Elevator health closed-loop evaluation system and method based on life and state of spare part

Abstract

The invention discloses an elevator health closed-loop evaluation system and method based on the service life and state of a spare part, and relates to the technical field of elevator operation data analysis and health evaluation. The elevator health closed-loop evaluation system and method based on the service life and the state of the spare part comprises the following steps that S1, an original data set of the whole operation process is constructed; s2, analyzing a cross-component offset relation between a portal crane action stage and a traction machine load stage on the basis of the original data set in the whole operation process; s3, constructing a weak coupling structure between the vibration behavior of the traction machine and the action stage of the brake by surrounding the cross-component offset mark and the multi-component segment features; s4, determining a key node and a propagation trend of a cross-component degradation link in combination with a link structure and landing behaviors; and S5, generating a maintenance strategy and work order information according to the degradation level and the monitoring adjustment action. The problem that secondary features in a degraded link are ignored for a long time due to the fact that a correlation structure is not constructed for weak coupling features of components is solved.

Description

Technical Field

[0001] This invention relates to the field of elevator operation data analysis and health assessment technology, specifically to an elevator health closed-loop assessment system and method based on spare parts lifespan and condition. Background Technology

[0002] With the widespread deployment of elevators in buildings, the frequency of equipment operation has increased significantly, and the opening and closing, starting and stopping, and various other action events generated during operation are gradually forming recordable operational data. The industry typically acquires event information and auxiliary status variables through basic sensing devices and operation log interfaces, and conducts maintenance management in conjunction with regular inspection mechanisms. To enhance the observability of the operation process, some systems have begun to build trend sequences or event logs based on historical operation records to assist in the analysis of changes in equipment operation.

[0003] For example, the invention patent with announcement number CN120632427B discloses a method for state perception and full life cycle health assessment of distribution network equipment. The method includes: acquiring relevant quantity data of transformers; constructing thermochemical degradation feature vectors and wet electrical degradation feature vectors, and normalizing each component; constructing expressions for the internal affine surface and the external linear surface of the transformer respectively, calculating the directed normalized distance from the normalized vector of thermochemical degradation features to the internal affine surface of the transformer, and simultaneously calculating the directed normalized distance from the normalized vector of wet electrical degradation features to the external linear surface of the transformer, and determining whether the current state of the transformer is on the risk side, calculating the risk vector and outputting it; introducing a non-isotropic Mahalanobis tensor metric to calculate the coupling strength between the thermochemical degradation and wet electrical degradation of the transformer, calculating the transformer health index and determining the health status, and judging the transformer status as qualified when the transformer health index reaches a preset health threshold.

[0004] For example, invention patent CN116150661B discloses an abnormality diagnosis method and device for elevator traction machines, which also involves electronic equipment. The method includes acquiring the high-frequency raw vibration signal of the elevator traction machine; obtaining a time-domain index value based on the high-frequency raw vibration signal; obtaining a time-domain health score based on the time-domain index value; the method further includes obtaining Intrinsic Mode Function (IMF) components based on the raw vibration signal; obtaining a Hilbert envelope spectrum signal based on the IMF components; calculating a frequency-domain health score by analyzing the Hilbert envelope spectrum signal; and averaging the time-domain and frequency-domain health scores to obtain the overall health score of the elevator traction machine. In this invention, the real-time health status of the elevator traction machine is determined by the overall health score, allowing staff to perform maintenance according to demand, thus avoiding waste of human resources.

[0005] Existing elevator operation status analysis methods largely rely on single-event records, fixed-cycle inspections, and independent judgments based on local data fragments. They lack a unified organizational format for multi-source operation records and have not established a mechanism for correlation analysis of operation events over time. Common methods typically rely on local thresholds and empirical rules for judgment, which are ill-suited for complex operating conditions or cross-cycle changes, leading to problems such as information dispersion, delayed status identification, and difficulty in continuously tracking trends. Furthermore, existing maintenance methods mostly use fixed plans or static cycles to determine maintenance times, failing to establish a dynamic evaluation system based on the entire operation chain. This makes it difficult to obtain continuous and systematic reference support for equipment health management.

[0006] To address the above issues, there is an urgent need for an elevator health closed-loop assessment system and method based on spare parts lifespan and condition. Summary of the Invention

[0007] Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an elevator health closed-loop assessment system and method based on spare parts lifespan and status, which solves the problem that the weak coupling characteristics of components are not associated with a structure, resulting in the long-term neglect of secondary characteristics in the degradation link.

[0008] Technical solution To achieve the above objectives, this invention provides the following technical solution: a closed-loop assessment method for elevator health based on spare parts lifespan and status, comprising the following steps: S1, collecting multi-source raw information generated during elevator operation, constructing a raw dataset for the entire operation process, and performing preprocessing on the raw dataset; S2, analyzing the cross-component offset relationship between the door operator action phase and the traction machine load phase based on the raw dataset, and performing offset marking and offset meaning transmission according to the offset trend; S3, constructing a weakly coupled structure between the traction machine vibration behavior and the brake action phase around the cross-component offset marking and multi-component segment features, classifying the coupling level according to the coupling change and synchronously coupling features to the link structure; S4, determining the key nodes and propagation direction of the cross-component degradation link by combining the link structure and floor behavior, classifying the degradation level according to the degradation progress, and generating monitoring and adjustment actions and task plans; S5, generating maintenance strategies and work order information based on the degradation level and monitoring and adjustment actions, and synchronizing the operating status after maintenance processing to the raw dataset for the entire operation process.

[0009] Furthermore, the specific steps for collecting multi-source raw information generated during elevator operation and constructing a raw dataset for the entire operation process are as follows: Collecting basic operating information generated by the elevator operation control system, door operator control board, traction machine main drive circuit current sensor, triaxial acceleration sensor, brake action detection sensor, leveling detection unit, and time management unit during elevator operation. This basic operating information includes: start and stop records of opening and closing actions, start and stop event records, brake engagement start and stop records, raw samples of current during the traction machine startup phase, raw sequences of triaxial acceleration during traction machine operation, leveling completion event records, and raw samples of vertical displacement. All information is accompanied by a timestamp. Simultaneously, calling external reference data corresponding to the elevator operating environment, including: floor location structure and shaft spatial distribution structure. The basic operating information is initially synchronized and integrated according to the time sequence of start and stop records, leveling records, and braking action records. A unified structure of action, time, vibration, and position information is established in conjunction with the external reference data to construct the raw dataset for the entire operation process.

[0010] Furthermore, the specific steps for preprocessing the entire raw dataset are as follows: A unified time axis is constructed using a dynamic time warping algorithm based on the timestamp alignment relationship between multi-source records; the sequences of running segments with different sampling frequencies are rearranged; the median absolute deviation method and the isolated forest algorithm are used to identify local and global abnormal segments and remove anomalies, combining the full-cycle behavior pattern; for time slice breaks occurring across component records, an exponentially weighted moving average algorithm and a neighborhood-based collaborative completion strategy are used to fill in missing segments; local weighted regression filtering and Kalman filtering algorithms are applied to the reconstructed continuous sequence based on running rhythm fluctuations to achieve trend fitting and noise suppression; numerical domain transformation is performed on the reconstructed sequence under the unified time axis, and the entire raw dataset of the entire running process is standardized and normalized.

[0011] Furthermore, the specific steps for analyzing the cross-component offset relationship between the gantry crane operation phase and the traction machine load phase based on the original dataset of the entire operation, and for executing offset marking and offset meaning transmission according to the offset trend, are as follows: Combining the start and end records of the opening and closing actions, read the start and end times of the current opening and closing action and perform time period conversion to form the gantry crane operation time; trace back to the previous opening and closing cycle according to the time sequence, and obtain the gantry crane operation time of the previous cycle from the gantry crane operation time record of the previous cycle; retrieve the original sample of the traction machine start-up current around the time slice corresponding to the traction machine start-up phase, and select the current peak segment at the moment of start-up to form the traction machine start-up current; locate the start-up phase of the previous start-stop cycle along the start-stop event record, and obtain the traction machine start-up current of the previous cycle from the traction machine start-up current record of the previous start-stop cycle; extract the time difference between two consecutive start-stop timestamps in the start-stop event record and complete the interval conversion to obtain the start-stop cycle time interval; extract the continuous operation cycle time based on the start and end records of the opening and closing actions. During the duration of the operation, the change in time span between adjacent cycles is calculated and compared with the time span threshold. The set of cycles where the change in time span is continuously less than the time span threshold is defined as the stable interval during normal operation. The lower bound of the time span is selected from the stable interval to form the phase offset protection constant. The change in gantry crane operation time is obtained by subtracting the gantry crane operation time of the previous cycle from the gantry crane operation time. The change in traction machine starting current is obtained by subtracting the traction machine starting current of the previous cycle from the traction machine starting current. The change in traction machine starting current is added to the phase offset protection constant to form the denominator structure. The change in gantry crane operation time is divided by the denominator structure to obtain the cross-component phase offset value. When the cross-component phase offset value increases or shows an abnormal ratio relationship in continuous operation cycles, the cross-component phase offset value is written to the offset buffer area, and the weak coupling modulation condition is recorded in the offset status field. When the cross-component phase offset value does not increase continuously or show an abnormal ratio relationship, the transmission to subsequent modules is stopped, and no adjustment operation is triggered.

[0012] Furthermore, the specific steps for constructing a weakly coupled structure between the traction machine vibration behavior and the brake action phase based on cross-component offset markers and multi-component segment features are as follows: extracting cross-component phase offset values ​​from the offset buffer; reading the start and end times of brake engagement in the current operating cycle by combining the brake engagement start and end records and performing time period conversion to form the brake engagement time; locating the acceleration data segment of the current cycle based on the original triaxial acceleration sequence during traction machine operation, selecting the amplitude peak segment in the segment and completing amplitude extraction to generate the traction machine vibration amplitude; reading the engagement duration segments and vibration amplitude segments of multiple adjacent cycles based on the brake engagement start and end records and the original triaxial acceleration sequence, and calculating the engagement amplitude of each segment. The amplitude of the continuous section change and the amplitude of the vibration amplitude segment change are combined, and the two types of amplitude changes are compared with the fluctuation amplitude threshold. The set of periods where the amplitude of change is less than the fluctuation amplitude threshold is defined as the historical stable period. A fixed reference value is selected from the value range corresponding to the historical stable period to form the coupling strength protection constant. The brake engagement time of each period in the historical operating window is multiplied by the vibration amplitude of the traction machine, the square is summed, and the square root of the sum is taken to form the joint energy term. The absolute value of the phase offset value across components is taken and added to the coupling strength protection constant to form the phase modulation denominator term. The reciprocal of the denominator term is taken to form the phase modulation coefficient. The joint energy term is multiplied by the phase modulation coefficient to obtain the weak coupling energy intensity value.

[0013] Furthermore, the specific steps for classifying coupling levels and synchronizing coupling features to the link structure based on coupling changes are as follows: The weak coupling energy intensity value is compared with a coupling threshold, which includes a first-level coupling threshold and a second-level coupling threshold; when the weak coupling energy intensity value is lower than the first-level coupling threshold, the weak coupling energy intensity value is written to the coupling buffer, and the link is maintained at the basic level; when the weak coupling energy intensity value is greater than or equal to the first-level coupling threshold but lower than the second-level coupling threshold, a concern marker is written to the link status table, and the weak coupling energy intensity value and corresponding operating cycle are recorded in the link trend structure; when the weak coupling energy intensity value is greater than or equal to the second-level coupling threshold, a link entry is registered in the key coupling structure, and the sensitivity parameter is updated in the link parameter table. Simultaneously, the weak coupling energy intensity value is written to the degradation evaluation path without triggering any execution layer operations.

[0014] Further, the specific steps for determining the key nodes and propagation path of cross-component degraded links by combining link structure and layer station behavior are as follows: Integrate the coupling structure into continuous degraded links in chronological order; locate the completion time of the current leveling action by combining the leveling completion event record, and select the displacement deviation of the corresponding time slice in the original vertical displacement sample to form the layer station leveling offset; retrieve the previous leveling completion time based on the leveling completion event record, and read the displacement deviation of the corresponding time in the original vertical displacement sample to form the previous cycle layer station leveling offset; extract two consecutive event timestamps associated with leveling based on the start / stop event record and perform interval conversion to generate the leveling cycle time interval; extract the weak coupling energy intensity value from the coupling buffer; subtract the previous cycle layer station leveling offset from the layer station leveling offset and divide by the leveling cycle time interval to obtain the leveling offset change rate; multiply the leveling offset change rate by the weak coupling energy intensity value to obtain the link degradation index value.

[0015] Furthermore, the specific steps for classifying degradation levels and generating monitoring and adjustment actions and task plans based on the degradation progress are as follows: The link degradation index value is compared with the degradation threshold, which includes a primary degradation threshold and a secondary degradation threshold; when the link degradation index value is less than the primary degradation threshold, the monitoring parameters are maintained, and no additional processing is performed; when the link degradation index value is greater than or equal to the primary degradation threshold and less than the secondary degradation threshold, the link is written into the early warning structure, the sampling frequency corresponding to the link is increased, an inspection task suggestion is generated, and the link degradation index value is written into the link record structure; when the link degradation index value is greater than or equal to the secondary degradation threshold, a maintenance work order is generated, a key monitoring strategy is initiated, the link random reset process is frozen, the link is written into the maintenance priority structure, and the link degradation index value is written into the closed-loop feedback path as input for the next operating cycle.

[0016] Furthermore, the specific steps for generating maintenance strategies and work order information based on degradation levels and monitoring adjustment actions, and synchronizing the operational status after maintenance to the original dataset of the entire operation process, are as follows: A maintenance strategy corresponding to the component is generated based on the causal node and propagation direction in the degradation link; the maintenance strategy is converted into maintenance plan parameters and work order information is generated, which is then pushed to the execution end through a task scheduling process; the monitoring frequency, key component acquisition granularity, and abnormal triggering conditions are adjusted according to changes in link status, and the adjusted parameters are written into the monitoring and control structure; after maintenance is completed, operational status data is collected and written into the original dataset of the entire operation process.

[0017] The second aspect of this invention provides an elevator health closed-loop assessment system based on spare parts lifespan and status, comprising: a component-wide data acquisition module, a cross-phase offset analysis module, a coupling feature recognition module, a link health assessment module, and a closed-loop strategy feedback module. The component-wide data acquisition module is used to collect multi-source raw information generated during elevator operation, construct a raw dataset for the entire operation process, and perform preprocessing on the raw dataset. The cross-phase offset analysis module is used to analyze the cross-component offset relationship between the door operator's action phase and the traction machine's load phase based on the raw dataset for the entire operation process, and to perform offset marking and offset meaning transmission according to the offset trend. The coupling feature identification module is used to construct a weakly coupled structure between the traction machine vibration behavior and the brake action stage based on cross-component offset markers and multi-component segment features. It classifies the coupling level according to the coupling changes and synchronously couples the features to the link structure. The link health assessment module is used to determine the key nodes and propagation direction of the cross-component degradation link by combining the link structure and the layer station behavior. It classifies the degradation level according to the degradation progress and generates monitoring and adjustment actions and task plans. The closed-loop strategy feedback module is used to generate maintenance strategies and work order information based on the degradation level and monitoring and adjustment actions, and synchronize the operation status after maintenance to the original dataset of the entire operation process.

[0018] Beneficial effects The present invention has the following beneficial effects: (1) This invention performs time alignment, anomaly removal, missing information completion and trend filtering on multi-source original information, and forms a unified sequence structure of gantry crane action time, traction machine start current, brake engagement time, vibration amplitude and floor leveling offset in the original dataset throughout the entire process, so that subsequent offset marking, coupling identification and degradation calculation can be carried out on the basis of complete, continuous and comparable time sequence.

[0019] (2) The present invention constructs cross-component phase offset values ​​based on the change in the gantry crane's operating time and the change in the traction machine's starting current, so that the micro-amplitude time offset and load fluctuation are expressed in a unified ratio structure, providing clear offset input conditions for the weakly coupled structure, and enabling the offset trend in subsequent cycles to be stably transmitted to the coupling analysis path.

[0020] (3) This invention generates weakly coupled energy intensity values ​​based on cross-component phase offset values, brake engagement time and traction machine vibration amplitude, so that mild disturbances between multiple components are expressed in the form of joint energy, and the level is classified by coupling threshold, so that the link structure has continuously updated coupling feature input, providing a continuous correlation basis for the construction of degraded links.

[0021] (4) The present invention constructs a link degradation index value based on the level station level offset change rate and weak coupling energy intensity value, and generates monitoring adjustment actions and maintenance strategies according to the degradation threshold, so that link-level changes can enter the closed-loop feedback loop in an exponential structure, and the maintenance plan, sampling frequency and monitoring conditions can be dynamically updated according to the degradation level.

[0022] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0023] Figure 1 This is a flowchart of the elevator health closed-loop assessment method based on spare parts life and condition according to the present invention. Figure 2 This is a structural diagram of the elevator health closed-loop assessment system based on spare parts life and condition according to the present invention. Figure 3 This is a trend chart of the link degradation index value of the present invention; Figure 4 This is a joint density diagram of coupling strength and offset rate of the present invention. Detailed Implementation

[0024] 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.

[0025] Please see Figures 1-4 This invention provides a technical solution: an elevator health closed-loop assessment method based on spare parts lifespan and status, comprising the following steps: S1, collecting multi-source raw information generated during elevator operation, constructing a raw dataset of the entire operation process, and performing preprocessing on the raw dataset; S2, analyzing the cross-component offset relationship between the door operator action stage and the traction machine load stage based on the raw dataset of the entire operation process, and performing offset marking and offset meaning transmission according to the offset trend; S3, constructing a weakly coupled structure between the traction machine vibration behavior and the brake action stage around the cross-component offset marking and multi-component segment features, classifying the coupling level according to the coupling change and synchronously coupling features to the link structure; S4, determining the key nodes and propagation direction of the cross-component degradation link by combining the link structure and floor behavior, classifying the degradation level according to the degradation progress and generating monitoring and adjustment actions and task plans; S5, generating maintenance strategies and work order information based on the degradation level and monitoring and adjustment actions, and synchronizing the operation status after maintenance processing to the raw dataset of the entire operation process.

[0026] Specifically, the steps for collecting multi-source raw information generated during elevator operation and constructing a raw dataset for the entire operation process are as follows: Collecting basic operational information generated by the elevator operation control system, door operator control board, traction machine main drive circuit current sensor, triaxial acceleration sensor, brake action detection sensor, leveling detection unit, and time management unit during elevator operation. This basic operational information includes start and stop records of opening and closing actions, start and stop event records, brake engagement start and stop records, raw current samples during the traction machine startup phase, raw triaxial acceleration sequences during traction machine operation, leveling completion event records, and raw vertical displacement samples. All records are accompanied by timestamps that can be used for time-series comparison and event localization, enabling a traceable relationship between the action phase, vibration phase, and leveling phase. Calling external reference data corresponding to the elevator operating environment, the external reference data package... The system includes the site location structure and shaft spatial distribution structure, which are used to provide a spatial correlation benchmark for action and location information during subsequent integration, enabling action events to be clearly mapped in the physical structure. The basic operation information is initially synchronized and integrated according to the time sequence of start-stop records, leveling records, and braking action records. By comparing various timestamps, an alignable time sequence chain is constructed, so that records with different sampling frequencies, different triggering mechanisms, and different event sources form continuous operation segments on a unified time axis. Combined with external reference data, a unified structure of action information, time information, vibration information, and location information is established, so that the duration of gantry crane action, traction machine start-up load, brake engagement section, acceleration peak section, and site leveling behavior can be organized and correlated according to the actual operation sequence, ultimately forming a complete original dataset covering each operation cycle and having complete event correlation relationships.

[0027] In this implementation plan, the original dataset of the entire operation has a continuous, complete, and alignable temporal foundation. Action information, time information, vibration information, and position information form a stable correlation in a unified structure. Key behaviors during elevator operation can be presented in the form of continuous segments, and the relationship between cross-component events is clearly expressed within a unified time axis, providing a directly referable operation sequence for offset marking, coupling identification, and degradation link analysis.

[0028] Specifically, the preprocessing steps for the raw dataset throughout the entire operation are as follows: A unified time axis is constructed using a dynamic time warping algorithm based on the timestamp alignment relationship between multi-source records. The operation segments with different sampling frequencies are rearranged to ensure that high-frequency vibration segments and low-frequency motion segments can form a continuous reference relationship on a unified timeline. Combining the full-cycle behavior pattern, the median absolute deviation method and the isolated forest algorithm are used to identify local and global anomalous segments and remove anomalies. By distinguishing between short-term mutations and cross-cycle offsets, normal fluctuations are avoided from being mistakenly included in subsequent analysis. For time slice breaks occurring in cross-component records, an exponentially weighted moving average algorithm and neighborhood-based collaborative completion are used. The strategy fills in missing segments, allowing the breaks caused by trigger intervals, transmission delays, and record omissions to reconstruct continuous segments in the time dimension. Local weighted regression filtering and Kalman filtering algorithms are applied to the reconstructed continuous sequence around the fluctuations in the running rhythm to achieve trend fitting and noise suppression, smoothing out high-frequency disturbances and measurement noise in the sequence. Numerical domain transformation is performed on the reconstructed sequence under a unified time axis. By scaling and normalizing different physical quantities, the dimensional differences of multiple components remain comparable in subsequent offset relationship construction and coupling analysis. Furthermore, the entire original dataset is standardized and normalized, enabling all sequences to serve as a unified input structure for subsequent modules.

[0029] In this implementation scheme, the original dataset throughout the entire process forms a continuous, smooth, and uninterrupted unified sequence in both the time and numerical dimensions. Local noise, cross-period mutations, and recording faults are eliminated and completed during the preprocessing process. Action behavior, vibration changes, and running rhythm remain stable within a unified time axis, providing a consistent, reliable, and directly callable temporal foundation for offset relationship construction, weak coupling identification, and degradation link calculation.

[0030] Specifically, based on the original dataset of the entire operation, the cross-component offset relationship between the gantry crane operation phase and the traction machine load phase is analyzed. The specific steps for offset marking and meaning transmission based on the offset trend are as follows: Combining the start and end records of the opening and closing actions, the start and end times of the current opening and closing action are read and time period conversion is performed to form the gantry crane operation time. Precise positioning of the continuous action segment ensures that the gantry crane operation behavior can be quantified on the time axis. Following the time sequence, the previous opening and closing cycle is traced back, and the gantry crane operation time of the previous cycle is obtained from the gantry crane operation time record of the previous cycle, so that the difference in action rhythm between the current cycle and the previous cycle can be presented in a directly comparable manner. The process revolves around the traction machine start-up phase. The process involves retrieving raw samples of the traction machine's startup current during time-slice retrieval, selecting the current peak segment at startup to form the traction machine's startup current, and ensuring accurate capture of load changes at startup through peak segment extraction. The startup phase of the previous start-stop cycle is located along the start-stop event records, and the startup current of the previous cycle is obtained from its startup current record, enabling a comparable difference in load changes between the two cycles. The time difference between two consecutive start-stop timestamps in the start-stop event records is calculated and interval conversion is performed to obtain the start-stop cycle time interval, used to describe the operating rhythm between the traction machine's load behavior and the gantry crane's actions. Finally, the duration of actions in continuous operating cycles is extracted by combining the start and end records of opening and closing actions, and adjacent cycles are calculated. The time span variation is compared with the time span threshold. By identifying the rhythm stability between cycles, a stable interval is established during normal operation. The lower bound of the time span is selected from the stable interval to form a phase offset protection constant, which is used to avoid calculation anomalies caused by the denominator approaching zero in subsequent ratio calculations. The gantry crane operation time variation is obtained by subtracting the gantry crane operation time of the previous cycle from the gantry crane operation time. The traction machine starting current variation is obtained by subtracting the traction machine starting current of the previous cycle from the traction machine starting current. The traction machine starting current variation is added to the phase offset protection constant to form the denominator structure. The cross-component phase offset value is obtained by dividing the gantry crane operation time variation by the denominator structure, so that the gantry crane operation rhythm variation and traction machine load variation can be quantified. The ratio relationship is presented; when the cross-component phase offset value increases or shows an abnormal ratio relationship in continuous operation cycles, the cross-component phase offset value is written into the offset buffer area, and the weak coupling modulation condition is recorded in the offset status field. The continuous increase is manifested as the ratio showing a continuous upward trend across multiple operation cycles, and the abnormal ratio relationship is manifested as the ratio being significantly higher than the reference level generated in the stable interval or a jump change between adjacent cycles, reflecting an abnormal synchronization relationship between the gantry crane's action rhythm offset and the traction machine's load fluctuation; when the cross-component phase offset value does not show a continuous increase or an abnormal ratio relationship, the transmission to subsequent modules is stopped, and the adjustment operation is not triggered, so that the offset relationship only enters the coupling analysis structure when it has real meaning.

[0031] The specific calculation method for the phase offset value across components is as follows: In the formula, Indicates the phase offset value across components. Indicates the time of the gantry crane's action. This indicates the gantry crane's operation time in the previous cycle. This indicates the starting current of the traction machine. This indicates the starting current of the traction machine in the previous cycle. Indicates the start-stop cycle time interval. This represents the phase offset protection constant.

[0032] In this implementation scheme, the changes in the gantry crane's operating rhythm and the changes in the traction machine's load form a quantifiable ratio structure in the time dimension. The synchronous offset between components is stably identified through continuous trends and abnormal ratio relationships. The meaning of the offset is clearly recorded in the offset buffer, providing directional offset input for the weakly coupled structure, so that subsequent coupling analysis and link construction have a traceable and interpretable offset basis.

[0033] Specifically, the steps for constructing a weakly coupled structure between the traction machine's vibration behavior and the brake's action phase, based on cross-component offset markers and multi-component segment features, are as follows: Extract cross-component phase offset values ​​from the offset buffer to enable the offset information to serve as modulation conditions for coupling construction; combine the brake engagement start and end records to read the engagement start and end times of the current operating cycle and perform time period conversion to form the brake engagement time, constructing a quantifiable engagement duration segment by clearly defining the action start and end points; locate the acceleration data segment of the current cycle based on the original triaxial acceleration sequence during traction machine operation, calculate the amplitude sequence of the instantaneous acceleration in the three axes using a vector synthesis method, so that vibration components in different directions are expressed in the same amplitude structure, select the amplitude peak segment in the segment and complete amplitude extraction to generate the traction machine vibration amplitude, so that the vibration intensity can reflect the operating disturbance situation with the peak value; read the engagement duration segments and vibration amplitude segments of multiple adjacent cycles based on the brake engagement start and end records and the original triaxial acceleration sequence, and identify the action duration and vibration amplitude over time by calculating the change amplitude between cycles. The stability of the interval dimension is compared with the fluctuation amplitude threshold. The set of cycles with all changes less than the fluctuation amplitude threshold is defined as the historical stable cycle, so that the truly stable cycle can be used as a reference benchmark. A fixed reference value is selected from the value range corresponding to the historical stable cycle to form a coupling strength protection constant, which is used to avoid abnormal amplification of coupling calculation under extremely small denominators. The brake engagement time of each cycle in the historical operating window is multiplied by the vibration amplitude of the traction machine, squared and summed. The joint energy term is formed by the square accumulation method, so that the synchronous disturbance of the two components in the same cycle can be expressed in the energy space in an accumulative way. The square root of the summation value is taken to keep the joint energy structure consistent with the dimension of the input quantity. The absolute value of the phase offset value across components is taken and added to the coupling strength protection constant to form a phase modulation denominator term. The larger the offset, the more significant the modulation effect. The reciprocal of the denominator term is taken to form the phase modulation coefficient. The joint energy term is multiplied by the phase modulation coefficient, so that the coupling of the action stage and the vibration energy coupling act together on the weak coupling energy intensity value, and finally form a coupling quantification result that can be used for link-level health assessment.

[0034] The specific calculation method for the weak coupling energy intensity value is as follows: In the formula, This represents the energy intensity value of weak coupling. Indicates the brake engagement time. Indicates the vibration amplitude of the traction machine. Indicates the phase offset value across components. This represents the coupling strength protection constant.

[0035] In this implementation scheme, the vibration behavior of the traction machine and the action phase of the brake form a quantifiable joint expression in the energy dimension. The change in vibration amplitude and the continuous change in pull establish a stable correlation in time. The phase shift across components forms a directional influence in the modulation structure. The weak coupling characteristics exhibit continuity, accumulation and interpretability within a multi-period window, providing a strength input that can reflect the actual operational correlation degree for coupling level classification and link degradation analysis.

[0036] Specifically, the steps for classifying coupling levels and synchronizing coupling characteristics to the link structure based on coupling changes are as follows: The weak coupling energy intensity value is compared with a coupling threshold, which includes a first-level coupling threshold and a second-level coupling threshold, used to establish clear boundaries under different coupling levels; when the weak coupling energy intensity value is lower than the first-level coupling threshold, it is written to the coupling buffer, maintaining the weak coupling behavior at a basic level in the link structure without generating additional markers; when the weak coupling energy intensity value is greater than or equal to the first-level coupling threshold but lower than the second-level coupling threshold, a concern marker is written to the link status table, ensuring the link's sustainability within the monitoring structure. Track attributes and record the weak coupling energy intensity value and corresponding operating cycle in the link trend structure, so that the coupling changes form a trajectory in the time series; when the weak coupling energy intensity value is greater than or equal to the secondary coupling threshold, register the link entry in the key coupling structure, so that the link obtains priority processing attributes in the overall operating structure, and update the sensitivity parameters in the link parameter table, so that subsequent degradation analysis can perform threshold judgment based on more sensitive parameters. At the same time, write the weak coupling energy intensity value into the degradation evaluation path so as to form a direct input in the calculation of the link degradation index value without triggering any execution layer operation, so that the coupling behavior maintains logical independence before degradation evaluation.

[0037] In this implementation scheme, the weak coupling energy intensity is expressed in a hierarchical manner within different threshold ranges. Coupling changes are accurately located in the link status table, link trend structure, and key coupling structure. Coupling behavior is clearly distinguished between the basic level, attention level, and key level. Coupling characteristics are entered into the link structure in a periodic recording manner, providing coupling inputs with continuity and sensitivity differences for the degradation assessment path, so that the link health analysis has traceable and distinguishable coupling basis.

[0038] Specifically, the key nodes and propagation paths of cross-component degradation links, determined by combining link structure and station behavior, are as follows: The coupled structure is integrated into a continuous degradation link in chronological order, ensuring that coupling changes in different operating cycles form a coherent link path on the time axis; the completion time of the current leveling action is located by combining the leveling completion event record, and the displacement deviation of the corresponding time slice is selected from the original vertical displacement sample to form the station leveling offset, allowing the station position error to be quantified at precise event nodes; the previous leveling completion time is retrieved based on the leveling completion event record, and the displacement deviation of the corresponding time is read from the original vertical displacement sample to form the station leveling offset of the previous cycle, ensuring that displacement changes in adjacent cycles are measured in the same dimension. The process involves comparing and contrasting the time stamps of two consecutive events associated with leveling based on start / stop event records, performing interval conversion to generate a leveling cycle time interval to describe the actual span of leveling station position changes in the time direction, extracting weak coupling energy intensity values ​​from the coupling buffer to make coupling behavior a transitive influence factor in the degraded link, subtracting the previous cycle leveling offset from the leveling offset and dividing by the leveling cycle time interval to form the leveling offset change rate, enabling the dynamic characterization of leveling station position change trends, and multiplying the leveling offset change rate by the weak coupling energy intensity value to obtain the link degradation index value, so that the position offset trend and coupling change together constitute a comprehensive quantitative expression of cross-component degradation, which is used as a direct entry point for subsequent degradation level classification.

[0039] The specific calculation method for the link degradation index is as follows: In the formula, This represents the link degradation index value. Indicates the leveling offset of the floor station. This indicates the leveling offset of the previous cycle's leveling station. Indicates the time interval of leveling cycle. This represents the energy intensity value of weak coupling.

[0040] Table 1 shows the link degradation index data table provided in this application embodiment. The layer leveling offset for degradation 1 is set to 1.12, the layer leveling offset of the previous cycle is set to 1.00, the leveling cycle time interval is set to 2.00, and the weak coupling energy intensity value is set to 4.60; the layer leveling offset for degradation 2 is set to 1.42, the layer leveling offset of the previous cycle is set to 1.15, the leveling cycle time interval is set to 2.00, and the weak coupling energy intensity value is set to 5.30; the layer leveling offset for degradation 3 is set to 1.50, and the layer leveling offset of the previous cycle is set to 1. .20, the leveling cycle time interval is set to 2.00, and the weak coupling energy intensity value is set to 6.00; the leveling offset of the degraded 4 layer is set to 1.75, the leveling offset of the previous layer is set to 1.10, the leveling cycle time interval is set to 2.00, and the weak coupling energy intensity value is set to 6.50; the leveling offset of the degraded 5 layer is set to 1.62, the leveling offset of the previous layer is set to 1.30, the leveling cycle time interval is set to 2.00, and the weak coupling energy intensity value is set to 5.10.

[0041] Table 1 Link Degradation Index Data Table like Figure 3 The figure shows a trend chart of the link degradation index values ​​provided in this application embodiment. According to the data in the image and table, the first-level degradation threshold is 1.00, the second-level degradation threshold is 1.50, and the link degradation index values ​​corresponding to the five links vary between 0.28 and 2.11, showing an overall trend of first increasing and then decreasing. The link degradation index value of degradation 1 is 0.28, significantly lower than the first-level degradation threshold; the link degradation index value of degradation 2 is 0.72, still below the first-level degradation threshold; the link degradation index value of degradation 3 is 0.90, close to the first-level degradation threshold, but still not reaching the threshold level; the link degradation index value of degradation 4 reaches 2.11, significantly exceeding both the first-level and second-level degradation thresholds; the link degradation index value of degradation 5 is 0.82, having fallen back below the first-level degradation threshold. This figure can be used to compare the relationship between the link degradation index values ​​of different numbers and the set thresholds, intuitively presenting the changing trend of degradation degree in each operating cycle.

[0042] like Figure 4The image shows a joint density map of coupling strength and offset rate provided in an embodiment of this application. Based on the data presented in the image, the weak coupling energy intensity value varies between 0.60 and 1.30, while the leveling offset rate of the strata is distributed between 0.10 and 0.30, forming an overall distribution trend that increases from the lower left to the upper right. Different color levels in the image correspond to different energy density levels; the darker the color level, the higher the energy density. A darker block appears at the offset rate of approximately 0.15, indicating a concentrated area of ​​local energy density. The remaining areas mainly show intermediate color levels, reflecting a relatively stable relationship between weak coupling energy and offset rate. This image can be used to visually demonstrate the corresponding distribution characteristics between offset rate and weak coupling energy, providing a visual reference for analyzing the correlation between their changes.

[0043] In this implementation plan, the trend of layer station location change and cross-component coupling behavior are expressed in a unified quantitative manner in the time dimension. Degradation signs show directional changes in continuous cycles. The link degradation index value reflects the combined effect of location offset and coupling strength in a joint manner. Degraded links have clear progressive characteristics on the time axis, providing a continuous and interpretable link-level measurement basis for degradation level classification and monitoring and adjustment actions.

[0044] Specifically, the steps for classifying degradation levels and generating monitoring and adjustment actions and task plans based on the degradation progress are as follows: The link degradation index value is compared with the degradation threshold, which consists of a primary degradation threshold and a secondary degradation threshold, used to establish clear boundary standards for different degrees of degradation; when the link degradation index value is less than the primary degradation threshold, monitoring parameters are maintained to ensure the link maintains its original observation strategy within the basic monitoring cycle; when the link degradation index value is greater than or equal to the primary degradation threshold but less than the secondary degradation threshold, the link is added to the early warning structure. The tracking capability for changes in operational status is enhanced by increasing the sampling frequency of the corresponding link, and inspection task suggestions are generated, enabling manual inspections to focus on potential degradation points, while simultaneously adjusting the link degradation... The degradation index value is written into the link record structure for subsequent trend comparison. When the link degradation index value is greater than or equal to the secondary degradation threshold, a maintenance work order is generated, enabling the maintenance process to start around the degraded link. The key monitoring strategy is activated to improve the observation accuracy of key components. At the same time, the link random reset process is frozen, and the periodic initialization action triggered within the stable range is suspended, so that the link is no longer automatically reset due to the rhythm self-recovery mechanism and the periodic initialization mechanism. This ensures that the degraded link maintains a continuous and complete trend record before the maintenance is completed. The link is written into the maintenance priority structure to form a priority processing order, and the link degradation index value is written into the closed-loop feedback path as the input for the next running cycle, so that the degradation progress can be continuously updated in the closed-loop structure.

[0045] In this implementation plan, degradation levels are clearly stratified within different threshold ranges. Monitoring strategies, sampling frequencies, and task plans can be structurally adjusted according to the degree of degradation. Early warning links and key links obtain clear tracking paths in the recording structure. The random reset process is suspended in the high-level degradation stage, so that the degradation links are continuously expressed. The link degradation index value forms a cyclically updatable input in the closed-loop path, providing a stable degradation basis for the generation of maintenance strategies.

[0046] Specifically, maintenance strategies and work order information are generated based on degradation levels and monitoring and adjustment actions. The operational status after maintenance is synchronized to the original dataset of the entire operation. The specific steps are as follows: Maintenance strategies corresponding to components are generated based on the root cause node and propagation direction in the degradation chain. By identifying the earliest offset node in the degradation chain and the path of the offset's outward spread, a strategy structure that can be directly used for task generation is formed. The maintenance strategy is then transformed into a minimal set of fields according to the correspondence between maintenance strategy and work order parameters, including component ID, priority, recommended replacement parts, recommended retest indicators, and retest frequency. This ensures that the work order generation process has a unified field system and is executable. The generated work order information is pushed to the execution end through a task scheduling process, enabling maintenance operations to proceed smoothly. The system can be configured based on the urgency and impact of the degraded links; it can adjust the monitoring frequency, the granularity of key component acquisition, and the abnormal triggering conditions according to changes in link status, enabling the monitoring structure to dynamically shrink or expand around the degree of degradation, and write updated parameters into the monitoring and control structure so that the next operating cycle can run under the new monitoring configuration; after maintenance is completed, it collects operating status data, and by re-acquiring the post-maintenance action behavior, vibration changes, load changes, and station behavior, it can re-enter the post-maintenance operating performance into the system's data infrastructure, write the operating status data into the original dataset of the entire operation, and enable the closed loop to re-establish input conditions in the next cycle, driving the continuous iteration of the offset analysis, coupling analysis, and degradation assessment structures.

[0047] In this implementation plan, the maintenance strategy and work order parameters form a minimal structure based on the degradation link. The monitoring frequency and collection granularity can be dynamically adjusted according to the degradation level. The maintenance execution results can be written back in real time in the original dataset throughout the operation. The closed-loop structure maintains continuity in periodic updates. The link status, monitoring strategy and maintenance tasks form a stable connection within the same data link, providing a continuously updated operational basis for subsequent offset analysis and coupling analysis.

[0048] like Figure 2The diagram shows the structure of the elevator health closed-loop assessment system based on spare parts lifespan and status provided in this application embodiment. This system applies an elevator health closed-loop assessment method based on spare parts lifespan and status, including: a component-wide acquisition module, a cross-phase offset analysis module, a coupling feature recognition module, a link health assessment module, and a closed-loop strategy feedback module. The component-wide acquisition module collects multi-source raw information generated during elevator operation. It forms a complete raw dataset of the entire operation process by performing time alignment and structural integration on motion records, vibration records, load records, and position records. Preprocessing is then performed on the raw dataset to ensure continuity and comparability of multi-dimensional information on a unified time axis. The cross-phase offset analysis module analyzes the cross-component offset relationship between the door operator's action phase and the traction machine's load phase based on the complete raw dataset of the entire operation process. It generates offset markers through the ratio structure formed by continuous change in action and load change, and transmits the offset meaning to subsequent structures, thus clearly defining the cross-component rhythm differences. Offset input; the coupling feature recognition module is used to construct a weakly coupled structure between the traction machine's vibration behavior and the brake's action stage based on cross-component offset markers and multi-component segment features. It identifies the synchronicity of vibration disturbance and pull-in changes through joint energy expression, classifies coupling levels according to coupling changes, and synchronously couples features to the link structure, so that the coupling behavior forms a clearly distinguishable level expression in the time dimension; the link health assessment module is used to determine the key nodes and propagation direction of the cross-component degraded link by combining the link structure and layer station behavior. It forms a link degradation index value through the position offset change trend and weak coupling strength, classifies degradation levels according to the degradation progress, and generates monitoring and adjustment actions and task plans, so that the degraded link has a clear direction of progress in periodic updates; the closed-loop strategy feedback module is used to generate maintenance strategies and work order information based on degradation levels and monitoring and adjustment actions. It realizes maintenance strategy mapping through a minimum work order parameter structure, and synchronizes the operating status after maintenance to the original dataset of the entire operation process, so that the collection, parsing, identification, evaluation and feedback form a continuously updated closed-loop control path.

[0049] In this implementation scheme, the whole-domain acquisition of components, cross-phase offset analysis, coupling feature identification, link health assessment and closed-loop strategy feedback form a coherent information flow path within the system. Multi-source data are integrated in a unified time axis, offset relationships and weak coupling behaviors are expressed in the structured link, the degree of degradation is presented in a hierarchical manner in quantitative indicators, and monitoring actions and maintenance strategies are continuously updated in the same link cycle, so that the entire health assessment process has the ability to be traceable, verifiable and closed-loop.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0051] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A closed-loop health assessment method for elevators based on spare parts lifespan and condition, characterized in that, Includes the following steps: S1: Collect multi-source raw information generated during elevator operation, construct a raw dataset of the entire operation process, and perform preprocessing on the raw dataset of the entire operation process; S2, based on the original dataset of the entire operation, analyzes the cross-component offset relationship between the gantry crane operation stage and the traction machine load stage, and performs offset marking and offset meaning transmission according to the offset trend; S3, construct a weakly coupled structure between the vibration behavior of the traction machine and the action stage of the brake based on the cross-component offset mark and the multi-component segment features, divide the coupling level according to the coupling change and synchronously couple the features to the link structure; S4, combining link structure and layer station behavior, determines the key nodes and propagation direction of cross-component degradation links, classifies degradation levels according to degradation progress, and generates monitoring and adjustment actions and task plans; S5 generates maintenance strategies and work order information based on the degradation level and monitoring adjustment actions, and synchronizes the operating status after maintenance to the original dataset of the entire operation process.

2. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for collecting multi-source raw information generated during elevator operation and constructing a raw dataset of the entire operation are as follows: The system collects basic operational information generated by the elevator operation control system, door operator control board, traction machine main drive circuit current sensor, triaxial acceleration sensor, brake action detection sensor, leveling detection unit, and time management unit during elevator operation. This basic operational information includes: start and stop records of opening and closing actions, start and stop event records, brake engagement start and stop records, raw samples of current during traction machine startup, raw sequences of triaxial acceleration during traction machine operation, leveling completion event records, and raw samples of vertical displacement. All information is timestamped. Simultaneously, external reference data corresponding to the elevator's operating environment is accessed, including: floor location structure and hoistway spatial distribution structure. The basic operational information is initially synchronized and integrated according to the chronological order of start and stop records, leveling records, and brake action records. Combined with the external reference data, a unified structure of action, time, vibration, and position information is established to construct a raw dataset for the entire operation process.

3. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for preprocessing the original dataset throughout the entire process are as follows: Based on the timestamp alignment relationship between multi-source records, a dynamic time warping algorithm is used to construct a unified time axis, and the sequence of running segments with different sampling frequencies is rearranged. Combining the full-cycle behavior pattern, the median absolute deviation method and the isolated forest algorithm are used to identify local and global abnormal segments and complete the anomaly removal. For time slice breaks that occur in cross-component records, the exponential weighted moving average algorithm and the neighborhood segment-based collaborative completion strategy are used to fill in the missing segments. Based on the running rhythm fluctuations, the reconstructed continuous sequence is subjected to the local weighted regression filtering algorithm and the Kalman filtering algorithm to complete the trend fitting and noise suppression. Under the unified time axis, the reconstructed sequence is subjected to numerical domain transformation, and the original dataset of the entire running process is standardized and normalized.

4. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for analyzing the cross-component offset relationship between the gantry crane's operation phase and the traction machine's load phase based on the original dataset of the entire operation process, and for executing offset marking and offset meaning transmission according to the offset trend, are as follows: Combine the start and end records of the opening and closing actions, read the start and end times of the current opening and closing action and perform time period conversion to form the gantry crane action time; backtrack to the previous opening and closing cycle according to the time sequence, and obtain the gantry crane action time of the previous cycle from the gantry crane action time record of the previous opening and closing cycle. The system retrieves the original current samples of the traction machine startup phase based on the time slice corresponding to the startup phase, and selects the current peak segment at the moment of startup to form the traction machine startup current. It locates the startup phase of the previous startup cycle along the startup and shutdown event records, and obtains the startup current of the previous cycle from the traction machine startup current record of the previous startup and shutdown cycle. It calculates the time difference between two consecutive startup and shutdown timestamps in the startup and shutdown event records and performs interval conversion to obtain the startup and shutdown cycle time interval. It extracts the duration of continuous operation cycles by combining the start and end records of startup and shutdown actions, calculates the time span change of adjacent cycles and compares it with the time span threshold, defines the set of cycles where the time span change is continuously less than the time span threshold as the stable interval during normal operation, and selects the lower bound of the time span from the stable interval to form the phase offset protection constant. The change in gantry crane operating time is obtained by subtracting the operating time of the gantry crane from the operating time of the previous cycle. The change in traction machine starting current is obtained by subtracting the starting current of the traction machine from the starting current of the previous cycle. The change in traction machine starting current is added to the phase offset protection constant to form the denominator structure. The phase offset value across components is obtained by dividing the change in gantry crane operating time by the denominator structure. When the cross-component phase offset value increases or exhibits an abnormal ratio during continuous operation, the cross-component phase offset value is written to the offset buffer, and the weak coupling modulation condition is recorded in the offset status field; when the cross-component phase offset value does not increase continuously or exhibit an abnormal ratio, the transmission to subsequent modules is stopped, and no adjustment operation is triggered.

5. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for constructing a weakly coupled structure between the traction machine's vibration behavior and the brake's actuation phase, based on cross-component offset markers and multi-component segment features, are as follows: Extract the cross-component phase offset value from the offset buffer; combine the brake engagement start and end records to read the current operating cycle engagement start time and engagement end time and perform time period conversion to form the brake engagement time; Locate the acceleration data segment of the current cycle based on the original triaxial acceleration sequence during the operation of the traction machine, select the amplitude peak segment in the segment, and extract the amplitude to generate the vibration amplitude of the traction machine; Based on the brake engagement start and end records and the original triaxial acceleration sequence, multiple adjacent cycles of engagement duration segments and vibration amplitude segments were read. The variation amplitude of the engagement duration segments and the variation amplitude of the vibration amplitude segments were calculated respectively. The two types of variation amplitudes were compared with the fluctuation amplitude threshold. The set of cycles in which the variation amplitude is less than the fluctuation amplitude threshold was defined as the historical stable cycle. Fixed reference values ​​were selected from the value range corresponding to the historical stable cycle to form the coupling strength protection constant. Multiply the brake engagement time of each cycle within the historical operating window by the vibration amplitude of the traction machine, square the result, and sum the results. Take the square root of the sum to form the joint energy term. Take the absolute value of the phase offset across components and add it to the coupling strength protection constant to form the phase modulation denominator term. Take the reciprocal of the denominator term to form the phase modulation coefficient. Multiply the joint energy term by the phase modulation coefficient to obtain the weak coupling energy intensity value.

6. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for classifying coupling levels based on coupling changes and synchronously coupling features to the link structure are as follows: The weak coupling energy intensity value is compared with the coupling threshold, which includes a first-level coupling threshold and a second-level coupling threshold. When the weak coupling energy intensity value is lower than the first-level coupling threshold, the weak coupling energy intensity value is written into the coupling buffer and the link is maintained at the basic level. When the weak coupling energy intensity value is greater than or equal to the first-level coupling threshold and lower than the second-level coupling threshold, write a concern mark in the link status table and record the weak coupling energy intensity value and corresponding running cycle in the link trend structure. When the weak coupling energy intensity value is greater than or equal to the secondary coupling threshold, the link entry is registered in the key coupling structure, the sensitivity parameter is updated in the link parameter table, and the weak coupling energy intensity value is written into the degradation evaluation path without triggering any execution layer operation.

7. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for determining the key nodes and propagation path of cross-component degradation links by combining link structure and layer station behavior are as follows: The coupled structure is integrated into a continuous degenerate link in chronological order; the completion time of the current leveling action is located by combining the leveling completion event record, and the displacement deviation of the corresponding time slice is selected from the original vertical displacement sample to form the leveling offset of the floor station; the completion time of the previous leveling is retrieved according to the leveling completion event record, and the displacement deviation of the corresponding time is read from the original vertical displacement sample to form the leveling offset of the previous cycle floor station; the timestamps of two consecutive events associated with leveling are extracted from the start and stop event records and interval conversion is performed to generate the leveling cycle time interval; the weak coupling energy intensity value is extracted from the coupling buffer area; The leveling offset rate is obtained by subtracting the previous leveling offset from the leveling offset and dividing by the leveling cycle time interval. The leveling offset rate is then multiplied by the weak coupling energy intensity value to obtain the link degradation index value.

8. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for classifying degradation levels based on the degradation progress and generating monitoring and adjustment actions and task plans are as follows: The link degradation index value is compared with the degradation threshold, which includes a primary degradation threshold and a secondary degradation threshold. When the link degradation index value is less than the primary degradation threshold, the monitoring parameters are maintained and no additional processing is performed. When the link degradation index value is greater than or equal to the first-level degradation threshold and less than the second-level degradation threshold, the link is written into the early warning structure, the sampling frequency of the link is increased, an inspection task suggestion is generated, and the link degradation index value is written into the link record structure. When the link degradation index value is greater than or equal to the second-level degradation threshold, a maintenance work order is generated, the key monitoring strategy is started, the link random reset process is frozen, the link is written into the maintenance priority structure, and the link degradation index value is written into the closed-loop feedback path as the input for the next operating cycle.

9. The elevator health closed-loop assessment method based on spare parts life and condition according to claim 1, characterized in that: The specific steps for generating maintenance strategies and work order information based on degradation levels and monitoring and adjustment actions, and synchronizing the operational status after maintenance to the original dataset of the entire operation process, are as follows: Based on the causal node and propagation direction in the degradation link, a maintenance strategy corresponding to the component is generated. The maintenance strategy is transformed into maintenance plan parameters and work order information is generated and pushed to the execution end through the task scheduling process. The monitoring frequency, key component collection granularity and abnormal triggering conditions are adjusted according to the link status changes, and the adjusted parameters are written into the monitoring and control structure. After the maintenance is completed, the operation status data is collected and written into the original dataset of the entire operation.

10. An elevator health closed-loop assessment system based on spare parts life and condition, employing the elevator health closed-loop assessment method based on spare parts life and condition as described in any one of claims 1-9, comprising: The module comprises a component-wide acquisition module, a cross-phase offset analysis module, a coupling feature recognition module, a link health assessment module, and a closed-loop strategy feedback module, characterized in that: The component global acquisition module is used to collect multi-source raw information generated during elevator operation, construct a raw dataset of the entire operation process, and perform preprocessing on the raw dataset of the entire operation process. The cross-phase offset analysis module is used to analyze the cross-component offset relationship between the gantry crane operation phase and the traction machine load phase based on the original dataset of the entire operation process, and to execute offset marking and offset meaning transmission according to the offset trend; The coupling feature identification module is used to construct a weak coupling structure between the vibration behavior of the traction machine and the action phase of the brake around the cross-component offset mark and the multi-component segment feature, and to divide the coupling level according to the coupling change and synchronously couple the features to the link structure. The link health assessment module is used to determine the key nodes and propagation direction of cross-component degradation links by combining the link structure and layer station behavior, classify the degradation level according to the degradation progress, and generate monitoring and adjustment actions and task plans. The closed-loop strategy feedback module is used to generate maintenance strategies and work order information based on the degradation level and monitoring adjustment actions, and to synchronize the operating status after maintenance to the original dataset of the entire operation process.

Images