operating route, the target historical axis data corresponding to that specific route can be extracted from the historical database. The current axis data can be compared with the target historical data using a random forest. If the deviation exceeds a preset threshold, it is determined to be an abnormal state.

[0034] One possible implementation of this application embodiment is that the fault sub-data includes the number of faults, fault location, fault cause, and fault area. When the current operating plan corresponds to a specific operating route, the ship's operating status within the current period is determined based on the current shafting data and the target historical shafting data, including: Filter the first historical shaft system data from the target historical shaft system data and determine the number of first historical shaft system data. The historical target working condition sub-data of the first historical shaft system data is consistent with the current working condition sub-data. If the number exceeds the threshold, the number of fault combinations in different first historical operation plans and the time series length corresponding to the number of fault combinations are determined. Based on the time series length and the number of fault combinations, the first historical operation plans are processed into unsupervised time series data to obtain fault event matrix data. Future fault matrix data is generated based on the fault event matrix data. The future fault matrix data is processed to determine the fault sub-data of the current ship in the current period, so as to determine the ship operation status corresponding to the current ship. If the number does not exceed the threshold, the operating condition sub-data and vibration sub-data of each target historical shafting data are analyzed to determine the influence curve of the operating condition sub-data on the vibration sub-data, and based on the current shafting data and the influence curve, the current ship operating status is determined.

[0035] The fault sub-data, a core component of the shafting data, records key information related to shafting faults in the ship's propulsion system. It serves as the core basis for analyzing historical fault patterns and predicting future faults, specifically including four categories of information: fault frequency, fault location, fault cause, and fault area. Specifically, the fault frequency is the total number of faults occurring in the ship's shafting system within any given period, a core indicator for measuring fault frequency. The fault location is the specific geographical location of the ship when the shafting fault occurred, usually represented by latitude and longitude coordinates or navigation route markers. The fault cause is the specific trigger that led to the shafting fault, such as "excessive speed fluctuations," "part aging and wear," "installation deviations," or "sudden changes in sea state causing load impacts." The fault area is the specific component or part of the shafting system where the fault occurred, such as "main shaft midsection," "coupling area," "bearing assembly," and "propeller connection area."

[0036] The operating condition sub-data includes load, and may also include speed, torque, power, etc.

[0037] Specifically, the current operating condition sub-data in the current shaft system data is read from the target historical shaft system database, which is pre-filtered according to a specific operating route. The data is then filtered according to certain criteria. These criteria include finding all historical target operating condition sub-data records that are substantially consistent with the current operating condition data during navigation on that route. Records meeting these criteria are categorized as the first historical shaft system data, and the number of these records is counted. The difference between the historical target operating condition sub-data and the current operating condition sub-data is considered substantially consistent if it falls within a preset range. This preset range can be set according to actual circumstances, and this application does not limit it.

[0038] Furthermore, the method for determining the ship's operating status when the number exceeds a threshold is as follows: All fault sub-data corresponding to the first historical operating plan are analyzed to identify different fault combinations. The frequency of occurrence of each fault combination (i.e., the number of fault combinations) is counted. Simultaneously, the time series length from occurrence to end of each fault combination in the corresponding historical voyage is extracted. An unsupervised clustering algorithm is used, with the number of fault combinations and time series length as the core dimensions, to classify and sort the fault data of the first historical operating plan. The sorted data is then used to construct a matrix according to the dimensions of "fault combination - time node - fault feature," resulting in fault event matrix data. Further, based on the fault event matrix data, a time-series pattern mining algorithm is used to mine fault evolution patterns, generating future fault matrix data containing possible faults within the current period. Then, outlier filtering and data standardization are performed on the future fault matrix data to extract structured fault event data. The occurrence frequency of fault events is counted, and combined with the predicted information of fault location, cause, and region, fault sub-data within the current period is formed. The ship's operating status is determined to be normal or abnormal based on the severity of the fault sub-data. Each fault sub-data corresponds to a judgment standard value. If any fault sub-data exceeds the corresponding judgment standard value, the ship's operating status is considered abnormal.

[0039] The method for determining the ship's operating status when the number of data points does not exceed a threshold is as follows: Correlation analysis can be performed on the operating condition sub-data and vibration sub-data in the historical shafting data for each target. A mathematical correlation model between the operating condition parameters and vibration parameters is established using regression analysis, quantifying the influence intensity of the operating condition sub-data on the vibration sub-data. Based on this influence intensity, an influence degree curve characterizing the correlation between the two is fitted with the change in the operating condition parameter as the horizontal axis and the corresponding change in the vibration parameter as the vertical axis. Subsequently, real-time operating condition sub-data and vibration sub-data are extracted from the current shafting data. The real-time vibration sub-data is compared with the theoretical vibration value under the corresponding real-time operating condition in the influence degree curve. If the real-time vibration sub-data is within the reasonable fluctuation range of the theoretical vibration value, the ship's operating status is determined to be normal; if it exceeds the reasonable fluctuation range, it is determined to be abnormal. The reasonable fluctuation range can be obtained based on the confidence interval of historical statistics.

[0040] One possible implementation of this application embodiment involves generating future fault matrix data based on fault event matrix data, processing the future fault matrix data, and determining the fault sub-data of the current ship within the current period, including: Time series pattern mining is performed on the fault event matrix data to identify and extract typical fault evolution sequence patterns; Perform similarity matching between the current shaft system data and the initial stages of the identified typical fault evolution sequence patterns; Based on the successfully matched typical fault evolution sequence patterns, the sequence patterns are extrapolated to generate future fault development prediction sequences, which constitute future fault matrix data. Data processing is performed on the data contained in the future fault matrix data to obtain fault event data. Based on the fault event data, the number of fault events occurring in the current period of the current ship is generated to obtain the fault sub-data of the current ship in the current period.

[0041] The fault event matrix data is a structured data representation used to encode the relationships between historical fault events. Rows represent different time segments, columns represent different fault event types, and the values ​​of matrix elements indicate the status of the event within a specific voyage / time period: occurrence = 1, no occurrence = 0.

[0042] Specifically, the constructed fault event matrix data is processed. This matrix can be viewed as a database composed of multiple historical fault sequences, each representing the fault evolution process of a single voyage. More specifically, the PrefixSpan time-series pattern mining algorithm can be used for in-depth analysis of the fault event matrix data. Using "fault combination type - fault characteristic change - time node" as the core dimensions, it mines recurring fault evolution paths with a frequency exceeding a preset frequency threshold. The mined paths are then filtered, removing sequences with excessive fluctuations and unstable evolution trends, retaining sequences with clear fault development stages and consistent characteristic change patterns, thus extracting typical fault evolution sequence patterns. The preset frequency threshold can be set according to actual conditions.

[0043] Furthermore, core feature data of the initial stage of typical fault evolution sequence patterns can be extracted, including the vibration sub-data features and working condition sub-data features of this stage. Corresponding real-time feature data can be extracted from the current shaft system data. The cosine similarity algorithm is used to calculate the similarity between the real-time feature data and the feature data of the initial stage of each typical pattern. The calculated similarity is compared with a preset similarity threshold (such as 0.8). If the similarity is higher than the threshold, it is determined that the match is successful.

[0044] For successfully matched typical fault evolution sequence patterns, predictive inference can be initiated. Specifically, based on the successfully matched typical fault evolution sequence patterns, and combined with the current vibration amplitude and real-time operating conditions of the current shaft system data, the extrapolation starting point is determined. An LSTM time-series prediction model is used to extend the trend of the typical pattern along the time dimension, predicting the development nodes of the fault, the evolution of the fault type, and the changing trends of fault characteristic parameters within the current period. This predictive information is organized according to the structure of the fault event matrix data to form the future fault matrix data. Here, the changing trend of fault characteristic parameters can be the rate of increase of vibration amplitude over time; the rows of the fault event matrix data correspond to predicted fault combinations, the columns correspond to time nodes, and the elements are predicted fault characteristic values.

[0045] Furthermore, the future fault matrix data containing the predicted sequence is analyzed and quantified. Specifically, each stage in the predicted sequence is transformed into specific fault event data, clarifying the fault type, component, and characterization parameters. For each predicted fault event, its possible severity level is estimated based on historical statistical models or physical models. Within the current diagnostic cycle, all predicted fault events are statistically analyzed to generate a prediction result, i.e., the number of fault events. This prediction result includes the total number of fault events expected to occur in future cycles.

[0046] One possible implementation of this application embodiment includes operating condition sub-data such as speed, torque, and power, and vibration sub-data such as torsional vibration, lateral vibration, and longitudinal vibration. The operating condition sub-data and vibration sub-data of each target historical shaft system data are analyzed to determine the influence curve of the operating condition sub-data on the vibration sub-data, including: Feature extraction is performed on the vibration sub-data of each target historical shaft system data to form a vibration feature dataset; For any vibration feature in the vibration feature dataset, a regression correlation model is established between the vibration feature and at least one working condition parameter in the normalized working condition data, and the quantitative influence coefficient of the vibration feature for each working condition parameter is calculated. Based on the quantified influence coefficient, the single-parameter influence curve of each working condition parameter on the vibration characteristics is generated by plotting the change value of a single working condition parameter on the x-axis and the relative change of the vibration characteristics caused by the single working condition parameter on the y-axis. The single-parameter influence curves of each working condition parameter are fused to obtain the influence curve characterizing the comprehensive influence of the working condition sub-data on the vibration sub-data.

[0047] For each target historical axis data set, three types of vibration sub-data sets—torsional vibration, lateral vibration, and longitudinal vibration—time-domain and frequency-domain features are extracted, including root mean square value, peak value, kurtosis, and waveform factor. Frequency-domain features include characteristic frequency amplitude, spectral energy ratio, and harmonic amplitude. All extracted vibration features are organized according to the dimension of "historical data sample - vibration type - feature index". Invalid features (such as features with constant values ​​and no fluctuations) are removed and integrated to form a structured vibration feature dataset.

[0048] Furthermore, the speed, torque, power, and load in the operating condition sub-data can be normalized using min-max normalization, mapping the data to the [0,1] interval to eliminate dimensional differences and obtain normalized operating condition data. For each vibration feature in the vibration feature dataset, a multiple linear regression model is established between that feature and the speed, torque, and power in the normalized operating condition data, with the vibration feature as the dependent variable and the single operating condition parameter as the independent variable. The regression coefficient is obtained through model training, and this regression coefficient is the quantitative influence coefficient of the corresponding operating condition parameter on the vibration feature, used to characterize the influence intensity of each unit change in the single operating condition parameter on the vibration feature.

[0049] Using the normalized change value of each operating condition parameter as the x-axis and the quantified influence coefficient corresponding to that operating condition parameter multiplied by the parameter change value, the relative change of vibration characteristics is obtained as the y-axis. Based on the corresponding data of multiple sets of "parameter change value - relative change of vibration characteristics", three single-parameter influence curves of speed on vibration characteristics, torque on vibration characteristics, and power on vibration characteristics are fitted and generated respectively, clearly showing the correlation between single operating condition parameters and vibration characteristics.

[0050] The weights are determined based on the absolute values ​​of the quantitative influence coefficients of each operating condition parameter on the vibration characteristics. A linear weighted fusion algorithm is used to superimpose the three single-parameter influence curves. During the fusion process, the vertical coordinate values ​​corresponding to each single-parameter curve under the same horizontal axis are calculated, and then the weighted sums are obtained to obtain the comprehensive vertical coordinate value. Based on the corresponding data of multiple sets of comprehensive changes and comprehensive relative changes, a smooth curve is fitted to generate, which is the curve representing the degree of influence of the operating condition sub-data on the comprehensive influence of the vibration sub-data. Among them, the larger the absolute value of the influence coefficient, the higher the weight; the horizontal axis represents the comprehensive change of the operating condition, and the vertical axis represents the comprehensive relative change of the vibration characteristics.

[0051] Step S104: If the operating route corresponding to the current operating plan is a non-specific operating route, then based on the historical shafting data, the current shafting data, and the current operating plan, determine the ship operating status corresponding to the current ship in the current cycle.

[0052] If the route is determined to be a non-specific operating route, a more general reasoning strategy can be adopted due to the lack of direct historical data for the same route. Specifically, a global ship propulsion system performance and vibration correlation model can be established based on all historical shafting data. Combined with the expected operating segments in the current operating plan, the operating load that the shafting may face in the current cycle can be predicted. The current shafting data can then be compared with the "expected state range" predicted based on the general model and the current plan to determine whether the current state is normal or abnormal.

[0053] One possible implementation of this application embodiment includes operating condition sub-data such as speed, torque, and power. Based on historical shafting data, current shafting data, and the current operating plan, the ship's operating status within the current cycle is determined, including: Based on historical shaft system data and current operating plans, determine the future operating condition sub-data within the current cycle; Determine the vibration reference range corresponding to the future operating condition sub-data; Based on the current shafting data and vibration reference range, determine the current shafting status of the ship.

[0054] Among them, the vibration reference range is the reasonable fluctuation range of vibration characteristic values ​​when the shaft system is running normally under a certain working condition in the future. It is the criterion for distinguishing between normal vibration and abnormal vibration.

[0055] The current operational plan is analyzed to extract the route and planned speed profile, and the sea state for the future time period of that route is obtained. An embedded "ship motion-propulsion performance" coupled simulation model is used within the electronic equipment. This model uses ship type, draft, and hull resistance characteristics as known parameters, and planned speed and sea state as inputs. By solving the mechanical equilibrium equations, it calculates in real-time the thrust required to overcome resistance and maintain speed, thereby deriving the main engine load and outputting the estimated speed, torque, and power sequence of the propulsion shaft system at each moment in the future voyage. This estimated time sequence is the future operating condition sub-data for the current period.

[0056] After obtaining the future operating condition sub-data sequence, it is possible to determine what vibration level is "normal" under this expected operating load. Specifically, data from all historical voyages can be retrieved from historical shaft system data and used as sample data to train a "operating condition-vibration mapping benchmark model." This model can employ Gaussian process regression or quantile regression, using historical operating conditions as input and corresponding historical vibration values ​​as output to establish a statistical relationship of "under what operating conditions, the vibration level typically falls within what range." After training, the future operating condition sub-data sequence predicted in step one is input into the model. The model outputs a corresponding statistically significant vibration prediction value and its normal fluctuation upper and lower limits for each future time point in the sequence. Connecting the upper and lower limits of all time points forms a dynamic envelope that varies over time; this envelope represents the vibration benchmark range corresponding to the future operating condition sub-data.

[0057] Furthermore, the system continuously receives current shafting data from sensors, particularly the vibrator data, and compares the currently measured vibration values ​​with a generated vibration reference range corresponding to the current moment in real time. The comparison logic is as follows: if the measured vibration value consistently and stably falls within the dynamic reference range, the ship's shafting status is determined to be "normal"; if the measured vibration value exceeds the reference range, especially exhibiting a continuous upward trend or high-level fluctuations, the status is determined to be "abnormal." The system records the characteristics, amplitude, and duration of the abnormal vibrations as a basis for early warning and preliminary diagnosis.

[0058] One possible implementation of this application embodiment involves determining future operating condition sub-data within the current period, including: The current operational plan's route is compared with the operational routes of all historical operational plans. Based on the similarity of the origin and destination ports, major turning points, and route length, one or more historical voyages with route similarity exceeding a preset similarity threshold are selected from the historical axis data and used as reference historical voyages. Extract historical operating condition sub-data of all reference historical voyages within the expected voyage time period corresponding to the current operation plan, and form a reference operating condition sequence set; Based on the reference operating condition sequence set, a time-series prediction model is used, combined with the current operating plan, to determine the future operating condition sub-data sequence of the ship's shafting in the future voyage within the current period, so as to obtain the future operating condition sub-data within the current period.

[0059] The origin and destination ports are the starting and ending ports of a route, and are fundamental elements for determining whether two routes are similar. Major turning points are key geographical locations along a route where ships need to change course; they are generally identified by latitude and longitude and are important indicators for judging route similarity.

[0060] Specifically, all historical flight plans are retrieved, and the metadata of each plan is extracted: origin and destination ports, major turning points, and route length. The current flight plan is parsed to obtain its corresponding origin and destination port, turning point sequence, and plan length, and the route similarity between the current plan and each historical plan is calculated. More specifically, the origin and destination ports are compared to see if they are completely identical. For the turning point sequence, the Dynamic Time Warping (DTW) algorithm is used to calculate the similarity of their spatial trajectories, and a comprehensive score is given based on the difference in route length. A preset similarity threshold is set, and all historical voyages with scores exceeding this threshold are marked. These voyages are considered to have high comparability in the navigation environment and are defined as reference historical voyages.

[0061] Once the reference historical voyages are determined, the corresponding data packets can be located and extracted from the historical shafting database. Specifically, the start and end times of the current cycle are used as key points to calculate the expected time labels for key points in the future voyage. For each reference historical voyage, the historical moments corresponding to these expected time labels are identified in its historical data, and historical operating condition sub-data within the time window before and after that moment are extracted. The operating condition sequence fragments extracted from all reference voyages and aligned with each stage of the future voyage are then compiled to form the reference operating condition sequence set. This reference operating condition sequence set reflects the actual load conditions experienced by the ship on similar routes and during the same time periods in the past.

[0062] First, the data from the reference operating condition sequence set is split chronologically into training and validation sets to train the LSTM time-series prediction model. After training, the model's prediction boundaries are adjusted based on constraints in the current operating plan, such as preset speed limits, load requirements, and the impact of expected sea states on the operating conditions. The trained model is then used to predict future voyages within the current period, obtaining predicted values ​​for engine speed, torque, and power at different time points. These predicted values ​​are arranged chronologically to form a future operating condition sub-data sequence. Finally, all data in the sequence is integrated to obtain the future operating condition sub-data for the current period.

[0063] One possible implementation of this application embodiment, determining the vibration reference range corresponding to future operating condition sub-data, includes: Historical vibration sub-data corresponding to the reference historical voyage is extracted from historical shaft system data. For each voyage condition included in the future operating condition sub-data, historical vibration characteristic data with the difference between the speed, torque, and load parameters and the voyage condition are all within the preset difference threshold. The historical vibration characteristic data obtained from the screening were statistically analyzed to obtain the sample mean and sample standard deviation of the vibration characteristics under each flight segment operating condition; Based on the changes in operating parameters in the future operating condition sub-data, operating condition correction factors are calculated. These correction factors include speed correction factor, load correction factor, and sea state correction factor. Specifically, the speed correction factor is the ratio of the speed of the future segment to the speed of the corresponding segment in the reference historical voyage; the load correction factor is the ratio of the load of the future segment to the load of the corresponding segment in the reference historical voyage; and the sea state correction factor is a correction percentage determined based on the difference between the future sea state class and the sea state class of the reference historical voyage. Based on the sample mean, sample standard deviation, and operating condition correction coefficient, the lower limit and upper limit of the vibration reference corresponding to the operating condition of each flight segment are calculated to obtain the vibration reference sub-range. The lower limit of the vibration reference is the sample mean multiplied by the minimum operating condition correction coefficient minus the correction value of the sample standard deviation, and the upper limit of the vibration reference is the sample mean multiplied by the maximum operating condition correction coefficient plus the correction value of the sample standard deviation. By integrating the vibration reference sub-ranges corresponding to the operating conditions of all flight segments within the current cycle, the complete vibration reference ranges corresponding to the future operating condition sub-data within the current cycle are obtained.

[0064] Specifically, historical vibration sub-data corresponding to all reference historical voyages can be extracted from historical shaft system data, and the future operating condition sub-data sequence can be divided into fixed time windows. For each time window, precise operating condition backtracking is performed in the historical data. More specifically, from the historical data of the corresponding segments of those reference voyages, all data points that meet the following conditions are selected: the differences between the recorded speed, torque, and load parameters and the operating conditions of the target segment are all within preset difference thresholds. The set of selected data points constitutes the historical vibration characteristic data usable under the operating conditions of the future segment.

[0065] For each future flight segment, statistical analysis is performed on the selected historical vibration characteristic data. For each vibration characteristic of interest, its sample mean and sample standard deviation in the sample set are calculated. This mean and standard deviation together define the historical "health center" and "normal distribution" of the vibration characteristic under the operating conditions of that specific flight segment.

[0066] Since future voyages and historical reference voyages will not be exactly the same in detail, dynamic adjustments can be made by introducing operating condition correction factors. These correction factors include engine speed correction factors, load correction factors, and sea state correction factors. Specifically, the engine speed correction factor is the average engine speed of the future voyage divided by the average engine speed of the corresponding historical voyage segment; the load correction factor is the average load of the future voyage divided by the average load of the corresponding historical voyage segment; and the sea state correction factor is the difference between the predicted sea state level of the future voyage and the actual sea state level experienced in the historical voyage, with a correction ratio determined through a predefined lookup table.

[0067] Using the above results, a dynamic vibration baseline sub-range is calculated for each future voyage's operating conditions. The specific calculation logic is as follows: Lower baseline = (Sample mean × Minimum operating condition correction factor) - k × Sample standard deviation; Upper baseline = (Sample mean × Maximum operating condition correction factor) + k × Sample standard deviation. Here, k is a constant, typically 2 or 3, representing a 2σ or 3σ confidence interval. The minimum and maximum operating condition correction factors are conservative overall estimates of the vibration influence direction, derived by comprehensively considering speed, load, and sea state coefficients. For example, if all corrections tend to increase vibration, the product is taken as a single correction factor; if the directions differ, the possible minimum and maximum effects are calculated separately. Through this calculation, the baseline range is not only based on historical statistics but also scaled and offset according to known differences in future operating conditions.

[0068] Furthermore, the vibration baseline sub-ranges calculated for all future flight segments within the current period are connected in chronological or flight sequence. Since each sub-range is associated with a specific future flight segment's operating condition prediction point, the integrated result is a dynamic envelope that fluctuates over time. This complete envelope represents the vibration baseline range corresponding to the complete future operating condition sub-data within the current period, describing the dynamic path of the vibration health level that should exist throughout the entire expected flight.

[0069] This application provides a method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms. By acquiring current and historical shafting data and operational plans, the method determines whether the current route is a specific route based on the historical operational plans. The routes are then categorized and processed. For specific routes, the corresponding historical shafting data is acquired and combined with the current shafting data to determine the ship's operational status. This utilizes the reference value of historical data for specific routes, enabling a more accurate assessment of the current ship's status for that route. For non-specific routes, the method comprehensively considers historical shafting data, current shafting data, and the current operational plan to determine the ship's operational status. This multi-dimensional analysis improves the accuracy and reliability of the diagnosis, helps to promptly detect anomalies in the ship's propulsion system, and ensures the safe operation of the ship.

[0070] The above embodiments introduce a method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms from the perspective of process flow. The following embodiments introduce a system for diagnosing the operational status of a ship propulsion system based on intelligent algorithms from the perspective of virtual modules or virtual units. For details, please refer to the following embodiments.

[0071] See Figure 2 The intelligent algorithm-based ship propulsion system operation status diagnosis system 20 may specifically include an intelligent algorithm-based ship propulsion system operation status diagnosis device 201 and an electronic device 202.

[0072] See Figure 3 One possible implementation of this application embodiment is that the ship propulsion system operation status diagnostic device 201 based on intelligent algorithms includes: The acquisition module 2011 is used to acquire the current shafting data and historical shafting data corresponding to the current vessel, and to acquire the current operation plan and historical operation plan corresponding to the current vessel. The current shafting data is the shafting data corresponding to the current moment, and the historical shafting data is the shafting data corresponding to each historical operation plan. The shafting data includes vibration sub-data, operating condition sub-data and fault sub-data. The current operation plan is the operation plan corresponding to the current vessel, and the operation plan includes the operation route. The route determination module 2012 is used to determine whether the route corresponding to the current operation plan is a specific operation route based on the historical operation plan. The specific operation route is the route whose historical operation count exceeds a preset threshold. The first state determination module 2013 is used to obtain the target historical shafting data corresponding to the specific operating route if the operating route corresponding to the current operating plan is a specific operating route, and determine the ship operating status corresponding to the current ship in the current period based on the current shafting data and the target historical shafting data. The ship operating status is either abnormal or normal. The second state determination module 2014 is used to determine the ship's operating status in the current cycle based on historical shafting data, current shafting data, and the current operating plan if the operating route corresponding to the current operating plan is a non-specific operating route.

[0073] In one possible implementation of this application embodiment, the fault sub-data includes the number of faults, fault location, fault cause, and fault area. When the current operating plan corresponds to a specific operating route, the first state determination module 2013, when determining the ship's operating state within the current cycle based on the current shafting data and the target historical shafting data, is specifically used for: Filter the first historical shaft system data from the target historical shaft system data and determine the number of first historical shaft system data. The historical target working condition sub-data of the first historical shaft system data is consistent with the current working condition sub-data. If the number exceeds the threshold, the number of fault combinations in different first historical operation plans and the time series length corresponding to the number of fault combinations are determined. Based on the time series length and the number of fault combinations, the first historical operation plans are processed into unsupervised time series data to obtain fault event matrix data. Future fault matrix data is generated based on the fault event matrix data. The future fault matrix data is processed to determine the fault sub-data of the current ship in the current period, so as to determine the ship operation status corresponding to the current ship. If the number does not exceed the threshold, the operating condition sub-data and vibration sub-data of each target historical shafting data are analyzed to determine the influence curve of the operating condition sub-data on the vibration sub-data, and based on the current shafting data and the influence curve, the current ship operating status is determined.

[0074] One possible implementation of this application embodiment is that the first state determination module 2013 generates future fault matrix data based on fault event matrix data, processes the future fault matrix data, and determines the fault sub-data of the current ship in the current period, including: Time series pattern mining is performed on the fault event matrix data to identify and extract typical fault evolution sequence patterns; Perform similarity matching between the current shaft system data and the initial stages of the identified typical fault evolution sequence patterns; Based on the successfully matched typical fault evolution sequence patterns, the sequence patterns are extrapolated to generate future fault development prediction sequences, which constitute future fault matrix data. Data processing is performed on the data contained in the future fault matrix data to obtain fault event data. Based on the fault event data, the number of fault events occurring in the current period of the current ship is generated to obtain the fault sub-data of the current ship in the current period.

[0075] In one possible implementation of this application embodiment, the operating condition sub-data includes speed, torque, and power, and the vibration sub-data includes torsional vibration, lateral vibration, and longitudinal vibration. The first state determination module 2013, when analyzing the operating condition sub-data and vibration sub-data of each target historical shaft system data to determine the influence curve of the operating condition sub-data on the vibration data, is specifically used for: Feature extraction is performed on the vibration sub-data of each target historical shaft system data to form a vibration feature dataset; For any vibration feature in the vibration feature dataset, a regression correlation model is established between the vibration feature and at least one working condition parameter in the normalized working condition data, and the quantitative influence coefficient of the vibration feature for each working condition parameter is calculated. Based on the quantified influence coefficient, the single-parameter influence curve of each working condition parameter on the vibration characteristics is generated by plotting the change value of a single working condition parameter on the x-axis and the relative change of the vibration characteristics caused by the single working condition parameter on the y-axis. The single-parameter influence curves of each working condition parameter are fused to obtain the influence curve characterizing the comprehensive influence of the working condition sub-data on the vibration sub-data.

[0076] In one possible implementation of this application embodiment, the operating condition sub-data includes speed, torque, and power. Based on historical shafting data, current shafting data, and the current operating plan, the second state determination module 2014, when determining the ship's operating state corresponding to the current ship within the current cycle, is specifically used for: Based on historical shaft system data and current operating plans, determine the future operating condition sub-data within the current cycle; Determine the vibration reference range corresponding to the future operating condition sub-data; Based on the current shafting data and vibration reference range, determine the current shafting status of the ship.

[0077] In one possible implementation of this application embodiment, the second state determination module 2014, when determining future operating condition sub-data within the current cycle, is specifically used for: The current operational plan's route is compared with the operational routes of all historical operational plans. Based on the similarity of the origin and destination ports, major turning points, and route length, one or more historical voyages with route similarity exceeding a preset similarity threshold are selected from the historical axis data and used as reference historical voyages. Extract historical operating condition sub-data of all reference historical voyages within the expected voyage time period corresponding to the current operation plan, and form a reference operating condition sequence set; Based on the reference operating condition sequence set, a time-series prediction model is used, combined with the current operating plan, to determine the future operating condition sub-data sequence of the ship's shafting in the future voyage within the current period, so as to obtain the future operating condition sub-data within the current period.

[0078] In one possible implementation of this application embodiment, when determining the vibration reference range corresponding to the future working condition sub-data, the second state determination module 2014 is specifically used for: Historical vibration sub-data corresponding to the reference historical voyage is extracted from historical shaft system data. For each voyage condition included in the future operating condition sub-data, historical vibration characteristic data with the difference between the speed, torque, and load parameters and the voyage condition are all within the preset difference threshold. The historical vibration characteristic data obtained from the screening were statistically analyzed to obtain the sample mean and sample standard deviation of the vibration characteristics under each flight segment operating condition; Based on the changes in operating parameters in the future operating condition sub-data, operating condition correction factors are calculated. These correction factors include speed correction factor, load correction factor, and sea state correction factor. Specifically, the speed correction factor is the ratio of the speed of the future segment to the speed of the corresponding segment in the reference historical voyage; the load correction factor is the ratio of the load of the future segment to the load of the corresponding segment in the reference historical voyage; and the sea state correction factor is a correction percentage determined based on the difference between the future sea state class and the sea state class of the reference historical voyage. Based on the sample mean, sample standard deviation, and operating condition correction coefficient, the lower limit and upper limit of the vibration reference corresponding to the operating condition of each flight segment are calculated to obtain the vibration reference sub-range. The lower limit of the vibration reference is the sample mean multiplied by the minimum operating condition correction coefficient minus the correction value of the sample standard deviation, and the upper limit of the vibration reference is the sample mean multiplied by the maximum operating condition correction coefficient plus the correction value of the sample standard deviation. By integrating the vibration reference sub-ranges corresponding to the operating conditions of all flight segments within the current cycle, the complete vibration reference ranges corresponding to the future operating condition sub-data within the current cycle are obtained.

[0079] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0080] This application also describes an electronic device from the perspective of a physical device, such as... Figure 4 As shown, Figure 4 The illustrated electronic device 202 includes a processor 2021 and a memory 2023. The processor 2021 and the memory 2023 are connected, for example, via a bus 2022. Optionally, the electronic device 202 may also include a transceiver 2024. It should be noted that in practical applications, the transceiver 2024 is not limited to one type, and the structure of this electronic device 202 does not constitute a limitation on the embodiments of this application.

[0081] The processor 2021 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), a FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 2021 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0082] Bus 2022 may include a pathway for transmitting information between the aforementioned components. Bus 2022 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 2022 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0083] The memory 2023 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0084] The memory 2023 stores application code that executes the scheme of this application, and its execution is controlled by the processor 2021. The processor 2021 executes the application code stored in the memory 2023 to implement the content shown in the foregoing method embodiments.

[0085] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers, and can also be servers, etc. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0086] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0087] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0088] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms, characterized in that, The method includes: Obtain the current shafting data and historical shafting data corresponding to the current vessel, and obtain the current operation plan and historical operation plan corresponding to the current vessel. The current shafting data is the shafting data corresponding to the current moment, and the historical shafting data is the shafting data corresponding to each historical operation plan. The shafting data includes vibration sub-data, operating condition sub-data, and fault sub-data. The current operation plan is the operation plan corresponding to the current vessel, and the operation plan includes the operation route. Based on the historical operation plan, determine whether the operation route corresponding to the current operation plan is a specific operation route, wherein the specific operation route is a route whose historical operation count exceeds a preset threshold; If the current operating plan corresponds to a specific operating route, then the target historical shafting data corresponding to the specific operating route is obtained, and the ship operating status corresponding to the current ship in the current period is determined based on the current shafting data and the target historical shafting data. The ship operating status is either an abnormal state or a normal state. If the route corresponding to the current operation plan is a non-specific route, then the ship operation status corresponding to the current ship in the current period is determined based on the historical shafting data, the current shafting data, and the current operation plan.

2. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 1, characterized in that, The fault sub-data includes the number of faults, fault location, fault cause, and fault area. When the current operating plan corresponds to a specific operating route, determining the ship's operating status within the current period based on the current shafting data and the target historical shafting data includes: First historical axle data is filtered from the target historical axle data, and the number of first historical axle data is determined. The historical target working condition sub-data of the first historical axle data is consistent with the current working condition sub-data. The working condition sub-data includes load. If the number exceeds a threshold, the number of fault combinations in different first historical operation plans and the time series length corresponding to each fault combination are determined. Based on the time series length and the number of fault combinations, the first historical operation plans are processed using unsupervised time series data to obtain fault event matrix data. Future fault matrix data is generated based on the fault event matrix data. The future fault matrix data is processed to determine the fault sub-data of the current ship in the current period, so as to determine the ship operation status corresponding to the current ship. If the number does not exceed the number threshold, the operating condition sub-data and vibration sub-data of each target historical shafting data are analyzed to determine the influence curve of the operating condition sub-data on the vibration sub-data, and the ship operation status corresponding to the current ship is determined based on the current shafting data and the influence curve.

3. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 2, characterized in that, The process of generating future fault matrix data based on the fault event matrix data, processing the future fault matrix data, and determining the fault sub-data of the current vessel in the current period includes: Time series pattern mining is performed on the fault event matrix data to identify and extract typical fault evolution sequence patterns; The current shaft system data is matched with the initial stage of the identified typical fault evolution sequence pattern; Based on the successfully matched typical fault evolution sequence patterns, sequence pattern extrapolation is performed to generate future fault development prediction sequences, which constitute the future fault matrix data. Data processing is performed on the data contained in the future fault matrix data to obtain fault event data. Based on the fault event data, the fault event occurrence volume of the current ship in the current period is generated to obtain the fault sub-data of the current ship in the current period.

4. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 2, characterized in that, The operating condition sub-data also includes speed, torque, and power; the vibration sub-data includes torsional vibration, lateral vibration, and longitudinal vibration. The analysis of the operating condition sub-data and vibration sub-data for each target historical shaft system data to determine the influence curve of the operating condition sub-data on the vibration sub-data includes: Feature extraction is performed on the vibration sub-data of each target historical shaft system data to form a vibration feature dataset; For any vibration feature in the vibration feature dataset, a regression correlation model is established between the vibration feature and at least one working condition parameter in the normalized working condition data, and the quantitative influence coefficient of the vibration feature for each working condition parameter is calculated. Based on the quantified influence coefficient, a single-parameter influence curve of each working condition parameter on the vibration characteristic is generated, with the change value of the single working condition parameter on the x-axis and the relative change of the vibration characteristic caused by the single working condition parameter on the y-axis. The single-parameter influence curves of each working condition parameter are fused to obtain the influence degree curve characterizing the comprehensive influence of the working condition sub-data on the vibration sub-data.

5. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 1, characterized in that, The operating condition sub-data includes speed, torque, and power. Determining the ship's operating status within the current period based on the historical shafting data, the current shafting data, and the current operating plan includes: Based on the historical shaft system data and the current operating plan, determine the future operating condition sub-data within the current cycle; Determine the vibration reference range corresponding to the future operating condition sub-data; Based on the current shafting data and the vibration reference range, the current shafting status of the ship is determined.

6. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 5, characterized in that, The determination of future operating condition sub-data within the current period includes: The current operating plan's route is compared with the operating routes of all historical operating plans. Based on the similarity of origin and destination ports, major turning points, and route length, one or more historical voyages with route similarity exceeding a preset similarity threshold are selected from the historical axis data as reference historical voyages. Extract historical operating condition sub-data of all reference historical voyages within the expected flight time period corresponding to the current operation plan to form a reference operating condition sequence set; Based on the reference operating condition sequence set, a time-series prediction model is used, combined with the current operating plan, to determine the future operating condition sub-data sequence of the ship shafting in the future voyage within the current period, so as to obtain the future operating condition sub-data within the current period.

7. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 6, characterized in that, Determining the vibration reference range corresponding to the future operating condition sub-data includes: Historical vibration sub-data corresponding to the reference historical voyage is extracted from the historical shaft system data. For each voyage condition included in the future operating condition sub-data, historical vibration characteristic data in which the differences in speed, torque, and load parameters with the voyage condition are all within a preset difference threshold are selected. The historical vibration characteristic data obtained from the screening were statistically analyzed to obtain the sample mean and sample standard deviation of the vibration characteristics under the operating conditions of each flight segment; Based on the changes in operating parameters of the future operating condition sub-data, operating condition correction coefficients are calculated. These correction coefficients include a speed correction coefficient, a load correction coefficient, and a sea state correction coefficient. The speed correction coefficient is the ratio of the speed of the future segment to the speed of the corresponding segment of the reference historical voyage. The load correction coefficient is the ratio of the load of the future segment to the load of the corresponding segment of the reference historical voyage. The sea state correction coefficient is a correction ratio determined based on the difference between the future sea state level and the sea state level of the reference historical voyage. Based on the sample mean, the sample standard deviation, and the operating condition correction coefficient, the vibration reference lower limit and reference upper limit corresponding to each operating condition of the flight segment are calculated to obtain the vibration reference sub-range. The vibration reference lower limit is the sample mean multiplied by the minimum operating condition correction coefficient minus the correction value of the sample standard deviation, and the vibration reference upper limit is the sample mean multiplied by the maximum operating condition correction coefficient plus the correction value of the sample standard deviation. By integrating the vibration reference sub-ranges corresponding to each flight segment operating condition within the current period, the complete vibration reference range corresponding to the future operating condition sub-data within the current period is obtained.

8. A ship propulsion system operation status diagnosis system based on intelligent algorithms, characterized in that, include: Intelligent algorithm-based diagnostic device for the operational status of ship propulsion systems; An electronic device, the electronic device being used to execute the intelligent algorithm-based ship propulsion system operation status diagnosis method according to any one of claims 1 to 7.

9. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 8, characterized in that, The intelligent algorithm-based ship propulsion system operation status diagnostic device includes: The acquisition module is used to acquire the current shafting data and historical shafting data corresponding to the current vessel, and to acquire the current operation plan and historical operation plan corresponding to the current vessel. The current shafting data is the shafting data corresponding to the current moment, and the historical shafting data is the shafting data corresponding to each historical operation plan. The shafting data includes vibration sub-data, operating condition sub-data, and fault sub-data. The current operation plan is the operation plan corresponding to the current vessel, and the operation plan includes the operation route. The route determination module is used to determine, based on the historical operation plan, whether the operation route corresponding to the current operation plan is a specific operation route, wherein the specific operation route is a route whose historical operation count exceeds a preset threshold. The first state determination module is used to obtain the target historical shafting data corresponding to the specific operating route if the operating route corresponding to the current operating plan is a specific operating route, and determine the ship operating status corresponding to the current ship in the current period based on the current shafting data and the target historical shafting data, wherein the ship operating status is an abnormal state or a normal state. The second state determination module is used to determine the ship operation status of the current ship in the current period based on the historical shafting data, the current shafting data, and the current operation plan if the operation route corresponding to the current operation plan is a non-specific operation route.

10. The method for diagnosing the operational status of a ship propulsion system based on intelligent algorithms according to claim 8 or 9, characterized in that, The electronic device includes: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, said at least one application being configured to: perform the intelligent algorithm-based ship propulsion system operation status diagnosis method according to any one of claims 1 to 7.

Images