Self-learning belt deviation adjustment method, device, equipment and storage medium

Abstract

The present application relates to the technical field of offset adjustment, and particularly relates to a self-learning belt offset adjustment method, device, equipment and storage medium, the method forms a self-learning closed loop of detection, adjustment, feedback and optimization through a complete process of real-time position detection, motor adjustment, reference compensation amount calculation, standard position optimization and positioning instruction generation; after each offset adjustment, not only is the current deviation corrected, but also adjustment experience is accumulated by updating and optimizing the standard position, so that subsequent adjustment is more in line with the actual running state of the belt; that is, the standard position is gradually corrected by the reference compensation amount, so as to ensure that the reference value is always matched with the current belt state, and the problem that a fixed reference value cannot adapt to slow time-varying characteristics is solved; and the self-learning mechanism replaces the traditional mode of relying on artificial preset parameters, reduces the dependence on fixed parameters, improves the intelligent level and anti-interference ability of belt offset adjustment, and is particularly suitable for dynamic interference adaptation in a high-speed running scene.

Description

Technical Field

[0001] This invention relates to the field of offset adjustment technology, and in particular to a self-learning belt offset adjustment method, apparatus, device, and storage medium. Background Technology

[0002] In the field of industrial automation, stationary corrugated box printing machines are key equipment in packaging production. The transmission system of their printing units directly affects printing accuracy and production efficiency. Currently, the industry mainly adopts two technical solutions: roller transmission and belt transmission. Among them, belt transmission has become the mainstream transmission method due to its significant advantages such as low energy consumption and strong transmission stability.

[0003] However, when the belt is running at high speed, typically at a linear speed of 200m / min, it is susceptible to multi-dimensional external disturbances, including dynamic tension fluctuations, frame mechanical vibrations, and changes in ambient temperature and humidity, which can cause belt misalignment. When the misalignment exceeds 2mm, it will directly lead to printing registration errors exceeding the standard.

[0004] In existing technologies, belt-correcting motor systems with PID closed-loop control are commonly used. These systems adjust the belt tension on one side in real time through feedback from displacement sensors. However, this control mode relies on fixed PID parameters and preset reference values, and lacks the ability to adapt to slow time-varying characteristics such as belt wear and environmental drift. Its stability tends to decrease after long-term operation.

[0005] It is evident that existing technologies still need improvement and enhancement. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a self-learning belt offset adjustment method that covers the entire process of detection, judgment, adjustment, update and positioning, avoids the discontinuity problem of only adjusting without calibration in traditional solutions, and ensures that the adjustment action is continuous and accurate.

[0007] The first aspect of this invention provides a self-learning belt offset adjustment method, wherein multiple offset detection probes are arranged on one side of the belt; the method includes: acquiring real-time status information of the multiple offset detection probes; confirming the real-time position of the belt based on the real-time status information, wherein the real-time position is a normal position or an offset position, and the offset position includes an offset direction and an offset degree; when the real-time position is an offset position, confirming a motor adjustment amount based on the offset degree and a preset offset adjustment range corresponding to the offset degree; generating a motor adjustment command based on the confirmed motor adjustment amount and a preset standard position; confirming a reference compensation amount based on the confirmed motor adjustment amount, and updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position; when the real-time position returns to the normal position, generating a motor positioning command based on the optimized standard position.

[0008] Optionally, in a first implementation of the first aspect of the present invention, five offset detection probes are provided on the left side of the belt, namely a first probe, a second probe, a third probe, a fourth probe, and a fifth probe. The five offset detection probes are arranged at equal intervals along the width direction of the belt, and the top detection surface of each probe is flush with the bottom mounting reference surface. The real-time position of the belt is confirmed based on real-time status information. The real-time position is either a normal position or an offset position. The offset position includes the offset direction and the offset degree, including: when the first probe and the second probe are in an active state, and the fourth probe and the fifth probe are in an off state, the real-time position of the belt is a normal position; when the first probe is in an active state, and... When the second, third, fourth, and fifth probes are off, the real-time position of the belt is a slight rightward offset. If, on top of this slight rightward offset, the first, second, third, fourth, and fifth probes are all off, the real-time position of the belt is a severe rightward offset. When the first, second, third, and fourth probes are active, and the fifth probe is off, the real-time position of the belt is a slight leftward offset. If, on top of this slight leftward offset, the first, second, third, fourth, and fifth probes are all active, the real-time position of the belt is a severe leftward offset.

[0009] Optionally, in a second implementation of the first aspect of the present invention, the step of determining the motor adjustment amount based on the degree of offset and a preset offset adjustment range corresponding to the degree of offset when the real-time position is an offset position includes: obtaining a preset offset adjustment range based on the degree of offset when the real-time position is an offset position, wherein the preset offset adjustment range includes a preset slight offset adjustment range and a preset severe offset adjustment range, and the preset severe offset adjustment range is greater than the preset slight offset adjustment range; determining the motor adjustment amount based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset, wherein the motor adjustment amount is any value within the preset offset adjustment range, and the motor adjustment amount includes a slight offset adjustment amount and a severe offset adjustment amount.

[0010] Optionally, in a third implementation of the first aspect of the present invention, the preset severe offset adjustment range is 0.8 mm to 1 mm, and the preset mild offset adjustment range is 0.2 mm to 0.4 mm.

[0011] Optionally, in a fourth implementation of the first aspect of the present invention, generating a motor adjustment command based on the confirmed motor adjustment amount and a preset standard position includes: when the offset position is a slight rightward offset or a severe rightward offset, confirming a motor adjustment position based on the confirmed motor adjustment amount and the preset standard position, wherein the motor adjustment position is the sum of the motor adjustment amount and the preset standard position, and generating a motor adjustment command based on the confirmed motor adjustment position; when the offset position is a slight leftward offset or a severe leftward offset, confirming a motor adjustment position based on the confirmed motor adjustment amount and the preset standard position, wherein the motor adjustment position is the difference between the motor adjustment amount and the preset standard position, and generating a motor adjustment command based on the confirmed motor adjustment position.

[0012] Optionally, in the fifth implementation of the first aspect of the present invention, the step of confirming the reference compensation amount based on the confirmed motor adjustment amount includes: when the offset position is a slight rightward offset, confirming a slight reference compensation amount based on the motor adjustment amount, wherein the value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount; when the offset position is a severe rightward offset, confirming a severe reference compensation amount based on the motor adjustment amount, wherein the value range of the severe reference compensation amount is slight offset adjustment amount - slight reference compensation amount < severe reference compensation amount < severe offset adjustment amount - slight reference compensation amount; when the offset position is a slight leftward offset, confirming a slight reference compensation amount based on the motor adjustment amount, wherein the value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount; when the offset position is a severe leftward offset, confirming a severe reference compensation amount based on the motor adjustment amount, wherein the value range of the severe reference compensation amount is slight offset adjustment amount - slight reference compensation amount < severe reference compensation amount < severe offset adjustment amount - slight reference compensation amount.

[0013] Optionally, in a sixth implementation of the first aspect of the present invention, the step of updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position includes: when the offset position is a slight rightward offset, updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the sum of the preset standard position and the slight reference compensation amount; when the offset position is a severe rightward offset, updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the sum of the preset standard position and the severe reference compensation amount; when the offset position is a slight leftward offset, updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the difference between the preset standard position and the slight reference compensation amount; and when the offset position is a severe leftward offset, updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the difference between the preset standard position and the severe reference compensation amount.

[0014] A second aspect of the present invention provides a self-learning belt offset adjustment device, comprising: an acquisition module for acquiring real-time status information of multiple offset detection probes; a position confirmation module for confirming the real-time position of the belt based on the real-time status information, wherein the real-time position is a normal position or an offset position, and the offset position includes an offset direction and an offset degree; an adjustment amount confirmation module for confirming a motor adjustment amount based on the offset degree and a preset offset adjustment range corresponding to the offset degree when the real-time position is an offset position; a first generation module for generating a motor adjustment command based on the confirmed motor adjustment amount and a preset standard position; a compensation amount confirmation module for confirming a reference compensation amount based on the confirmed motor adjustment amount and updating the preset standard position based on the confirmed reference compensation amount to obtain an optimized standard position; and a second generation module for generating a motor positioning command based on the optimized standard position when the real-time position returns to the normal position.

[0015] A third aspect of the present invention provides a self-learning belt offset adjustment device, the self-learning belt offset adjustment device comprising: a memory and at least one processor, the memory storing instructions; the at least one processor calling the instructions in the memory to cause the self-learning belt offset adjustment device to perform the steps of the self-learning belt offset adjustment method described in any of the preceding claims.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the self-learning belt offset adjustment method described in any of the preceding claims.

[0017] In the technical solution of this invention, a complete process of real-time position detection, motor adjustment, reference compensation calculation, standard position optimization, and positioning command generation is formed, creating a self-learning closed loop of detection, adjustment, feedback, and optimization. After each offset adjustment, not only is the current correction completed, but adjustment experience is also accumulated by updating and optimizing the standard position, making subsequent adjustments more in line with the actual operating state of the belt. That is, the standard position is gradually corrected and optimized through the reference compensation amount, ensuring that the reference value always matches the current belt state, solving the problem that fixed reference values ​​cannot adapt to slow time-varying characteristics, and significantly improving long-term operational stability. Moreover, by replacing the traditional mode that relies on manually preset parameters with a self-learning mechanism, the dependence on fixed parameters is reduced, and the intelligence level and anti-interference ability of belt offset adjustment are improved, which is especially suitable for dynamic interference adaptation in high-speed operating scenarios. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the self-learning belt offset adjustment method provided in this embodiment of the invention;

[0019] Figure 2This is a schematic diagram of the structure of the self-learning belt offset adjustment device provided in an embodiment of the present invention;

[0020] Figure 3 This is a schematic diagram of the structure of the self-learning belt offset adjustment device provided in an embodiment of the present invention. Detailed Implementation

[0021] This invention provides a self-learning belt offset adjustment method, apparatus, device, and storage medium. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0022] In an embodiment of the present invention, the belt detection mechanism for the transmission system includes five offset detection probes, which are disposed on the left side of the belt along the belt's running direction. The five offset detection probes are designated as a first probe, a second probe, a third probe, a fourth probe, and a fifth probe. The five offset detection probes are arranged at equal intervals along the belt width direction, with the top detection surface of each probe flush with the bottom mounting reference surface to ensure consistent detection height and avoid detection deviations due to installation errors. By detecting the combination of the active and off states of the five probes, the real-time position of the belt is determined, providing a direct basis for subsequent motor adjustments and a precise position feedback foundation for the self-learning belt offset adjustment method. This is particularly suitable for dynamic offset monitoring requirements in high-speed operating scenarios. Specifically, when a probe is in an active state, it indicates that the probe is covered by the belt; when a probe is in an off state, it indicates that the probe is not covered by the belt.

[0023] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the self-learning belt offset adjustment method in this invention includes:

[0024] 101. Obtain real-time status information from multiple offset detection probes;

[0025] In this embodiment, the real-time working status of multiple offset detection probes on one side of the belt is collected, including active and off states, to provide raw data for position determination.

[0026] 102. Confirm the real-time position of the belt based on real-time status information. The real-time position is either the normal position or the offset position. The offset position includes the offset direction and the offset degree.

[0027] In this embodiment, the current position of the belt is determined based on the probe status. If it is in an off-center position, the direction of the offset, such as left or right offset, and the degree of offset, such as slight or severe offset, are further determined based on the working status of the offset detection probe.

[0028] 103. When the real-time position is an offset position, the motor adjustment amount is confirmed based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset;

[0029] In this embodiment, the specific adjustment range of the correction motor, i.e., the motor adjustment amount, is determined based on the offset position, the degree of offset, and the preset corresponding adjustment range.

[0030] 104. Generate motor adjustment instructions based on the confirmed motor adjustment amount and preset standard position;

[0031] In this embodiment, a command to drive the motor is generated based on a preset standard position, i.e., the initial reference position of the motor, and combined with the motor adjustment amount.

[0032] 105. Based on the confirmed motor adjustment amount, confirm the reference compensation amount, and update the preset standard position based on the confirmed reference compensation amount to obtain the optimized standard position;

[0033] In this embodiment, a reference compensation amount is calculated based on the motor adjustment amount, and the preset standard position is updated based on the calculated reference compensation amount to obtain an optimized standard position that adapts to the current belt state. The updating of the reference compensation amount is the key difference from traditional PID control: the traditional solution uses a fixed reference, while this technical solution dynamically optimizes the reference through the reference compensation amount, achieving a self-learning effect where the reference is more in line with reality with each adjustment; and it avoids the discontinuity problem of adjusting without calibration in the traditional solution, ensuring that the adjustment action is continuous and accurate, and improving the long-term operational stability of the equipment.

[0034] 106. When the real-time position returns to the normal position, generate a motor positioning command based on the optimized standard position;

[0035] In this embodiment, when the belt returns from the offset position to the normal position, that is, when the first and second probes are active and the fourth and fifth probes are off, a motor positioning command is generated based on the optimized standard position to drive the correction motor to position to the optimized standard position. At the same time, the system records the positioning residual and dynamically updates the learning rate of the reference compensation amount according to the recorded positioning residual. For example, the learning rate is reduced when the residuals are continuously small, and increased when they are large, thus completing one closed-loop adjustment.

[0036] This application discloses a self-learning belt offset adjustment method. Through a complete process of real-time position detection, motor adjustment, reference compensation calculation, standard position optimization, and positioning command generation, a self-learning closed loop of detection, adjustment, feedback, and optimization is formed. After each offset adjustment, not only is the current correction completed, but adjustment experience is also accumulated by updating and optimizing the standard position, making subsequent adjustments more closely match the actual operating state of the belt. Specifically, the standard position is gradually corrected and optimized through reference compensation, ensuring that the reference value always matches the current belt state. This solves the problem that fixed reference values ​​cannot adapt to slow time-varying characteristics, significantly improving long-term operational stability. Furthermore, by replacing the traditional mode that relies on manually preset parameters with a self-learning mechanism, the dependence on fixed parameters is reduced, improving the intelligence level and anti-interference capability of belt offset adjustment, making it particularly suitable for adapting to dynamic interference in high-speed operating scenarios.

[0037] In this embodiment of the invention, the real-time position of the belt is confirmed based on real-time status information. The real-time position is either a normal position or an offset position. The offset position includes the offset direction and the offset degree, including:

[0038] 201. When the first and second probes are active and the fourth and fifth probes are off, the real-time position of the belt is the normal position.

[0039] 202. When the first probe is active and the second, third, fourth and fifth probes are off, the real-time position of the belt is the offset position, specifically a slight rightward offset; if the first, second, third, fourth and fifth probes are all off on top of the slight rightward offset, then the real-time position of the belt is a severe rightward offset.

[0040] 203. When the first, second, third, and fourth probes are active and the fifth probe is off, the real-time position of the belt is the offset position, specifically a slight leftward offset; if the first, second, third, fourth, and fifth probes are all active on top of the slight leftward offset, then the real-time position of the belt is a severe leftward offset.

[0041] In this embodiment, five offset detection probes are arranged sequentially as the first to the fifth probes along the width of the belt, with equal spacing between adjacent probes. This ensures that the degree of offset corresponds linearly to the change in the number of activated probes, avoiding misjudgment of the degree of offset due to uneven spacing.

[0042] The relative position of each probe's detection surface to the belt edge defines its working state: the active state corresponds to the belt edge covering the detection surface, and the deactivated state corresponds to the belt edge detaching from the detection surface.

[0043] The offset direction is defined by the relative motion relationship between the probe array and the belt: right offset indicates that the belt moves away from the probe array, and left offset indicates that the belt moves closer to the probe array. This definition is directly mapped to the adjustment direction of the correction motor.

[0044] In this embodiment, based on the multi-dimensional state combination of five probes, a three-level quantitative distinction is achieved between normal position, slight deviation and severe deviation, overcoming the limitation of traditional single probe or dual probe systems that can only achieve binary direction judgment, and improving the accuracy of deviation degree recognition.

[0045] In this embodiment of the invention, when the real-time position is an offset position, determining the motor adjustment amount based on the degree of offset and a preset offset adjustment range corresponding to the degree of offset includes:

[0046] 301. When the real-time position is an offset position, a preset offset adjustment range is obtained based on the degree of offset. The preset offset adjustment range includes a preset slight offset adjustment range and a preset severe offset adjustment range. The preset severe offset adjustment range is greater than the preset slight offset adjustment range.

[0047] In this embodiment, preset mild offset adjustment ranges and preset severe offset adjustment ranges are set for different degrees of offset, and the value of the severe adjustment range is greater than that of the mild adjustment range. When the belt deviates slightly, the amount of belt deviation is small, so the mild adjustment range needs to be small and precise. When the belt deviates significantly, the amount of belt deviation is large, so the severe adjustment range needs to be large and effective. This avoids the situation where a small adjustment amount cannot correct the severe offset or a large adjustment amount causes the mild offset to be overcorrected, thereby improving the success rate of a single adjustment.

[0048] 302. Based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset, determine the motor adjustment amount, wherein the motor adjustment amount is any value within the preset offset adjustment range, and the motor adjustment amount includes the slight offset adjustment amount and the severe offset adjustment amount;

[0049] In this embodiment, the motor adjustment amount is any value within a preset adjustment range, i.e., a slight deviation corresponds to a slight deviation adjustment amount, and a severe deviation corresponds to a severe deviation adjustment amount. In practical applications, the adjustment amount can be dynamically selected based on the belt material and tension characteristics. For example, rubber belts have high elasticity, so the slight deviation adjustment amount can be taken as the upper limit of the range, such as 0.4mm; canvas belts have high rigidity, so the lower limit of the range can be taken, such as 0.2mm. That is, the selection of the motor adjustment amount can be adapted to different working conditions without redesigning the algorithm due to changes in belt material and running speed, thus reducing equipment adaptation costs.

[0050] In this embodiment, the preset severe offset adjustment range is 0.8mm to 1mm, and the preset mild offset adjustment range is 0.2mm to 0.4mm. The mild adjustment range can correct slight deviations that are within tolerance, while the severe adjustment range can correct severe deviations that are close to tolerance. Both ensure that the deviation amount after adjustment is ≤2mm, directly avoiding excessive registration errors caused by insufficient adjustment and improving the pass rate of printed products. Moreover, the 0.2-1mm adjustment range covers the belt deviation control requirements of mainstream fixed corrugated carton printing machines, eliminating the need to reset values ​​for specific models and shortening the on-site debugging cycle.

[0051] In this embodiment of the invention, generating the motor adjustment command based on the confirmed motor adjustment amount and the preset standard position includes:

[0052] 401. When the offset position is a slight rightward offset or a severe rightward offset, the motor adjustment position is confirmed based on the confirmed motor adjustment amount and the preset standard position. The motor adjustment position is the sum of the motor adjustment amount and the preset standard position. A motor adjustment command is generated according to the confirmed motor adjustment position.

[0053] 402. When the offset position is a slight leftward offset or a severe leftward offset, the motor adjustment position is confirmed based on the confirmed motor adjustment amount and the preset standard position. The motor adjustment position is the difference between the motor adjustment amount and the preset standard position. A motor adjustment command is generated based on the confirmed motor adjustment position.

[0054] In this embodiment, the preset standard position refers to the initial reference position of the correction motor, which is usually calibrated and set according to the center position of the belt during equipment installation. For right-side offset, whether it is a slight or severe right-side offset, the motor adjustment position = preset standard position + motor adjustment amount, that is, the motor moves away from the belt center, and the belt tension is adjusted to push the belt back to the left. For left-side offset, whether it is a slight or severe left-side offset, the motor adjustment position = preset standard position - motor adjustment amount, that is, the motor moves closer to the belt center, and the belt back to the right. Finally, based on the calculated motor adjustment position, a command to drive the motor is generated. The format of the generated motor adjustment command must match the motor type, such as the step number-direction command for a stepper motor and the position pulse command for a servo motor.

[0055] In this embodiment, a deterministic mapping relationship between the offset direction and the motor adjustment direction is established through a mapping algorithm of right offset incremental compensation and left offset decrement compensation, eliminating the risk of reverse offset caused by misjudgment of direction logic in traditional systems; the logic of this mapping algorithm has linear interpretability and can complete the adjustment calculation in microseconds, meeting the real-time adjustment requirements of high-speed belt operation scenarios.

[0056] In this embodiment of the invention, the step of confirming the benchmark compensation amount based on the confirmed motor adjustment amount includes:

[0057] 501. When the offset position is a slight rightward offset, a slight reference compensation amount is determined based on the motor adjustment amount. The value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount.

[0058] 502. When the offset position is a severe rightward offset, the severe reference compensation amount is determined based on the motor adjustment amount. The value range of the severe reference compensation amount is: slight offset adjustment amount - slight reference compensation amount < severe reference compensation amount < severe offset adjustment amount - slight reference compensation amount.

[0059] 503. When the offset position is a slight leftward offset, a slight reference compensation amount is determined based on the motor adjustment amount. The value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount.

[0060] 504. When the offset position is a severe leftward offset, the severe reference compensation amount is determined based on the motor adjustment amount. The value range of the severe reference compensation amount is: slight offset adjustment amount - slight reference compensation amount < severe reference compensation amount < severe offset adjustment amount - slight reference compensation amount.

[0061] In this embodiment, the left and right offset compensation values ​​are symmetrical, differing only in the direction coefficient during the update of the preset standard position: left offset compensation corresponds to a negative direction coefficient, and right offset compensation corresponds to a positive direction coefficient. The reference compensation value is defined as a parameter for fine-tuning the preset standard position, and its value must satisfy: the reference compensation value < the motor adjustment value corresponding to the degree of offset. If the reference compensation value ≥ the motor adjustment value, it will cause excessive offset of the reference in the next adjustment. Taking a slight adjustment of 0.4mm, a slight compensation value ΔC_light = 0.2mm, and a severe adjustment of 1mm as an example, the severe compensation... The compensation amount ΔC_heavy has a range of 0.2mm < ΔC_heavy < 0.8mm. This range is determined by the formula ΔC_light < ΔC_heavy < (heavy adjustment amount - ΔC_light). This ensures the baseline adjustment range after heavy offset while avoiding baseline jumps caused by disconnection from the light compensation amount. The baseline compensation amount needs to be dynamically adjusted according to the belt offset return effect: when the belt returns to center stably after light offset adjustment, the light compensation amount takes the midpoint of the range, such as 0.2mm; when there are still slight fluctuations after centering, the upper limit of the range is taken to further optimize the baseline.

[0062] In this embodiment, by limiting the range of values ​​between the reference compensation amount and the motor adjustment amount and the value linked to the slight compensation amount, the reference drift caused by excessively large reference compensation amount or the ineffective compensation caused by excessively small reference compensation amount is prevented, thus ensuring the effectiveness and appropriateness of the reference update. The reference compensation amount serves as the core parameter carrier of the self-learning closed loop. By optimizing the preset standard position through the compensation amount after each adjustment, the reference value gradually adapts to the actual operating state of the belt, thereby realizing the dynamic calibration function.

[0063] In this embodiment of the invention, updating the preset standard position based on the confirmed benchmark compensation amount to obtain the optimized standard position includes:

[0064] 601. When the offset position is a slight rightward offset, the preset standard position is updated based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the sum of the preset standard position and the slight reference compensation amount.

[0065] 602. When the offset position is a severe rightward offset, the preset standard position is updated based on the confirmed benchmark compensation amount to obtain an optimized standard position, wherein the optimized standard position is the sum of the preset standard position and the severe benchmark compensation amount;

[0066] 603. When the offset position is a slight leftward offset, the preset standard position is updated based on the confirmed reference compensation amount to obtain an optimized standard position, wherein the optimized standard position is the difference between the preset standard position and the slight reference compensation amount.

[0067] 604. When the offset position is a severe leftward offset, the preset standard position is updated based on the confirmed benchmark compensation amount to obtain an optimized standard position, wherein the optimized standard position is the difference between the preset standard position and the severe benchmark compensation amount.

[0068] In this embodiment, the calculation rule for the optimized standard position is defined as follows: For slight right-side offset and severe right-side offset, the optimized standard position = preset standard position + corresponding reference compensation amount, which is adjusted by shifting the reference to the right to adapt to the stable operating state after the belt deviates to the right; For slight left-side offset and severe left-side offset, the optimized standard position = preset standard position - corresponding reference compensation amount, which is adjusted by shifting the reference to the left to adapt to the stable operating state after the belt deviates to the left.

[0069] The updated optimized standard position serves as the reference value for the next belt offset adjustment, achieving iterative self-optimization of the reference. The update of the optimized standard position is synchronized in real time with the generation of the motor adjustment command. That is, the reference value is updated synchronously when the motor adjustment action is started, ensuring that the belt is accurately positioned to the updated reference position when it returns to the correct position, avoiding positioning deviations caused by the adjustment process and the reference update being out of sync.

[0070] In this embodiment, the technical problem that traditional fixed references cannot adapt to slow time-varying characteristics is solved by the mechanism of synchronously updating the reference value (standard position) with each offset adjustment: Specifically, when the belt wears down due to long-term operation and the tension changes, the reference value can be gradually corrected by accumulating multiple compensation updates without manual recalibration; the optimized reference value is dynamically adapted to the actual operating state of the belt, which improves the accuracy of motor adjustment calculation for subsequent offset adjustments and reduces the number of adjustment iterations; the reference self-optimization mechanism extends the equipment maintenance cycle to 1 to 3 months, reduces downtime for calibration, and improves production continuity.

[0071] The self-learning belt offset adjustment method in the embodiments of the present invention has been described above. The self-learning belt offset adjustment device in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the self-learning belt offset adjustment device in this invention includes:

[0072] The acquisition module 701 is used to acquire real-time status information of multiple offset detection probes;

[0073] The position confirmation module 702 is used to confirm the real-time position of the belt based on real-time status information. The real-time position is either a normal position or an offset position, and the offset position includes the offset direction and the offset degree.

[0074] The adjustment amount confirmation module 703 is used to confirm the motor adjustment amount based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset when the real-time position is an offset position;

[0075] The first generation module 704 is used to generate motor adjustment instructions based on the confirmed motor adjustment amount and the preset standard position;

[0076] The compensation amount confirmation module 705 is used to confirm the reference compensation amount based on the confirmed motor adjustment amount, and update the preset standard position based on the confirmed reference compensation amount to obtain the optimized standard position.

[0077] The second generation module 706 is used to generate motor positioning commands based on the optimized standard position when the real-time position returns to the normal position.

[0078] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0079] above Figure 2 The self-learning belt offset adjustment device in this embodiment of the invention is described in detail from the perspective of modular functional entities. The self-learning belt offset adjustment device in this embodiment of the invention is described in detail below from the perspective of hardware processing.

[0080] Figure 3This is a schematic diagram of the structure of a self-learning belt offset adjustment device 800 provided in an embodiment of the present invention. The self-learning belt offset adjustment device 800 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 810 (e.g., one or more processors) and a memory 820, and one or more storage media 830 (e.g., one or more mass storage devices) storing application programs 833 or data 832. The memory 820 and storage media 830 can be temporary or persistent storage. The program stored in the storage media 830 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the self-learning belt offset adjustment device 800. Furthermore, the processor 810 may be configured to communicate with the storage media 830 and execute the series of instruction operations in the storage media 830 on the self-learning belt offset adjustment device 800 to implement the steps of the self-learning belt offset adjustment method provided in the above-described method embodiments.

[0081] The self-learning belt offset adjustment device 800 may also include one or more power supplies 840, one or more wired or wireless network interfaces 850, one or more input / output interfaces 860, and / or one or more operating systems 831, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The illustrated structure of the self-learning belt offset adjustment device does not constitute a limitation on the self-learning belt offset adjustment device, which may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0082] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of the self-learning belt offset adjustment method.

[0083] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0084] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A self-learning belt offset adjustment method, characterized in that, Five offset detection probes are installed on one side of the belt, namely, probe 1, probe 2, probe 3, probe 4, and probe 5. The five offset detection probes are arranged at equal intervals along the width direction of the belt, and the top detection surface of each probe is flush with the bottom mounting reference surface; the method includes: Acquire real-time status information from multiple offset detection probes; The real-time position of the belt is confirmed based on real-time status information. This real-time position is either a normal position or an off-center position, where the off-center position includes both the direction and degree of deviation. Specifically, when the first and second probes are active, and the fourth and fifth probes are off, the belt's real-time position is normal. When the first probe is active, and the second, third, fourth, and fifth probes are off, the belt's real-time position is off-center, specifically a slight rightward deviation. If, in addition to a slight rightward deviation, all four probes are off, the belt's real-time position is severely off-center. When the first, second, third, and fourth probes are active, and the fifth probe is off, the belt's real-time position is off-center, specifically a slight leftward deviation. If, in addition to a slight leftward deviation, all four probes are active, the belt's real-time position is severely off-center. When the real-time position is an offset position, the motor adjustment amount is determined based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset. Specifically, when the real-time position is an offset position, the preset offset adjustment range is obtained based on the degree of offset. The preset offset adjustment range includes a preset slight offset adjustment range and a preset severe offset adjustment range, where the preset severe offset adjustment range is greater than the preset slight offset adjustment range. Based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset, the motor adjustment amount is determined. The motor adjustment amount is any value within the preset offset adjustment range, and the motor adjustment amount includes slight offset adjustment amounts and severe offset adjustment amounts. Motor adjustment commands are generated based on the confirmed motor adjustment amount and preset standard position; The reference compensation amount is determined based on the confirmed motor adjustment amount, and the preset standard position is updated based on the confirmed reference compensation amount to obtain the optimized standard position. When the real-time position returns to the normal position, a motor positioning command is generated based on the optimized standard position.

2. The self-learning belt offset adjustment method according to claim 1, characterized in that, The preset heavy offset adjustment range is 0.8mm to 1mm, and the preset light offset adjustment range is 0.2mm to 0.4mm.

3. The self-learning belt offset adjustment method according to claim 1, characterized in that, The generation of motor adjustment commands based on the confirmed motor adjustment amount and preset standard position includes: When the offset position is a slight rightward offset or a severe rightward offset, the motor adjustment position is confirmed based on the confirmed motor adjustment amount and the preset standard position. The motor adjustment position is the sum of the motor adjustment amount and the preset standard position, and a motor adjustment command is generated according to the confirmed motor adjustment position. When the offset position is a slight leftward offset or a severe leftward offset, the motor adjustment position is confirmed based on the confirmed motor adjustment amount and the preset standard position. The motor adjustment position is the difference between the motor adjustment amount and the preset standard position, and a motor adjustment command is generated according to the confirmed motor adjustment position.

4. The self-learning belt offset adjustment method according to claim 1, characterized in that, The determination of the benchmark compensation amount based on the confirmed motor adjustment amount includes: When the offset position is a slight rightward offset, a slight reference compensation amount is determined based on the motor adjustment amount. The value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount. When the offset position is a severe rightward offset, the severe reference compensation amount is determined based on the motor adjustment amount. The value range of the severe reference compensation amount is: mild offset adjustment amount - mild reference compensation amount < severe reference compensation amount < severe offset adjustment amount - mild reference compensation amount. When the offset position is a slight leftward offset, a slight reference compensation amount is determined based on the motor adjustment amount. The value range of the slight reference compensation amount is 0 < slight reference compensation amount < slight offset adjustment amount. When the offset position is a severe leftward offset, the severe reference compensation amount is determined based on the motor adjustment amount. The value range of the severe reference compensation amount is: slight offset adjustment amount - slight reference compensation amount < severe reference compensation amount < severe offset adjustment amount - slight reference compensation amount.

5. The self-learning belt offset adjustment method according to claim 4, characterized in that, The step of updating the preset standard position based on the confirmed benchmark compensation amount to obtain the optimized standard position includes: When the offset position is a slight rightward offset, the preset standard position is updated based on the confirmed reference compensation amount to obtain an optimized standard position, which is the sum of the preset standard position and the slight reference compensation amount. When the offset position is a severe rightward offset, the preset standard position is updated based on the confirmed benchmark compensation amount to obtain an optimized standard position, which is the sum of the preset standard position and the severe benchmark compensation amount. When the offset position is a slight leftward offset, the preset standard position is updated based on the confirmed reference compensation amount to obtain an optimized standard position, which is the difference between the preset standard position and the slight reference compensation amount. When the offset position is a severe leftward offset, the preset standard position is updated based on the confirmed baseline compensation amount to obtain an optimized standard position, which is the difference between the preset standard position and the severe baseline compensation amount.

6. A self-learning belt offset adjustment device, characterized in that, include: The acquisition module is used to acquire real-time status information of multiple offset detection probes; The position confirmation module is used to confirm the real-time position of the belt based on real-time status information. The real-time position is either a normal position or an off-center position, and the off-center position includes the direction and degree of offset. Specifically, when the first and second probes are active, and the fourth and fifth probes are off, the real-time position of the belt is the normal position. When the first probe is active, and the second, third, fourth, and fifth probes are off, the real-time position of the belt is an off-center position, specifically a slight rightward offset. If, in addition to a slight rightward offset, the first, second, third, fourth, and fifth probes are all off, the real-time position of the belt is a severe rightward offset. When the first, second, third, and fourth probes are active, and the fifth probe is off, the real-time position of the belt is an off-center position, specifically a slight leftward offset. If, in addition to a slight leftward offset, the first, second, third, fourth, and fifth probes are all active, the real-time position of the belt is a severe leftward offset. The adjustment amount confirmation module is used to confirm the motor adjustment amount based on the degree of offset and a preset offset adjustment range corresponding to the degree of offset when the real-time position is an offset position. Specifically, when the real-time position is an offset position, the preset offset adjustment range is obtained based on the degree of offset. The preset offset adjustment range includes a preset slight offset adjustment range and a preset severe offset adjustment range, wherein the preset severe offset adjustment range is greater than the preset slight offset adjustment range. Based on the degree of offset and the preset offset adjustment range corresponding to the degree of offset, the motor adjustment amount is confirmed. The motor adjustment amount is any value within the preset offset adjustment range, and the motor adjustment amount includes slight offset adjustment amounts and severe offset adjustment amounts. The first generation module is used to generate motor adjustment instructions based on the confirmed motor adjustment amount and the preset standard position; The compensation amount confirmation module is used to confirm the reference compensation amount based on the confirmed motor adjustment amount, and update the preset standard position based on the confirmed reference compensation amount to obtain the optimized standard position. The second generation module is used to generate motor positioning commands based on the optimized standard position when the real-time position returns to the normal position.

7. A self-learning belt offset adjustment device, characterized in that, The self-learning belt offset adjustment device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the self-learning belt offset adjustment device to perform the steps of the self-learning belt offset adjustment method as described in any one of claims 1-5.

8. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the self-learning belt offset adjustment method as described in any one of claims 1-5.

Images