A method and apparatus for predicting the lifecycle of a slide module based on a servo system
By using servo drive to automatically identify slide parameters and employing sensorless diagnostic technology, the problems of incorrect parameter configuration, low detection efficiency, and inaccurate life prediction in servo slide maintenance have been solved. This enables automated quality inspection and accurate life prediction of slide modules, reducing maintenance costs and downtime risks.
Patent Information
- Application Number
- CN202611342032.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-31
- Publication Date
- 2026-11-10
AI Technical Summary
Existing servo slide maintenance solutions suffer from problems such as high parameter configuration error rate, low detection efficiency, low life prediction accuracy, inaccurate maintenance cycle, high hardware cost, and inaccurate fault diagnosis, which lead to positioning deviations, frequent downtime, and high maintenance costs.
A servo system-based slide module lifecycle prediction method is adopted. By automatically identifying slide parameters through servo drivers, automated quality inspection and multi-dimensional loss measurement are achieved. Combined with adaptive load correction and sensorless mechanical diagnosis, fault root cause location and automatic maintenance data update are realized.
It enables plug-and-play functionality, automated quality inspection, and accurate lifespan prediction for the slide module, reducing manual debugging and hardware procurement costs, improving equipment stability and maintenance efficiency, and minimizing downtime risks.
Smart Images

Figure CN123067338A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of predictive maintenance technology for the lifecycle of servo motion control slide modules, and relates to a method for predicting the lifecycle of a slide module based on a servo system and a slide module lifecycle prediction device using the servo system-based slide module lifecycle prediction method. Background Technology
[0002] The slide module is a core component of linear drive automation, consisting of linear guides, ball screws, servo motors, bearings, and a base. Its operational accuracy and stability directly determine the production quality and continuous production capacity of the automated production line. Due to long-term reciprocating motion, variable loads, continuous friction and vibration, and temperature changes, the slide is highly susceptible to failures such as guide wear, ball screw fatigue, bearing aging, widening transmission clearance, and loosening and breakage of the synchronous belt. Sudden failures and shutdowns will result in production line interruptions, material waste, and high downtime maintenance costs.
[0003] Existing servo slide maintenance solutions have six inherent technical defects:
[0004] 1. The mechanical parameters of the slide table are entirely configured manually item by item. The manual selection of parameters such as lead, stroke, reduction ratio, and motor direction is prone to errors, which can lead to accidents such as positioning errors, equipment collisions, and control parameter mismatches.
[0005] 2. There is no automated closed-loop detection algorithm for the smoothness of the slide table movement. The industry generally relies on external oscilloscopes to manually observe the torque waveform, which results in low detection efficiency, inconsistent judgment standards, and the inability to automate the batch quality inspection before leaving the factory.
[0006] 3. The equipment life measurement dimension is singular, only counting the time the driver is powered on and running, and cannot combine encoder feedback to reflect the actual mechanical wear of the lead screw and guide rail, resulting in extremely low life prediction accuracy.
[0007] 4. The overcurrent and overload protection of the driver uses a fixed threshold, which cannot adapt to the differences between new and old machines and light and heavy load workstations. Light load workstations are prone to false alarms, and hidden wear under heavy load cannot be identified in advance.
[0008] 5. The maintenance cycle and maintenance records rely entirely on manual operation. After maintenance is completed, the timer must be manually reset and the next maintenance node must be manually set. The maintenance cycle cannot be automatically iterated according to the actual wear and tear of the equipment.
[0009] 6. Mechanical health monitoring relies on external acceleration and tension sensors, which result in high costs for hardware procurement, wiring, and installation. It is also unsuitable for small installation spaces, and industrial dust and oil can easily damage external sensors, making mass deployment extremely cost-effective. Summary of the Invention
[0010] The purpose of this invention is to address the above-mentioned problems by providing a method for predicting the lifecycle of a slide module based on a servo system.
[0011] The purpose of this invention is to address the above-mentioned problems by providing a slide module lifecycle prediction device that employs a servo system-based slide module lifecycle prediction method.
[0012] To achieve the above objectives, the present invention adopts the following technical solutions:
[0013] A method for predicting the lifecycle of a slide module based on a servo system includes the following steps:
[0014] S1. Parameter self-identification: The servo driver reads the unique device identification code of the slide module upon power-up, retrieves the complete set of slide mechanical parameters bound to the identification code, and automatically completes parameter backfilling, eliminating the need for manual parameter configuration.
[0015] S2. Automated smoothness inspection: The servo driver's built-in state machine executes the reciprocating self-inspection motion of the slide module, collects the servo driver's electrical signals and compares them with the judgment threshold, automatically outputs the inspection results, and synchronously records the slide position corresponding to the fault when there is an abnormality.
[0016] S3. Multi-dimensional life cycle loss measurement: Parallel collection of three types of loss indicators: total running mileage of the slide table, cumulative running days, and total rotations of the servo drive. Each indicator is matched with an independent maintenance early warning threshold. If any indicator exceeds the limit, a graded maintenance early warning will be triggered.
[0017] S4. Adaptive load correction lifespan: The servo drive self-learns the dedicated steady-state load benchmark model of the slide, and classifies multiple abnormal load conditions based on the benchmark model. It then uses the mechanical loss coefficient to weighted correct the predicted value of the remaining lifespan of the slide.
[0018] S5. Sensorless Three-in-One Mechanical Deterioration Diagnosis: Reuses native servo current and torque feedback signals, without adding external sensors, and performs friction inference, vibration inference, and synchronous belt tension inference in parallel; when the smoothness inspection determines an abnormality, it links the three types of inference results to locate the root cause of mechanical failure and outputs maintenance suggestions;
[0019] S6, Maintenance Closed-Loop Iteration: All wear, diagnostic, and alarm data are stored in the servo drive's non-volatile storage medium; maintenance thresholds are automatically updated and maintenance logs are archived after maintenance is completed; factory initialization commands can clear all wear data and maintenance records with one click.
[0020] S7, Three-layer linkage control: A two-way data interaction channel between the servo driver and the host computer is established through the EtherCAT bus. The host computer integrates parameter configuration, quality inspection analysis, life visualization and fault log management functions to realize integrated control of slide table factory quality inspection, online monitoring and life cycle prediction.
[0021] In the above-mentioned method for predicting the life cycle of a slide module based on a servo system, in S1, the complete set of slide mechanical parameters includes product series, product model, lead, stroke, reduction ratio, motor mounting direction, special code, slide speed limit, return to zero direction, and production batch number.
[0022] The unique device identification code is a factory-preset independent production batch number identifier for the slide table. After the driver verifies that the identification code is valid, it reads all parameters in batches and completes automatic backfilling, thus realizing plug-and-play functionality for the slide table.
[0023] In the above-mentioned method for predicting the lifecycle of a slide module based on a servo system, step S2 includes the following steps:
[0024] A1. The servo driver adaptively matches the position, speed, acceleration / deceleration, and dwell time within a segment of the reciprocating motion according to the slide model, and also supports the upper computer to customize and modify the diagnostic motion parameters.
[0025] A2. Users can preset the upper and lower limits of current as diagnostic criteria. When the diagnostic mode is running, the host computer will display real-time current and position data in a pop-up window.
[0026] A3. The test mode will automatically exit after the diagnosis is completed, and the OK / NG quality inspection conclusion will be output by comparing the current data; in the NG state, the coordinate position of the slide table corresponding to the abnormal torque will be accurately recorded.
[0027] In the above-mentioned method for predicting the life cycle of a slide module based on a servo system, in S3, under the servo enabled state, the servo driver reads the position increment change in real time, calculates the total absolute running stroke of the slide by accumulating the absolute value, automatically calculates the next maintenance mileage according to the maintenance cycle, and displays the running mileage in real time on the upper computer intelligent diagnostic layer. When the running mileage > the next maintenance mileage, the servo driver triggers a mileage reached warning.
[0028] The servo drive continuously tracks and counts the total running time of the device. Based on the maintenance cycle, it automatically sets the next maintenance time. When the total time exceeds the next maintenance time, the servo drive triggers a timeout warning.
[0029] The total number of rotations of the servo drive is calculated by combining the encoder resolution, lead screw, and reduction ratio. Based on the rotation replacement cycle, the number of replacement rotations is automatically calculated. When the total number of rotations exceeds the number of replacement rotations, the servo drive triggers a rotation count warning.
[0030] In the above-mentioned method for predicting the lifecycle of a slide module based on a servo system, step S5 includes the following steps:
[0031] B1. Friction Fault Prediction: Collect independent torque-speed operation data of the slide table in both forward and reverse directions and fit characteristic curves to separate three types of friction parameters: static friction, Coulomb friction, and viscous friction; normalize all friction values to a percentage of rated torque and set three-level judgment thresholds: upper limit, reference, and lower limit; friction values higher than the upper limit indicate lubrication drying or excessive preload; friction values lower than the lower limit indicate excessive guide rail clearance or ball detachment; excessive difference in friction values between forward and reverse strokes indicates uneven wear of one side of the guide rail or assembly tilt; friction parameters can be identified online during operation at any point and in smoothness self-check conditions.
[0032] B2. Vibration Fault Prediction; Separate Acquisition of Dynamic and Static Dual-Condition Data: Split vibration data into two sets: static state when the servo drive is stopped and dynamic state when the servo drive is running. Static vibration reflects the rigidity of the base and the health of the bearings; running vibration reflects the aging of the lead screw and synchronous belt drive; Based on the FFT spectrum analysis of current high-frequency ripple, extract two major indicators: vibration frequency and vibration level, and set vibration level threshold; Capture high-frequency micro-amplitude vibrations that cannot be identified manually, and provide early warning of structural loosening, ball bearing aging, and coupling coaxiality deviation before abnormal noise and accuracy decay occur;
[0033] B3. Tension Inference: Establish a linear mapping model between the starting static friction torque and the physical tension of the synchronous belt, and convert the actual physical tension of the synchronous belt by the identified static friction percentage; adopt a two-level judgment mechanism: the percentage threshold judges the tension decay trend, and the absolute tension value judges the belt fatigue failure; when the static friction value drops to the preset threshold, the synchronous belt is judged to be loose; link the tension diagnosis results with multi-dimensional life measurement data, and synchronously output belt replacement reminders in advance to avoid tooth skipping, belt breakage and downtime failures.
[0034] In the above-mentioned method for predicting the lifecycle of a slide module based on a servo system, step S5 further includes the following steps:
[0035] Multi-diagnostic module linkage root cause determination: preset fault priority order: synchronous belt tension fault > single-sided guide rail friction wear fault > high-frequency vibration structure loosening fault; when the friction, vibration and tension modules trigger alarms at the same time, the fault with the highest priority is output as the only root cause, and corresponding operation and maintenance rectification suggestions are pushed at the same time.
[0036] In the above-mentioned method for predicting the life cycle of a slide module based on a servo system, in step S6, the non-volatile storage medium is the E2 storage area built into the servo driver; when the device is powered off or the software is reset, the real-time refreshed life parameters and diagnostic alarm data are automatically written to the E2 storage to ensure continuous data without loss; the system configures independent fault codes for mileage exceeding limits, time exceeding limits, number of laps exceeding limits, abnormal smoothness, and abnormal load, so as to realize graded and accurate alarms, and can store the 10 most recent historical warnings, which include the real-time display of the warning occurrence time, warning code, and mileage at the time of the warning;
[0037] After maintenance is completed, the system automatically iterates and updates the maintenance threshold for the next stage, archives and stores maintenance time and maintenance records; when it receives the factory initialization reset command, it clears the total running mileage, total motor revolutions, cumulative running days, all maintenance logs, and resets the initial maintenance threshold.
[0038] In the above-mentioned method for predicting the life cycle of a slide module based on a servo system, the bidirectional interaction logic of the EtherCAT bus in S7 is as follows: the servo driver periodically uploads real-time torque, current, position, life statistics, alarm status at all levels, and friction / vibration / tension characteristic parameters to the host computer; the host computer sends custom diagnostic test parameters, maintenance thresholds, equipment relearning instructions, and factory reset instructions to the driver.
[0039] The upper computer intelligent diagnostic layer integrates three interactive interfaces: a slide data configuration interface for automatic parameter configuration, a diagnostic test interface for automatic operation and real-time waveform display to judge smoothness, and a predictive maintenance interface for storing mileage, time, track and maintenance records, thus enabling digital management of the entire process from factory quality inspection, online monitoring and maintenance replacement.
[0040] A slide module lifecycle prediction device employing a servo system-based slide module lifecycle prediction method includes a bottom firmware layer, a bus interaction layer, and a host computer intelligent diagnostic layer.
[0041] The underlying firmware layer is deployed on the servo driver MCU and includes a built-in parameter self-identification module, a smoothness self-test state machine module, a multi-dimensional life measurement module, an adaptive current self-learning module, a sensorless three-in-one mechanical diagnostic module, and an E2 storage maintenance iteration module.
[0042] The bus interaction layer uses the EtherCAT real-time bus.
[0043] The host computer intelligent diagnostic layer integrates a parameter visualization configuration interface, an automatic quality inspection waveform analysis interface, a maintenance data visualization interface, and a fault log management interface.
[0044] In the aforementioned slide module life cycle prediction device based on a servo system-based slide module life cycle prediction method, the sensorless three-in-one mechanical diagnostic module includes a friction identification submodule, an FFT vibration analysis submodule, and a synchronous belt tension mapping submodule. The three submodules share the original data of servo driver current and torque for parallel processing, and the output data is uniformly pushed to the smoothness quality inspection module and the life measurement module for linkage analysis.
[0045] Compared with existing technologies, the advantages of this invention are:
[0046] 1. The method automatically reads all mechanical parameters through the unique identification code of the slide table, eliminating the need for manual configuration; it automatically completes smoothness detection and locates abnormal points based on the built-in self-test process of the servo drive; it collects multiple types of loss indicators in parallel to achieve graded maintenance early warning; it adaptively corrects the remaining lifespan through the equipment's dedicated load benchmark; it reuses the servo's native electrical signals to achieve sensorless multi-dimensional mechanical degradation diagnosis and link to locate the root cause of the fault; it relies on the drive's storage to achieve automatic updates of maintenance data and factory initialization; and it uses the EtherCAT bus to connect the drive and the host computer to achieve full-process management and control. This integrated solution does not require additional sensor hardware, unifies the factory quality inspection judgment standards, distinguishes between instantaneous impact and long-term mechanical wear, and achieves early fault tracing and prediction, forming a digital closed loop from factory inspection, online status monitoring to life cycle maintenance, significantly reducing the comprehensive operation and maintenance costs caused by manual debugging, hardware procurement, and sudden downtime.
[0047] 2. The identification code binding parameter range and verification reading logic of the parameter self-identification process are limited. The factory batch number of the slide table is used as a unique identification identifier. After the verification is passed, all relevant control parameters of the whole machine are read in batches and automatically filled back. This structure solves the problem of easy error in manually entering all parameters of the traditional slide table. Different specifications of slide tables can be adapted as soon as they are powered on, realizing plug and play of the equipment. It saves the debugging time of manually checking and modifying parameters, and avoids positioning deviation and collision failure caused by parameter mismatch from the source.
[0048] 3. The entire process logic of automated quality inspection for smoothness is refined. The servo driver can adaptively match the entire set of self-inspection motion parameters according to the model, and also supports the upper computer to customize the detection conditions. The running status data is displayed in real time during the detection process. After the detection is completed, standardized quality inspection conclusions are automatically output and the location of abnormality is accurately marked. This solution solves the problems of low detection efficiency and inconsistent judgment standards caused by relying on external oscilloscopes to manually observe waveforms. It is suitable for batch automated quality inspection scenarios, which facilitates the quick location of the fault area corresponding to the mechanical jamming of the slide table and simplifies the process of checking defective products before leaving the factory.
[0049] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description
[0050] Figure 1 This is a schematic diagram of the overall three-layer system architecture of the present invention;
[0051] Figure 2 This is a schematic diagram of the fully automated smoothness diagnostic test execution process of the present invention;
[0052] Figure 3 This is a block diagram of the multi-dimensional life cycle loss measurement logic for the slide table of the present invention;
[0053] Figure 4 This is a flowchart of the adaptive load grading health early warning determination process of the present invention;
[0054] Figure 5 This is a block diagram of the sensorless friction, vibration, and tension integrated mechanical deterioration diagnosis module of the present invention. Detailed Implementation
[0055] like Figures 1-5 As shown, a method for predicting the lifecycle of a slide module based on a servo system includes the following steps:
[0056] S1. Parameter self-identification: The servo driver reads the unique device identification code of the slide module upon power-up, retrieves the complete set of slide mechanical parameters bound to the identification code, and automatically completes parameter backfilling, eliminating the need for manual parameter configuration.
[0057] S2. Automated smoothness inspection: The servo driver's built-in state machine executes the reciprocating self-inspection motion of the slide module, collects the servo driver's electrical signals and compares them with the judgment threshold, automatically outputs the inspection results, and synchronously records the slide position corresponding to the fault when there is an abnormality.
[0058] S3. Multi-dimensional life cycle loss measurement: Parallel collection of three types of loss indicators: total running mileage of the slide table, cumulative running days, and total rotations of the servo drive. Each indicator is matched with an independent maintenance early warning threshold. If any indicator exceeds the limit, a graded maintenance early warning will be triggered.
[0059] S4. Adaptive load correction lifespan: The servo drive self-learns the dedicated steady-state load benchmark model of the slide, and classifies multiple abnormal load conditions based on the benchmark model. It then uses the mechanical loss coefficient to weighted correct the predicted value of the remaining lifespan of the slide.
[0060] S5. Sensorless Three-in-One Mechanical Deterioration Diagnosis: Reuses native servo current and torque feedback signals, without adding external sensors, and performs friction inference, vibration inference, and synchronous belt tension inference in parallel; when the smoothness inspection determines an abnormality, it links the three types of inference results to locate the root cause of mechanical failure and outputs maintenance suggestions;
[0061] S6, Maintenance Closed-Loop Iteration: All wear, diagnostic, and alarm data are stored in the servo drive's non-volatile storage medium; maintenance thresholds are automatically updated and maintenance logs are archived after maintenance is completed; factory initialization commands can clear all wear data and maintenance records with one click.
[0062] S7, Three-layer linkage control: A two-way data interaction channel between the servo driver and the host computer is established through the EtherCAT bus. The host computer integrates parameter configuration, quality inspection analysis, life visualization and fault log management functions to realize integrated control of slide table factory quality inspection, online monitoring and life cycle prediction.
[0063] In this invention, the method automatically reads all mechanical parameters through the unique identification code of the slide table, eliminating the need for manual configuration; it automatically completes smoothness detection and locates abnormal points based on the built-in self-test process of the servo driver; it collects multiple types of loss indicators in parallel to achieve graded maintenance early warning; it adaptively corrects the remaining lifespan through the equipment's dedicated load benchmark; it reuses the servo's native electrical signals to achieve sensorless multi-dimensional mechanical degradation diagnosis and link to locate the root cause of the fault; it relies on the driver's storage to achieve automatic updates of maintenance data and factory initialization; and it uses the EtherCAT bus to connect the driver and the host computer to achieve full-process management and control. This integrated solution does not require additional sensor hardware, unifies the factory quality inspection judgment standards, distinguishes between instantaneous impact and long-term mechanical wear, and achieves early fault tracing and prediction, forming a digital closed loop from factory inspection, online status monitoring to life-cycle maintenance, significantly reducing the comprehensive operation and maintenance costs caused by manual debugging, hardware procurement, and sudden downtime.
[0064] Specifically, in S1, the complete set of slide table mechanical parameters includes product series, product model, lead, stroke, reduction ratio, motor mounting direction, special code, slide table speed limit, zero return direction, and production batch number;
[0065] The unique device identification code is a factory-preset independent production batch number identifier for the slide table. After the driver verifies the validity of the identification code, it reads all parameters in batches and automatically fills them in, thus achieving plug-and-play functionality for the slide table. The identification code binding parameter range and verification reading logic in the parameter self-identification process are limited. The slide table's factory batch number serves as the unique identification identifier. After successful verification, all relevant control parameters of the entire machine are read in batches and automatically filled in. This structure solves the problem of errors caused by manual entry of all parameters for traditional slide tables. Different specifications of slide tables can be adapted upon power-on, achieving plug-and-play functionality and saving the debugging time of manual parameter verification and modification. It avoids positioning deviations and collision failures caused by parameter mismatches from the source.
[0066] Specifically, S2 includes the following steps:
[0067] A1. The servo driver adaptively matches the position, speed, acceleration / deceleration, and dwell time within a segment of the reciprocating motion according to the slide model, and also supports the upper computer to customize and modify the diagnostic motion parameters.
[0068] A2. Users can preset the upper and lower limits of current as diagnostic criteria. When the diagnostic mode is running, the host computer will display real-time current and position data in a pop-up window.
[0069] A3. Upon completion of the diagnostic process, the system automatically exits the test mode and outputs an OK / NG quality inspection conclusion based on the compared current data. In the NG state, the system accurately records the coordinate position of the slide table corresponding to the torque abnormality. The system refines the logic of the entire automated quality inspection process for smoothness. The servo driver can adaptively match the entire set of self-test motion parameters according to the model, and also supports the upper computer to customize the test conditions. The system displays the running status data in real time during the test, and automatically outputs standardized quality inspection conclusions and accurately marks the location of the abnormality upon completion of the test. This solution solves the problems of low detection efficiency and inconsistent judgment standards caused by the traditional reliance on external oscilloscopes for manual waveform observation. It is suitable for batch automated quality inspection scenarios, facilitates the rapid location of the fault area corresponding to the mechanical jamming of the slide table, and simplifies the process of troubleshooting defective products.
[0070] Specifically, in S3, when the servo is enabled, the servo driver reads the position increment change in real time, calculates the total absolute travel of the slide by accumulating the absolute values, and automatically calculates the next maintenance mileage according to the maintenance cycle. The host computer intelligent diagnostic layer displays the running mileage in real time. When the running mileage > the next maintenance mileage, the servo driver triggers a mileage reached warning.
[0071] The servo drive continuously tracks and counts the total running time of the device. Based on the maintenance cycle, it automatically sets the next maintenance time. When the total time exceeds the next maintenance time, the servo drive triggers a timeout warning.
[0072] By combining encoder resolution, lead screw lead, and reduction ratio, the total number of rotations of the servo drive is converted from the running mileage. Based on the rotation replacement cycle, the number of replacement rotations is automatically calculated. When the total number of rotations exceeds the number of replacement rotations, the servo drive triggers a rotation count warning. The implementation logic limits the independent collection and early warning of three types of wear indicators. These three types of wear data are obtained based on position increment, system timing, and mechanical parameter conversion, respectively corresponding to different maintenance scenarios: lubrication maintenance, overall machine maintenance, and transmission component replacement. This solution abandons the single lifespan measurement mode that only counts power-on time. It quantifies the degree of equipment wear from the perspective of actual mechanical wear, and pushes corresponding maintenance reminders according to different scenarios. This avoids the deviation in maintenance timing caused by relying solely on power-on time to predict lifespan, and proactively avoids the risk of fatigue failure and downtime of transmission components.
[0073] Specifically, step S4 is implemented as follows:
[0074] The servo drive continuously samples and learns under standard no-load / rated steady-state load conditions, automatically filtering out interference data from sudden acceleration / deceleration, instantaneous impact, and sudden load changes, and establishing a dedicated reference load current model for the device. After major equipment overhaul or replacement of transmission components, the reference model can be updated again by receiving a relearning command from the host computer.
[0075] Based on the reference current, three levels of abnormal operating conditions are divided: instantaneous overload, continuous heavy load, and low load. The duration of each abnormal condition is statistically analyzed. The static life prediction duration is corrected by weighting the mechanical loss coefficients corresponding to different operating conditions.
[0076] Specifically, S5 includes the following steps:
[0077] B1. Friction Fault Prediction: Collect independent torque-speed operation data of the slide table in both forward and reverse directions and fit characteristic curves to separate three types of friction parameters: static friction, Coulomb friction, and viscous friction; normalize all friction values to a percentage of rated torque and set three-level judgment thresholds: upper limit, reference, and lower limit; friction values higher than the upper limit indicate lubrication drying or excessive preload; friction values lower than the lower limit indicate excessive guide rail clearance or ball detachment; excessive difference in friction values between forward and reverse strokes indicates uneven wear of one side of the guide rail or assembly tilt; friction parameters can be identified online during operation at any point and in smoothness self-check conditions.
[0078] B2. Vibration Fault Prediction; Separate Acquisition of Dynamic and Static Dual-Condition Data: Split vibration data into two sets: static state when the servo drive is stopped and dynamic state when the servo drive is running. Static vibration reflects the rigidity of the base and the health of the bearings; running vibration reflects the aging of the lead screw and synchronous belt drive; Based on the FFT spectrum analysis of current high-frequency ripple, extract two major indicators: vibration frequency and vibration level, and set vibration level threshold; Capture high-frequency micro-amplitude vibrations that cannot be identified manually, and provide early warning of structural loosening, ball bearing aging, and coupling coaxiality deviation before abnormal noise and accuracy decay occur;
[0079] B3. Tension Inference: A linear mapping model between the static friction torque and the physical tension of the synchronous belt is established. The actual physical tension of the synchronous belt is calculated by converting the identified static friction percentage. A two-level judgment mechanism is adopted: the percentage threshold judges the tension decay trend, and the absolute tension value judges belt fatigue failure. When the static friction value drops to a preset threshold, the synchronous belt is judged to be loose. The tension diagnosis results are linked with multi-dimensional life measurement data to synchronously output belt replacement reminders in advance, avoiding tooth skipping, belt breakage and downtime. The complete implementation logic of three types of sensorless diagnosis of friction, vibration and tension is refined. Multiple friction parameters are separated by forward and reverse stroke curves and wear faults are judged by multi-level thresholds. The current spectrum is analyzed by splitting dynamic and static working conditions to identify aging defects of the base and transmission. The mapping relationship between static friction and synchronous belt tension is established to judge belt loosening failure, and the diagnosis results can be linked with the life measurement system for early warning. This structure fully reuses the servo native feedback signal, without the need to install additional acceleration and tension sensors, reducing hardware wiring and procurement costs. It can identify early mechanical deterioration that is difficult to detect manually online and distinguish different types of hidden faults such as lubrication, clearance and transmission belt.
[0080] Specifically, S5 also includes the following steps:
[0081] Multi-diagnostic module linkage root cause determination: A preset fault priority order is established: synchronous belt tension fault > unilateral guide rail friction wear fault > high-frequency vibration structural loosening fault. When the friction, vibration, and tension modules trigger alarms simultaneously, the fault with the highest priority is output as the sole root cause, and corresponding maintenance and rectification suggestions are pushed simultaneously. A fault priority determination rule is added for simultaneous alarms from multiple diagnostic modules. When multiple mechanical anomalies trigger alarms simultaneously, a single fault root cause is output according to the preset priority, and a corresponding maintenance solution is pushed. This design solves the problem of chaotic alarm information and the inability of maintenance personnel to quickly pinpoint the core fault when multiple faults occur concurrently, simplifies the fault investigation sequence, shortens production line downtime for handling faults, and improves equipment maintenance efficiency.
[0082] Specifically, in S6, the non-volatile storage medium is the E2 storage area built into the servo driver; when the device is powered off or the software is reset, the real-time refreshed life parameters and diagnostic alarm data are automatically written to the E2 storage to ensure continuous data without loss; the system configures independent fault codes for mileage exceeding limits, time exceeding limits, lap count exceeding limits, abnormal smoothness, and abnormal load, so as to realize hierarchical and accurate alarms, and can store the 10 most recent historical warnings (maintenance records). The historical warnings include information such as the real-time display of the warning occurrence time, warning code, and mileage at the time of the warning;
[0083] After maintenance is completed, the system automatically iterates and updates the maintenance thresholds for the next stage, archives and stores maintenance time and records. Upon receiving a factory initialization reset command, it clears the total mileage, total motor revolutions, cumulative operating days, and all maintenance logs, and resets the initial maintenance thresholds. The limited driver's built-in E2 storage medium ensures the persistence of all maintenance data; data is not lost after power failure or reset. Different anomaly types correspond to independent fault codes for tiered alarms. Maintenance thresholds are automatically updated and records are archived upon completion of maintenance, and one-click clearing of all operating and maintenance data is supported when the entire machine leaves the factory. This solution solves the problems of traditional manual registration of maintenance records, manual modification of maintenance cycles, and the need to clear data line by line upon equipment reset, ensuring complete and traceable operational data throughout the entire lifecycle, while also meeting the needs of both factory initialization and routine on-site maintenance.
[0084] Specifically, in S7, the bidirectional interaction logic of the EtherCAT bus is as follows: the servo driver periodically uploads real-time torque, current, position, life statistics, alarm status at all levels, and friction / vibration / tension characteristic parameters to the host computer; the host computer sends custom diagnostic test parameters, maintenance thresholds, equipment relearning instructions, and factory reset instructions to the driver.
[0085] The upper-level intelligent diagnostic layer integrates three interactive interfaces: a slide table data configuration interface for automatic parameter configuration, a diagnostic testing interface for automatic operation and real-time waveform display to judge smoothness, and a predictive maintenance interface for storing mileage, time, track width, and maintenance records. This establishes a complete digital management system for the entire process, from factory inspection and online monitoring to maintenance and replacement. The system refines the bidirectional interaction content of the EtherCAT bus and integrates the interface functions with the upper-level computer. The driver synchronously uploads full-dimensional characteristic data of operation, diagnosis, and alarms to the upper-level computer, while the upper-level computer issues various control, calibration, and reset commands. The upper-level computer integrates functions such as slide table data configuration, diagnostic testing, and predictive maintenance. This architecture establishes a complete digital link for the slide table's factory inspection, online monitoring, maintenance, and replacement process, enabling linkage between the underlying control algorithm and the upper-level visual operation and maintenance platform. This allows maintenance personnel to intuitively view the equipment's health status and issue batch testing and maintenance configuration commands.
[0086] like Figures 1-5 As shown, a slide module lifecycle prediction device using a servo system-based slide module lifecycle prediction method includes a bottom firmware layer, a bus interaction layer, and a host computer intelligent diagnostic layer.
[0087] The underlying firmware layer is deployed on the servo driver MCU and includes a built-in parameter self-identification module, a smoothness self-test state machine module, a multi-dimensional life measurement module, an adaptive current self-learning module, a sensorless three-in-one mechanical diagnostic module, and an E2 storage maintenance iteration module.
[0088] The bus interaction layer uses the EtherCAT real-time bus.
[0089] The host computer intelligent diagnostic layer integrates a parameter visualization configuration interface, an automatic quality inspection waveform analysis interface, a maintenance data visualization interface, and a fault log management interface. It is a three-layer architecture device that limits the execution of the aforementioned prediction method. The bottom-layer driver MCU integrates all self-developed algorithm modules, the middle layer uses a standard EtherCAT bus for data interaction, and the host computer is equipped with a matching visualization management interface. This device lightweightly deploys all detection, lifespan calculation, and fault diagnosis logic onto the existing servo driver hardware, eliminating the need for additional edge control devices. It is directly compatible with existing automated production line bus architectures, has a low modification and adaptation threshold, and is easy to deploy in large quantities.
[0090] Specifically, the sensorless three-in-one mechanical diagnostic module includes a friction identification submodule, an FFT vibration analysis submodule, and a synchronous belt tension mapping submodule. These three submodules share the same raw current and torque acquisition data from the servo driver, performing parallel processing. The output data is then uniformly pushed to the smoothness inspection module and the life measurement module for joint analysis. The three-in-one mechanical diagnostic module within the device is further refined, with the friction, vibration, and tension submodules sharing the same set of raw motor acquisition data for parallel processing. The diagnostic output data is synchronously pushed to the smoothness detection and life measurement modules for joint analysis. This module architecture reduces the resource consumption of repeatedly acquiring electrical signals for multiple diagnostic types, lowers the computing power consumption of the driver, and enables multi-dimensional data collaborative analysis of smoothness anomalies, life loss, and mechanical degradation, improving the completeness and accuracy of the overall machine health status assessment.
[0091] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.
Claims
1. A method for predicting the lifecycle of a slide module based on a servo system, characterized in that, Includes the following steps: S1. Parameter self-identification: The servo driver reads the unique device identification code of the slide module upon power-up, retrieves the complete set of slide mechanical parameters bound to the identification code, and automatically completes parameter backfilling, eliminating the need for manual parameter configuration. S2. Automated smoothness inspection: The servo driver's built-in state machine executes the reciprocating self-inspection motion of the slide module, collects the servo driver's electrical signals and compares them with the judgment threshold, automatically outputs the inspection results, and synchronously records the slide position corresponding to the fault when there is an abnormality. S3. Multi-dimensional life cycle loss measurement: Parallel collection of three types of loss indicators: total running mileage of the slide table, cumulative running days, and total rotations of the servo drive. Each indicator is matched with an independent maintenance early warning threshold. If any indicator exceeds the limit, a graded maintenance early warning will be triggered. S4. Adaptive load correction lifespan: The servo drive self-learns the dedicated steady-state load benchmark model of the slide, and classifies multiple abnormal load conditions based on the benchmark model. It then uses the mechanical loss coefficient to weighted correct the predicted value of the remaining lifespan of the slide. S5. Sensorless Three-in-One Mechanical Deterioration Diagnosis: Reuses native servo current and torque feedback signals, without adding external sensors, and performs friction inference, vibration inference, and synchronous belt tension inference in parallel; when the smoothness inspection determines an abnormality, it links the three types of inference results to locate the root cause of mechanical failure and outputs maintenance suggestions; S6, Maintenance Closed-Loop Iteration: All wear, diagnostic, and alarm data are stored in the servo drive's non-volatile storage medium; maintenance thresholds are automatically updated and maintenance logs are archived after maintenance is completed; factory initialization commands can clear all wear data and maintenance records with one click. S7, Three-layer linkage control: A two-way data interaction channel between the servo driver and the host computer is established through the EtherCAT bus. The host computer integrates parameter configuration, quality inspection analysis, life visualization and fault log management functions to realize integrated control of slide table factory quality inspection, online monitoring and life cycle prediction.
2. The method for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, In S1, the complete set of slide table mechanical parameters includes product series, product model, lead, stroke, reduction ratio, motor mounting direction, special code, slide table speed limit, zero return direction, and production batch number; The unique device identification code is a factory-preset independent production batch number identifier for the slide table. After the driver verifies that the identification code is valid, it reads all parameters in batches and completes automatic backfilling, thus realizing plug-and-play functionality for the slide table.
3. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, S2 includes the following steps: A1. The servo driver adaptively matches the position, speed, acceleration / deceleration, and dwell time within a segment of the reciprocating motion according to the slide model, and also supports the upper computer to customize and modify the diagnostic motion parameters. A2. Users can preset the upper and lower limits of current as diagnostic criteria. When the diagnostic mode is running, the host computer will display real-time current and position data in a pop-up window. A3. The test mode will automatically exit after the diagnosis is completed, and the OK / NG quality inspection conclusion will be output by comparing the current data; in the NG state, the coordinate position of the slide table corresponding to the abnormal torque will be accurately recorded.
4. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, In S3, when the servo is enabled, the servo driver reads the position increment change in real time, calculates the total absolute travel of the slide by accumulating the absolute values, and automatically calculates the next maintenance mileage according to the maintenance cycle. The host computer intelligent diagnostic layer displays the running mileage in real time. When the running mileage > the next maintenance mileage, the servo driver triggers a mileage reached warning. The servo drive continuously tracks and counts the total running time of the device. Based on the maintenance cycle, it automatically sets the next maintenance time. When the total time exceeds the next maintenance time, the servo drive triggers a timeout warning. The total number of rotations of the servo drive is calculated by combining the encoder resolution, lead screw, and reduction ratio. Based on the rotation replacement cycle, the number of replacement rotations is automatically calculated. When the total number of rotations exceeds the number of replacement rotations, the servo drive triggers a rotation count warning.
5. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, In S5, the following steps are included: B1. Friction Fault Prediction: Collect independent torque-speed operation data of the slide table in both forward and reverse directions and fit characteristic curves to separate three types of friction parameters: static friction, Coulomb friction, and viscous friction; normalize all friction values to a percentage of rated torque and set three-level judgment thresholds: upper limit, reference, and lower limit; friction values higher than the upper limit indicate lubrication drying or excessive preload; friction values lower than the lower limit indicate excessive guide rail clearance or ball detachment; excessive difference in friction values between forward and reverse strokes indicates uneven wear of one side of the guide rail or assembly tilt; friction parameters can be identified online during operation at any point and in smoothness self-check conditions. B2. Vibration Fault Prediction; Separate Acquisition of Dynamic and Static Dual-Condition Data: Split vibration data into two sets: static state when the servo drive is stopped and dynamic state when the servo drive is running. Static vibration reflects the rigidity of the base and the health of the bearings; running vibration reflects the aging of the lead screw and synchronous belt drive; Based on the FFT spectrum analysis of current high-frequency ripple, extract two major indicators: vibration frequency and vibration level, and set vibration level threshold; Capture high-frequency micro-amplitude vibrations that cannot be identified manually, and provide early warning of structural loosening, ball bearing aging, and coupling coaxiality deviation before abnormal noise and accuracy decay occur; B3. Tension Inference: Establish a linear mapping model between the starting static friction torque and the physical tension of the synchronous belt, and convert the actual physical tension of the synchronous belt by the identified static friction percentage; adopt a two-level judgment mechanism: the percentage threshold judges the tension decay trend, and the absolute tension value judges the belt fatigue failure; when the static friction value drops to the preset threshold, the synchronous belt is judged to be loose; link the tension diagnosis results with multi-dimensional life measurement data, and synchronously output belt replacement reminders in advance to avoid tooth skipping, belt breakage and downtime failures.
6. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 5, characterized in that, S5 also includes the following steps: Multi-diagnostic module linkage root cause determination: preset fault priority order: synchronous belt tension fault > single-sided guide rail friction wear fault > high-frequency vibration structure loosening fault; when the friction, vibration and tension modules trigger alarms at the same time, the fault with the highest priority is output as the only root cause, and corresponding operation and maintenance rectification suggestions are pushed at the same time.
7. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, In S6, the non-volatile storage medium is the E2 storage area built into the servo driver; when the device is powered off or the software is reset, the real-time refreshed life parameters and diagnostic alarm data are automatically written to the E2 storage to ensure continuous data without loss; the system configures independent fault codes for mileage exceeding limits, time exceeding limits, lap count exceeding limits, abnormal smoothness, and abnormal load, so as to realize graded and accurate alarms, and can store the 10 most recent historical warnings, which include the real-time display of the warning occurrence time, warning code, and mileage at the time of the warning; After maintenance is completed, the system automatically iterates and updates the maintenance threshold for the next stage, archives and stores maintenance time and maintenance records; when it receives the factory initialization reset command, it clears the total running mileage, total motor revolutions, cumulative running days, all maintenance logs, and resets the initial maintenance threshold.
8. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 1, characterized in that, In S7, the bidirectional interaction logic of the EtherCAT bus is as follows: the servo driver periodically uploads real-time torque, current, position, life statistics, alarm status at all levels, and friction / vibration / tension characteristic parameters to the host computer; the host computer sends custom diagnostic test parameters, maintenance thresholds, equipment relearning instructions, and factory reset instructions to the driver. The upper computer intelligent diagnostic layer integrates three interactive interfaces: a slide data configuration interface for automatic parameter configuration, a diagnostic test interface for automatic operation and real-time waveform display to judge smoothness, and a predictive maintenance interface for storing mileage, time, track and maintenance records, thus enabling digital management of the entire process from factory quality inspection, online monitoring and maintenance replacement.
9. A slide module lifecycle prediction device employing the slide module lifecycle prediction method based on a servo system as described in any one of claims 1-8, characterized in that, This includes the underlying firmware layer, the bus interaction layer, and the host computer intelligent diagnostic layer. The underlying firmware layer is deployed on the servo driver MCU and includes a built-in parameter self-identification module, a smoothness self-test state machine module, a multi-dimensional life measurement module, an adaptive current self-learning module, a sensorless three-in-one mechanical diagnostic module, and an E2 storage maintenance iteration module. The bus interaction layer uses the EtherCAT real-time bus. The host computer intelligent diagnostic layer integrates a parameter visualization configuration interface, an automatic quality inspection waveform analysis interface, a maintenance data visualization interface, and a fault log management interface.
10. The method and apparatus for predicting the lifecycle of a slide module based on a servo system according to claim 9, characterized in that, The sensorless three-in-one mechanical diagnostic module includes a friction identification submodule, an FFT vibration analysis submodule, and a synchronous belt tension mapping submodule. The three submodules share the raw data of servo driver current and torque for parallel processing, and the output data is uniformly pushed to the smoothness quality inspection module and the life measurement module for joint analysis.
