A method and system for port railway equipment life cycle management based on one code per item

Abstract

This invention relates to the field of port railway equipment management technology, and in particular to a method and system for full lifecycle management of port railway equipment based on a unique identifier code. The method generates a unique identifier code for each port railway equipment and maps it to static attribute information to establish an initial equipment file. This code is used to collect real-time status data and maintenance records during the operational phase to form a dynamically updated file. Based on the temporal evolution characteristics in the dynamically updated file, the current lifecycle stage is determined and a stage identifier is generated. According to the stage identifier, the corresponding management strategy is extracted from a preset lifecycle management strategy library. Management instructions are generated based on the dynamically updated file and sent to the equipment management terminal for intervention. The execution results are collected and written back to the dynamically updated file. This achieves dynamic closed-loop control and precise operation and maintenance management of the entire lifecycle of port railway equipment, improving equipment management efficiency and decision-making accuracy.

Description

Technical Field

[0001] This invention relates to the field of port railway equipment management technology, and in particular to a method and system for full life-cycle management of port railway equipment based on one item, one code. Background Technology

[0002] Port railway equipment operates under high loads, strong vibrations, and high dust levels, directly impacting port throughput efficiency and railway operational safety. Current technology largely relies on traditional equipment management methods, focusing on equipment ledgers, maintenance plans, and fault reporting. Equipment management departments typically establish internal files based on equipment type or location, with files primarily consisting of static forms covering basic information such as equipment model, installation date, and line affiliation. This information remains largely fixed after equipment commissioning, showing little correlation with subsequent technical upgrades, component replacements, and maintenance history. Inspection records, lubrication and maintenance records, and fault handling forms are often filled out separately by different personnel, resulting in multiple independent data tables. The lack of a unified information indexing mechanism spanning the entire equipment lifecycle makes it difficult to effectively connect the traces left by the same equipment at different stages in a timeline. Equipment condition assessments and maintenance decisions often depend on the personal experience of on-site personnel and periodic manual inspections. Maintenance strategies primarily rely on scheduled and reactive maintenance. While condition monitoring data is collected, it is only used for current fault diagnosis, failing to provide a systematic analysis from the perspective of the overall equipment evolution and phase transitions. Funding and spare parts arrangements are mostly submitted in accordance with the annual plan, which fails to respond to the differentiated needs of individual equipment at different stages. This results in some equipment being over-maintained while others are in disrepair, and the allocation of management resources is not precise enough.

[0003] Information gaps between management stages result in equipment records exhibiting a clear tendency towards staticity. The actual value of these records is concentrated in the asset inventory phase, making it difficult to support refined decision-making during the operational and deterioration phases. Maintenance schedules based on manual experience and fixed cycles fail to reflect the gradual deterioration of equipment performance in a timely manner. Potential faults often only surface after degradation accumulates to a threshold, at which point the maintenance window is already very limited, leading to a persistently high frequency of unplanned downtime. Furthermore, the vast amount of historical data existing in the form of tables and text reports lacks the ability to dynamically integrate it with the current state of the equipment. Management relies on outcome-based statistics rather than process analysis, often failing to provide sufficient decision-making support when equipment enters the performance decline or even scrapping / replacement phase, resulting in a disconnect between management and the actual condition of the equipment. Summary of the Invention

[0004] This invention provides a method and system for full lifecycle management of port railway equipment based on one item, one code, which can solve the problems in the prior art.

[0005] A first aspect of this invention provides a method for full lifecycle management of port railway equipment based on a unique item code, comprising:

[0006] A unique identifier code is generated for the port railway equipment, and the unique identifier code is associated with the static attribute information of the port railway equipment to generate the initial equipment file;

[0007] The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identification code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file.

[0008] Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated.

[0009] Based on the stage identifier, the management strategy corresponding to the life cycle stage is extracted from the preset life cycle management strategy library. The management instructions for the port railway equipment are generated by combining the dynamic update file. The management instructions are issued to the equipment management terminal to perform the corresponding intervention operation and collect the execution results. The dynamic update file is updated based on the execution results.

[0010] A unique identifier code is generated for each port railway equipment, and an association mapping is established between the unique identifier code and the static attribute information of the port railway equipment to generate an initial equipment file, including:

[0011] Obtain the equipment type characteristics and individual equipment characteristics of the port railway equipment, determine the type identifier bit segment and attribute mapping bit segment based on the equipment type characteristics, and generate an encoding rule template;

[0012] Based on the individual characteristics of the device, individual distinguishing information is filled into the attribute mapping bit field of the encoding rule template to generate the unique identifier code;

[0013] Parse the structured and unstructured attribute fields in the static attribute information;

[0014] Establish an index mapping relationship between the unique identifier code and the structured attribute field, and after converting the unstructured attribute field into a standardized attribute descriptor, establish an association mapping relationship with the unique identifier code to obtain an identifier attribute mapping table;

[0015] The unique identifier code, the identifier attribute mapping table, and the static attribute information are written into a preset file storage structure to generate the device initial file.

[0016] The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identifier code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file, including:

[0017] Based on the unique identifier code, the storage location of the initial equipment file is retrieved in the equipment file storage system, and the data collection range and collection frequency constraints corresponding to the port railway equipment are determined according to the static attribute information in the initial equipment file.

[0018] Based on the data collection range and the collection frequency constraints, real-time status data of the port railway equipment is obtained from the monitoring terminal, and maintenance record data associated with the unique identifier code is extracted from the maintenance management system.

[0019] The real-time status data and the maintenance record data are timestamped to obtain timestamped data;

[0020] Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined.

[0021] The timestamp data is written to the storage location corresponding to the initial file of the device according to the writing order to obtain the dynamically updated file.

[0022] Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined, including:

[0023] Read the timestamp sequence of the stored data from the storage location corresponding to the initial file of the device, and extract the timestamp information from the timestamp data to obtain the timestamp to be written;

[0024] Based on the timestamp sequence, a time continuity feature vector containing time interval distribution patterns and data density distribution patterns is constructed, and abnormal time interval segments in the timestamp sequence are identified according to the time interval distribution patterns.

[0025] Associate and match the timestamp to be written with the abnormal time interval segment to determine whether the timestamp to be written belongs to the time filling range of the abnormal time interval segment;

[0026] When it is determined that the timestamp to be written belongs to the time filling range, the relative position weight of the timestamp to be written in the abnormal time interval segment is calculated according to the data density distribution pattern, and the insertion position is determined and an insertion position identifier is generated based on the relative position weight in the abnormal time interval segment.

[0027] When it is determined that the timestamp to be written does not belong to the time filling range, the append position is determined according to the time increment direction of the timestamp sequence and an append position identifier is generated.

[0028] The writing order of the timestamp data in the storage location corresponding to the device's initial file is determined based on the insertion position identifier or the append position identifier.

[0029] Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated, including:

[0030] Based on the dynamically updated archive, the time series of real-time status data and maintenance record data are determined to obtain a multidimensional time series dataset.

[0031] The degradation trend of the multidimensional time series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector.

[0032] The comprehensive degradation feature vector is matched with a preset life cycle stage feature library to obtain a similarity matching result. The life cycle stage feature library includes features of the commissioning stage, stable operation stage, performance degradation stage, and near-expiration stage.

[0033] The confidence level of the port railway equipment at each stage of its life cycle is calculated based on the similarity matching results.

[0034] The life cycle stage with the highest attribution confidence is selected as the current life cycle stage of the port railway equipment, and a corresponding stage identifier is generated based on the current life cycle stage.

[0035] The degradation trend of the multidimensional time-series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector, including:

[0036] The performance index time series data in the multidimensional time series dataset are segmented by a sliding window, and the rate of change and fluctuation amplitude of the performance index are calculated within each sliding window;

[0037] A performance degradation rate curve is constructed based on the rate of change and the fluctuation amplitude. The performance degradation mode of the port railway equipment is determined based on the slope change points and inflection point distribution of the performance degradation rate curve.

[0038] Event density analysis was performed on the maintenance event time series data in the multidimensional time series dataset to statistically analyze the frequency of maintenance events and the distribution of maintenance types within a unit time window;

[0039] Based on the frequency of occurrence and the distribution of maintenance types, a maintenance intensity evolution curve is constructed. Based on the peak change trend and periodic characteristics of the maintenance intensity evolution curve, the maintenance demand change pattern of the port railway equipment is determined.

[0040] By calculating the correlation strength between the performance degradation rate curve and the maintenance intensity evolution curve within the corresponding time interval, the causal response pattern between performance degradation and maintenance requirements can be identified.

[0041] The performance degradation pattern and the maintenance requirement change pattern are weighted and fused according to the causal response pattern to generate the comprehensive degradation feature vector.

[0042] Based on the stage identifier, management strategies corresponding to the lifecycle stage are extracted from a preset lifecycle management strategy library. These strategies are then combined with the dynamically updated archive to generate management instructions for the port railway equipment. These instructions are then issued to the equipment management terminal to execute corresponding intervention operations and collect execution results. The dynamically updated archive is updated based on the execution results, including:

[0043] Based on the stage identifier, the corresponding management strategy is retrieved from the lifecycle management strategy library, and the comprehensive degradation feature vector and historical intervention records are extracted from the dynamically updated archive.

[0044] Based on the historical intervention records, the execution effect and time decay characteristics of historical intervention actions are analyzed. Based on the execution effect and time decay characteristics, the intervention actions in the management strategy are prioritized to obtain the priority ranking result.

[0045] Target intervention actions are selected based on the priority ranking, and management instructions are generated by combining the device status parameters and resource constraints in the dynamically updated archive.

[0046] The management command is sent to the device management terminal, which then performs the corresponding intervention operation and collects the execution results.

[0047] The post-intervention equipment response data in the execution results are compared and analyzed with the comprehensive degradation feature vector to calculate quantitative indicators of the impact of the intervention operation on the performance degradation mode and the maintenance requirement change mode.

[0048] An intervention effect evaluation record is generated based on the aforementioned quantitative indicators, and the dynamic update archive is updated based on the execution results and the intervention effect evaluation record.

[0049] A second aspect of this invention provides a port railway equipment lifecycle management system based on a unique item and code, comprising:

[0050] The filing unit is used to generate a unique identification code for port railway equipment and establish an association mapping between the unique identification code and the static attribute information of the port railway equipment to generate the initial equipment file.

[0051] The file update unit is used to obtain the real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase through the unique identifier code, and write the real-time status data and the maintenance record data into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file;

[0052] The stage determination unit is used to determine the current life cycle stage of the port railway equipment based on the temporal evolution characteristics in the dynamically updated archive, and generate a stage identifier.

[0053] The strategy execution unit is used to extract the management strategy corresponding to the life cycle stage from the preset life cycle management strategy library according to the stage identifier, generate management instructions for the port railway equipment in combination with the dynamic update file, issue the management instructions to the equipment management terminal to perform corresponding intervention operations and collect the execution results, and update the dynamic update file based on the execution results.

[0054] A third aspect of the present invention provides an electronic device, comprising:

[0055] processor;

[0056] Memory used to store processor-executable instructions;

[0057] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0058] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0059] This invention integrates static attributes, operating status, and maintenance records into a single dynamic archive by using a unique identifier code to run through the entire lifecycle of equipment data. This completely connects all data links from initial filing to disposal, forming a complete and traceable archive chain, and providing a unified and accurate data foundation for equipment management.

[0060] This invention automatically determines the current life cycle stage of port railway equipment based on temporal evolution characteristics, freeing stage division from subjective biases caused by human experience and enabling objective and continuous tracking of equipment status. The corresponding management strategies in the stage identifier trigger strategy library can provide differentiated intervention plans for different degrees of aging, degradation patterns, and remaining life characteristics, thereby executing the correct maintenance actions at the right time and improving the pertinence and effectiveness of maintenance strategies.

[0061] This invention forms a closed loop of decision-making, execution, and feedback by updating dynamic archives after management commands are executed, ensuring that every intervention is based on evidence. The archive data is continuously enriched as the equipment's lifecycle progresses, allowing managers to query historical trajectories and current status at any stage, providing a comprehensive overview of the health of all port and railway equipment. The closed-loop mechanism can promptly identify misjudgments and loss of control, automatically correcting subsequent strategy choices.

[0062] This invention reduces reliance on manual inspections and paper-based records, minimizes information loss during data collection and transfer, and significantly lowers equipment management costs. Phased maintenance plans extend equipment service life, reduce production losses caused by unplanned downtime, and ensure port railway equipment maintains optimal operating conditions for extended periods, providing solid support for equipment safety and operational efficiency. Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating the method for full lifecycle management of port railway equipment based on one item, one code, according to an embodiment of the present invention.

[0064] Figure 2 This is a schematic diagram of the process for generating the initial file of the device according to an embodiment of the present invention;

[0065] Figure 3 This is a schematic diagram of the performance degradation rate curve according to an embodiment of the present invention;

[0066] Figure 4 This is a schematic diagram of the maintenance intensity evolution curve according to an embodiment of the present invention. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0069] Figure 1 This is a flowchart illustrating the port railway equipment lifecycle management method based on one item, one code, according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0070] A unique identifier code is generated for the port railway equipment, and the unique identifier code is associated with the static attribute information of the port railway equipment to generate the initial equipment file;

[0071] The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identification code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file.

[0072] Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated.

[0073] Based on the stage identifier, the management strategy corresponding to the life cycle stage is extracted from the preset life cycle management strategy library. The management instructions for the port railway equipment are generated by combining the dynamic update file. The management instructions are issued to the equipment management terminal to perform the corresponding intervention operation and collect the execution results. The dynamic update file is updated based on the execution results.

[0074] Figure 2 This is a schematic diagram illustrating the process of generating the initial device file according to an embodiment of the present invention. Figure 2 As shown, a unique identifier code is generated for the port railway equipment, and the unique identifier code is associated with the static attribute information of the port railway equipment to generate the initial equipment file, including:

[0075] Obtain the equipment type characteristics and individual equipment characteristics of the port railway equipment, determine the type identifier bit segment and attribute mapping bit segment based on the equipment type characteristics, and generate an encoding rule template;

[0076] Based on the individual characteristics of the device, individual distinguishing information is filled into the attribute mapping bit field of the encoding rule template to generate the unique identifier code;

[0077] Parse the structured and unstructured attribute fields in the static attribute information;

[0078] Establish an index mapping relationship between the unique identifier code and the structured attribute field, and after converting the unstructured attribute field into a standardized attribute descriptor, establish an association mapping relationship with the unique identifier code to obtain an identifier attribute mapping table;

[0079] The unique identifier code, the identifier attribute mapping table, and the static attribute information are written into a preset file storage structure to generate the device initial file.

[0080] Obtaining the equipment type characteristics and individual equipment characteristics of port railway equipment is the starting point for generating unique identification codes. Equipment type characteristics cover the major category and functional type to which the equipment belongs, such as common equipment in port railway scenarios like rail cranes, shunting locomotives, freight cars, turnout switching devices, and signal interlocking equipment. Different types of equipment have fundamental differences in management logic, maintenance cycles, and failure modes, thus requiring differentiation in the coding structure. Individual equipment characteristics are the basis for distinguishing different individuals within the same type of equipment, including information such as the factory serial number, installation location number, date of commissioning, and port area section to which it belongs. Based on the equipment type characteristics, the division method of the type identifier bit field and attribute mapping bit field is determined, forming a coding rule template. The type identifier bit field is used to carry the identification information of the major category and subcategory of equipment, while the attribute mapping bit field reserves field space for filling individual differentiation information. The bit field division of the coding rule template must balance scalability and uniqueness to ensure that the coding space will not conflict or be exhausted in scenarios where the number of port railway equipment expands significantly.

[0081] Based on the individual characteristics of the equipment, unique identifier codes are generated by filling the attribute mapping segments in the coding rule template with individual distinguishing information. The filling process follows preset segment alignment rules. For fields with insufficient length, padding is used to maintain consistent segment width. For fields containing special characters, standardized escaping is performed to ensure that the overall code conforms to the established character set constraints. The generated unique identifier code is globally unique, and no two pieces of equipment can share the same code within the same port railway management system. The unique identifier code can be attached to the equipment body in the form of a QR code, RFID tag, or barcode, facilitating on-site scanning and data association.

[0082] When processing static attribute information, it is necessary to separate structured and unstructured attribute fields. Structured attribute fields refer to attributes with clear data types, fixed formats, and predefined field names, such as equipment rated load (unit: tons), track gauge (unit: millimeters), manufacturer code, year of manufacture, and design service life (unit: years). These fields can be directly incorporated into relational data tables for storage and retrieval. Unstructured attribute fields refer to attribute descriptions existing in the form of free text, drawings, technical specifications, photographs, etc., such as equipment installation and commissioning instructions, original factory inspection reports, and equipment appearance images. These fields cannot be directly stored in a fixed format and require further standardization processing. The parsing process scans the raw attribute information using predefined field recognition rules, classifying identifiable fields into structured categories and classifying content that cannot match the fixed format into unstructured categories.

[0083] When establishing an index mapping relationship between unique identifier codes and structured attribute fields, the unique identifier code is used as the primary key. The field names and values ​​of each structured attribute field are written into the index mapping table as key-value pairs. This index mapping relationship supports fast forward queries (retrieving equipment attributes via unique identifier codes) and reverse queries (locating the corresponding equipment code via specific attribute values), meeting the dual needs of batch filtering and precise positioning in port and railway equipment management. For unstructured attribute fields, they are converted into standardized attribute descriptors before establishing an association mapping relationship with the unique identifier code. The conversion process for standardized attribute descriptors includes: extracting key attribute tags from free text content to form a structured set of descriptive tags; extracting metadata information such as file type, version number, file size, and storage path from drawings and image files to form file descriptors; and extracting document number, version information, and document summary from technical specification documents to form document descriptors. Through these conversions, the core information of unstructured attribute fields is saved in a searchable and associative standardized form, while the original unstructured content is retained in the storage system via reference links, ensuring the integrity of the original information is not lost. The index mapping relationship of structured attribute fields is integrated with the association mapping relationship of unstructured attribute fields to form a complete identifier attribute mapping table.

[0084] The structural design of the identifier attribute mapping table needs to consider the compatibility of subsequent dynamic data writing and query expansion. At the static attribute level, the identifier attribute mapping table records the inherent attributes of the device during the factory and initial installation stages, which in principle will not change during the device's life cycle. At the extension interface level, the identifier attribute mapping table reserves an interface for association with dynamically updated files, using a unique identifier code as the link, so that the real-time status data and maintenance record data generated in the subsequent operation stage can form a complete data loop with the static attribute information.

[0085] The unique identifier code, identifier attribute mapping table, and static attribute information are written into a preset file storage structure to generate the initial equipment file. The file storage structure adopts a hierarchical design: the top layer uses the unique identifier code as the index node, the middle layer stores the identifier attribute mapping table, and the bottom layer stores the original static attribute information and unstructured raw files. After the write operation is completed, the file storage structure is checked for integrity, verifying the uniqueness constraint of the unique identifier code, the compliance of the data type of the structured attribute fields, and the validity of the storage path of the unstructured attribute fields. If the check passes, the initial equipment file enters a valid state, becoming the basic data source for subsequent data collection and lifecycle stage determination. For files that fail the check, abnormal field information is recorded and a manual review process is triggered to ensure that the data quality of the initial file meets the basic requirements of full lifecycle management. In practical scenarios with a large number of port railway equipment, the batch generation of initial equipment files can be achieved through the equipment information import interface, supporting batch parsing, encoding generation, and file writing of attribute information for multiple devices, significantly improving equipment filing efficiency.

[0086] The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identifier code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file, including:

[0087] Based on the unique identifier code, the storage location of the initial equipment file is retrieved in the equipment file storage system, and the data collection range and collection frequency constraints corresponding to the port railway equipment are determined according to the static attribute information in the initial equipment file.

[0088] Based on the data collection range and the collection frequency constraints, real-time status data of the port railway equipment is obtained from the monitoring terminal, and maintenance record data associated with the unique identifier code is extracted from the maintenance management system.

[0089] The real-time status data and the maintenance record data are timestamped to obtain timestamped data;

[0090] Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined.

[0091] The timestamp data is written to the storage location corresponding to the initial file of the device according to the writing order to obtain the dynamically updated file.

[0092] When retrieving the storage location of the initial file of a device in the device file storage system based on its unique identifier, the unique identifier is used as the index key. An exact match query is performed in the index table of the device file storage system to locate the file storage node corresponding to the port railway equipment. Since a mapping relationship between the unique identifier and the storage location is established when the initial file of each port railway device is generated, the retrieval process can be completed in constant time complexity, avoiding the decrease in retrieval efficiency caused by the increase in the number of devices.

[0093] After locating the storage location, the static attribute information in the initial equipment file is further read to determine the data acquisition range and frequency constraints for the port railway equipment. The static attribute information includes equipment type, model, installation location, design operating parameters, and a list of key monitoring indicators. Different types of port railway equipment correspond to different data acquisition ranges. For example, track scale equipment requires the collection of load cell signals, track deformation, and ambient temperature and humidity data, while electric locomotive equipment requires the collection of traction current, braking pressure, and axle temperature data. The acquisition frequency constraints are determined based on the equipment's operating condition sensitivity and historical fault modes. For high-frequency dynamic parameters such as vibration signals, the acquisition frequency constraints are higher; for parameters with slow changes, such as lubricating oil level, the constraints are relatively lower. By matching the static attribute information with a preset acquisition rule base, an acquisition configuration scheme for the equipment can be automatically generated, including the lower limit of the acquisition frequency for each monitoring indicator. With upper limit of sampling frequency And the priority weights of various data sources.

[0094] Based on the defined data acquisition range and frequency constraints, real-time status data of the port railway equipment is acquired from monitoring terminals. These terminals are deployed at key locations on the equipment and convert the physical quantities collected by sensors into digital signals via wired or wireless communication, then upload the data. When acquiring real-time status data, a unique identifier code serves as the equipment's identity credential to initiate a data request to the monitoring terminal. The monitoring terminal selects the corresponding data channel based on the acquisition range parameters carried in the request and controls the data upload interval according to the acquisition frequency constraints, ensuring that the uploaded data volume meets the accuracy requirements of status monitoring without exceeding the storage and transmission bandwidth limits. Simultaneously, maintenance record data associated with the unique identifier code is extracted from the maintenance management system. This maintenance record data includes structured and unstructured content such as inspection records, fault reports, parts replacement records, lubrication and maintenance records, and maintenance acceptance reports. All of this data is stored in the maintenance management system using the unique identifier code as the associated field; therefore, a precise matching query can be used to fully extract the entire maintenance history of the equipment.

[0095] The acquired real-time status data and maintenance record data are timestamped to obtain timestamped data. The timestamping uses a unified Coordinated Universal Time (UTC) format with millisecond-level precision to ensure comparability of data from different data sources on the timeline. For real-time status data, the timestamp records the physical moment when the sensor completes a sampling cycle, not the moment the data arrives at the storage system, thus eliminating the impact of network transmission delays on timing accuracy. For maintenance record data, the timestamp records the actual moment the maintenance operation occurs. If the maintenance record already contains an operation time field, it is directly extracted and uniformly converted to UTC format; if the maintenance record only contains date information and lacks a specific moment, the start time of that date is used as the default timestamp, and the time precision level is marked as "day-level" in the data quality label field to distinguish between time-series data of different precisions during subsequent analysis.

[0096] Based on the time continuity relationship between the timestamp information in the time stamp data and the data already stored in the device's initial file, the writing order of the time stamp data in the storage location is determined. The data already stored in the device's initial file is arranged in ascending order of timestamps, forming a continuous time-series data chain. The timestamp set of newly acquired time-marked data is compared with the timestamp set of stored data to identify three scenarios: First, the timestamp of the new data is later than the latest timestamp of the stored data, which is a sequential append scenario, and is directly inserted at the end of the time-series data chain in ascending order of time. Second, the timestamp of the new data overlaps with the stored data, which is a data supplementation or retransmission scenario. It is necessary to accurately locate the insertion position according to the timestamp within the overlapping range, and perform deduplication verification on the data with duplicate timestamps. If the data content is consistent, the duplicates are discarded. If the data content is inconsistent, the two copies of the data are retained and marked as "conflict pending verification". Third, the timestamp of the new data is earlier than the earliest timestamp of the stored data, which is a historical data backfilling scenario. It is inserted at the beginning of the time-series data chain in ascending order of time, and a time-series continuity and integrity check is triggered to assess whether there is a time gap between the backfilled data and the existing data. If a gap exists, a missing data placeholder record is generated in the gap interval to maintain the structural integrity of the time-series data chain.

[0097] Through the aforementioned time continuity analysis, the specific writing order of each time-marked data entry in the storage location is determined, forming an ordered sequence of write operations. The time-marked data is written one by one to the storage location corresponding to the device's initial file according to the writing order. The writing process employs a transaction mechanism to ensure atomicity, guaranteeing that each write operation either succeeds entirely or rolls back entirely, avoiding inconsistencies in the file data due to write interruptions. After each batch of data is written, the "latest data timestamp" and "total number of data records" fields in the file metadata are updated to quickly determine the file's time-series coverage during subsequent retrieval. Once all write operations are completed, the device's initial file is transformed into a dynamically updated file containing real-time status data and maintenance records from the operational phase. This file continuously accumulates time-series information throughout the device's entire lifecycle, indexed by a unique identifier, providing a complete data foundation for subsequent lifecycle phase determination.

[0098] In actual port railway operations, monitoring terminals of different devices may experience data breaks due to network fluctuations or device offline. By setting the maximum allowable break duration threshold in the acquisition frequency constraint, abnormal break points can be automatically identified and a supplementary acquisition request can be triggered. The supplementary acquisition data will be correctly returned to its original position according to the aforementioned time continuity rules, thereby ensuring the integrity and coherence of the time-series data chain in the dynamic update archive and laying a reliable data foundation for the accurate assessment of equipment health status.

[0099] Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined, including:

[0100] Read the timestamp sequence of the stored data from the storage location corresponding to the initial file of the device, and extract the timestamp information from the timestamp data to obtain the timestamp to be written;

[0101] Based on the timestamp sequence, a time continuity feature vector containing time interval distribution patterns and data density distribution patterns is constructed, and abnormal time interval segments in the timestamp sequence are identified according to the time interval distribution patterns.

[0102] Associate and match the timestamp to be written with the abnormal time interval segment to determine whether the timestamp to be written belongs to the time filling range of the abnormal time interval segment;

[0103] When it is determined that the timestamp to be written belongs to the time filling range, the relative position weight of the timestamp to be written in the abnormal time interval segment is calculated according to the data density distribution pattern, and the insertion position is determined and an insertion position identifier is generated based on the relative position weight in the abnormal time interval segment.

[0104] When it is determined that the timestamp to be written does not belong to the time filling range, the append position is determined according to the time increment direction of the timestamp sequence and an append position identifier is generated.

[0105] The writing order of the timestamp data in the storage location corresponding to the device's initial file is determined based on the insertion position identifier or the append position identifier.

[0106] The system reads the timestamp sequence of stored data from the storage location corresponding to the device's initial file. This timestamp sequence records the timestamps attached to historical data writes and is arranged in the order of data writing. At the same time, it extracts the timestamp information from the timestamp data to be written to obtain the timestamp to be written. Timestamp to be written This reflects the actual time of the business transaction corresponding to the timestamp data, not the time the data arrives at the storage end; there is a delay difference between the two. For port railway equipment, due to network interruptions and equipment offline issues at the on-site data acquisition terminals, the deviation between the data arrival time and the data generation time is quite common in engineering practice. Therefore, it is necessary to determine the data writing order based on timestamp information rather than the order of arrival.

[0107] Based on the read timestamp sequence, a time continuity feature vector is constructed that includes a time interval distribution pattern and a data density distribution pattern. The time interval distribution pattern is calculated by determining the interval between adjacent timestamps. (in The distribution pattern of the timestamp sequence is described by the index number of adjacent timestamp pairs. Under normal operating conditions, the data acquisition interval of port railway equipment should be maintained within a certain range. If the interval between two adjacent timestamps... Significantly larger than the average interval of the entire sequence If a certain threshold is reached, the segment is identified as an anomalous time interval segment, indicating that there is missing data or delayed data arrival within that time range. The threshold for identifying anomalous time interval segments can be determined by combining the statistical characteristics of historical data, for example, when... Exceed with standard deviation When the sum is a multiple of several times, the corresponding segment is marked as an abnormal segment. The data density distribution pattern describes the density of data points in each time period, which is used to calculate the relative position weight of the data to be written in the abnormal segment, providing a basis for the precise location of the insertion position.

[0108] timestamp to be written The identified abnormal time interval segments are then matched and associated. Specifically, for each abnormal time interval segment, its start timestamp is recorded as follows: The end timestamp is recorded as The time filling range is the interval. . judge Does it meet the requirements? If the conditions are met, the timestamp to be written is determined to belong to the time filling range of the abnormal time interval segment, indicating that the data to be written is supplementary data for the missing time period; if the above conditions are not met for all abnormal time interval segments, the timestamp to be written is determined not to belong to any time filling range, indicating that the data is newly generated time series data.

[0109] When it is determined that a timestamp needs to be written... Belongs to a certain abnormal time interval segment When filling the time range, calculate based on the data density distribution pattern. Relative position weight within this segment Relative position weights Taking all factors into consideration The absolute position on the timeline and the distribution density of existing data points within that segment. If there is already a small amount of supplementary data within that segment, then the time stamp positions of these supplementary data points must be considered, and the segment must be determined in ascending order of time. The specific location where it should be inserted. The calculation formula is This value reflects The normalized time position within the abnormal segment. Combining the density distribution of existing data points within the segment, the optimal insertion point satisfying the time continuity constraint is found along the time-increasing direction, and a corresponding insertion position identifier is generated. Insertion position marker The specific insertion index in the target storage location is recorded, so that after the insertion operation is completed, the timestamp sequence in the storage location still maintains strict time-incrementing order, thereby ensuring data consistency for subsequent time series analysis and lifecycle stage determination.

[0110] When it is determined that a timestamp needs to be written... If a data point does not fall within any time padding range, it indicates that its timestamp is later than the maximum value in the current timestamp sequence, or although it is earlier than some existing timestamps, it does not fall into any abnormal segment. For the former, it is processed according to the increasing time direction of the timestamp sequence. Append to the end of the sequence and generate an append position identifier. Add location markers This points to the position after the last piece of data in the storage location. For the latter, the time-incrementing direction is also followed; the first data in the timestamp sequence that is greater than... The timestamp location is used as the target location for appending and an append location identifier is generated. Append operations do not involve adjustments to the existing data storage structure, and their execution efficiency is higher than that of insert operations. They are suitable for scenarios where data arrives in order and normally.

[0111] Based on the generated insertion position identifier Or add location markers Finally, the writing order of the timestamp data in the corresponding storage location of the device's initial file is determined, if an insertion position identifier is used. Then the time stamp data will be written. The pointer points to the specified location, and the existing data after that location is shifted one storage unit to the right to complete the ordered insertion; if an append position identifier is used... Then the time stamp data will be written directly. The system can reach the specified end or target location without moving existing data. After writing, the timestamp sequence in the storage location maintains temporal continuity and order, providing an accurate and complete data foundation for subsequent determination of the lifecycle stage of port railway equipment based on temporal evolution characteristics. This mechanism effectively solves the problem of out-of-order data writing caused by network fluctuations and acquisition delays in complex operating environments of port railway equipment, ensuring the consistency and reliability of data temporal sequence in dynamically updated archives.

[0112] Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated, including:

[0113] Based on the dynamically updated archive, the time series of real-time status data and maintenance record data are determined to obtain a multidimensional time series dataset.

[0114] The degradation trend of the multidimensional time series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector.

[0115] The comprehensive degradation feature vector is matched with a preset life cycle stage feature library to obtain a similarity matching result. The life cycle stage feature library includes features of the commissioning stage, stable operation stage, performance degradation stage, and near-expiration stage.

[0116] The confidence level of the port railway equipment at each stage of its life cycle is calculated based on the similarity matching results.

[0117] The life cycle stage with the highest attribution confidence is selected as the current life cycle stage of the port railway equipment, and a corresponding stage identifier is generated based on the current life cycle stage.

[0118] When determining the current lifecycle stage of port railway equipment based on dynamically updated archives, it is first necessary to extract real-time status data and maintenance record data from the archives. These two types of data are then aligned and arranged along a timeline to construct a multidimensional time-series dataset. Real-time status data includes multiple sensor-collected dimensions such as vibration amplitude, temperature readings, current fluctuations, operating speed, and braking response time. Maintenance record data includes discrete record fields such as fault type codes, maintenance man-hours, replaced parts categories, and maintenance intervals. Aligning these two types of data on a unified timeline creates a multidimensional time-series dataset where each time step contains both status and maintenance dimensions, providing a complete input foundation for subsequent degradation trend extraction.

[0119] When extracting degradation trends from multidimensional time-series datasets, for real-time state data, a sliding window approach is used to calculate the trend slope and variance of the numerical sequences for each state dimension, identifying whether there are monotonically decreasing or increasing degradation signals in each dimension. Specifically, a fixed window length is gradually slid across the time series, fitting a linear trend to the numerical sequences within the window, extracting the slope as an estimate of the degradation rate for that dimension within the window, and simultaneously calculating the mean squared error of the values ​​within the window as a volatility intensity index. For maintenance record data, the frequency of fault occurrence, the trend of changes in average maintenance man-hours, and the recurrence interval of similar faults are statistically analyzed per unit time to identify whether maintenance needs show an increasing trend, thereby extracting the maintenance need change pattern. The performance degradation pattern is jointly described by the degradation rate vector and volatility intensity vector of each state dimension, and the maintenance need change pattern is composed of indicators such as the fault frequency change rate, the average maintenance man-hour change rate, and the recurrence interval shortening rate.

[0120] The performance degradation pattern is fused with the maintenance requirement change pattern to generate a comprehensive degradation feature vector. The fusion process involves weighted concatenation: the performance degradation mode vector is processed using weighted concatenation. Vector of Changes in Maintenance Requirements Assign weight coefficients respectively and ,in The weight values ​​are pre-configured based on the device type. The calculation method for the comprehensive degradation feature vector is as follows: ,in This indicates a weighted concatenation operation, which involves scaling two vectors according to their weights and then concatenating them sequentially into a higher-dimensional feature vector. This is useful for vibration-sensitive equipment (such as axles, bearings, and gearboxes). Take a larger value to highlight the contribution of the state degradation signal; for equipment that relies on manual inspection (such as turnout switching mechanisms). Larger values ​​are used to increase the weight of changes in the maintenance records. The fused composite degradation feature vector is uniformly normalized to a unit length to eliminate matching bias caused by different units.

[0121] The pre-defined lifecycle stage feature library contains four types of stage feature prototypes: commissioning stage features, stable operation stage features, performance degradation stage features, and near-scrap stage features. The commissioning stage feature prototype corresponds to the break-in period characteristics of equipment when it is first put into use, characterized by brief fluctuations in some state dimensions followed by stabilization, and maintenance records mainly consisting of routine checks. The stable operation stage feature prototype corresponds to equipment in normal working condition, with the trend slope of each state dimension close to zero, low variance in fluctuation, and infrequent failures. The performance degradation stage feature prototype corresponds to equipment performance beginning to decline significantly, with the degradation rate of key dimensions such as vibration and temperature remaining negative, increased failure frequency, and increased maintenance hours. The near-scrap stage feature prototype corresponds to equipment performance that has severely deteriorated, with multiple state dimensions simultaneously exhibiting rapid degradation, extremely high failure frequency, and extremely short intervals between repeated failures. Each stage feature prototype is associated with a reference vector of the same dimension as the comprehensive degradation feature vector. It means that among them These correspond to the four stages mentioned above.

[0122] The comprehensive degradation feature vector Reference vectors for each stage in the lifecycle stage feature library Similarity matching is performed, using cosine similarity as the matching metric. Cosine similarity at each stage The calculation is as follows:

[0123] ;

[0124] in Let be the inner product of two vectors. and Each of them Norm. The range of cosine similarity values ​​is... The closer the value is to 1, the more similar the comprehensive degradation feature vector is to the reference vector of that stage. Similarity values ​​for the four stages are obtained. Then, similarity is converted into attribution confidence using softmax normalization. The calculation method is as follows:

[0125] ;

[0126] in This is a temperature coefficient used to control the smoothness of the confidence level distribution. When the value is smaller, the confidence distribution is more concentrated; when the value is larger, the distribution is smoother. The default value is determined based on historical device data and is usually set to [value]. to Between. Confidence level of attribution at each stage. The sum is 1, each The representative equipment belongs to the first Probability estimates for each life cycle stage.

[0127] After obtaining the attribution confidence scores for the four lifecycle stages, select... The highest stage is used as the current lifecycle stage of the device. If the maximum confidence value exceeds a preset threshold (e.g., 0.6), the stage is directly confirmed as the current stage. If the maximum confidence value does not exceed the threshold, it is determined to be a stage transition state. The stage identifiers with the highest and second highest confidence values ​​are recorded, and intervention suggestions for both stages are considered in the management strategy to avoid management response delays due to unclear stage determination. Based on the confirmed current lifecycle stage, a corresponding stage identifier is generated and written to the dynamic update file. The stage identifier adopts a structured encoding format, including the device's unique identifier code, stage category code (corresponding to four categories: commissioning, stable operation, performance degradation, and nearing scrap), determination timestamp, and the highest assigned confidence value, which facilitates direct parsing and calling by the subsequent management strategy extraction module.

[0128] After the stage identifier is generated, a file update operation is triggered simultaneously. The comprehensive degradation feature vector of this stage determination, the confidence distribution of each stage's attribution, and the final determination result are all recorded in the stage determination history field of the dynamically updated file, forming a complete stage evolution trajectory. This historical trajectory can be used for continuous optimization of the determination model. By comparing the results of manual review with the results of automatic determination, the reference vectors of each stage in the life cycle stage feature library are continuously corrected to improve the matching accuracy.

[0129] The degradation trend of the multidimensional time-series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector, including:

[0130] The performance index time series data in the multidimensional time series dataset are segmented by a sliding window, and the rate of change and fluctuation amplitude of the performance index are calculated within each sliding window;

[0131] A performance degradation rate curve is constructed based on the rate of change and the fluctuation amplitude. The performance degradation mode of the port railway equipment is determined based on the slope change points and inflection point distribution of the performance degradation rate curve.

[0132] Event density analysis was performed on the maintenance event time series data in the multidimensional time series dataset to statistically analyze the frequency of maintenance events and the distribution of maintenance types within a unit time window;

[0133] Based on the frequency of occurrence and the distribution of maintenance types, a maintenance intensity evolution curve is constructed. Based on the peak change trend and periodic characteristics of the maintenance intensity evolution curve, the maintenance demand change pattern of the port railway equipment is determined.

[0134] By calculating the correlation strength between the performance degradation rate curve and the maintenance intensity evolution curve within the corresponding time interval, the causal response pattern between performance degradation and maintenance requirements can be identified.

[0135] The performance degradation pattern and the maintenance requirement change pattern are weighted and fused according to the causal response pattern to generate the comprehensive degradation feature vector.

[0136] For performance index time-series data in multidimensional time-series datasets, a sliding window segmentation method is used to structure the historical observation sequences of each dimension's performance index. A fixed-length time window is set, and a certain proportion of overlapping areas is retained between adjacent windows to avoid information truncation at window boundaries. Within each sliding window, the rate of change and fluctuation amplitude of the performance index within the window's coverage period are calculated. The rate of change reflects the trend direction and speed of change of the performance index between the start and end times of the window, and is obtained by dividing the difference between the mean at the end of the window and the mean at the beginning by the window duration. The fluctuation amplitude characterizes the dispersion of the performance index within the window, and is quantified using the range or standard deviation of all sampled values ​​within the window. For port railway equipment, typical performance indicators include traction motor current, braking response time, track geometric deviation, and root mean square value of bearing vibration acceleration. The rate of change and fluctuation amplitude of different indicators within the same sliding window together constitute the performance state description vector of that window.

[0137] The change rate sequence calculated by each sliding window is arranged in chronological order to form a performance degradation rate curve, which shows the performance evolution trajectory of the equipment from the time it is put into operation to the present moment on the time axis.

[0138] Figure 3 This is a schematic diagram of the performance degradation rate curve according to an embodiment of the present invention. Figure 3As shown, the curve illustrates the evolution of the rate of change of performance indicators of port railway equipment over time within a sliding window. The horizontal axis represents the window numbers arranged chronologically, and the vertical axis represents the calculated rate of change of performance indicators within each window, expressed in monthly increments. The curve shows an overall trend from a relatively flat initial state to a significant upward trend later, reflecting a gradually accelerating rate of equipment performance degradation. The data points on the curve correspond to the calculated rate of change for each sliding window, with adjacent points connected by solid lines. The change in curvature visually represents the shift in the acceleration of degradation. Sections with significantly increased slopes can be observed in the middle and later parts of the curve; these locations correspond to the nodes where the equipment enters the accelerated degradation and rapid failure stages. A second derivative curve, represented by a dashed line, is also superimposed on the figure to help identify moments of significant slope changes. When the absolute value of the second derivative exceeds a preset threshold, it is determined to be a slope change point. These points appear as local peaks with large fluctuations in the figure, providing a quantitative basis for the division of degradation stages.

[0139] The performance degradation rate curve is analyzed to detect slope change points, identifying moments when the curve's slope changes significantly. These moments correspond to acceleration or deceleration points in the equipment's performance degradation rate. A difference-based second derivative estimation method is used for slope change point detection; when the absolute value of the second derivative exceeds a preset threshold, that moment is considered a slope change point. Simultaneously, inflection point distribution analysis is performed on the performance degradation rate curve. Inflection points, where the curve changes from convex to concave or from concave to convex, indicate a change in the direction of degradation acceleration. Combining the location distribution of slope change points and the density of inflection points, equipment performance degradation is categorized into different modes, such as initial slow degradation, mid-term accelerated degradation, and late-term rapid failure. A structured description of the performance degradation mode, i.e., a performance degradation mode vector, is constructed using characteristic parameters of each stage (including start and end times, average rate of change, and mean curve curvature). .

[0140] When performing event density analysis on maintenance event time-series data in a multidimensional time-series dataset, a fixed time window (the length of which is consistent with the time scale of the sliding window in performance analysis) is used as the statistical unit. The frequency and type distribution of maintenance events within each time window are statistically analyzed. Maintenance types typically include preventative maintenance, fault repair, planned overhaul, and ad-hoc inspections. Within each time window, the occurrence frequency of each type of maintenance event is counted, and the proportion of each type of maintenance event to the total maintenance events in that window is calculated, resulting in a maintenance type distribution vector. The frequency and type distribution together describe the intensity of maintenance demand and the structure of maintenance behavior within that time window.

[0141] The frequency of maintenance events in each time window is arranged chronologically to form a maintenance intensity evolution curve. Peak value trend analysis is performed on this curve, local maxima are extracted, and the time interval and amplitude variation trends between adjacent peaks are statistically analyzed. If the peak amplitude increases over time, it indicates a continuous increase in equipment maintenance demand; if the peak interval gradually shortens, it indicates that maintenance events are occurring more frequently. Periodic characteristic analysis is performed on the maintenance intensity evolution curve, using the autocorrelation function method to estimate the main periodic components of the curve and identify periodic patterns in maintenance demand (such as periodic patterns related to seasonal operating conditions and scheduled overhauls). Combining the peak value trend and periodic characteristics, the maintenance demand variation pattern of port railway equipment is determined, and a maintenance demand variation pattern vector is established. The structured representation includes components such as maintenance frequency growth rate, mean peak amplitude, main period length, and period stability coefficient.

[0142] Figure 4 This is a schematic diagram of the maintenance intensity evolution curve according to an embodiment of the present invention. Figure 4 The graph illustrates the variation in the frequency of maintenance events on port railway equipment within a unit time window over time. The horizontal axis represents months, and the vertical axis represents the total number of maintenance events per month. The curve shows an overall fluctuating upward trend, indicating that the demand for equipment maintenance gradually increases over time. The peak points in the graph clearly reflect the periods of concentrated maintenance activity, and the time intervals between adjacent peaks are relatively stable, showing a certain periodicity. In the early stages, the peak amplitudes of the curve are smaller and the intervals are longer, while in the middle and later stages, the peak amplitudes gradually increase, and the intervals between adjacent peaks shorten, indicating that maintenance events occur more frequently and their intensity continuously increases. This curve has a clear temporal correspondence with the performance degradation rate curve; the peak of maintenance intensity usually appears after the peak of performance degradation. This lag relationship provides intuitive temporal evidence for subsequent causal response analysis.

[0143] After identifying the performance degradation pattern and the maintenance demand change pattern, the correlation strength between the performance degradation rate curve and the maintenance intensity evolution curve is calculated within the corresponding time interval. The correlation strength is calculated using cross-correlation analysis: a cross-correlation function is calculated for the numerical sequences of the two curves within the same time interval. The peak position of the cross-correlation function reflects the time lag between the two curves, and the peak amplitude reflects the degree of correlation. When the peak of the maintenance intensity curve appears after the peak of the performance degradation rate curve at a certain time lag, and the correlation coefficient between the two is high, a causal relationship of "performance degradation driving maintenance demand response" can be determined. Conversely, if the maintenance intensity peak appears before the performance degradation rate peak, there is a preventative maintenance effect on performance degradation. Through statistical analysis of the correlation strength within different time intervals, the dominant causal response pattern between performance degradation and maintenance demand is identified, and the degree of correlation in each time interval is quantified using a causal response strength coefficient.

[0144] Determine the performance degradation mode vector based on the causal response pattern. Vector of Changes in Maintenance Requirements The fusion weights vary. When the causal response strength is high, it indicates strong information complementarity between the two modes, and in this case, the weight of the cross-terms in the fusion should be appropriately increased. When the causal response strength is low, the two modes are relatively independent, and the fusion weights tend to reflect the proportions of their independent contributions. Fusion weight coefficients and The corresponding contribution ratios of the performance degradation mode vector and the maintenance requirement change mode vector to the comprehensive degradation feature vector are respectively satisfied. Its specific value is adaptively adjusted within a preset range based on the magnitude of the causal response intensity coefficient. (Comprehensive degradation feature vector) Through the and The result is obtained by weighted linear fusion, i.e. The resulting comprehensive degradation feature vector It carries both degradation trajectory information at the equipment performance level and demand evolution information at the maintenance behavior level, as well as the causal coupling relationship between the two, which can more comprehensively depict the overall degradation state of port railway equipment at the current moment and provide high-quality feature input for subsequent life cycle stage determination.

[0145] Based on the stage identifier, management strategies corresponding to the lifecycle stage are extracted from a preset lifecycle management strategy library. These strategies are then combined with the dynamically updated archive to generate management instructions for the port railway equipment. These instructions are then issued to the equipment management terminal to execute corresponding intervention operations and collect execution results. The dynamically updated archive is updated based on the execution results, including:

[0146] Based on the stage identifier, the corresponding management strategy is retrieved from the lifecycle management strategy library, and the comprehensive degradation feature vector and historical intervention records are extracted from the dynamically updated archive.

[0147] Based on the historical intervention records, the execution effect and time decay characteristics of historical intervention actions are analyzed. Based on the execution effect and time decay characteristics, the intervention actions in the management strategy are prioritized to obtain the priority ranking result.

[0148] Target intervention actions are selected based on the priority ranking, and management instructions are generated by combining the device status parameters and resource constraints in the dynamically updated archive.

[0149] The management command is sent to the device management terminal, which then performs the corresponding intervention operation and collects the execution results.

[0150] The post-intervention equipment response data in the execution results are compared and analyzed with the comprehensive degradation feature vector to calculate quantitative indicators of the impact of the intervention operation on the performance degradation mode and the maintenance requirement change mode.

[0151] An intervention effect evaluation record is generated based on the aforementioned quantitative indicators, and the dynamic update archive is updated based on the execution results and the intervention effect evaluation record.

[0152] When performing a search in the lifecycle management strategy library based on the stage identifier, the stage identifier is used as the index key to traverse the pre-configured strategy entries for each stage in the strategy library and match the corresponding management strategy set. The management strategy set contains multiple types of intervention action templates, such as periodic inspection instructions, lubrication replenishment instructions, component replacement instructions, load reduction operation instructions, and scrapping disposal instructions. The types of intervention actions and triggering conditions corresponding to different lifecycle stages are significantly different. At the same time, a comprehensive degradation feature vector and historical intervention records are extracted from the dynamically updated archives. The comprehensive degradation feature vector reflects the current overall degradation status of the equipment, while the historical intervention records contain the action type, execution time, executing entity, snapshots of state parameters before and after execution, and operator feedback information for each past intervention operation. These records form the data basis for subsequent priority ranking.

[0153] The analysis of historical intervention records is divided into two dimensions: execution effect analysis and time-effect decay characteristic analysis. Execution effect analysis quantifies the degree to which various intervention actions inhibit performance degradation trends by comparing the changes in equipment status parameters before and after each intervention. Specifically, it is reflected in the magnitude and duration of the recovery of key equipment performance indicators after the intervention. Time-effect decay characteristic analysis focuses on the decay pattern of intervention effects over time, that is, the rate at which the performance improvement benefits brought by a certain intervention gradually weaken within subsequent time windows. For time-effect decay characteristics, a time-effect decay coefficient is introduced. Describes the rate of decay of the intervention's effectiveness per unit of time. A higher score indicates a lower sustainability of the intervention's effectiveness. Based on the execution effectiveness score and the time-lapse coefficient, a comprehensive priority score is calculated for each intervention action in the management strategy. The calculation method is as follows: ,in The average score of historical performance. This refers to the time interval since the last similar intervention. This item is used to correct for the impact of time decay on the urgency of current interventions. (By...) All candidate intervention actions are ranked from highest to lowest priority to obtain the priority ranking results.

[0154] Based on the priority ranking, the highest-ranked intervention action is selected as the target intervention action. During the management instruction generation phase, the target intervention action needs to be parameterized and instantiated by combining the equipment status parameters and resource constraints in the dynamically updated archive. Equipment status parameters include current load rate, operating temperature, vibration intensity, and cumulative runtime, while resource constraints include the number of available maintenance personnel, spare parts inventory status, and the current equipment work plan window. If resource constraints are not met, the priority ranking can be shifted to the next best intervention action to ensure the management instruction is operable in actual execution. The content structure of the management instruction includes the instruction number, the unique identifier of the target equipment, the intervention action type, execution parameters, execution time limit, priority level, and a description of the expected target state. After forming a structured instruction message, it is sent to the equipment management terminal.

[0155] After management instructions are sent to the equipment management terminal via the communication interface, the terminal dispatches the corresponding operators or automated execution units to complete the intervention operation based on the instructions. During the intervention operation, the terminal continuously collects real-time status data of the equipment. After the operation is completed, the terminal sends back the execution result data, which includes the actual completion time of the intervention operation, operator confirmation information, the collected values ​​of various status parameters after the intervention, and on-site anomaly records. If the intervention operation fails to be executed on time due to temporary resource unavailability or equipment failure upgrades, the terminal also needs to send back the reason for non-execution and the current equipment status so that the scheduling process can be retried later.

[0156] When comparing and analyzing the post-intervention equipment response data and the comprehensive degradation feature vector in the execution results, the pre-intervention comprehensive degradation feature vector is used as the benchmark. The equipment status data collected after the intervention is remapped to the feature space, and quantitative indicators of the impact of the intervention operation on the performance degradation mode and the maintenance requirement change mode are calculated. Quantitative indicators of the impact on performance degradation mode. Defined as the displacement magnitude of the performance degradation mode vector in the feature space before and after intervention. A positive value indicates that the intervention effectively suppressed the performance degradation trend. Values ​​close to zero or negative indicate that the intervention effect is not significant or even shows a reverse shift. Quantitative indicators of the impact of changing maintenance demand patterns. Defined as the degree of difference in the pattern vector of maintenance demand changes before and after intervention. A larger value indicates that the intervention has significantly altered the distribution of subsequent maintenance needs, for example, transforming previously high-frequency emergency repair needs into low-frequency planned maintenance needs. (Comprehensive impact quantification indicators) Through weighted fusion and get, ,in The weight is affected by performance degradation. To maintain the weight of demand impact, the sum of the two is 1, and the specific value is configured according to the emphasis on performance stability and maintenance cost at different life cycle stages.

[0157] Based on the quantitative indicators of impact , , The intervention effect evaluation record is generated by providing complete information on the implementation results. This record includes the type of intervention, implementation time, quantitative impact indicators, effect level rating (divided into four levels: significantly effective, generally effective, insufficient effect, and negative effect), and a recommended follow-up period. The effect level rating is based on... Based on the threshold range, when Exceeding the preset high efficiency threshold The time evaluation was deemed significantly effective. Located at the intermediate threshold and The evaluation is generally valid when the time interval is between. Below However, a value greater than zero is considered insufficient in effectiveness. When the value is negative, it is rated as a negative effect and an alert is triggered.

[0158] When updating the dynamic update file based on the execution results and intervention effect evaluation records, the post-intervention equipment status data, management instruction details, execution result data, and intervention effect evaluation records are all written to the corresponding storage location of the dynamic update file, and the historical intervention record entries in the file are refreshed. If the current intervention effect evaluation level is insufficient or negative, the strategy feedback mechanism is triggered, and the execution effect score of this intervention action is calculated. By reducing the weight of intervention actions in the historical records, subsequent priority ranking can adaptively reduce the ranking weight of such intervention actions, thereby prioritizing intervention methods with better historical effects in the generation of subsequent management instructions, forming a closed-loop self-optimizing management process.

[0159] A second aspect of this invention provides a port railway equipment lifecycle management system based on a unique item and code, comprising:

[0160] The filing unit is used to generate a unique identification code for port railway equipment and establish an association mapping between the unique identification code and the static attribute information of the port railway equipment to generate the initial equipment file.

[0161] The file update unit is used to obtain the real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase through the unique identifier code, and write the real-time status data and the maintenance record data into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file;

[0162] The stage determination unit is used to determine the current life cycle stage of the port railway equipment based on the temporal evolution characteristics in the dynamically updated archive, and generate a stage identifier.

[0163] The strategy execution unit is used to extract the management strategy corresponding to the life cycle stage from the preset life cycle management strategy library according to the stage identifier, generate management instructions for the port railway equipment in combination with the dynamic update file, issue the management instructions to the equipment management terminal to perform corresponding intervention operations and collect the execution results, and update the dynamic update file based on the execution results.

[0164] A third aspect of the present invention provides an electronic device, comprising:

[0165] processor;

[0166] Memory used to store processor-executable instructions;

[0167] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0168] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0169] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for full lifecycle management of port railway equipment based on one item, one code, characterized in that: include: A unique identifier code is generated for the port railway equipment, and the unique identifier code is associated with the static attribute information of the port railway equipment to generate the initial equipment file; The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identification code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file. Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated. Based on the stage identifier, the management strategy corresponding to the life cycle stage is extracted from the preset life cycle management strategy library. The management instructions for the port railway equipment are generated by combining the dynamic update file. The management instructions are issued to the equipment management terminal to perform the corresponding intervention operation and collect the execution results. The dynamic update file is updated based on the execution results.

2. The method according to claim 1, characterized in that, A unique identifier code is generated for each port railway equipment, and an association mapping is established between the unique identifier code and the static attribute information of the port railway equipment to generate an initial equipment file, including: Obtain the equipment type characteristics and individual equipment characteristics of the port railway equipment, determine the type identifier bit segment and attribute mapping bit segment based on the equipment type characteristics, and generate an encoding rule template; Based on the individual characteristics of the device, individual distinguishing information is filled into the attribute mapping bit field of the encoding rule template to generate the unique identifier code; Parse the structured and unstructured attribute fields in the static attribute information; Establish an index mapping relationship between the unique identifier code and the structured attribute field, and after converting the unstructured attribute field into a standardized attribute descriptor, establish an association mapping relationship with the unique identifier code to obtain an identifier attribute mapping table; The unique identifier code, the identifier attribute mapping table, and the static attribute information are written into a preset file storage structure to generate the device initial file.

3. The method according to claim 1, characterized in that, The real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase are obtained through the unique identifier code, and the real-time status data and maintenance record data are written into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file, including: Based on the unique identifier code, the storage location of the initial equipment file is retrieved in the equipment file storage system, and the data collection range and collection frequency constraints corresponding to the port railway equipment are determined according to the static attribute information in the initial equipment file. Based on the data collection range and the collection frequency constraints, real-time status data of the port railway equipment is obtained from the monitoring terminal, and maintenance record data associated with the unique identifier code is extracted from the maintenance management system. The real-time status data and the maintenance record data are timestamped to obtain timestamped data; Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined. The timestamp data is written to the storage location corresponding to the initial file of the device according to the writing order to obtain the dynamically updated file.

4. The method according to claim 3, characterized in that, Based on the temporal continuity between the timestamp information in the timestamp data and the data already stored in the device initial file, the writing order of the timestamp data in the storage location corresponding to the device initial file is determined, including: Read the timestamp sequence of the stored data from the storage location corresponding to the initial file of the device, and extract the timestamp information from the timestamp data to obtain the timestamp to be written; Based on the timestamp sequence, a time continuity feature vector containing time interval distribution patterns and data density distribution patterns is constructed, and abnormal time interval segments in the timestamp sequence are identified according to the time interval distribution patterns. Associate and match the timestamp to be written with the abnormal time interval segment to determine whether the timestamp to be written belongs to the time filling range of the abnormal time interval segment; When it is determined that the timestamp to be written belongs to the time filling range, the relative position weight of the timestamp to be written in the abnormal time interval segment is calculated according to the data density distribution pattern, and the insertion position is determined and an insertion position identifier is generated based on the relative position weight in the abnormal time interval segment. When it is determined that the timestamp to be written does not belong to the time filling range, the append position is determined according to the time increment direction of the timestamp sequence and an append position identifier is generated. The writing order of the timestamp data in the storage location corresponding to the device's initial file is determined based on the insertion position identifier or the append position identifier.

5. The method according to claim 1, characterized in that, Based on the temporal evolution characteristics in the dynamically updated archive, the current life cycle stage of the port railway equipment is determined, and a stage identifier is generated, including: Based on the dynamically updated archive, the time series of real-time status data and maintenance record data are determined to obtain a multidimensional time series dataset. The degradation trend of the multidimensional time series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector. The comprehensive degradation feature vector is matched with a preset life cycle stage feature library to obtain a similarity matching result. The life cycle stage feature library includes features of the commissioning stage, stable operation stage, performance degradation stage, and near-expiration stage. The confidence level of the port railway equipment at each stage of its life cycle is calculated based on the similarity matching results. The life cycle stage with the highest attribution confidence is selected as the current life cycle stage of the port railway equipment, and a corresponding stage identifier is generated based on the current life cycle stage.

6. The method according to claim 5, characterized in that, The degradation trend of the multidimensional time-series dataset is extracted to identify the performance degradation mode and maintenance demand change mode of the port railway equipment. The performance degradation mode and the maintenance demand change mode are then fused to generate a comprehensive degradation feature vector, including: The performance index time series data in the multidimensional time series dataset are segmented by a sliding window, and the rate of change and fluctuation amplitude of the performance index are calculated within each sliding window; A performance degradation rate curve is constructed based on the rate of change and the fluctuation amplitude. The performance degradation mode of the port railway equipment is determined based on the slope change points and inflection point distribution of the performance degradation rate curve. Event density analysis was performed on the maintenance event time series data in the multidimensional time series dataset to statistically analyze the frequency of maintenance events and the distribution of maintenance types within a unit time window; Based on the frequency of occurrence and the distribution of maintenance types, a maintenance intensity evolution curve is constructed. Based on the peak change trend and periodic characteristics of the maintenance intensity evolution curve, the maintenance demand change pattern of the port railway equipment is determined. By calculating the correlation strength between the performance degradation rate curve and the maintenance intensity evolution curve within the corresponding time interval, the causal response pattern between performance degradation and maintenance requirements can be identified. The performance degradation pattern and the maintenance requirement change pattern are weighted and fused according to the causal response pattern to generate the comprehensive degradation feature vector.

7. The method according to claim 1, characterized in that, Based on the stage identifier, management strategies corresponding to the lifecycle stage are extracted from a preset lifecycle management strategy library. These strategies are then combined with the dynamically updated archive to generate management instructions for the port railway equipment. These instructions are then issued to the equipment management terminal to execute corresponding intervention operations and collect execution results. The dynamically updated archive is updated based on the execution results, including: Based on the stage identifier, the corresponding management strategy is retrieved from the lifecycle management strategy library, and the comprehensive degradation feature vector and historical intervention records are extracted from the dynamically updated archive. Based on the historical intervention records, the execution effect and time decay characteristics of historical intervention actions are analyzed. Based on the execution effect and time decay characteristics, the intervention actions in the management strategy are prioritized to obtain the priority ranking result. Target intervention actions are selected based on the priority ranking, and management instructions are generated by combining the device status parameters and resource constraints in the dynamically updated archive. The management command is sent to the device management terminal, which then performs the corresponding intervention operation and collects the execution results. The post-intervention equipment response data in the execution results are compared and analyzed with the comprehensive degradation feature vector to calculate quantitative indicators of the impact of the intervention operation on the performance degradation mode and the maintenance requirement change mode. An intervention effect evaluation record is generated based on the aforementioned quantitative indicators, and the dynamic update archive is updated based on the execution results and the intervention effect evaluation record.

8. A port railway equipment lifecycle management system based on one item, one code, used to implement the method as described in any one of claims 1-7, characterized in that, include: The filing unit is used to generate a unique identification code for port railway equipment and establish an association mapping between the unique identification code and the static attribute information of the port railway equipment to generate the initial equipment file. The file update unit is used to obtain the real-time status data and maintenance record data of the corresponding port railway equipment during the operation phase through the unique identifier code, and write the real-time status data and the maintenance record data into the storage location corresponding to the initial file of the equipment to obtain a dynamically updated file; The stage determination unit is used to determine the current life cycle stage of the port railway equipment based on the temporal evolution characteristics in the dynamically updated archive, and generate a stage identifier. The strategy execution unit is used to extract the management strategy corresponding to the life cycle stage from the preset life cycle management strategy library according to the stage identifier, generate management instructions for the port railway equipment in combination with the dynamic update file, issue the management instructions to the equipment management terminal to perform corresponding intervention operations and collect the execution results, and update the dynamic update file based on the execution results.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images