A 3D printer drag chain line fault prediction method and system

Abstract

The embodiment of the application provides a 3D printer drag chain line fault prediction method and system, relates to the 3D printer drag chain line fault prediction field, and the method comprises the following steps: acquiring the downtime length of a 3D printer; when the downtime length is greater than a preset stress relaxation time threshold, identifying that a long interval downtime event exists; in response to the long interval downtime event, acquiring state data of the drag chain line; according to the current state data and the downtime state data, acquiring a historical risk degree index at the time of downtime and a current risk degree index at the current moment; calculating the difference between the current risk degree index and the historical risk degree index; when the apparent state of the 3D printer drag chain line appears abnormal rollback, according to the working parameter before downtime and the downtime length, the recoverable risk degree value is calculated; the irreversible damage base index is acquired; according to the irreversible damage base index and the current risk degree index, the double-branch prediction drag chain line fault is adopted. The application can improve the accuracy and reliability of fault prediction.

Description

Technical Field

[0001] This application relates to the field of 3D printer drag chain line fault prediction, and more specifically, to a method and system for 3D printer drag chain line fault prediction. Background Technology

[0002] In the field of 3D printer cable conveyor failure prediction, to ensure continuous and stable operation of the equipment, existing technologies typically perform a self-test procedure before each printing job begins. This involves monitoring the cable conveyor's condition during low-speed trial operation and comparing it with historical trend data to assess the current health of the cable conveyor and predict its remaining lifespan. If the assessed value is within a safe threshold, the system allows a new printing job to be executed. This conventional prediction process based on historical trend extrapolation can effectively capture the monotonic cumulative characteristics of wear and tear under continuous operating conditions.

[0003] However, in mass production or studio applications, printers are often not operating continuously. After completing a full-sheet model with high-intensity printing, the equipment typically experiences downtime ranging from several hours to several days. In such long-interval downtime scenarios, after experiencing the high temperatures and micro-cracks generated by the high-frequency bending in the early stages, the polymer material of the cable chain sheath undergoes sufficient stress relaxation and elastic recovery during the downtime. This allows some micro-cracks to close and causes macroscopic parameters such as overall bending stiffness to appear to recover to levels better than before the downtime.

[0004] When the system performs its initial self-check after a shutdown and connects the current health assessment results with those before the shutdown, this apparent state regression can severely contradict the accumulation of irreversible internal fatigue damage. Because existing trend-capturing logic is typically based on the assumption that wear increases monotonically with usage frequency, it's difficult to distinguish between the recoverable components caused by the elastic recovery of the sheath and the residual internal damage. Therefore, it often tends to directly fit the current assessment value into the long-term wear trend model. This results in a downward pseudo-improvement inflection point on the state evolution curve, smoothing out and lowering the overall wear accumulation rate. After long-interval shutdowns during batch printing tasks, the inability to distinguish between the apparent state improvement caused by the elastic recovery of the cable conveyor sheath and the continued accumulation of irreversible internal damage leads to a pseudo-improvement inflection point in the state evolution trend. This causes the fault prediction model to underestimate the long-term wear rate, ultimately resulting in a delayed triggering of fault warnings during the next high-intensity task.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application discloses a method and system for predicting faults in a 3D printer cable conveyor system. The aim is to solve the problem in the prior art where, after long-interval downtime, the apparent improvement caused by the elastic recovery of the cable conveyor sheath is not distinguished from the continuous accumulation of irreversible internal damage, leading to the introduction of a pseudo-improvement inflection point in the state evolution trend. This causes the fault prediction model to underestimate the long-term wear rate, ultimately resulting in a delayed triggering of fault warnings in the next high-intensity task.

[0007] In a first aspect, this application discloses a method for predicting faults in a 3D printer's cable conveyor belt, the method comprising: Get the downtime of the 3D printer, which is the time interval between the current downtime and the last downtime. When the downtime exceeds a preset stress relaxation time threshold, a long-interval downtime event is identified. In response to the long-interval shutdown event, the status data of the 3D printer cable chain is acquired. The status data includes current status data and shutdown status data. The current status data includes current current data, current vibration data and current ambient temperature. The shutdown status data includes shutdown current data, shutdown vibration data and shutdown ambient temperature. Based on the current status data and shutdown status data, the risk index prediction model is used to obtain the historical risk index at the time of shutdown and the current risk index at the current moment. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer cable conveyor. The risk index is used to characterize the magnitude of the cable conveyor failure. Calculate the difference between the current risk level indicator and the historical risk level indicator; When the difference is negative and the absolute value exceeds the preset risk threshold, it is determined that the 3D printer's drag chain has an apparent state regression abnormality. In response to the apparent state regression anomaly, the recoverability risk value is calculated based on the working parameters before the shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature and total cumulative bending count. Add the current risk level index to the recoverable risk level value to obtain the modified irreversible damage baseline index; Based on the modified irreversible damage baseline index and the current risk index, a two-branch prediction method is used to predict 3D printer drag chain failures. The two branches include a long-term damage trend branch and an immediate safety determination branch.

[0008] Furthermore, according to the above method, the steps for predicting 3D printer drag chain line failures using a dual-branch approach, based on the modified irreversible damage substrate index and the current risk level index, include: In the long-term damage trend branch, the degradation rate and remaining lifetime are calculated based on the modified irreversible damage base index to predict 3D printer drag chain failures. In the immediate safety assessment branch, based on the current risk level index and the expected wear increment of the current task, it is predicted whether the current printing task can be started, in order to predict 3D printer drag chain failure.

[0009] More specifically, in some implementation schemes, the steps in the immediate security determination branch that predict whether the current printing task can be started, based on the current risk level indicators and the expected wear increment of the current task, include: Parse the code file of the current printing task to extract the trajectory length, motion trajectory complexity, and average movement speed of this printing task; Based on the trajectory length, the motion trajectory complexity, and the average moving speed, calculate the expected wear increment caused by this task to the cable conveyor. The expected increase in wear and tear is converted into a risk level indicator equivalent. Add the current risk level indicator to the equivalent risk level indicator to obtain the predicted peak value; The predicted peak value is compared with the preset break line. If the predicted peak value is lower than the break line, the current printing task is predicted to start. If the predicted peak value is not lower than the break line, the current printing task is predicted not to start.

[0010] Preferably, according to the above method, the step of calculating the expected wear increment of the cable conveyor caused by this task based on the trajectory length, motion trajectory complexity, and average moving speed includes: Obtain the unit cycle reference wear amount, which is the wear amount of the cable chain in one standard cycle calibrated using historical tests; Based on the complexity of the motion trajectory and the average moving speed, the trajectory complexity coefficient and speed influence coefficient are obtained by looking up the table. Based on the unit cyclic reference wear amount, the trajectory complexity coefficient, the speed influence coefficient, and the trajectory length, calculate the expected wear increment caused by this task to the cable chain, where the expected wear increment = unit cyclic reference wear amount * trajectory complexity coefficient * speed influence coefficient * trajectory length.

[0011] Based on the above, this application further proposes that, according to the above method, the steps of obtaining the historical risk index at the time of shutdown and the current risk index at the current moment using a risk index prediction model based on the current status data and shutdown status data include: A nonlinear compensation curve between temperature and material reference stiffness is established based on the Arrhenius equation. The current current data, current vibration data, shutdown current data, and shutdown vibration data are normalized to equivalent data at a standard reference temperature of 25 degrees Celsius to obtain the current equivalent current data, current equivalent vibration data, shutdown equivalent current data, and shutdown equivalent vibration data. Fourier transforms are performed on the current equivalent vibration data and the shutdown equivalent vibration data respectively, and the spectral energy integral value of the preset frequency band is extracted as the current dynamic vibration feature and the shutdown dynamic vibration feature. The current equivalent current data and the current dynamic vibration characteristics are input into the risk index prediction model to obtain the current risk index. The shutdown equivalent current data and the shutdown dynamic vibration characteristics are input into the risk index prediction model to obtain the historical risk index.

[0012] In some preferred embodiments, according to the above method, the step of responding to an apparent state regression anomaly and calculating a recoverable risk value based on the operating parameters before shutdown and the shutdown duration, wherein the operating parameters include continuous operating duration, average ambient temperature, and total cumulative bending count, includes: In response to apparent state rollback anomalies, obtain the continuous working time before shutdown, average ambient temperature, total cumulative number of bends, and shutdown time; Calculate the maximum recoverable amplitude coefficient based on the continuous working time, the average ambient temperature, and the total number of cumulative bends; The recoverable risk level is calculated using the exponential decay formula, where H = K_max * (1 - exp(− t_stop / tau)), H is the recoverable risk level, K_max is the maximum recoverable magnitude coefficient, t_stop is the downtime, and tau is the characteristic constant of the cable chain material.

[0013] Building upon this, this application further proposes that, following the step of adding the current risk level index to the recoverable risk level index to obtain the corrected irreversible damage baseline index according to the above method, the method further includes: Compare the modified irreversible damage baseline index with the historical risk index; If the modified irreversible damage baseline index is less than the historical risk index, then the modified irreversible damage baseline index is forcibly assigned the value of the historical risk index plus a constant preset based on the cumulative number of bends, so as to satisfy the law of monotonically increasing fatigue damage over time.

[0014] To improve the solution, based on the above method, the step of using a dual-branch method to predict 3D printer drag chain line failures, after modifying the irreversible damage substrate index and the current risk level index, also includes: After the print job restarts, the real-time risk index is continuously calculated and recorded to obtain the risk index curve. Calculate the first derivative of the risk index curve and monitor its upward slope; When the first derivative falls back and becomes parallel to the historical degradation slope before shutdown, the degradation period is determined to have ended, and the true risk index at the end of the degradation period is recorded. The true risk level index is compared with the modified irreversible damage baseline index, and the deviation value is calculated. If the deviation value exceeds the preset deviation threshold, the Kalman filter gain algorithm or gradient descent method is used to perform secondary retrospective correction on the irreversible damage baseline index in the historical database.

[0015] In practical applications, according to the above method, the steps for acquiring the status data of the 3D printer's cable chain in response to long-interval downtime events include: The current and shutdown current data of the stepper motor are obtained by collecting phase current waveform data of the stepper motor through current transformers and voltage transformers. Vibration signals during the movement of the cable conveyor are collected using a triaxial accelerometer to obtain current vibration data and vibration data during shutdown. Temperature data of the cable carrier is collected by thermocouples to obtain the current ambient temperature and the shutdown ambient temperature.

[0016] This technical solution enables the collection of multi-dimensional state data through various sensors, providing comprehensive and accurate input for the risk index prediction model, thereby improving the reliability of fault prediction.

[0017] Secondly, this application also discloses a 3D printer cable chain failure prediction system, which includes: The downtime acquisition module is used to acquire the downtime of the 3D printer, which is the time interval between the current downtime and the last downtime. The long-interval downtime event identification module is used to identify long-interval downtime events when the downtime duration exceeds a preset stress relaxation time threshold. The status data acquisition module is used to acquire the status data of the 3D printer drag chain line in response to the long-interval shutdown event. The status data includes current status data and shutdown status data. The current status data includes current current data, current vibration data and current ambient temperature. The shutdown status data includes shutdown current data, shutdown vibration data and shutdown ambient temperature. The risk index acquisition module is used to obtain the historical risk index at the time of shutdown and the current risk index at the current moment based on the current status data and shutdown status data, using the risk index prediction model. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer drag chain line. The difference calculation module is used to calculate the difference between the current risk level indicator and the historical risk level indicator; The apparent state regression anomaly judgment module is used to determine that the 3D printer drag chain has an apparent state regression anomaly when the difference is negative and the absolute value exceeds the preset risk threshold. The recoverable risk value calculation module is used to calculate the recoverable risk value in response to the apparent state rollback anomaly, based on the working parameters before shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature and total number of cumulative bends. The module for obtaining the base index of corrected irreversible damage is used to add the current risk index to the recoverable risk value to obtain the base index of corrected irreversible damage. The prediction module is used to predict 3D printer drag chain failures using a dual-branch approach based on the modified irreversible damage baseline index and the current risk level index. The dual-branch approach includes a long-term damage trend branch and an immediate safety determination branch.

[0018] This technical solution provides a system for implementing the aforementioned fault prediction method. Through modular design, the system functions are clear, easy to deploy and maintain, thereby improving the automation and intelligence level of fault prediction.

[0019] The 3D printer cable conveyor failure prediction method disclosed in this application identifies long-interval shutdown events when the shutdown duration exceeds a preset stress relaxation time threshold by acquiring the 3D printer's downtime. In response to this event, the system acquires the current and shutdown status data of the cable conveyor and uses a risk index prediction model to obtain historical and current risk indices. By calculating the difference between the two, a negative difference exceeding the preset risk threshold is identified as an apparent state regression anomaly. For this anomaly, the method calculates a recoverable risk value based on the pre-shutdown operating parameters and shutdown duration, and adds it to the current risk index to obtain a corrected irreversible damage baseline index. Finally, based on the corrected irreversible damage baseline index and the current risk index, a dual-branch approach (long-term damage trend branch and immediate safety judgment branch) is used to predict cable conveyor failures.

[0020] This method effectively solves the problem in existing technologies where, after long-interval downtime, the apparent improvement caused by the elastic recovery of the cable conveyor sheath cannot be distinguished from the continuous accumulation of irreversible internal damage. This leads to the introduction of a false improvement inflection point in the state evolution trend, causing the fault prediction model to underestimate the long-term wear rate and ultimately resulting in delayed triggering of fault warnings during the next high-intensity task. By introducing an identification mechanism for "apparent state regression anomalies" and the calculation of "recoverable risk level," this application can accurately distinguish between the elastic recovery of the cable conveyor material and actual irreversible damage, avoiding the "false improvement inflection point" caused by apparent improvement in traditional prediction models. In addition, the introduction of "corrected irreversible damage baseline index" ensures the monotonicity of fatigue damage accumulation, making long-term damage trend prediction more accurate. The dual-branch prediction mechanism takes into account both long-term health assessment and immediate task safety judgment, significantly improving the accuracy and reliability of fault prediction, effectively avoiding equipment failures and production interruptions caused by inaccurate predictions, and has significant technological progress and practical value.

[0021] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0022] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0023] Figure 1 A flowchart illustrating a 3D printer drag chain line fault prediction method provided in one embodiment of this application; Figure 2 A flowchart illustrating a prediction of whether the current print job can be started, provided as an embodiment of this application; Figure 3 This is a schematic diagram of a process for calculating the expected wear increment, provided as an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] See Figure 1 , Figure 1 This is a flowchart illustrating a 3D printer cable chain fault prediction method according to an embodiment of this application. The 3D printer cable chain fault prediction method provided in this embodiment includes, but is not limited to, steps S110 to S190, which are described below.

[0026] Step S110: Obtain the downtime of the 3D printer. The downtime is the time interval between the current downtime and the last downtime. Step S120: When the downtime exceeds a preset stress relaxation time threshold, a long-interval downtime event is identified; Step S130: In response to a long-interval shutdown event, acquire the status data of the 3D printer's drag chain line. The status data includes current status data and shutdown status data. The current status data includes current current data, current vibration data, and current ambient temperature. The shutdown status data includes shutdown current data, shutdown vibration data, and shutdown ambient temperature. Step S140: Based on the current status data and shutdown status data, use the risk index prediction model to obtain the historical risk index at the time of shutdown and the current risk index at the current moment. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer cable chain. The risk index is used to characterize the magnitude of the cable chain failure. Step S150: Calculate the difference between the current risk level index and the historical risk level index; Step S160: When the difference is negative and the absolute value exceeds the preset risk threshold, it is determined that the 3D printer's drag chain has an apparent state regression anomaly; Step S170: In response to the apparent state rollback anomaly, calculate the recoverable risk value based on the working parameters before shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature and total number of cumulative bends. Step S180: Add the current risk level index to the recoverable risk level value to obtain the corrected irreversible damage baseline index; Step S190: Based on the modified irreversible damage baseline index and the current risk level index, a dual-branch prediction method is used to predict 3D printer drag chain line failures. The dual branches include a long-term damage trend branch and an immediate safety judgment branch.

[0027] It's important to note that "status data" is crucial information for assessing the health of the cable conveyor line, including "current status data" and "shutdown status data." Current status data reflects the cable conveyor line's operating condition at the current moment, including "current current data," "current vibration data," and "current ambient temperature." Shutdown status data records the cable conveyor line's state at the moment of shutdown, including "shutdown current data," "shutdown vibration data," and "shutdown ambient temperature." These data collectively provide the basis for calculating the risk index. The "risk index" is a quantifiable indicator of the severity of a cable conveyor line failure; a higher value indicates a greater risk of failure. This index is calculated using a "risk index prediction model," which is trained based on historical status data.

[0028] "Apparent condition regression anomaly" refers to an improvement in the apparent health of a cable conveyor after shutdown, but this improvement is not a reduction in actual damage; rather, it is a result of material elastic recovery. This anomaly is determined by comparing the difference between the current risk level index and the historical risk level index. When the difference is negative and the absolute value exceeds the "preset risk level threshold," an apparent condition regression anomaly is considered to exist. The "recoverable risk level value" is an indicator that quantifies the degree of this apparent condition regression, reflecting the degree of health that can be restored due to stress relaxation and elastic recovery. The calculation of this value depends on the "operating parameters before shutdown," including "continuous operating time," "average ambient temperature," and "total number of cumulative bends," which collectively affect the fatigue damage degree and stress relaxation potential of the cable conveyor.

[0029] The "corrected irreversible damage baseline index" is one of the core concepts of this application. By adding the current risk index to the recoverable risk index, it aims to eliminate the influence of apparent state regression and more accurately reflect the accumulation of irreversible fatigue damage inside the cable conveyor line. Finally, the "dual-branch prediction" mechanism uses the corrected irreversible damage baseline index and the current risk index to comprehensively predict cable conveyor line failures from two dimensions: long-term damage trend and immediate safety assessment.

[0030] Firstly, there are several ways to obtain the downtime of a 3D printer. For example, it can be achieved by integrating a timer module into the 3D printer's control system. This timer module starts at the beginning of each printing job and stops at the end. When the printer is not in operation, the timer continuously records the time. When the printer restarts, the system can read the recorded downtime and use it as the downtime. Another method is to indirectly obtain the downtime by monitoring the on / off status of the 3D printer's main power supply. When the main power is disconnected, the system records the disconnection time; when the main power is reconnected, the system records the reconnection time, and calculates the interval between the two reconnections as the downtime.

[0031] Secondly, when the downtime exceeds a preset stress relaxation time threshold, it is necessary to identify whether a long-interval downtime event exists. This can be achieved by setting a comparator module in the control system. This module compares the acquired downtime with the preset stress relaxation time threshold. If the downtime exceeds the threshold, the system will trigger a long-interval downtime event signal. For example, the preset stress relaxation time threshold can be set based on the characteristics of the cable chain material and experimental data, such as 2 hours. When the downtime reaches or exceeds 2 hours, the system will identify it as a long-interval downtime event.

[0032] Next, in response to a long-interval shutdown event, it is necessary to acquire the status data of the 3D printer's cable conveyor. This status data includes current status data and shutdown status data. Current status data includes current current data, current vibration data, and current ambient temperature; shutdown status data includes shutdown current data, shutdown vibration data, and shutdown ambient temperature. This data can be acquired using various sensors. For example, current data can be acquired by installing current sensors or Hall effect sensors on the stepper motor drive circuit; vibration data can be acquired by installing accelerometers at key locations on the cable conveyor; and ambient temperature data can be acquired by installing thermocouples or temperature sensors inside the 3D printer or near the cable conveyor. When a shutdown event occurs, the system records the current, vibration, and ambient temperature at that time as shutdown status data. After the printer restarts and recognizes the long-interval shutdown event, the system again acquires the current current, vibration, and ambient temperature as current status data.

[0033] Then, based on the current state data and shutdown state data, the risk index prediction model is used to obtain the historical risk index at the time of shutdown and the current risk index at the current moment. The risk index prediction model is a model built based on the state data of the 3D printer's cable conveyor line, predicting the risk index to characterize the magnitude of cable conveyor line failures. This model can be a machine learning model, such as a support vector machine, neural network, or random forest, trained using historical data. After obtaining the current state data and shutdown state data, these data are input into the pre-trained risk index prediction model. The model will output the historical risk index at the time of shutdown and the current risk index at the current moment. For example, the model can receive current, vibration, and temperature data as input and output a value between 0 and 1, where 0 represents no fault and 1 represents a complete failure.

[0034] Next, it is necessary to calculate the difference between the current risk level indicator and the historical risk level indicator. This can be achieved using a simple subtraction module. This module takes the current risk level indicator and the historical risk level indicator as input and outputs the difference between them. For example, if the current risk level indicator is 0.3 and the historical risk level indicator is 0.4, then the difference is -0.1.

[0035] Next, when the difference is negative and its absolute value exceeds a preset risk threshold, the system determines that the 3D printer's cable conveyor has experienced an apparent state regression anomaly. This can be achieved using a comparator module and an absolute value calculation module. First, the absolute value of the difference is calculated. Then, the difference is compared to zero to determine if it is negative. Simultaneously, the absolute value of the difference is compared to a preset risk threshold. If the difference is negative and its absolute value is greater than the preset risk threshold (e.g., 0.05), the system determines that an apparent state regression anomaly exists.

[0036] In response to an apparent state regression anomaly, the system calculates a recoverable risk value based on the pre-stop operating parameters and stoppage duration. Operating parameters include continuous operating time, average ambient temperature, and total cumulative bends. These parameters are continuously recorded during 3D printer operation. For example, continuous operating time can be recorded using a timer, average ambient temperature can be calculated using the average value from a temperature sensor during operation, and the total cumulative bends can be estimated using data from a cable bending sensor or stepper motor encoder. Upon determining that an apparent state regression anomaly exists, the system reads these historical operating parameters and the current stoppage duration, and calculates the recoverable risk value using a preset formula or lookup table. For example, an empirical formula or lookup table based on these parameters can be created to output the corresponding recoverable risk value.

[0037] Then, the current risk level indicator is added to the recoverable risk level value to obtain the modified irreversible damage baseline indicator. This can be achieved through an addition module. This module takes the current risk level indicator and the calculated recoverable risk level value as input and outputs the sum of the two, i.e., the modified irreversible damage baseline indicator. For example, if the current risk level indicator is 0.3 and the recoverable risk level value is 0.08, then the modified irreversible damage baseline indicator is 0.38.

[0038] Finally, based on the corrected irreversible damage baseline index and the current risk level index, a two-branch method is used to predict 3D printer cable conveyor failures. The two branches include a long-term damage trend branch and an immediate safety determination branch. This can be implemented through a decision module. This module takes the corrected irreversible damage baseline index and the current risk level index as input and activates the long-term damage trend branch and the immediate safety determination branch according to preset logic. The long-term damage trend branch uses the corrected irreversible damage baseline index to predict the long-term degradation trend and remaining lifespan of the cable conveyor, while the immediate safety determination branch uses the current risk level index combined with the expected wear increment of the current task to determine whether it is safe to start the current printing task.

[0039] Compared to the closest existing technologies, this application has the advantage that existing technologies, which are typically based on historical trend extrapolation, struggle to effectively address the apparent state regression problem caused by long-interval downtime and tend to underestimate the true wear rate. In contrast, this application, through a series of innovative steps including introducing a stress relaxation time threshold, judging apparent state regression anomalies, calculating recoverable risk, and correcting irreversible damage baseline indices, can accurately distinguish between recoverable components and irreversible damage, thus providing more accurate and reliable fault prediction. This method not only improves the timeliness of fault warnings, avoiding equipment failures and production interruptions due to inaccurate predictions, but also provides a more scientific basis for the maintenance and management of 3D printers, significantly improving equipment operating efficiency and safety.

[0040] Specifically, the steps for predicting 3D printer drag chain line failures using a dual-branch approach, based on the modified irreversible damage baseline index and the current risk level index, include: In the long-term damage trend branch, the degradation rate and remaining lifetime are calculated based on the modified irreversible damage base index to predict 3D printer drag chain failures. In the immediate safety assessment branch, based on the current risk level index and the expected wear increment of the current task, it is predicted whether the current printing task can be started, in order to predict 3D printer drag chain failure.

[0041] The long-term damage trend branch primarily focuses on the cumulative fatigue damage and aging process of the cable conveyor material. The modified irreversible damage baseline index is considered the actual cumulative damage level of the cable conveyor after experiencing shutdown stress relaxation; it reflects the irreversible structural degradation of the cable conveyor. Based on this index, the degradation rate of the cable conveyor can be calculated, i.e., the rate at which damage increases with time or number of uses. Furthermore, by analyzing the degradation rate, the remaining lifespan of the cable conveyor can be predicted, thus achieving long-term trend prediction of 3D printer cable conveyor failures.

[0042] The immediate safety assessment branch focuses on evaluating the immediate safety of the cable conveyor line performing a specific printing task at the current moment. The current risk level indicator characterizes the instantaneous health of the cable conveyor line in its current operating state. The expected wear increment for the current task refers to the additional wear or damage that the upcoming printing task may cause to the cable conveyor line. By combining the current risk level indicator with the expected wear increment for the current task, the potential risk peak of the cable conveyor line after completing the task can be predicted, thereby determining whether it is safe to start the current printing task.

[0043] Through the above technical solutions, this application provides a more refined and multi-dimensional cable conveyor failure prediction strategy. The introduction of the long-term damage trend branch makes the prediction of cable conveyor lifespan more accurate, facilitating the development of preventative maintenance plans and avoiding sudden failures caused by cumulative damage. The setting of the immediate safety judgment branch effectively solves the problem of assessing the immediate safety of the cable conveyor before starting a new task, avoiding the risk of equipment damage or production interruption that may result from blindly starting a task when the cable conveyor is in a critical state. This dual-branch collaborative prediction significantly improves the comprehensiveness, accuracy, and practicality of 3D printer cable conveyor failure prediction, thereby improving equipment operational reliability and production efficiency.

[0044] Specifically, in the aforementioned real-time security determination branch, the step of predicting whether the current printing task can be started based on the current risk level index and the expected wear increment of the current task can be further refined into the following specific operations: like Figure 2 , Figure 2 This is a flowchart illustrating a method for predicting whether a current print job can be started, provided as an embodiment of this application. In the immediate security determination branch, the steps for predicting whether a current print job can be started, based on the current risk level index and the expected wear increment of the current job, include, but are not limited to, steps S210 to S250: Step S210: Parse the code file of the current printing task and extract the trajectory length, motion trajectory complexity and average movement speed of this printing task; Step S220: Based on the trajectory length, motion trajectory complexity, and average moving speed, calculate the expected wear increment caused by this task to the cable chain. Step S230: Convert the expected wear increment into a risk level index equivalent; Step S240: Add the current risk level indicator to the risk level indicator equivalent to obtain the predicted peak value; Step S250: Compare the predicted peak value with the preset break red line. If the predicted peak value is lower than the break red line, the current printing task can be started. If the predicted peak value is not lower than the break red line, the current printing task cannot be started.

[0045] Specifically, parsing the code file of the current printing task refers to analyzing the G-code or other control files of the 3D printer's upcoming printing task to identify and extract key parameters related to the movement of the cable conveyor. These parameters include the trajectory length of this printing task, i.e., the total distance the cable conveyor is expected to travel during the entire printing process; the trajectory complexity, which characterizes the degree of complex movements such as bending and twisting that the cable conveyor undergoes during movement; and the average movement speed, i.e., the average running speed of the cable conveyor during the task. These parameters form the basis for assessing the wear and stress that the cable conveyor may experience in the upcoming task.

[0046] The calculation of the expected wear increment on the cable conveyor based on the trajectory length, the trajectory complexity, and the average moving speed refers to quantifying the additional wear that the printing task is expected to cause to the cable conveyor using a pre-established wear model or empirical formula, based on the extracted task parameters. This expected wear increment reflects the degree of physical degradation that the cable conveyor may experience after performing a specific task.

[0047] Furthermore, the expected wear increment is converted into a risk index equivalent. The purpose is to map the physical wear amount to a unified risk index system so that it can be directly compared and superimposed with the current risk index. This conversion is usually achieved through a preset conversion function or lookup table to ensure that damage information from different dimensions can be uniformly quantified.

[0048] Therefore, the current risk level indicator is added to the equivalent risk level indicator to obtain the predicted peak value. This predicted peak value represents the highest possible level of risk level the cable conveyor line can reach after completing the current printing task. This peak value comprehensively considers both the current health status of the cable conveyor line and the additional burden imposed on it by the upcoming task.

[0049] Finally, the predicted peak value is compared with a preset breakage threshold. The breakage threshold is a preset safety threshold representing the maximum risk the cable conveyor can withstand; exceeding this threshold significantly increases the risk of malfunction or breakage. If the predicted peak value is below the breakage threshold, the current printing task is predicted to start, indicating that the cable conveyor remains within a safe range after completing the task. Conversely, if the predicted peak value is not lower than the breakage threshold, the current printing task is predicted not to start to avoid potential malfunctions and safety risks.

[0050] Through the above technical solution, this application enables accurate real-time safety assessment of 3D printer cable conveyors. By combining the current risk level index with the expected wear increment for a specific task and comparing it with the breakage threshold, the system can predict in advance whether the cable conveyor will reach a dangerous state after performing the current task. This allows users or the system to make informed decisions before starting a task, effectively avoiding printing interruptions, equipment damage, or safety accidents caused by cable conveyor malfunctions, significantly improving the operational reliability and production efficiency of 3D printers. This solution provides 3D printer operators with a reliable basis for real-time safety judgment, ensuring the safety of the equipment before each task starts.

[0051] Specifically, in the aforementioned real-time safety determination branch, when predicting whether the current printing task can be started based on the current risk level index and the expected wear increment of the current task, the calculation of the expected wear increment caused by this task to the cable conveyor can be further refined.

[0052] like Figure 3 As shown, Figure 3 This is a flowchart illustrating the calculation of expected wear increment in one embodiment of this application. Based on trajectory length, motion trajectory complexity, and average moving speed, the steps for calculating the expected wear increment caused to the cable conveyor by this task include, but are not limited to, steps S310 to S330: Step S310: Obtain the unit cycle reference wear amount, which is the wear amount of the cable chain in one standard cycle calibrated using historical tests; Step S320: Based on the trajectory complexity and average moving speed, look up the table to obtain the trajectory complexity coefficient and speed influence coefficient; Step S330: Calculate the expected wear increment caused by this task to the cable chain based on the unit cycle reference wear amount, trajectory complexity coefficient, speed influence coefficient and trajectory length, where the expected wear increment = unit cycle reference wear amount * trajectory complexity coefficient * speed influence coefficient * trajectory length.

[0053] The unit cycle reference wear amount refers to the wear amount of the cable conveyor under specific standard cycle conditions, obtained through historical tests and calibrations. It serves as a fundamental quantitative indicator of the inherent wear characteristics of the cable conveyor material. In practical applications, this value can be obtained through fatigue life tests and friction and wear tests on the cable conveyor, and recorded in a database for future reference. The trajectory complexity coefficient characterizes the impact of the complexity of the printing task's motion trajectory on the cable conveyor wear. For example, frequent starts and stops, sharp turns, and multi-axis linkages increase cable conveyor wear. This coefficient can be obtained by consulting a pre-established coefficient table based on the preset motion trajectory complexity level. The speed influence coefficient quantifies the impact of the average moving speed of the cable conveyor on wear. Generally, higher operating speeds result in greater impact and friction, thus accelerating wear. This coefficient can also be determined by consulting a preset speed-influence coefficient lookup table.

[0054] This application's solution constructs a refined model for calculating the expected wear increment by introducing unit cycle baseline wear, trajectory complexity coefficient, and speed influence coefficient, combined with trajectory length. Specifically, the unit cycle baseline wear provides the basic wear characteristics of the cable conveyor material under standard operating conditions, serving as the starting point for calculation. The trajectory complexity coefficient and speed influence coefficient quantify the additional impact of the printing task's motion trajectory complexity and average movement speed on the cable conveyor wear, respectively. These coefficients are obtained through table lookup, ensuring accurate capture of wear characteristics under different operating conditions. Finally, by multiplying these parameters, the material characteristics, motion mode, and operating speed of the cable conveyor can be comprehensively considered, thereby accurately estimating the expected wear increment caused by the task to the cable conveyor. This calculation method avoids the errors that may be introduced by simple linear models, making the assessment of cable conveyor damage more realistic.

[0055] Specifically, the steps described above, based on current status data and shutdown status data, and using a risk index prediction model to obtain the historical risk index at the time of shutdown and the current risk index at the current moment, include: A nonlinear compensation curve between temperature and material reference stiffness is established based on the Arrhenius equation. The current current data, current vibration data, shutdown current data, and shutdown vibration data are normalized to equivalent data at a standard reference temperature of 25 degrees Celsius to obtain the current equivalent current data, current equivalent vibration data, shutdown equivalent current data, and shutdown equivalent vibration data. Fourier transforms are performed on the current equivalent vibration data and the shutdown equivalent vibration data respectively, and the spectral energy integral value of the preset frequency band is extracted as the current dynamic vibration feature and the shutdown dynamic vibration feature. The current equivalent current data and the current dynamic vibration characteristics are input into the risk index prediction model to obtain the current risk index. The shutdown equivalent current data and the shutdown dynamic vibration characteristics are input into the risk index prediction model to obtain the historical risk index.

[0056] Specifically, in acquiring historical risk indicators at the time of shutdown and current risk indicators at the current moment, the first step is to establish a nonlinear compensation curve between temperature and material reference stiffness based on the Arrhenius equation. The Arrhenius equation, typically used to describe the relationship between chemical reaction rates and temperature, is here applied to characterize the influence of temperature on the performance of cable chain materials (e.g., reference stiffness). By establishing this nonlinear compensation curve, current current data, current vibration data, shutdown current data, and shutdown vibration data collected at different ambient temperatures can be normalized to equivalent data at a standard reference temperature (e.g., 25 degrees Celsius), thereby obtaining current equivalent current data, current equivalent vibration data, shutdown equivalent current data, and shutdown equivalent vibration data. This step aims to eliminate the influence of ambient temperature changes on sensor data, ensuring the comparability and consistency of data under different operating conditions.

[0057] Furthermore, Fourier transforms are performed on both the current equivalent vibration data and the shutdown equivalent vibration data. Fourier transform converts the vibration signal in the time domain into a frequency domain signal, thereby revealing the distribution of vibration energy at different frequencies. By extracting the spectral energy integral value of a preset frequency band, it can be used as the current dynamic vibration characteristic and the shutdown dynamic vibration characteristic. These preset frequency bands are typically associated with the characteristic vibration frequencies generated by the cable conveyor under specific fault modes (such as wear, fatigue, or loosening), and changes in their spectral energy integral values ​​can directly reflect the degree of damage to the cable conveyor.

[0058] Subsequently, the current equivalent current data and the current dynamic vibration characteristics are input into the risk index prediction model to obtain the current risk index. Simultaneously, the shutdown equivalent current data and the shutdown dynamic vibration characteristics are input into the risk index prediction model to obtain the historical risk index at the time of shutdown. The risk index prediction model is a trained model that can output a risk index characterizing the magnitude of the cable chain fault based on the processed input current data and dynamic vibration characteristics.

[0059] The above technical solution effectively eliminates the interference of ambient temperature changes on the acquisition and analysis of cable conveyor condition data, improving the accuracy and consistency of the data. Furthermore, by performing frequency domain analysis on vibration data and extracting key dynamic characteristics, the risk index prediction model can more sensitively and accurately capture early damage and degradation trends of the cable conveyor. Therefore, this application provides more reliable and refined risk indicators, offering a solid data foundation for subsequent fault prediction and maintenance decisions, and significantly improving the accuracy and robustness of fault prediction for 3D printer cable conveyors.

[0060] Specifically, the steps described above for responding to apparent state regression anomalies and calculating the recoverable risk value based on the operating parameters before shutdown and the shutdown duration can be further refined.

[0061] The step of responding to the apparent state regression anomaly and calculating the recoverable risk value based on the operating parameters before shutdown and the shutdown duration, wherein the operating parameters include continuous operating duration, average ambient temperature, and total cumulative bending count, includes: In response to apparent state rollback anomalies, obtain the continuous working time before shutdown, average ambient temperature, total cumulative number of bends, and shutdown time; The maximum recoverable amplitude coefficient is calculated based on the continuous working time, the average ambient temperature, and the total number of cumulative bends. The recoverable risk level is calculated according to the exponential decay formula, where H = K_max * (1 - exp(− t_stop / tau)), H is the recoverable risk level, K_max is the maximum recoverable magnitude coefficient, t_stop is the downtime, and tau is the characteristic constant of the cable chain material.

[0062] Among them, "apparent state regression anomaly" refers to a situation where, after a long period of shutdown, the current risk index of the 3D printer's drag chain line shows a negative change compared to the historical risk index at the time of shutdown, and the amount of change exceeds the preset threshold. This indicates that the drag chain line may have undergone stress relaxation or self-repair during the shutdown period, resulting in some recovery of its apparent damage state.

[0063] The "operating parameters" specifically include continuous operating time, average ambient temperature, and total cumulative number of bends. These parameters characterize the stress history and environmental conditions experienced by the cable conveyor before shutdown and are key factors affecting its stress relaxation and damage recovery capabilities.

[0064] The "Recoverable Risk Value" H is used to quantify the degree of damage recovered by the cable conveyor line during downtime due to stress relaxation and other effects, i.e., the amount of risk reduction.

[0065] The "maximum recoverable amplitude coefficient" K_max is calculated based on the operating parameters before shutdown (continuous working time, average ambient temperature, and total number of cumulative bends). It represents the maximum risk level that the cable conveyor can theoretically recover under a given working history. This coefficient reflects the inherent characteristics of the cable conveyor material and the degree of fatigue accumulation under specific operating conditions.

[0066] "Stop duration" t_stop refers to the time interval between the last stop of the 3D printer and the current moment, and is a direct time variable that affects the degree of stress relaxation.

[0067] The “tau” is a constant related to the material properties of the cable chain, which characterizes the time scale of the material’s stress relaxation or damage recovery process.

[0068] The "exponential decay formula" H = K_max * (1 - exp(− t_stop / tau)) is a commonly used model to describe the nonlinear behavior of materials during stress relaxation or recovery. This formula shows that the recoverable risk value H gradually increases with the downtime t_stop, but its growth rate gradually slows down and eventually approaches the maximum recoverable magnitude coefficient K_max.

[0069] Through the above technical solution, this application can accurately quantify the stress relaxation and damage recovery effects of 3D printer cable conveyors during long-interval downtime. Traditional fault prediction methods often ignore the material self-healing phenomenon during downtime, which may lead to misjudgment of the actual damage state of the cable conveyor. This application introduces a recoverable risk value and calculates it based on the operating parameters before downtime and the downtime duration, making the assessment of the cable conveyor's damage state more consistent with reality. This not only improves the accuracy of fault prediction and avoids premature maintenance or replacement due to ignoring the recovery effect, but also enables more rational use of the remaining lifespan of the cable conveyor, thereby reducing operating costs and improving the reliability and economy of the equipment.

[0070] In some embodiments described above in this application, a corrected irreversible damage baseline index is obtained by adding the current risk level index to the recoverable risk level index. However, in practical applications, due to measurement errors, model limitations, or the influence of specific operating conditions, the calculated corrected irreversible damage baseline index may, in some cases, be lower than the historical risk level index at the time of shutdown. If this problem is not addressed, the assessment results of fatigue damage will not conform to its monotonically increasing physical law over time, thereby affecting the accuracy and reliability of subsequent failure prediction.

[0071] In response, this application further proposes that after adding the current risk level index to the recoverable risk level index to obtain the corrected irreversible damage baseline index, it also includes: Compare the modified irreversible damage baseline index with the historical risk index; If the modified irreversible damage baseline index is less than the historical risk index, then the modified irreversible damage baseline index is forcibly assigned the value of the historical risk index plus a constant preset based on the cumulative number of bends, so as to satisfy the law of monotonically increasing fatigue damage over time.

[0072] Specifically, after obtaining the corrected irreversible damage baseline index, it needs to be compared with the historical risk index at the time of shutdown. The historical risk index represents the level of irreversible damage accumulated by the cable conveyor before shutdown. If the calculated corrected irreversible damage baseline index is lower than this historical risk index, it indicates that the current damage assessment results contradict the physical characteristics of fatigue damage. In this case, to ensure the rationality and physical consistency of the damage assessment, the corrected irreversible damage baseline index needs to be forcibly corrected. Specifically, the corrected irreversible damage baseline index will be reassigned as the sum of the historical risk index and a preset constant based on the cumulative number of bends. This preset constant aims to reflect that even after shutdown and recovery, the cable conveyor will still accumulate a small amount of irreversible fatigue damage in subsequent operation, thus ensuring that the corrected irreversible damage baseline index always shows a monotonically increasing trend, consistent with the objective laws of material fatigue damage.

[0073] Through the above technical solution, this application can effectively avoid the non-physical decline of irreversible damage indicators caused by calculation errors or model limitations, thereby ensuring the physical rationality and monotonically increasing nature of fatigue damage assessment results. As a result, the accuracy and reliability of cable chain failure prediction are significantly improved, avoiding underestimation of the actual damage level of the cable chain and providing a more precise guarantee for the safe operation of 3D printers.

[0074] In some embodiments, this application further proposes, after the above-described step of using a dual-branch prediction method for 3D printer drag chain line failure based on a modified irreversible damage baseline index and a current risk index, the following steps are also included: After the print job restarts, the real-time risk level index is continuously calculated and recorded to obtain the risk level index curve; Calculate the first derivative of the risk index curve and monitor its upward slope; When the first derivative falls back and becomes parallel to the historical degradation slope before shutdown, the degradation period is determined to have ended, and the true risk index at the end of the degradation period is recorded. The actual risk level index is compared with the modified irreversible damage baseline index, and the deviation value is calculated; If the deviation value exceeds the preset deviation threshold, the Kalman filter gain algorithm or gradient descent method is used to perform secondary retrospective correction on the irreversible damage baseline index in the historical database.

[0075] Specifically, after a printing job restarts, the real-time risk index refers to the risk index calculated in real time using a risk index prediction model after the 3D printer is restarted, by continuously collecting status data of the cable chain (such as current data, vibration data, and ambient temperature). These real-time calculated risk indices are continuously recorded, forming a risk index curve reflecting the change of the cable chain's status over time. This curve can intuitively show the dynamic changes of the cable chain after restarting. The calculation of the first derivative of the risk index curve aims to quantify the rate of change of the risk index over time, i.e., the degradation rate of the cable chain. By monitoring its rising slope, the escalation trend of cable chain damage can be tracked in real time. The steeper the rising slope, the faster the cable chain degrades.

[0076] In practical applications, the end of the degradation period can be determined when the first derivative falls back and becomes parallel to the historical degradation slope before shutdown. The degradation period refers to the stage where the performance of the cable conveyor gradually transitions from a recovery state to a stable degradation state after a long-interval shutdown. The historical degradation slope before shutdown is the average degradation rate calculated based on risk index data during normal operation of the cable conveyor, representing the inherent degradation trend of the cable conveyor under stable operating conditions. When the real-time degradation slope (first derivative) becomes parallel to the historical degradation slope, it indicates that the cable conveyor has essentially completed its degradation period adjustment, and its degradation behavior has recovered to a stable pattern similar to that before shutdown. At this point, the true risk index at the end of the degradation period is recorded; this index is considered the actual initial risk level of the cable conveyor in the stable degradation phase. Furthermore, this true risk index is compared with the previously calculated corrected irreversible damage baseline index to calculate the deviation between the two. This deviation reflects the difference between the initial correction value and the actual stable state. If the deviation exceeds a preset deviation threshold, it indicates a significant deviation in the initial corrected irreversible damage baseline index, requiring correction. In this case, the Kalman filter gain algorithm or gradient descent method can be used to perform a secondary retrospective correction of the corrected irreversible damage baseline index in the historical database. The Kalman filter gain algorithm optimizes the state estimate by fusing real-time observation data and the prediction model; the gradient descent method iteratively adjusts parameters to make the predicted value closer to the true value. Both methods can be used to dynamically adjust and optimize the baseline index in historical data, making it more accurately reflect the actual damage level of the cable conveyor.

[0077] Through the above technical solutions, this application can significantly improve the accuracy and robustness of 3D printer cable conveyor failure prediction. Especially after long-interval downtime, this solution can dynamically monitor and correct the initial irreversible damage baseline indicators, effectively avoiding prediction deviations caused by dynamic changes in the decay period. Therefore, the prediction results of the long-term damage trend branch will more closely resemble the actual degradation trajectory of the cable conveyor, thus providing users with more reliable maintenance and replacement recommendations. Furthermore, by introducing a Kalman filter gain algorithm or gradient descent method for secondary retrospective correction, the prediction model of this application possesses adaptive and learning capabilities, continuously optimizing the damage baseline indicators in the historical database based on actual operating data, further enhancing the intelligence level and long-term stability of the prediction system.

[0078] In some embodiments, the step of acquiring status data of the 3D printer's cable chain in response to a long-interval downtime event may include the following: In response to the aforementioned long-interval downtime event, the steps for obtaining the status data of the 3D printer's cable conveyor include: The current and shutdown current data of the stepper motor are obtained by collecting phase current waveform data of the stepper motor through current transformers and voltage transformers. Vibration signals during the movement of the cable conveyor are collected using a triaxial accelerometer to obtain current vibration data and vibration data during shutdown. Temperature data of the cable carrier is collected by thermocouples to obtain the current ambient temperature and the shutdown ambient temperature.

[0079] Current transformers and voltage transformers are used to non-contactly measure the phase current waveform data of the stepper motor in both operating and stopped states. The phase current waveform data reflects the stepper motor's load condition, operational smoothness, and the presence of abnormal resistance, thus indirectly characterizing the operating status of the cable conveyor. Specifically, the current data refers to the current data collected during normal operation of the 3D printer, while the shutdown current data is collected when the 3D printer is stopped, with the purpose of establishing a baseline or making state comparisons.

[0080] Furthermore, a triaxial accelerometer is configured to monitor vibration signals of the cable conveyor in real time during its movement. Vibration signals are a key indicator for assessing the health of mechanical components; changes in their spectrum and amplitude can reveal potential faults such as wear, loosening, jamming, or fatigue in the cable conveyor. Specifically, current vibration data refers to vibration signals collected during normal operation of the 3D printer, while shutdown vibration data is collected when the 3D printer is stopped, with the aim of analyzing changes in its state before and after shutdown.

[0081] In addition, thermocouples are used to accurately measure the ambient temperature of the cable carrier. Ambient temperature is a crucial factor affecting material properties and the lifespan of mechanical components, especially for plastic cable carriers, where excessively high or low temperatures can accelerate aging or embrittlement. In practical applications, the current ambient temperature refers to temperature data collected during normal 3D printer operation, while the shutdown ambient temperature is temperature data collected when the 3D printer is stopped, with the aim of assessing the impact of the environment on the cable carrier's condition.

[0082] The aforementioned technical solution enables multi-dimensional and high-precision acquisition of status data for 3D printer cable conveyors. Specifically, by comprehensively monitoring electrical parameters, mechanical vibration, and ambient temperature, subtle changes in the cable conveyor during operation and shutdown can be fully captured, providing richer and more accurate raw data for cable conveyor fault prediction. This comprehensive data acquisition method significantly improves the accuracy and reliability of the risk index prediction model, thereby enhancing the effectiveness of the entire fault prediction method. It helps to promptly detect potential cable conveyor anomalies and avoid misjudgments or omissions caused by incomplete or inaccurate data.

[0083] This application also discloses a 3D printer drag chain fault prediction system, including: The downtime acquisition module is used to acquire the downtime of the 3D printer, wherein the downtime is the time interval between the current downtime and the last downtime. The long-interval downtime event identification module is used to identify the existence of a long-interval downtime event when the downtime duration exceeds a preset stress relaxation time threshold. The status data acquisition module is used to acquire the status data of the 3D printer drag chain line in response to the long-interval shutdown event. The status data includes current status data and shutdown status data. The current status data includes current current data, current vibration data and current ambient temperature. The shutdown status data includes shutdown current data, shutdown vibration data and shutdown ambient temperature. The risk index acquisition module is used to acquire the historical risk index at the time of shutdown and the current risk index at the current moment based on the current status data and shutdown status data, using the risk index prediction model. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer drag chain line. The difference calculation module is used to calculate the difference between the current risk level indicator and the historical risk level indicator; The apparent state regression anomaly judgment module is used to determine that the 3D printer's drag chain has an apparent state regression anomaly when the difference is negative and the absolute value exceeds a preset risk threshold. The recoverable risk value calculation module is used to calculate the recoverable risk value in response to the apparent state rollback anomaly, based on the working parameters before shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature and total number of cumulative bends. The module for obtaining the modified irreversible damage baseline index is used to add the current risk index to the recoverable risk index to obtain the modified irreversible damage baseline index; The prediction module is used to predict 3D printer drag chain line failures using a dual-branch approach based on the corrected irreversible damage baseline index and the current risk level index. The dual-branch approach includes a long-term damage trend branch and an immediate safety determination branch.

[0084] The 3D printer cable conveyor fault prediction system proposed in this application aims to effectively resolve the contradiction between apparent state regression caused by material stress relaxation and the accumulation of actual internal fatigue damage after long-interval shutdowns of 3D printer cable conveyors through modular design. The system, through a shutdown duration acquisition module, a long-interval shutdown event identification module, and a status data acquisition module, can accurately capture and respond to long-interval shutdown events and comprehensively collect the operating status data of the cable conveyor. Subsequently, a risk index acquisition module, a difference calculation module, and an apparent state regression anomaly judgment module work together to accurately identify whether there are apparent state regression anomalies in the cable conveyor. Once an anomaly is identified, a recoverable risk value calculation module and a corrected irreversible damage baseline index acquisition module intervene to quantify and correct the apparent improvement, thereby obtaining a more realistic irreversible damage baseline index. Finally, the prediction module uses the corrected index and the current risk index, through a dual-branch prediction mechanism, to comprehensively and accurately predict cable conveyor faults from both long-term trend and immediate safety dimensions. This effectively avoids the problem of delayed fault warning triggering in traditional methods, significantly improving the reliability and safety of 3D printer operation.

[0085] The 3D printer cable conveyor fault prediction system proposed in this application represents a significant advancement compared to existing technologies. Traditional systems, when dealing with fault prediction after long-interval shutdowns of 3D printer cable conveyors, often struggle to distinguish between the apparent improvement caused by material stress relaxation and the accumulation of actual internal fatigue damage. This introduces false improvement inflection points into the state evolution trend, leading the fault prediction model to underestimate the long-term wear rate and ultimately causing a delay in triggering fault warnings.

[0086] The above provides a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined in this application.

Claims

1. A method for predicting faults in a 3D printer's cable conveyor belt, characterized in that, include: Obtain the downtime of the 3D printer, where the downtime is the time interval between the current downtime and the last downtime. When the downtime exceeds a preset stress relaxation time threshold, a long-interval downtime event is identified. In response to the long-interval shutdown event, the status data of the 3D printer cable chain is acquired, wherein the status data includes current status data and shutdown status data, the current status data includes current current data, current vibration data and current ambient temperature, and the shutdown status data includes shutdown current data, shutdown vibration data and shutdown ambient temperature; Based on the current status data and shutdown status data, the historical risk index at the time of shutdown and the current risk index at the current moment are obtained using the risk index prediction model. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer cable conveyor. The risk index is used to characterize the magnitude of the cable conveyor failure. Calculate the difference between the current risk level index and the historical risk level index; When the difference is negative and its absolute value exceeds a preset risk threshold, it is determined that the 3D printer's drag chain has experienced an apparent state regression anomaly. In response to the apparent state regression anomaly, a recoverable risk value is calculated based on the working parameters before shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature, and total cumulative bending count. The current risk level index is added to the recoverable risk level value to obtain the corrected irreversible damage baseline index; Based on the modified irreversible damage baseline index and the current risk index, a dual-branch method is used to predict 3D printer drag chain line failures. The dual branches include a long-term damage trend branch and an immediate safety judgment branch.

2. The method according to claim 1, characterized in that, The step of predicting 3D printer drag chain line failures using a dual-branch approach based on the corrected irreversible damage substrate index and the current risk index includes: In the long-term damage trend branch, the degradation rate and remaining lifetime are calculated based on the modified irreversible damage baseline index to predict 3D printer drag chain failures; In the immediate safety determination branch, based on the current risk level index and the expected wear increment of the current task, it is predicted whether the current printing task can be started, in order to predict 3D printer drag chain failure.

3. The method according to claim 2, characterized in that, In the immediate security determination branch, the step of predicting whether the current printing task can be started based on the current risk level index and the expected wear increment of the current task includes: Parse the code file of the current printing task to extract the trajectory length, motion trajectory complexity, and average movement speed of this printing task; Based on the trajectory length, the motion trajectory complexity, and the average moving speed, calculate the expected wear increment on the cable conveyor caused by this task; The expected wear increment is converted into a risk level index equivalent; The predicted peak value is obtained by adding the current risk level index to the risk level index equivalent. The predicted peak value is compared with a preset breakage red line. If the predicted peak value is lower than the breakage red line, it is predicted that the current printing task can be started. If the predicted peak value is not lower than the breakage red line, it is predicted that the current printing task cannot be started.

4. The method according to claim 3, characterized in that, The step of calculating the expected wear increment on the cable conveyor caused by this task based on the trajectory length, the motion trajectory complexity, and the average moving speed includes: Obtain the unit cycle reference wear amount, which is the wear amount of the cable chain in one standard cycle calibrated using historical tests; Based on the motion trajectory complexity and the average moving speed, the trajectory complexity coefficient and speed influence coefficient are obtained by looking up a table; Based on the unit cycle reference wear amount, the trajectory complexity coefficient, the speed influence coefficient, and the trajectory length, the expected wear increment caused by this task to the cable chain is calculated, wherein the expected wear increment = unit cycle reference wear amount * trajectory complexity coefficient * speed influence coefficient * trajectory length.

5. The method according to claim 1, characterized in that, The steps of obtaining the historical risk index at the time of shutdown and the current risk index at the current moment using a risk index prediction model based on the current status data and shutdown status data include: A nonlinear compensation curve between temperature and material reference stiffness is established based on the Arrhenius equation. The current current data, current vibration data, shutdown current data, and shutdown vibration data are normalized to equivalent data at a standard reference temperature of 25 degrees Celsius to obtain the current equivalent current data, current equivalent vibration data, shutdown equivalent current data, and shutdown equivalent vibration data. Fourier transforms are performed on the current equivalent vibration data and the shutdown equivalent vibration data respectively, and the spectral energy integral value of the preset frequency band is extracted as the current dynamic vibration feature and the shutdown dynamic vibration feature; The current equivalent current data and the current dynamic vibration characteristics are input into the risk index prediction model to obtain the current risk index. The shutdown equivalent current data and the shutdown dynamic vibration characteristics are input into the risk index prediction model to obtain the historical risk index.

6. The method according to claim 1, characterized in that, The step of responding to the apparent state regression anomaly and calculating the recoverable risk value based on the operating parameters before shutdown and the shutdown duration, wherein the operating parameters include continuous operating duration, average ambient temperature, and total cumulative bending count, includes: In response to apparent state rollback anomalies, obtain the continuous working time before shutdown, average ambient temperature, total cumulative number of bends, and shutdown time; Calculate the maximum recoverable amplitude coefficient based on the continuous working time, the average ambient temperature, and the total number of cumulative bends; The recoverable risk level is calculated according to the exponential decay formula, where H = K_max * (1 - exp(− t_stop / tau)), H is the recoverable risk level, K_max is the maximum recoverable magnitude coefficient, t_stop is the downtime, and tau is the characteristic constant of the cable chain material.

7. The method according to claim 1, characterized in that, The step of adding the current risk index to the recoverable risk index to obtain the corrected irreversible damage baseline index further includes: The modified irreversible damage baseline index is compared with the historical risk index; If the modified irreversible damage baseline index is less than the historical risk index, then the modified irreversible damage baseline index is forcibly assigned the value of the historical risk index plus a constant preset based on the cumulative number of bends, so as to satisfy the law of monotonically increasing fatigue damage over time.

8. The method according to claim 1, characterized in that, Following the step of predicting 3D printer drag chain line failures using a dual-branch approach based on the corrected irreversible damage substrate index and the current risk index, the method further includes: After the print job restarts, the real-time risk level index is continuously calculated and recorded to obtain the risk level index curve; Calculate the first derivative of the risk index curve and monitor its upward slope; When the first derivative falls back and becomes parallel to the historical degradation slope before shutdown, the degradation period is determined to have ended, and the true risk index at the end of the degradation period is recorded. The actual risk level index is compared with the modified irreversible damage baseline index, and the deviation value is calculated; If the deviation value exceeds the preset deviation threshold, the Kalman filter gain algorithm or gradient descent method is used to perform secondary retrospective correction on the irreversible damage baseline index in the historical database.

9. The method according to claim 1, characterized in that, In response to the long-interval shutdown event, the steps for acquiring the status data of the 3D printer's cable chain include: The phase current waveform data of the stepper motor is collected by current transformers and voltage transformers to obtain current data and shutdown current data; Vibration signals during the movement of the cable conveyor are collected using a triaxial accelerometer to obtain current vibration data and vibration data during shutdown. Temperature data of the cable carrier is collected by thermocouples to obtain the current ambient temperature and the shutdown ambient temperature.

10. A 3D printer cable chain fault prediction system, characterized in that, include: The downtime acquisition module is used to acquire the downtime of the 3D printer, wherein the downtime is the time interval between the current downtime and the last downtime. The long-interval downtime event identification module is used to identify the existence of a long-interval downtime event when the downtime duration exceeds a preset stress relaxation time threshold. The status data acquisition module is used to acquire the status data of the 3D printer drag chain line in response to the long-interval shutdown event. The status data includes current status data and shutdown status data. The current status data includes current current data, current vibration data and current ambient temperature. The shutdown status data includes shutdown current data, shutdown vibration data and shutdown ambient temperature. The risk index acquisition module is used to acquire the historical risk index at the time of shutdown and the current risk index at the current moment based on the current status data and shutdown status data, using the risk index prediction model. The risk index prediction model is a model for predicting risk index based on the status data of the 3D printer drag chain line. The difference calculation module is used to calculate the difference between the current risk level indicator and the historical risk level indicator; The apparent state regression anomaly judgment module is used to determine that the 3D printer's drag chain has an apparent state regression anomaly when the difference is negative and the absolute value exceeds a preset risk threshold. The recoverable risk value calculation module is used to calculate the recoverable risk value in response to the apparent state rollback anomaly, based on the working parameters before shutdown and the shutdown duration. The working parameters include continuous working duration, average ambient temperature and total number of cumulative bends. The module for obtaining the modified irreversible damage baseline index is used to add the current risk index to the recoverable risk index to obtain the modified irreversible damage baseline index; The prediction module is used to predict 3D printer drag chain line failures using a dual-branch approach based on the corrected irreversible damage baseline index and the current risk level index. The dual-branch approach includes a long-term damage trend branch and an immediate safety determination branch.

Images