A predictive maintenance method, apparatus, device, medium and program product of a magnetic drive conveying system

Abstract

The application relates to the technical field of magnetic drive conveying, in particular to a predictive maintenance method, device, equipment, medium and program product of a magnetic drive conveying system. The method comprises the following steps: obtaining running data from a shared memory of a target system, the shared memory comprising a real-time execution shared memory, a scheduling shared memory and a log shared memory; constructing an object-level health vector according to the running data, the object-level health vector being used for representing the health state of a single running object; processing the object-level health vectors of multiple historical periods based on a pre-trained time sequence convolution network model to obtain the risk type and the risk level of each running object; generating and executing an active degradation control action according to the risk type and the risk level until the risk state of the running object meets a preset safe operation constraint condition. By using the method, passive response can be changed into active maintenance, and the safety of the system can be improved.

Description

Technical Field

[0001] This application relates to the field of magnetic drive conveying technology, and in particular to a predictive maintenance method, apparatus, equipment, medium, and program product for a magnetic drive conveying system. Background Technology

[0002] Magnetic drive conveyor systems are intelligent conveying devices that achieve non-contact driving based on the principle of electromagnetic coupling. Their basic components include a stator coil array and a mover (or trolley) with permanent magnets. When the stator is energized, it generates a traveling wave magnetic field, driving the mover along a guide rail. By controlling the amplitude and phase of the stator's current, the position and speed of the mover can be closed-loop regulated. Compared to traditional mechanical transmission methods (such as belt conveyors, chain conveyors, and ball screws), magnetic drive conveyor systems offer significant advantages such as zero-contact wear, high motion accuracy, high acceleration, excellent cleanliness, and flexible, reconfigurable paths. Based on these characteristics, magnetic drive conveyor systems have been widely used in industrial production scenarios with high requirements for cleanliness and motion accuracy, such as new energy battery assembly, semiconductor wafer transfer, pharmaceutical filling, 3C electronic product assembly, and food packaging.

[0003] In typical magnetic drive conveyor systems, a hierarchical control architecture is usually adopted to meet multiple requirements of real-time performance, determinism, and operational flexibility. The bottom layer is the real-time execution control layer, primarily responsible for hard real-time tasks such as mover position closed-loop control, speed planning, interpolation calculations, driver status acquisition, synchronous zone motion coordination, and low-level input / output linkage. Its control cycle is typically in the millisecond range to ensure the smoothness of mover movement and positioning accuracy. The middle layer is the scheduling and decision-making layer, responsible for workstation flow logic, process route scheduling, path switching management, synchronous zone queuing and release control, RFID card reading confirmation, trolley number reconfirmation, and coordination between the mobile platform or shuttle section. This layer needs to balance real-time response with the flexibility of complex decision-making. The top layer is the host computer and service layer, mainly used for functions such as operation log recording, alarm push notifications, system parameter management, operation status monitoring, and human-machine configuration interaction.

[0004] Building upon the aforementioned layered architecture, an increasing number of magnetic drive conveyor systems are adopting shared memory as the core mechanism for inter-module data interaction. Shared memory is an efficient inter-process communication method that allows multiple modules to simultaneously read and write to the same physical memory region, avoiding data copying and network protocol stack overhead, and meeting the requirements of magnetic drive conveyor systems for low-latency, high-frequency data exchange.

[0005] However, current maintenance methods for magnetic drive conveyor systems have the following technical problems:

[0006] Although existing systems can record a large amount of status and alarm information, their processing methods are mostly still in the passive stage of alarming after a fault occurs, which has serious lag and needs to be optimized. Summary of the Invention

[0007] Therefore, it is necessary to provide a predictive maintenance method, apparatus, computer equipment, computer-readable storage medium, and computer program product for a magnetic drive conveyor system that can shift from passive response to proactive maintenance and improve system safety, in order to address the above-mentioned technical problems.

[0008] In a first aspect, this application provides a predictive maintenance method for a magnetic drive conveyor system. The method includes:

[0009] Runtime data is obtained from the shared memory of the target system, wherein the shared memory includes real-time execution shared memory, scheduling shared memory, and log shared memory;

[0010] Based on the operational data, an object-level health vector is constructed, which is used to characterize the health status of a single running object.

[0011] The object-level health vectors from multiple historical periods are processed using a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object.

[0012] Based on the risk type and risk level, generate and execute proactive degradation control actions until the risk status of the running object meets the preset safe operation constraints.

[0013] In one embodiment, the running object includes at least one of a mover, a road segment, a synchronization zone, a card reader link, a controller boundary handover object, and a mobile platform or shuttle segment object. The step of constructing an object-level health vector based on the running data, wherein the object-level health vector characterizes the health status of a single running object, includes:

[0014] Multiple object-level health vectors within the same target area are aggregated to obtain a region-level health vector, which is used to characterize the overall health status of multiple running objects within the target area.

[0015] All the regional health vectors are aggregated to obtain a global health vector, which is used to characterize the overall health status of the target system.

[0016] In one embodiment, the pre-trained temporal convolutional network model processes the object-level health vectors from multiple historical periods to obtain the risk type and risk level of each running object, including:

[0017] Obtain the probability vector output by the temporal convolutional network model, which includes normal states and various abnormal types;

[0018] The risk type of the running object is determined based on the anomaly type corresponding to the maximum probability value in the probability vector;

[0019] The risk score corresponding to the risk type is determined based on the probability value of the normal state, and the risk level is determined based on the comparison result between the risk score and the preset threshold.

[0020] In one embodiment, generating and executing proactive degradation control actions based on the risk type and risk level until the risk status of the running object meets preset safe operation constraints includes:

[0021] Based on the risk type, the risk level, the process priority of the operating object, and the current context information, a preset strategy mapping relationship is matched to generate the corresponding proactive degradation control action;

[0022] The active degradation control actions include at least one of the following: reducing the target speed, restricting entry into the synchronization zone, triggering secondary card reading, guiding to the maintenance station, setting path flow restriction, and performing local isolation.

[0023] In one embodiment, generating and executing proactive degradation control actions based on the risk type and risk level until the risk status of the running object meets preset safe operation constraints includes:

[0024] In response to the risk level reaching or exceeding the preset moderately abnormal level for multiple consecutive prediction periods, the active downgrade control action is executed.

[0025] During the prediction period following the execution of the proactive downgrade control action, if the risk level continuously decreases to a mildly abnormal level or a normal level, the proactive downgrade control action is gradually lifted.

[0026] In one embodiment, when the running object includes an RFID card reader link object, the step of constructing an object-level health vector based on the running data, wherein the object-level health vector is used to characterize the health status of a single running object, includes:

[0027] Obtain the operational data of the RFID card reader link object, and calculate the RFID health score based on the operational data. The operational data includes card reading success rate, average reread rate, and frequency of reconfirmation of vehicle number.

[0028] In response to the RFID health score being lower than a preset threshold, the active degradation control action is executed. The active degradation control action includes increasing the card reading trigger frequency, enabling secondary card reading, or adding vehicle number context consistency verification.

[0029] Secondly, this application also provides a predictive maintenance device for a magnetic drive conveyor system. The device includes:

[0030] The data acquisition module is used to obtain runtime data from the shared memory of the target system, wherein the shared memory includes real-time execution shared memory, scheduling shared memory and log shared memory;

[0031] A vector construction module is used to construct an object-level health vector based on the running data, wherein the object-level health vector is used to characterize the health status of a single running object;

[0032] The risk prediction module is used to process the object-level health vectors of multiple historical periods based on a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object.

[0033] The maintenance execution module is used to generate and execute proactive degradation control actions based on the risk type and risk level until the risk status of the running object meets the preset safe operation constraints.

[0034] In one embodiment, the running object includes at least one of a mover, a road segment, a synchronization zone, a card reader link, a controller boundary handover object, and a mobile platform or shuttle segment object, and the vector construction module is further used for:

[0035] Multiple object-level health vectors within the same target area are aggregated to obtain a region-level health vector, which is used to characterize the overall health status of multiple running objects within the target area.

[0036] All the regional health vectors are aggregated to obtain a global health vector, which is used to characterize the overall health status of the target system.

[0037] In one embodiment, the risk prediction module is further configured to:

[0038] Obtain the probability vector output by the temporal convolutional network model, which includes normal states and various abnormal types;

[0039] The risk type of the running object is determined based on the anomaly type corresponding to the maximum probability value in the probability vector;

[0040] The risk score corresponding to the risk type is determined based on the probability value of the normal state, and the risk level is determined based on the comparison result between the risk score and the preset threshold.

[0041] In one embodiment, the maintenance execution module is further configured to:

[0042] Based on the risk type, the risk level, the process priority of the operating object, and the current context information, a preset strategy mapping relationship is matched to generate the corresponding proactive degradation control action;

[0043] The active degradation control actions include at least one of the following: reducing the target speed, restricting entry into the synchronization zone, triggering secondary card reading, guiding to the maintenance station, setting path flow restriction, and performing local isolation.

[0044] In one embodiment, the maintenance execution module is further configured to:

[0045] In response to the risk level reaching or exceeding the preset moderately abnormal level for multiple consecutive prediction periods, the active downgrade control action is executed.

[0046] During the prediction period following the execution of the proactive downgrade control action, if the risk level continuously decreases to a mildly abnormal level or a normal level, the proactive downgrade control action is gradually lifted.

[0047] In one embodiment, when the running object includes an RFID reader link object, the vector construction module is further configured to:

[0048] Obtain the operational data of the RFID card reader link object, and calculate the RFID health score based on the operational data. The operational data includes card reading success rate, average reread rate, and frequency of reconfirmation of vehicle number.

[0049] In response to the RFID health score being lower than a preset threshold, the active degradation control action is executed. The active degradation control action includes increasing the card reading trigger frequency, enabling secondary card reading, or adding vehicle number context consistency verification.

[0050] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a predictive maintenance method for a magnetic drive conveyor system as described in any embodiment of the first aspect.

[0051] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of a predictive maintenance method for a magnetic drive conveyor system as described in any embodiment of the first aspect.

[0052] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of a predictive maintenance method for a magnetic drive conveyor system as described in any embodiment of the first aspect.

[0053] The aforementioned predictive maintenance method, apparatus, computer equipment, storage medium, and computer program product for a magnetic drive conveyor system, derived from the technical features in the embodiments, can achieve the following beneficial effects to address the technical problems in the background art:

[0054] This embodiment provides a predictive maintenance method for a magnetic drive conveyor system. In implementation, by directly acquiring operational data from shared memory, it fully utilizes the existing data resources of the magnetic drive conveyor system, eliminating the need for additional sensors or significant system architecture modifications, resulting in low engineering implementation costs and low barriers to entry. By constructing object-level health vectors, the original operational parameters are transformed into structured health status representations, providing standardized input for subsequent predictive models. Based on this, a temporal convolutional network model is used to process health vectors from multiple historical periods, capturing the temporal dependencies of operational data. This enables early identification of progressive risks such as degradation of various objects (movers, road segments, synchronization zones, and RFID links), regional congestion, and synchronization anomalies, upgrading traditional passive alarms after faults to proactive prediction before faults occur. By generating and executing proactive degradation control actions based on risk type and level, a complete automated closed loop from risk discovery to intervention is formed, effectively reducing the risk of equipment downtime and production line chain congestion. Furthermore, by continuously assessing the risk status of the operating objects until the safe operating constraints are met, the loss of production line capacity due to overly conservative control is avoided. While ensuring equipment safety, the efficient operation of the production line is maintained, and the intelligent maintenance level of the system and the overall capacity utilization rate are improved. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1This is a schematic diagram of the first process of a predictive maintenance method for a magnetic drive conveyor system in one embodiment;

[0057] Figure 2 This is a second flowchart illustrating a predictive maintenance method for a magnetically driven conveyor system in another embodiment;

[0058] Figure 3 This is a schematic diagram of the third process of a predictive maintenance method for a magnetic drive conveyor system in another embodiment;

[0059] Figure 4 This is a schematic diagram of the fourth process of a predictive maintenance method for a magnetic drive conveyor system in another embodiment;

[0060] Figure 5 This is a schematic diagram of the fifth process of a predictive maintenance method for a magnetic drive conveyor system in another embodiment;

[0061] Figure 6 This is a sixth flowchart of a predictive maintenance method for a magnetic drive conveyor system in another embodiment;

[0062] Figure 7 This is a structural block diagram of a predictive maintenance device for a magnetic drive conveyor system in one embodiment;

[0063] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0064] To make the objectives, technical solutions, 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.

[0065] In one embodiment, such as Figure 1 As shown, a predictive maintenance method for a magnetic drive conveyor system is provided. This embodiment illustrates the application of this method to a terminal within the magnetic drive conveyor system. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0066] Step 102: Obtain runtime data from the shared memory of the target system.

[0067] The shared memory includes real-time execution shared memory, scheduling shared memory, and log shared memory. Specifically, real-time execution shared memory provides millisecond-level operational data such as position, speed, acceleration, following error, driver status, and synchronization motion status. Scheduling shared memory provides scheduling data such as vehicle target, road segment occupancy, waiting count, workstation flow, release status, route switching results, and vehicle number reconfirmation results. The log shared area provides log data such as errors, warnings, event timestamps, and abnormal growth trends.

[0068] Step 104: Based on the running data, construct an object-level health vector, which is used to characterize the health status of a single running object.

[0069] For example, the terminal can be used to aggregate the collected raw data according to object dimensions and time windows to construct quantifiable health characteristics. Preferably, the sampling period is set to... The length of the sliding window is Each window contains One sampling point.

[0070] Step 106: Process the object-level health vectors from multiple historical periods based on a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object.

[0071] Among them, the temporal convolutional network model can refer to a convolutional neural network architecture specifically designed for processing time series data. Its core feature is the use of a structural design that combines causal convolution and dilated convolution.

[0072] Among them, risk type is used to characterize the specific abnormal mode category currently faced by the operating object. In this embodiment, based on the most common failure modes with progressive degradation characteristics in the actual operation of magnetic drive conveyor systems, the risk types are divided into the following four categories: The first category is drive degradation anomalies, which mainly reflect the decline in the electromagnetic performance of the mover or driver. Typical manifestations include continuously increasing following error, aggravated speed fluctuations, and sluggish acceleration response. This type of anomaly is usually related to stator coil aging, permanent magnet demagnetization, or performance degradation of driver power devices. The second category is path or area congestion anomalies, which mainly reflect poor traffic flow on the conveyor path. Typical manifestations include persistently high road occupancy rate, increased mover waiting time ratio, and drastic fluctuations in queue length. This type of anomaly is usually related to upstream workstations. The first category is related to rhythm fluctuations, unreasonable path switching strategies, or improper release control parameter settings; the second category is synchronization zone anomalies, which mainly reflect the coordination problems of multiple movers in synchronous motion mode. Typical manifestations include excessive synchronization following error, increased binding failure rate, and synchronization zone queuing backlog. This type of anomaly is usually related to synchronization control parameter mismatch, mechanical installation deviation, or sensor signal noise; the third category is RFID link degradation anomalies, which mainly reflect the decline in communication quality of the radio frequency identification link. Typical manifestations include decreased card reading success rate, increased rereading times, and frequent reconfirmation of vehicle number. This type of anomaly is usually related to antenna installation misalignment, tag contamination, environmental electromagnetic interference, or aging of radio frequency modules.

[0073] The risk level can be used to quantitatively assess the severity of abnormal states of operating objects, providing a basis for the differentiated selection of subsequent proactive degradation control actions. In this embodiment, the system first calculates the risk score based on the 5-dimensional probability vector output by the TCN model. Specifically, since the first dimension p0 in the probability vector represents the probability that the object is in a normal state, the risk score of the object is defined as R = 1 - p0, that is, the probability value of the object deviating from the normal state, with a score range of 0 to 1.

[0074] For example, based on the above risk scores, the system can classify risks into four levels according to preset thresholds: Level L0 corresponds to 0≤R<0.30, indicating that the system is healthy and its operating status is within the normal fluctuation range, requiring no intervention; Level L1 corresponds to 0.30≤R<0.55, indicating a slight anomaly, where the system enters observation mode, only recording the abnormal trend but not performing proactive degradation actions; Level L2 corresponds to 0.55≤R<0.80, indicating a moderate anomaly, where the system performs flexible degradation control actions, that is, preserving production capacity as much as possible while ensuring safety; Level L3 corresponds to 0.80≤R≤1.00, indicating a severe anomaly, where the system performs strong protection actions, including partial isolation, protective shutdown, or triggering manual confirmation. The above thresholds can be flexibly adjusted according to the different emphasis requirements of safety and production capacity of specific production lines, which will not be elaborated here.

[0075] Step 108: Based on the risk type and the risk level, generate and execute proactive degradation control actions until the risk status of the running object meets the preset safe operation constraints.

[0076] Among them, proactive degradation control actions refer to the intervention instructions that the system actively issues to the scheduling and execution layers after predicting that there is an abnormal trend in the running object. Its purpose is to take preventive measures before the fault actually occurs, control the risk within an acceptable range, and at the same time avoid causing excessive disturbance to the normal production process.

[0077] The preset safe operation constraints can refer to a set of judgment criteria used to determine whether the system has recovered to an acceptable operating state through active degradation control. In this invention, the constraints are not a single static indicator, but a composite judgment condition that integrates risk level recovery and duration of continuous stability.

[0078] For example, after executing proactive degradation control actions, the system can continue to assess the risk level of the target object in each subsequent forecast cycle. The system determines that the risk status of the operating object meets the safe operation constraints when the following conditions are simultaneously met: first, the risk level of the target object in the current forecast cycle has decreased to L1 (mild anomaly) or L0 (normal); second, the aforementioned low-risk status remains stable over multiple consecutive forecast cycles without repeated jumps. After meeting these conditions, the system initiates a recovery process, gradually releasing the previously applied proactive degradation control actions, such as gradually restoring the target speed of speed-limited objects to the normal set value, or gradually lifting traffic restrictions on paths. If the risk level rises again during the release process, the system immediately reapplies degradation control to ensure that safety always takes priority. This mechanism effectively avoids the "bounce effect" of risk rebound caused by withdrawing degradation measures too quickly, and also prevents capacity loss caused by the system maintaining overly conservative control for a long period due to a single, occasional fluctuation.

[0079] In the above-mentioned predictive maintenance method for a magnetic drive conveyor system, reasonable derivation based on the technical features in the embodiments achieves the beneficial effect of solving the technical problems raised in the background art:

[0080] This embodiment provides a predictive maintenance method for a magnetic drive conveyor system. In implementation, by directly acquiring operational data from shared memory, it fully utilizes the existing data resources of the magnetic drive conveyor system, eliminating the need for additional sensors or significant system architecture modifications, resulting in low engineering implementation costs and low barriers to entry. By constructing object-level health vectors, the original operational parameters are transformed into structured health status representations, providing standardized input for subsequent predictive models. Based on this, a temporal convolutional network model is used to process health vectors from multiple historical periods, capturing the temporal dependencies of operational data. This enables early identification of progressive risks such as degradation of various objects (movers, road segments, synchronization zones, and RFID links), regional congestion, and synchronization anomalies, upgrading traditional passive alarms after faults to proactive prediction before faults occur. By generating and executing proactive degradation control actions based on risk type and level, a complete automated closed loop from risk discovery to intervention is formed, effectively reducing the risk of equipment downtime and production line chain congestion. Furthermore, by continuously assessing the risk status of the operating objects until the safe operating constraints are met, the loss of production line capacity due to overly conservative control is avoided. While ensuring equipment safety, the efficient operation of the production line is maintained, and the intelligent maintenance level of the system and the overall capacity utilization rate are improved.

[0081] In one embodiment, it can be as follows Figure 2 As shown, the operating object includes at least one of the following: a moving vehicle, a road segment, a synchronization zone, a card reader link, a controller boundary handover object, and a mobile platform or shuttle segment object. Step 104 includes:

[0082] Step 202: Aggregate multiple object-level health vectors within the same target area to obtain a region-level health vector. The region-level health vector is used to characterize the overall health status of multiple running objects within the target area.

[0083] Step 204: Aggregate all the regional health vectors to obtain a global health vector, which is used to characterize the overall health status of the target system.

[0084] In this embodiment, by aggregating object-level health vectors into regional-level health vectors and global health vectors, a multi-level health assessment system is constructed, ranging from a single operating object to a local area and then to the entire system. This enables the system to not only perceive anomalies in individual actuators or devices, but also to identify regional congestion and systemic degradation trends, providing a more comprehensive basis for predictive maintenance and improving the macroscopic and accurate nature of risk perception.

[0085] In one embodiment, it can be as follows Figure 3 As shown, step 106 includes:

[0086] Step 302: Obtain the probability vector output by the temporal convolutional network model, which includes normal states and various abnormal types.

[0087] Step 304: Determine the risk type of the running object based on the anomaly type corresponding to the maximum probability value in the probability vector.

[0088] Step 306: Determine the risk score corresponding to the risk type based on the probability value of the normal state, and determine the risk level based on the comparison result of the risk score and the preset threshold.

[0089] For example, in one specific embodiment, the structure of the TCN model provided in this application can be as follows: The input layer receives a temporal input matrix consisting of 50 historical windows (each window contains 16-dimensional object-level health features); the first layer is a causal one-dimensional convolutional layer with a kernel size of 3, an output channel number of 32, and a dilation coefficient of 1; the second layer is a residual convolutional block with a kernel size of 3, a channel number of 32, and a dilation coefficient of 2; the third layer is a residual convolutional block with a kernel size of 3, a channel number of 16, and a dilation coefficient of 4; then, the temporal dimension is compressed through a global average pooling layer; finally, a 5-dimensional Softmax probability vector is output by a fully connected layer. The five dimensions of this probability vector correspond to normal state, driving degradation anomalies, path or area congestion anomalies, synchronization zone anomalies, and RFID link degradation anomalies, respectively.

[0090] In this embodiment, a probability vector containing normal states and various anomaly types is output through a temporal convolutional network model. The risk type is determined based on the maximum probability value, and the risk score is calculated and the risk level is divided based on the probability value of the normal state. This enables accurate identification and hierarchical quantification of various progressive anomalies (such as drive degradation, regional congestion, synchronization anomalies, and RFID link degradation), providing a clear and quantifiable decision basis for subsequent differentiated proactive degradation control.

[0091] In one embodiment, it can be as follows Figure 4 As shown, step 108 includes:

[0092] Step 402: Based on the risk type, the risk level, the process priority of the running object, and the current context information, match the preset strategy mapping relationship to generate the corresponding active degradation control action.

[0093] The active degradation control actions include at least one of the following: reducing the target speed, restricting entry into the synchronization zone, triggering secondary card reading, guiding to the maintenance station, setting path flow restriction, and performing local isolation.

[0094] In this embodiment, by comprehensively considering risk type, risk level, process priority and context information, a preset strategy mapping relationship is matched to generate targeted proactive degradation control actions, thereby achieving differentiated graded intervention. At the same time, a variety of optional actions, including rate limiting, access prohibition, supplementary reading, guidance, flow limiting and isolation, are provided, enabling the system to flexibly select the optimal mitigation strategy according to abnormal scenarios, effectively curbing the deterioration of risks while maximizing the retention of production line throughput capacity.

[0095] In one embodiment, it can be as follows Figure 5 As shown, step 108 includes:

[0096] Step 502: In response to the risk level reaching or exceeding the preset moderate abnormality level in multiple consecutive prediction periods, the active downgrade control action is executed.

[0097] Step 504: During the prediction period after the execution of the active downgrade control action, if the risk level continuously decreases to the mildly abnormal level or the normal level, the active downgrade control action is gradually lifted.

[0098] In this embodiment, by setting a condition that active downgrading is only triggered when the risk level reaches moderate or above for multiple consecutive prediction cycles, the false triggering caused by single prediction jitter is effectively suppressed, improving the stability and reliability of control. At the same time, after the downgrading action is executed, the control is gradually released according to the continuous change of the risk level, avoiding the risk rebound caused by the downgrading measures being withdrawn too quickly or the production line efficiency being affected by the withdrawal being too slow. A flexible balance is achieved between downgrading intervention and recovery withdrawal, ensuring safety while taking into account the continuity of system operation and optimal production capacity.

[0099] In one embodiment, it can be as follows Figure 6 As shown, when the running object includes an RFID card reader link object, step 104 includes:

[0100] Step 602: Obtain the operating data of the RFID card reading link object, and calculate the RFID health score based on the operating data. The operating data includes the card reading success rate, average reread rate, and frequency of reconfirmation of the vehicle number.

[0101] Step 604: In response to the RFID health score being lower than a preset threshold, execute the active degradation control action, which includes increasing the card reading trigger frequency, enabling secondary card reading, or adding a vehicle number context consistency check.

[0102] In this embodiment, addressing the pain point that RFID card reading links in magnetic drive conveyor systems are susceptible to electromagnetic interference, tag damage, or installation deviations, leading to card reading failures or incorrect numbering, a comprehensive assessment of RFID health scores is constructed to evaluate card reading success rate, average reread rate, and the frequency of reconfirmation of vehicle numbers. This enables a specific quantitative perception of RFID link degradation risks. When the health score falls below a threshold, flexible mitigation actions such as supplementary reading, secondary card reading, and contextual consistency verification are triggered, rather than directly classifying the object as having an incorrect number or triggering a shutdown. This effectively reduces the risk of material traceability interruptions and production line stoppages caused by occasional RFID anomalies, and improves the system's fault tolerance and self-healing capabilities against RFID link degradation.

[0103] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0104] Based on the same inventive concept, this application also provides a predictive maintenance device for a magnetic drive conveyor system for implementing the predictive maintenance method for a magnetic drive conveyor system described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the predictive maintenance device for a magnetic drive conveyor system provided below can be found in the limitations of the predictive maintenance method for a magnetic drive conveyor system described above, and will not be repeated here.

[0105] In one embodiment, such as Figure 7 As shown, a predictive maintenance device for a magnetic drive conveyor system is provided, comprising: a data acquisition module, a vector construction module, a risk prediction module, and a maintenance execution module, wherein:

[0106] The data acquisition module is used to obtain runtime data from the shared memory of the target system, wherein the shared memory includes real-time execution shared memory, scheduling shared memory and log shared memory;

[0107] A vector construction module is used to construct an object-level health vector based on the running data, wherein the object-level health vector is used to characterize the health status of a single running object;

[0108] The risk prediction module is used to process the object-level health vectors of multiple historical periods based on a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object.

[0109] The maintenance execution module is used to generate and execute proactive degradation control actions based on the risk type and risk level until the risk status of the running object meets the preset safe operation constraints.

[0110] In one embodiment, the running object includes at least one of a mover, a road segment, a synchronization zone, a card reader link, a controller boundary handover object, and a mobile platform or shuttle segment object, and the vector construction module is further used for:

[0111] Multiple object-level health vectors within the same target area are aggregated to obtain a region-level health vector, which is used to characterize the overall health status of multiple running objects within the target area.

[0112] All the regional health vectors are aggregated to obtain a global health vector, which is used to characterize the overall health status of the target system.

[0113] In one embodiment, the risk prediction module is further configured to:

[0114] Obtain the probability vector output by the temporal convolutional network model, which includes normal states and various abnormal types;

[0115] The risk type of the running object is determined based on the anomaly type corresponding to the maximum probability value in the probability vector;

[0116] The risk score corresponding to the risk type is determined based on the probability value of the normal state, and the risk level is determined based on the comparison result between the risk score and the preset threshold.

[0117] In one embodiment, the maintenance execution module is further configured to:

[0118] Based on the risk type, the risk level, the process priority of the operating object, and the current context information, a preset strategy mapping relationship is matched to generate the corresponding proactive degradation control action;

[0119] The active degradation control actions include at least one of the following: reducing the target speed, restricting entry into the synchronization zone, triggering secondary card reading, guiding to the maintenance station, setting path flow restriction, and performing local isolation.

[0120] In one embodiment, the maintenance execution module is further configured to:

[0121] In response to the risk level reaching or exceeding the preset moderately abnormal level for multiple consecutive prediction periods, the active downgrade control action is executed.

[0122] During the prediction period following the execution of the proactive downgrade control action, if the risk level continuously decreases to a mildly abnormal level or a normal level, the proactive downgrade control action is gradually lifted.

[0123] In one embodiment, when the running object includes an RFID reader link object, the vector construction module is further configured to:

[0124] Obtain the operational data of the RFID card reader link object, and calculate the RFID health score based on the operational data. The operational data includes card reading success rate, average reread rate, and frequency of reconfirmation of vehicle number.

[0125] In response to the RFID health score being lower than a preset threshold, the active degradation control action is executed. The active degradation control action includes increasing the card reading trigger frequency, enabling secondary card reading, or adding vehicle number context consistency verification.

[0126] The modules in the predictive maintenance device of the aforementioned magnetic drive conveyor system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0127] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a predictive maintenance method for a magnetic drive conveyor system. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0128] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0129] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0130] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0131] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0132] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0133] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0134] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0135] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A predictive maintenance method for a magnetically driven conveyor system, characterized in that, The method includes: Runtime data is obtained from the shared memory of the target system, wherein the shared memory includes real-time execution shared memory, scheduling shared memory, and log shared memory; Based on the operational data, an object-level health vector is constructed, which is used to characterize the health status of a single running object. The object-level health vectors from multiple historical periods are processed using a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object. Based on the risk type and risk level, generate and execute proactive degradation control actions until the risk status of the running object meets the preset safe operation constraints.

2. The method according to claim 1, characterized in that, The operating objects include at least one of the following: mover, road segment, synchronization zone, card reader link, controller boundary handover object, and mobile platform or shuttle segment object. The step of constructing an object-level health vector based on the operating data, wherein the object-level health vector is used to characterize the health status of a single operating object, includes: Multiple object-level health vectors within the same target area are aggregated to obtain a region-level health vector, which is used to characterize the overall health status of multiple running objects within the target area. All the regional health vectors are aggregated to obtain a global health vector, which is used to characterize the overall health status of the target system.

3. The method according to claim 2, characterized in that, The pre-trained temporal convolutional network model processes the object-level health vectors from multiple historical periods to obtain the risk type and risk level of each running object, including: Obtain the probability vector output by the temporal convolutional network model, which includes normal states and various abnormal types; The risk type of the running object is determined based on the anomaly type corresponding to the maximum probability value in the probability vector; The risk score corresponding to the risk type is determined based on the probability value of the normal state, and the risk level is determined based on the comparison result between the risk score and the preset threshold.

4. The method according to claim 1, characterized in that, The step of generating and executing proactive degradation control actions based on the risk type and risk level until the risk status of the operating object meets the preset safe operation constraints includes: Based on the risk type, the risk level, the process priority of the operating object, and the current context information, a preset strategy mapping relationship is matched to generate the corresponding proactive degradation control action; The active degradation control actions include at least one of the following: reducing the target speed, restricting entry into the synchronization zone, triggering secondary card reading, guiding to the maintenance station, setting path flow restriction, and performing local isolation.

5. The method according to any one of claims 1 to 4, characterized in that, The step of generating and executing proactive degradation control actions based on the risk type and risk level until the risk status of the operating object meets the preset safe operation constraints includes: In response to the risk level reaching or exceeding the preset moderately abnormal level for multiple consecutive prediction periods, the active downgrade control action is executed. During the prediction period following the execution of the proactive downgrade control action, if the risk level continuously decreases to a mildly abnormal level or a normal level, the proactive downgrade control action is gradually lifted.

6. The method according to claim 1, characterized in that, When the running object includes an RFID card reader link object, the step of constructing an object-level health vector based on the running data, wherein the object-level health vector is used to characterize the health status of a single running object, includes: Obtain the operational data of the RFID card reader link object, and calculate the RFID health score based on the operational data. The operational data includes card reading success rate, average reread rate, and frequency of reconfirmation of vehicle number. In response to the RFID health score being lower than a preset threshold, the active degradation control action is executed. The active degradation control action includes increasing the card reading trigger frequency, enabling secondary card reading, or adding vehicle number context consistency verification.

7. A predictive maintenance device for a magnetic drive conveyor system, characterized in that, The device includes: The data acquisition module is used to obtain runtime data from the shared memory of the target system, wherein the shared memory includes real-time execution shared memory, scheduling shared memory and log shared memory; A vector construction module is used to construct an object-level health vector based on the running data, wherein the object-level health vector is used to characterize the health status of a single running object; The risk prediction module is used to process the object-level health vectors of multiple historical periods based on a pre-trained temporal convolutional network model to obtain the risk type and risk level of each running object. The maintenance execution module is used to generate and execute proactive degradation control actions based on the risk type and risk level until the risk status of the running object meets the preset safe operation constraints.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images