A device life prediction method, apparatus, device, and storage medium

Abstract

The present application relates to the technical field of equipment life prediction, and particularly relates to an equipment life prediction method, device, equipment and storage medium; the equipment life prediction method comprises the following steps: analyzing elevator rotation speed data, mechanical coupling relationship, electric appliance coupling relationship and fault statistical coupling relationship to obtain directed network topology and topology orientation data; predicting the elevator sound data, the topology orientation data and the directed network topology according to a preset fault mode library to obtain real-time predicted residual life; generating equipment life prediction information according to the real-time predicted residual life; the present application predicts the residual life of the elevator by fusing multiple data, and provides support for elevator operation and maintenance and whole life cycle management.

Description

Technical Field

[0001] This invention relates to the field of equipment life prediction technology, and in particular to a method, apparatus, device and storage medium for predicting equipment life. Background Technology

[0002] In the field of equipment life prediction technology, existing methods generally suffer from shortcomings such as insufficient data collection targeting, lack of consideration for system coupling relationships, and simplistic prediction logic. For example, in the elevator life prediction scenario, it is difficult to meet the health management needs of the elevator throughout its entire life cycle: on the one hand, most solutions do not combine data collection based on the differentiated elevator operating status, but only obtain a single parameter (such as rotational speed) at a fixed frequency, which cannot reflect the differences in elevator life loss at different stages such as stationary, transitional, and stable, resulting in low data effectiveness; on the other hand, they ignore the coupling relationship between the elevator's mechanical system (such as traction machine and guide rail) and electrical system (such as frequency converter and motor), and do not incorporate historical fault statistics, making it impossible to quantify the impact of multi-system interactions on elevator life; in addition, some Some existing solutions rely solely on single data points to predict elevator lifespan, failing to establish a correlation between multi-dimensional data and failure modes. This leads to significant prediction biases and poor practicality, hindering elevator maintenance decision-making. The core technological advantage of this solution lies in phased data acquisition, multi-dimensional coupling relationship analysis, and failure mode library matching. Scenario migration can be achieved as long as the target equipment meets the following three conditions: the equipment operation has clearly defined phases (e.g., startup, stabilization, shutdown, corresponding to the elevator's operational stages); basic data reflecting component status, such as "sound and speed," can be collected, and the equipment exhibits "mechanical and electrical coupling relationships"; and historical failure data allows for the construction of a "failure mode library" (e.g., the correspondence between "a certain data feature and the remaining lifespan of a certain component"). Based on this, this solution can broadly cover electromechanical equipment in industries such as industrial, transportation, civil, and energy sectors, providing accurate lifespan prediction support for the entire lifecycle maintenance of equipment. Summary of the Invention

[0003] In order to overcome the shortcomings of the prior art, the present invention aims to provide a method, apparatus, device and storage medium for predicting device lifespan.

[0004] The first aspect of this invention provides a method for predicting equipment lifespan, comprising: acquiring elevator sound data and analyzing the elevator sound data according to a preset time period and a preset timestamp to obtain the elevator operating stage; collecting elevator speed data according to the elevator operating stage and a preset iterative stopping condition; acquiring mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships; analyzing the elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data; predicting the elevator sound data, topology guidance data, and directed network topology according to a preset fault mode library to obtain a real-time predicted remaining lifespan; and generating equipment lifespan prediction information based on the real-time predicted remaining lifespan.

[0005] Furthermore, the step of analyzing elevator sound data according to a preset time period and a preset timestamp to obtain the elevator operation stage includes: obtaining a phase marker signal according to the time period; converting the phase marker signal according to a preset segmentation start point to obtain a phase reference signal; aligning the elevator sound data according to the timestamp and the phase reference signal to obtain aligned signal data; and analyzing the aligned signal data according to a preset sound fluctuation threshold to obtain the elevator operation stage.

[0006] Further, the step of collecting elevator speed data based on the elevator operation stage and preset iterative stopping conditions includes: when the elevator operation stage is the elevator stationary stage, collecting stationary stage speed data according to a preset first continuous collection time and a preset first speed threshold; when the elevator operation stage is the elevator operation transition stage, collecting transition stage speed data according to a preset second continuous collection time and a preset second speed threshold; when the elevator operation stage is the elevator operation stable stage, collecting stable stage speed data according to a preset third continuous collection time and a preset speed threshold range; determining whether the elevator operation stage meets the iterative stopping conditions; if the elevator operation stage does not meet the iterative stopping conditions, returning to the process of acquiring elevator sound data until the elevator operation stage meets the iterative stopping conditions, then acquiring all types of speed data currently collected to obtain elevator speed data.

[0007] Furthermore, the analysis of elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data includes: performing trend analysis on elevator speed data according to preset fault evolution laws to obtain attenuation factors and fault chain analysis paths; updating preset coupling type weights according to attenuation factors and fault chain analysis paths to obtain update type weights; and analyzing update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data.

[0008] Furthermore, the analysis of update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data includes: constructing a directed network topology based on preset topology construction conditions, update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships; obtaining traction system evaluation loss indicators and determining whether the traction system evaluation loss indicators are greater than or equal to preset loss indicator thresholds; if the traction system evaluation loss indicators are greater than or equal to the loss indicator thresholds, then performing structural analysis on the directed network topology to obtain topology guidance data.

[0009] Furthermore, the step of predicting elevator sound data, topology guidance data, and directed network topology based on a preset fault mode library to obtain real-time predicted remaining life includes: constructing a curve to be analyzed based on elevator sound data; mapping and comparing the curve to be analyzed with a preset sound standard curve to obtain a mapping normalization distance; extracting features from the elevator sound data based on the mapping normalization distance to obtain frequency amplitude features and deviation rate features; and predicting topology guidance data, directed network topology, frequency amplitude features, and deviation rate features based on the fault mode library to obtain real-time predicted remaining life.

[0010] Furthermore, the step of generating equipment life prediction information based on real-time predicted remaining life includes: obtaining the predicted remaining life of the previous month, and performing deviation analysis on the real-time predicted remaining life based on the predicted remaining life of the previous month to obtain a deviation index; obtaining the cumulative running time of the elevator, and determining whether the cumulative running time has reached the preset fatigue test duration; when the cumulative running time reaches the fatigue test duration, generating equipment life prediction information based on the deviation index and the real-time predicted remaining life.

[0011] Furthermore, an equipment life prediction device includes: a first analysis module for acquiring elevator sound data and analyzing the elevator sound data according to a preset time period and a preset timestamp to obtain the elevator operation stage; a first data acquisition module for acquiring elevator speed data according to the elevator operation stage and a preset iterative stopping condition; a second data acquisition module for acquiring mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships; a second analysis module for analyzing the elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data; a prediction module for predicting the elevator sound data, topology guidance data, and directed network topology according to a preset fault mode library to obtain a real-time predicted remaining lifespan; and an information generation module for generating equipment life prediction information based on the real-time predicted remaining lifespan.

[0012] Furthermore, a device lifetime prediction device includes: a memory and at least one processor, the memory storing instructions; at least one processor invokes the instructions in the memory to cause the device lifetime prediction device to perform the steps of a device lifetime prediction method as described in any one of the above descriptions.

[0013] Furthermore, a computer-readable storage medium stores instructions that, when executed by a processor, implement the steps of a device lifetime prediction method as described in any of the preceding claims.

[0014] In the technical solution of this invention, elevator sound data is first acquired, and then combined with time period, timestamp, and elevator sound data to accurately analyze the elevator operation stage, laying the scenario foundation for subsequent data collection. Next, elevator speed data is collected based on the elevator operation stage and iterative stopping conditions, specifically acquiring information for each stage to ensure complete data coverage and high effectiveness, avoiding blind collection. Simultaneously, three types of coupling relationships are comprehensively acquired, and combined with speed data analysis to form directed network topology and topology guidance data, quantifying subsystem interactions, clarifying fault paths and core nodes, and providing system logic support for prediction. Finally, based on a fault mode library, sound data and topology data are fused to predict real-time remaining lifespan and generate information, taking into account both real-time status and system logic, improving prediction accuracy. The overall solution not only solves the problems of low data effectiveness, lack of coupling consideration, and large prediction deviations in traditional methods, but also provides clear lifespan nodes and decision-making basis for operation and maintenance, helping to plan maintenance in advance, reducing the risk of sudden failures, and effectively ensuring the health management of the elevator throughout its entire life cycle. Attached Figure Description

[0015] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a first flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 2 This is a second flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 3 This is a third flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 4 This is a fourth flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 5 A fifth flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 6 A sixth flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 7 A seventh flowchart of a device life prediction method provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of a device for predicting equipment lifespan according to an embodiment of the present invention; Figure 9 This is a schematic diagram of the structure of a device for predicting device lifespan, provided in an embodiment of the present invention. Detailed Implementation

[0016] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” or “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0017] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the equipment life prediction method of the present invention includes: 101. Acquire elevator sound data and analyze the elevator sound data according to the preset time period and preset timestamp to obtain the elevator operation stage; In this embodiment, elevator sound data is first acquired, and then combined with time period (e.g., 10ms / time) and timestamp (e.g., "05s-5s after start" and "35s-8s before stop"), the sound fluctuation threshold (e.g., fluctuation ≤5dB in the stationary stage, 5-20dB in the transition stage, and 7dB≤8dB in the stable stage) is analyzed to accurately divide the three major operating stages of stationary, transition, and stable operation. This lays the foundation for subsequent differentiated collection of rotation speed data and ensures that the data can reflect the differences in lifespan loss at different stages. 102. Elevator speed data are collected based on the elevator operation stage and preset iterative stopping conditions; In this embodiment, based on the elevator's operating stage and combined with iterative stopping conditions, rotational speed data is collected. This allows for the targeted acquisition of rotational speed information for different stages of the elevator (stationary, transitional, and stable), accurately reflecting the lifespan characteristics of each stage and avoiding the problem of low effectiveness caused by blind data collection. At the same time, iterative stopping conditions ensure that the collected data covers the entire operating cycle, providing comprehensive and high-quality data support for subsequent multi-system coupling analysis and lifespan prediction, thereby improving the accuracy of equipment lifespan prediction and maintenance decisions. 103. Obtain mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships; 104. Analyze elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data; In this embodiment, the three types of coupling relationships of elevator mechanics, electrical system, and fault statistics are first comprehensively acquired. Then, the directed network topology and topology guidance data are obtained by combining rotation speed data analysis. The impact of the interaction of each subsystem on the lifespan can be quantified, the fault propagation path and core nodes can be identified, and accurate system logic support can be provided for subsequent lifespan prediction. This avoids misjudgment based on single data and improves the comprehensiveness and reliability of equipment lifespan prediction. 105. Based on a preset fault mode library, predict elevator sound data, topology guidance data, and directed network topology to obtain real-time predicted remaining lifespan. 106. Generate equipment life prediction information based on real-time predicted remaining life; In this embodiment, based on the fault mode library, elevator sound data (reflecting real-time status), topology guidance data (clarifying fault paths), and directed network topology (quantifying system coupling) are integrated to make predictions and accurately output the real-time remaining lifespan. Based on this, lifespan prediction information is generated, which allows the prediction to take into account both real-time status and system logic, thus improving accuracy. At the same time, the output information can provide clear lifespan nodes and decision-making basis for operation and maintenance, help to plan maintenance in advance, reduce the risk of sudden failures, and ensure the healthy management of the elevator throughout its entire life cycle. In this embodiment, elevator sound data is first acquired, and combined with time period, timestamp, and elevator sound data to accurately analyze the elevator operation stage, laying the scenario foundation for subsequent data collection. Then, elevator speed data is collected based on the elevator operation stage and iteration stopping conditions, selectively acquiring information for each stage to ensure complete data coverage and high effectiveness, avoiding blind collection. At the same time, three types of coupling relationships are comprehensively acquired, and combined with speed data analysis to form directed network topology and topology guidance data, quantifying subsystem interactions, clarifying fault paths and core nodes, and providing system logic support for prediction. Finally, based on the fault mode library, sound data and topology data are fused to predict real-time remaining lifespan and generate information, taking into account both real-time status and system logic, improving prediction accuracy. The overall solution not only solves the problems of low data effectiveness, lack of coupling consideration, and large prediction deviation in traditional methods, but also provides clear lifespan nodes and decision-making basis for operation and maintenance, helping to plan maintenance in advance, reducing the risk of sudden failures, and effectively ensuring the health management of the elevator throughout its entire life cycle.

[0018] Please see Figure 2 A second embodiment of a device life prediction method according to the present invention includes: 201. Obtain the phase marker signal based on the time period; In this embodiment, (if the bearing rotation cycle is 0.4 seconds, then a phase marker signal is acquired every 0.4 seconds) to obtain the phase marker signal. The core function of this step is to "precisely bind the rotational motion of the elevator mechanical component (bearing) with time" to provide a time reference for the mechanical motion dimension for subsequent segmented alignment of sound data. For example, one rotation of the bearing corresponds to the elevator traction machine completing one complete action. The time of occurrence of the phase marker signal (such as the 10.000th second) can be used as the time anchor point of the starting point of this action, solving the problem that sound data only relies on timestamps (which may have system clock errors) and establishing a dual reference of mechanical motion and time. 202. Convert the phase marker signal according to the preset segmentation start point to obtain the phase reference signal; In this embodiment, the phase mark signal is converted at the segment start point (e.g., setting every 5 phase marks as a segment, corresponding to 5 rotations of the bearing, approximately 2 seconds), to obtain a phase reference signal. The core is to convert the discrete phase mark signal into a continuous segmented time reference. This can be achieved through a signal processing algorithm (e.g., counting the number of phase marks, generating a segment start signal each time a preset number is reached). For example, the preset segment start point is every 3 phase marks. When the 3rd, 6th, and 9th phase marks are detected, a phase reference signal is generated, marked as "Segment 1 Start," "Segment 2 Start," etc. This step converts the periodicity of the mechanical motion (phase marks) into the periodicity of the data segments. 203. Align the elevator sound data according to the timestamp and phase reference signal to obtain aligned signal data; In this embodiment, the specific logic of the alignment process is to take the moment of the phase reference signal as the starting point of the segmentation, and to extract the sound data into segments synchronized with the phase reference signal by comparing the timestamps. By aligning with the phase reference signal, it is ensured that each segment of sound data corresponds to the complete motion cycle of the elevator mechanical component, so that the sound characteristics can accurately reflect the mechanical state within the cycle (such as the change in sound amplitude during the bearing's three rotations, which corresponds to the bearing's wear). 204. Analyze the alignment signal data according to the preset sound fluctuation threshold to obtain the elevator operation stage; In this embodiment, the specific operation involves extracting the amplitude fluctuation characteristics of each segment of the aligned sound data (such as calculating the amplitude standard deviation, the difference between the maximum and minimum values ​​within a segment of data), comparing it with the sound fluctuation threshold, and determining the elevator operating stage. For example, if the amplitude fluctuation of a segment of aligned signal data is 3dB, ≤5dB, it is determined to be a stationary stage; if the fluctuation is 12dB, within the 5dB-20dB range, it is determined to be a transition stage. The purpose of this step is to transform the aligned sound data into a classification basis for the elevator operating state, providing a foundation for subsequent processes such as collecting speed data in stages and constructing the curve to be analyzed. Only by accurately identifying the operating stage can the characteristic data of the corresponding stage be collected in a targeted manner, avoiding analysis errors caused by the mixing of cross-stage data. In this embodiment, a phase marker signal is first acquired based on the time period to precisely bind the bearing rotation motion to time, providing a time reference for the mechanical motion dimension for subsequent audio data segmentation and alignment. Then, the phase marker signal is converted into a phase reference signal according to the segment start point, transforming the periodicity of mechanical motion into the periodicity of data segmentation. Next, the elevator audio data is aligned using timestamps and phase reference signals to ensure that each segment of audio data corresponds to a complete motion cycle of the mechanical component. Finally, the aligned signal data is analyzed based on the audio fluctuation threshold to determine the elevator operation stage. This improves the synchronization and segmentation rationality of audio data and mechanical motion, accurately identifies the operation stage, and provides a reliable basis for subsequent staged data collection and curve construction, effectively supporting equipment life prediction. It also adapts to multi-system coupled analysis, promotes the integration of multiple data and processes, and improves the overall prediction effect.

[0019] Please see Figure 3 A third embodiment of a device life prediction method according to the present invention includes: 301. When the elevator is in a stationary phase during its operation, the stationary phase speed data is collected based on the preset first continuous acquisition time and the preset first speed threshold. In this embodiment, the rotational speed is close to zero during the stationary phase, mainly reflecting the basic wear of components under idling conditions (such as static friction of bearings). The data collection is based on a first continuous collection time and a first rotational speed threshold. The former ensures that the collection time is sufficient to reflect static wear (e.g., 30 seconds to avoid the randomness of instantaneous data), while the latter limits the data range (e.g., ≤5 rpm to exclude minute instantaneous rotational speeds caused by external interference). The implementation logic is as follows: when the elevator is determined to be in a stationary phase (e.g., sound fluctuation ≤5dB, no obvious operating noise), the data collection device is activated, and rotational speed data is collected continuously for 30 seconds. Only data ≤5 rpm is retained, and outliers exceeding the threshold are filtered out to obtain the rotational speed data during the stationary phase. This data can be used to analyze the degree of static wear of the bearings (e.g., the larger the range of rotational speed fluctuations when stationary, the more severe the bearing wear). 302. When the elevator operation phase is the elevator operation transition phase, the transition phase speed data is collected according to the preset second continuous acquisition time and the preset second speed threshold. In this embodiment, the data collection is based on a second continuous data collection time (e.g., 10 seconds, covering the complete acceleration and deceleration process of starting and stopping) and a second speed threshold (e.g., 5rpm-1500rpm, matching the range of speed change from low to high or from high to low during the transition phase). The implementation logic is as follows: when the elevator is determined to be in a transition phase (such as sound fluctuation of 5dB-20dB, accompanied by obvious acceleration or deceleration noise), the speed data is continuously collected within 10 seconds, and only the data of 5-1500rpm is retained (excluding 0rpm when stationary and constant speed when stable), and finally the speed data of the transition phase is obtained. This data can be used to analyze the component wear under dynamic load (such as whether the speed rise is smooth during acceleration, if there is a stutter, it may indicate wear of the traction machine gearbox). 303. When the elevator is in a stable operating phase, the stable operating speed data is collected according to the preset third continuous acquisition time and the preset speed threshold range. In this embodiment, the data collection is based on the third continuous data collection time (e.g., 60 seconds, to ensure coverage of a sufficiently long stable operating cycle) and the speed threshold range (e.g., 1480rpm-1520rpm, matching the elevator's rated speed range, allowing for minor fluctuations). The implementation logic is as follows: when the elevator is determined to be in a stable phase (e.g., 7dB-8dB, with uniform operating noise), speed data is collected continuously for 60 seconds, and only the data from 1480rpm to 1520rpm is retained (excluding non-constant speeds during the transition phase). Finally, the stable phase speed data is obtained. This data is key to analyzing the continuous wear and tear of the elevator's core components (such as the motor and traction machine), such as whether the average speed during the stable phase decreases month by month (reflecting motor aging) or whether the fluctuation range expands (reflecting bearing wear). 304. Determine whether the iterative stopping condition is met during the elevator operation phase; In this embodiment, the iteration stopping condition is the elevator operation stage, which is the stable operation stage of the elevator, to avoid the one-sidedness of analysis caused by collecting data from only a single stage. 305. If the iteration stop condition is not met during the elevator operation phase, return to the process of acquiring elevator sound data until the iteration stop condition is met during the elevator operation phase. Then, acquire all types of speed data collected so as to obtain elevator speed data. In this embodiment, if the stopping condition is not met (e.g., only the static and transition phase data are collected, and the stable phase data is missing), the process returns to the elevator sound data acquisition stage, re-determines the operating phase, and collects the corresponding speed data; if the stopping condition is met, the speed data of the three types of phases are integrated to form complete elevator speed data, covering all phases of a complete elevator operation, ensuring that subsequent analysis can fully reflect the lifespan loss characteristics of each phase. In this embodiment, a differentiated speed acquisition strategy is formulated for different operating stages of the elevator. The speed data collected during the stationary stage accurately reflects the static wear of the bearings; the speed data collected during the transition stage captures the dynamic wear of components such as the traction machine gearbox; and the speed data collected during the stable stage monitors the continuous wear of core components such as the motor, improving the relevance and effectiveness of the data. At the same time, the stable stage is used as the iteration stop condition to ensure that all three types of data are collected, avoiding one-sided analysis. Overall, this solution can provide high-quality, multi-dimensional data support for subsequent elevator life loss analysis, help accurately trace faults, provide differentiated basis for operation and maintenance, provide early warning of fault risks, avoid over- or under-maintenance, reduce operation and maintenance costs and risks, and effectively support equipment life prediction and health management.

[0020] Please see Figure 4 A fourth embodiment of a device life prediction method according to the present invention includes: 401. Perform trend analysis on elevator speed data based on the preset fault evolution law to obtain the attenuation factor and fault chain analysis path; In this embodiment, the fault evolution pattern originates from a summary of massive fault cases in the elevator industry and research on the failure mechanisms of core components, covering the development paths and data characteristics of typical faults in mechanical and electrical systems. For example, the evolution pattern of motor winding aging is manifested as "decreased speed stability → increased risk of sudden speed drop under load," corresponding to characteristics such as "intermittent abnormal fluctuations" and "speed below standard value under rated load" in the speed data. These patterns are solidified in the database, providing clear judgment criteria for speed data analysis. By selecting speed data during the stable operation phase of the elevator and using linear regression or exponential fitting models, the attenuation rate of speed per unit time is calculated to obtain the attenuation factor. For example, traction machine rated speed data for three consecutive months is collected. If the average speed is 1500 rpm in the first month, 1495 rpm in the second month, and 1488 rpm in the third month, a decay factor of 4 rpm / month is obtained through linear fitting, quantifying the rate of speed performance degradation and indirectly reflecting the wear and tear of components (such as bearings and motors). Combined with the fault evolution law, the abnormal characteristics of the speed data are traced and correlated. For example, when the speed data shows "increased fluctuations and intermittent pauses", and historical data shows that this feature often appears before the abnormal current of the traction machine, and then further triggers the overload alarm of the control cabinet, the fault chain analysis path is identified as "traction machine bearing wear → abnormal traction machine speed → abnormal electrical load of control cabinet", clarifying the path of fault propagation from the mechanical system to the electrical system. 402. Update the preset coupling type weights based on the attenuation factor and the fault chain analysis path to obtain the updated type weights; In this embodiment, the coupling type weights are set based on the functional importance and theoretical coupling strength of each elevator subsystem. The weights are adjusted by combining attenuation factors (quantifying the rate of component performance degradation) and fault chain analysis paths (clarifying the fault propagation logic). For example, when speed decay is dominated by mechanical wear, the mechanical coupling weight is increased; when the fault chain is dominated by electrical fault propagation, the electrical coupling weight is increased. This ensures that the weights accurately match the actual fault state and operating conditions of the elevator, avoiding the shortcomings of traditional fixed weights that are detached from reality. This dynamic update mechanism allows the coupling type weights to reflect the real impact strength of the coupling relationship between each subsystem in real time, providing accurate data support for subsequent directed network topology construction. This enables the topology to accurately characterize the effect of multi-system coupling on elevator lifespan, providing a reliable basis for lifespan prediction and effectively improving the accuracy and adaptability of equipment lifespan prediction. 403. Analyze the update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistics coupling relationships to obtain directed network topology and topology guidance data; In this embodiment, based on a large number of fault cases and component failure mechanisms, a decay factor for quantitative performance degradation is fitted using stable phase rotational speed data. Combined with fault evolution patterns, abnormal characteristics are traced to clarify the fault propagation chain path. Then, based on this, the coupling type weights are dynamically updated to accurately match the actual fault state and operating conditions, truly reflecting the coupling influence intensity of the subsystems. Finally, the updated weights and multiple coupling relationships are combined to construct a directed network topology and guidance data, comprehensively and accurately depicting the effect of multi-system coupling on elevator lifespan, providing reliable support for subsequent equipment lifespan prediction.

[0021] Please see Figure 5 A fifth embodiment of a device life prediction method according to the present invention includes: 501. A directed network topology is constructed based on preset topology construction conditions, update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistics coupling relationships; In this embodiment, the directed network topology construction process is as follows: Node selection for directed network topology: The elevator's subsystems are used as network nodes; these subsystems are the core components for the elevator to perform its functions, and changes in the state of each subsystem will affect the overall lifespan of the elevator; elevator subsystems (such as traction system, door system, control cabinet system, safety system, guide system, balancing system, etc.). Directed edge determination: Mechanical coupling relationship: For example, there is mechanical vibration transmission between the traction system (A) and the door system (E). This transmission will affect the life of the door system. Therefore, a directed edge A→E is constructed, and the weight (e.g., 0.3) represents the degree of impact of vibration transmission on the life of E. Electrical coupling relationship: Like the control cabinet system (B) and the traction system (A), if the control cabinet fails (e.g., abnormal output current), it will accelerate the life loss of the traction system. Therefore, a directed edge B→A is constructed, and the weight (e.g., 0.2) reflects the degree to which the failure of B accelerates the life loss of A. Fault statistical coupling relationship: Based on historical fault data statistics, if the failure of a certain subsystem (e.g., safety system C) often triggers the failure of another subsystem (e.g., guidance system D), a directed edge C→D is constructed to reflect the impact of the statistical law of fault propagation on life. Based on the topology construction conditions (such as ensuring that nodes cover key subsystems and edges can accurately reflect the direction and strength of coupling relationships), the nodes, directed edges and corresponding weights defined above are integrated to construct a weighted directed network topology. This topology can comprehensively and intuitively present the coupling relationships between various elevator subsystems and the impact of these relationships on the elevator's lifespan. 502. Obtain the traction system evaluation loss index and determine whether the traction system evaluation loss index is greater than or equal to the preset loss index threshold. In this embodiment, the traction system is the core power system of the elevator, and its wear and tear is crucial to the overall lifespan of the elevator. The traction system wear assessment index (such as P(A)=0.5) is calculated by comprehensively considering factors such as the wear of the traction machine, the fatigue state of the bearings, and the wear of the rope grooves of the traction sheave, and is obtained through professional testing methods (such as vibration detection, temperature monitoring, wear measurement, etc.) and assessment models. It is used to quantify the wear and tear of the traction system. 503. If the loss index of the traction system is greater than or equal to the loss index threshold, then perform structural analysis on the directed network topology to obtain topology guidance data. In this embodiment, a loss index threshold (e.g., 0.4) is used. When the loss index of the traction system is ≥ this threshold, it indicates that the loss of the traction system has reached a certain level, and its coupling relationship with other subsystems will have a more significant impact on the overall lifespan of the elevator. At this time, structural analysis is performed on the previously constructed directed network topology, such as analyzing the critical path (i.e., the path where faults or losses are most easily propagated) and the centrality of nodes (which subsystems are in a core position in the coupling relationship and have a greater impact on the overall lifespan), to obtain topology-guided data. This data can clarify how the coupling effect between the subsystems guides the trend of elevator lifespan changes under the current loss state of the traction system, providing key structural basis for subsequent lifespan prediction. In this embodiment, a weighted directed network topology is constructed, selecting key elevator subsystems as nodes and using mechanical coupling, electrical coupling, and fault statistics coupling as directed edges with corresponding weights. This comprehensively and intuitively presents the impact of coupling between subsystems on elevator lifespan. Simultaneously, focusing on the core traction system, and triggering analysis based on whether its assessment loss index reaches a threshold, structural analysis of the network topology, including critical path and node centrality, is performed. The resulting topology-guided data clarifies the guiding trend of subsystem coupling on elevator lifespan under the current traction system loss state, providing accurate structural basis for subsequent lifespan prediction. This effectively integrates multi-system coupling relationships with the impact of core subsystem loss, improving the comprehensiveness and accuracy of equipment lifespan prediction and providing strong support for elevator operation and maintenance decisions.

[0022] Please see Figure 6 The sixth embodiment of a device life prediction method in this invention includes: 601. The curve to be analyzed is constructed based on the elevator sound data; In this embodiment, sound data from multiple stages of the elevator is collected by an acoustic sensor to construct a curve to be analyzed. The disordered data is transformed into visual frequency and amplitude features. This can capture subtle abnormalities of components in real time, laying the foundation for subsequent comparison, feature extraction and life prediction, improving prediction accuracy, and also helping to quickly locate faults, reduce maintenance costs, and promote the transformation of elevator operation and maintenance towards proactivity. 602. Map and compare the curve to be analyzed with the preset sound standard curve to obtain the mapping normalization distance; In this embodiment, by dynamically adjusting the alignment paths of the two curves, the minimum cumulative distance between the curve to be analyzed and the sound standard curve is found, i.e., the mapping normalization distance. The smaller the distance, the closer the actual sound is to the standard state (e.g., a mapping normalization distance ≤ 0.05 is normal, and a mapping normalization distance > 0.1 is abnormal), accurately quantifying the difference between the two. The sound standard curve is obtained from the bearing sound standard database: using an online monitoring platform, for different models of rolling bearings, the operating sound data of the bearings under different speeds and loads are collected at key noise detection points to establish a standard sound curve for each bearing; the sound standard curve is the standard sound curve for each bearing. 603. Extract features from elevator sound data based on the mapped normalized distance to obtain frequency amplitude features and deviation rate features; In this embodiment, guided by the mapping normalization distance, priority is given to focusing on frequency bands with significant differences in elevator sound data (e.g., when the mapping normalization distance > 0.1, the difference in the 150Hz-200Hz frequency band accounts for 60%), and two types of core features are extracted: Frequency amplitude features: including the "peak amplitude", "amplitude fluctuation range", and "peak frequency" of the frequency band; for example, the peak amplitude of the 150Hz frequency band increases from the standard value of 0.1Pa to 0.3Pa, and the fluctuation range expands from ±0.02Pa to ±0.08Pa. These parameters directly reflect the degree of bearing wear (the more severe the wear, the larger the peak amplitude and fluctuation range); Deviation rate features: based on the amplitude of the frequency band corresponding to the standard curve, the "average deviation rate" and "maximum deviation rate" of the actual frequency amplitude are calculated; for example, the actual frequency amplitude of the 150Hz frequency band is 0.3Pa, the standard frequency amplitude is 0.1Pa, and the maximum deviation rate = (0.3-0.1) / 0.1×100%=200%. The higher the deviation rate, the higher the risk of component failure. 604. Based on the fault mode library, predict the topology-guided data, directed network topology, frequency amplitude characteristics, and deviation rate characteristics to obtain the real-time predicted remaining lifetime. In this embodiment, the fault mode library integrates a massive number of fault cases and failure mechanisms of core components in the elevator industry, storing the correspondence between feature combinations and remaining lifespan. During prediction, multi-dimensional data is integrated into a "feature combination package," and the most matching entry (e.g., a matching degree of 92%) is retrieved from the library using a similarity algorithm (e.g., cosine similarity), and the real-time predicted remaining lifespan is output. If the matching degree is low (e.g., <70%, or if the feature combination does not completely cover the entries in the library), the core nodes in the topology guidance data (e.g., the traction machine is a core component) are combined for correction to ensure the reliability of the prediction results. In this embodiment, multi-stage sound data is first collected to construct the curve to be analyzed, transforming disordered data into visual features and capturing subtle anomalies in components in real time, laying the foundation for subsequent analysis. Then, through dynamic alignment, the curve to be analyzed is compared with the standard sound curve to accurately quantify the differences. Next, the frequency amplitude and deviation rate features are extracted by focusing on the difference frequency bands, intuitively reflecting component wear and failure risk. Finally, by combining the fault mode library and topology data, the remaining lifespan is predicted through multi-dimensional matching, and corrections can be made when the matching degree is low to ensure reliability. Overall, fault signs can be captured in advance, improving the accuracy of equipment lifespan prediction, helping to quickly locate faults, reducing operation and maintenance detection time and costs, promoting elevator operation and maintenance from passive response to proactive prevention, and ensuring the safe and stable operation of elevators.

[0023] Please see Figure 7 The seventh embodiment of a device life prediction method in this invention includes: 701. Obtain the predicted remaining lifetime from the previous month, and perform a deviation analysis on the real-time predicted remaining lifetime based on the predicted remaining lifetime from the previous month to obtain the deviation index. In this embodiment, the remaining life prediction results for the same elevator last month are first extracted from the system's historical database, and then the numerical difference between the two is compared with the current real-time prediction results. Next, the deviation level is determined by referring to the preset deviation level standards (e.g., difference ≤ 1 month is excellent, 1-2 months is qualified, and > 2 months is noteworthy). Finally, if the deviation exceeds the qualified range, possible causes such as changes in operating conditions and data acquisition interference are analyzed to form a deviation index. 702. Obtain the cumulative running time of the elevator and determine whether the cumulative running time has reached the preset fatigue test duration; In this embodiment, the duration of the elevator in normal passenger or freight operation is filtered from the elevator control system's operation log, and invalid durations such as standby, maintenance, and malfunction shutdown are excluded (e.g., if the elevator runs from 9:00 to 21:00 every day, 28 days a month, after 15 months, the cumulative running time = 12 × 28 × 15 = 5040 hours). 703. When the cumulative running time reaches the fatigue test duration, equipment life prediction information is generated based on the deviation index and the real-time predicted remaining life. In this embodiment, the fatigue test duration is set based on the elevator type (e.g., residential elevator, shopping mall elevator, medical elevator) and the characteristics of core components (e.g., traction machine, door system). Residential elevators operate for an average of 6 hours per day, with a fatigue test duration of 8000 hours; shopping mall elevators operate for an average of 12 hours per day, with a fatigue test duration of 6000 hours (fatigue accumulates faster under high-frequency operation, requiring a lower duration standard), ensuring that the duration standard aligns with the actual usage scenarios of the elevators. If the cumulative operating time is greater than or equal to the fatigue test duration, it indicates that the predicted data has covered the complete fatigue cycle and possesses statistical validity; if the standard is not met, the final prediction information is not generated temporarily, and the message "Data collection needs to continue until the cumulative duration reaches the standard; the current prediction is for reference only" is displayed. In this embodiment, during the deviation analysis stage, the prediction results of the same elevator from the previous month are extracted, compared with the real-time prediction difference, and the deviation level is determined according to a preset level. If the deviation exceeds the acceptable range, the cause is traced back to the working conditions, data collection, etc., effectively avoiding the random error of a single prediction and improving the stability and reliability of the prediction. In the cumulative running time judgment stage, the normal running time of the elevator is accurately screened and invalid time is excluded. In addition, the fatigue test time is set differently according to the elevator type and the characteristics of core components to ensure that the prediction is based on complete fatigue cycle data and avoids distortion caused by insufficient samples. When the cumulative time reaches the target, prediction information is generated, which not only provides clear time nodes and targeted suggestions for operation and maintenance, reducing ineffective operation and maintenance costs and the risk of sudden failures, but also adapts to the needs of different scenarios. It also lays the foundation for the industry to form a replicable standardized evaluation framework, comprehensively ensuring the accuracy, practicality and industry adaptability of equipment life prediction.

[0024] The above describes a method for predicting equipment lifespan according to an embodiment of the present invention. The following describes a device for predicting equipment lifespan according to an embodiment of the present invention. Please refer to [link / reference]. Figure 8 One embodiment of the device life prediction device of the present invention includes: The first analysis module is used to acquire elevator sound data and analyze the elevator sound data according to a preset time period and a preset timestamp to obtain the elevator operation stage. The first data acquisition module is used to collect elevator speed data according to the elevator operation stage and preset iterative stopping conditions; The second data acquisition module is used to acquire mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships. The second analysis module is used to analyze elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data. The prediction module is used to predict elevator sound data, topology guidance data and directed network topology based on a preset fault mode library in order to obtain real-time predicted remaining lifespan. The information generation module is used to generate equipment life prediction information based on real-time predicted remaining life. In this embodiment, elevator sound data is first acquired, and combined with time period, timestamp, and elevator sound data to accurately analyze the elevator operation stage, laying the scenario foundation for subsequent data collection. Then, elevator speed data is collected based on the elevator operation stage and iteration stopping conditions, selectively acquiring information for each stage to ensure complete data coverage and high effectiveness, avoiding blind collection. At the same time, three types of coupling relationships are comprehensively acquired, and combined with speed data analysis to form directed network topology and topology guidance data, quantifying subsystem interactions, clarifying fault paths and core nodes, and providing system logic support for prediction. Finally, based on the fault mode library, sound data and topology data are fused to predict real-time remaining lifespan and generate information, taking into account both real-time status and system logic, improving prediction accuracy. The overall solution not only solves the problems of low data effectiveness, lack of coupling consideration, and large prediction deviation in traditional methods, but also provides clear lifespan nodes and decision-making basis for operation and maintenance, helping to plan maintenance in advance, reducing the risk of sudden failures, and effectively ensuring the health management of the elevator throughout its entire life cycle.

[0025] Figure 9 This is a schematic diagram of the structure of a device lifetime prediction device 900 provided in an embodiment of the present invention. This device lifetime prediction device 900 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 910 (e.g., one or more processors) and a memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) for storing application programs 933 or data 932. The memory 920 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the device lifetime prediction device 900. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute a series of instruction operations on the storage media 930 on the device lifetime prediction device 900 to implement the steps of the device lifetime prediction method provided in the above-described method embodiments.

[0026] A device lifespan prediction device 900 may further include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating devices 931, such as Windows Server, MacOSX, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 9The illustrated device life prediction device structure does not constitute a limitation on a device life prediction device, and may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0027] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a device lifetime prediction method.

[0028] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the device or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0029] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0030] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting equipment lifespan, characterized in that, include: Acquire elevator sound data and analyze it according to a preset time period and a preset timestamp to determine the elevator operation stage; Elevator speed data is collected based on the elevator operation stage and preset iterative stopping conditions; Obtain mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships; The elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships are analyzed to obtain the directed network topology and topology guidance data. Based on a pre-defined fault mode library, elevator sound data, topology guidance data, and directed network topology are used to predict the remaining lifespan in real time. Equipment life prediction information is generated based on real-time prediction of remaining life.

2. The equipment life prediction method as described in claim 1, characterized in that, The process of analyzing elevator sound data based on a preset time period and a preset timestamp to determine the elevator operation stage includes: The phase marker signal is obtained based on the time period; The phase marker signal is converted according to the preset segmentation start point to obtain the phase reference signal; The elevator sound data is aligned based on the timestamp and phase reference signal to obtain aligned signal data; The alignment signal data is analyzed based on a preset sound fluctuation threshold to determine the elevator operation stage.

3. The equipment life prediction method as described in claim 1, characterized in that, The elevator speed data collected based on the elevator operation stage and preset iterative stopping conditions includes: When the elevator is in a stationary phase during its operation, the stationary phase speed data is collected based on the preset first continuous acquisition time and the preset first speed threshold. When the elevator operation phase is the elevator operation transition phase, the transition phase speed data is collected according to the preset second continuous acquisition time and the preset second speed threshold. When the elevator is in a stable operating phase, the stable phase speed data is collected according to the preset third continuous collection time and the preset speed threshold range. Determine whether the iterative stopping condition is met during the elevator's operation phase; If the iteration stop condition is not met during the elevator operation phase, the process returns to acquiring elevator sound data until the iteration stop condition is met during the elevator operation phase. Then, all types of speed data collected at the moment are acquired to obtain the elevator speed data.

4. The equipment life prediction method as described in claim 1, characterized in that, The analysis of elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships yields directed network topology and topology guidance data, including: Based on the preset fault evolution law, trend analysis is performed on the elevator speed data to obtain the attenuation factor and fault chain analysis path; The preset coupling type weights are updated based on the attenuation factor and the fault chain analysis path to obtain the updated type weights. The update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistics coupling relationships are analyzed to obtain directed network topology and topology guidance data.

5. The equipment life prediction method as described in claim 4, characterized in that, The analysis of update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology-oriented data includes: A directed network topology is constructed based on preset topology construction conditions, update type weights, mechanical coupling relationships, electrical coupling relationships, and fault statistics coupling relationships. Obtain the traction system evaluation loss index and determine whether the traction system evaluation loss index is greater than or equal to the preset loss index threshold. If the loss index of the traction system is greater than or equal to the loss index threshold, then structural analysis of the directed network topology is performed to obtain topology guidance data.

6. The equipment life prediction method as described in claim 1, characterized in that, The step of predicting elevator sound data, topology guidance data, and directed network topology based on a preset fault mode library to obtain a real-time predicted remaining lifespan includes: The curve to be analyzed is constructed based on the elevator sound data; The curve to be analyzed is mapped and compared with the preset sound standard curve to obtain the mapping normalization distance; Feature extraction is performed on elevator sound data based on the mapped normalized distance to obtain frequency amplitude features and deviation rate features; Based on the failure mode library, predictions are made using topology-guided data, directed network topology, frequency amplitude characteristics, and deviation rate characteristics to obtain real-time predicted remaining lifetime.

7. The equipment life prediction method as described in claim 1, characterized in that, The process of generating equipment lifespan prediction information based on real-time predicted remaining lifespan includes: Obtain the predicted remaining lifetime from the previous month, and perform a deviation analysis on the real-time predicted remaining lifetime based on the predicted remaining lifetime from the previous month to obtain the deviation index. Obtain the cumulative running time of the elevator and determine whether the cumulative running time has reached the preset fatigue test duration; When the cumulative running time reaches the fatigue test duration, equipment life prediction information is generated based on the deviation index and the real-time predicted remaining life.

8. A device for predicting equipment lifespan, characterized in that, include: The first analysis module is used to acquire elevator sound data and analyze the elevator sound data according to a preset time period and a preset timestamp to obtain the elevator operation stage. The first data acquisition module is used to collect elevator speed data according to the elevator operation stage and preset iterative stopping conditions; The second data acquisition module is used to acquire mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships. The second analysis module is used to analyze elevator speed data, mechanical coupling relationships, electrical coupling relationships, and fault statistical coupling relationships to obtain directed network topology and topology guidance data. The prediction module is used to predict elevator sound data, topology guidance data and directed network topology based on a preset fault mode library in order to obtain real-time predicted remaining lifespan. The information generation module is used to generate equipment life prediction information based on real-time predicted remaining life.

9. A device for predicting equipment lifespan, characterized in that, include: A memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the device lifetime prediction device to perform the steps of a device lifetime prediction method as claimed in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the various steps of the device lifetime prediction method as described in any one of claims 1-7.

Images