Cooperative control method of photovoltaic intelligent manufacturing equipment production line

Abstract

The invention relates to the technical field of control or regulation systems, and discloses a cooperative control method for a photovoltaic intelligent manufacturing equipment production line, and the method comprises the steps: a control unit collects the stock and change rate of materials in a physical cache region in real time; establishing and mapping the physical cache region into a virtual viscoelastic dynamic model with non-Newtonian fluid characteristics; calculating a virtual elastic restoring force enabling the stock to return to a balance point and a virtual viscous damping force preventing the stock state from changing based on the model; wherein an asymmetric anisotropic damping generation strategy is executed, and a damping coefficient is dynamically split according to a material flowing trend; and finally, superposing the virtual adjustment correction after vector synthesis to the basic transmission speed to generate a dynamic speed instruction, and by constructing a virtual dynamic field with rheological characteristics and an asymmetric damping mechanism, the problem of nonlinear cascade oscillation in discrete logistics transmission is solved, and differential self-adaptive suppression of accumulation and evacuation risks is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of photovoltaic intelligent manufacturing equipment line's cooperative control method, belong to control or regulating system technical field. BACKGROUND

[0002] Current photovoltaic cell and component precision manufacturing production line is usually composed of multiple process equipment through physical buffer zone in series, and the existing stage system control generally follows the single machine independent operation to cooperate with the physical buffer coupling discrete event driven mode, each process unit is independently operated according to its own set beat, and the material transfer between devices depends on the high and low threshold signals of the material inventory of the physical buffer zone, which triggers the start-stop or multi-section speed switching of the upstream equipment. This control strategy uses the logical decoupling characteristic to maintain the basic logistics transmission of the production line under low speed or low coupling condition. The industry tries to introduce a cooperative mechanism based on data analysis, for example, the Chinese patent application with publication number CN118536953A discloses a production line production cooperation method, system, electronic device and medium. The scheme matches the production task attributes with the preset over-production rate, combines the upstream task breakage rate and the excellent rate interval to determine the over-production reason, realizes the cooperation and correction at the production task level, and the essence of the scheme is based on the discrete statistical data macro management logic, focuses on production task accounting, post-tracing and state confirmation, and the static or quasi-static cooperation based on statistical interval cannot penetrate to the bottom of the device dynamics control. In the face of high-speed transmission of ultra-thin sheets in photovoltaic production line, the method cannot respond to the sudden change of logistics flow in milliseconds, lacks real-time suppression mechanism for physical inertia and mechanical shock, and the micro transmission level of the production line is still in a rigid confrontation state.

[0003] In the face of high-throughput, ultra-thin substrate and high requirements for continuous rheological properties in modern manufacturing environment, the discrete threshold rigid control paradigm has limitations in dynamics. The existing technology regards the continuous flow characteristic material flow as a discrete particle set, and uses finite state machine switch logic to constrain the fluid characteristic flow dynamics. This control method ignores the fluid characteristics and wave conduction characteristics presented by the stage system under high-speed running state. The existing technology has the following disadvantages: discrete control leads to nonlinear cascade oscillation, because there is a lag in the mechanical inertia and control response of the upstream equipment, simple start-stop or step speed instruction makes the logistics speed switch sharply between full load and starvation state, the system lacks internal damping mechanism to absorb and dissipate disturbance energy, and extreme limit cycle oscillation leads to frequent changes in transmission speed; disturbance amplification effect, small beat fluctuation of downstream process is amplified to a large speed adjustment of upstream equipment through physical buffer zone threshold quantization processing, so that the system is difficult to maintain global dynamic balance under long-chain coupling.

[0004] Therefore, how to break through the traditional discrete threshold control rigid constraint, and construct a flexible logical connection mechanism with adaptive buffer and energy dissipation capability by converting the rigid physical connection between devices into a flexible logical connection mechanism, to realize the continuous modulation and dynamic suppression of logistics fluctuation in multi-stage serial discrete manufacturing system, has become a technical problem to be solved by the present application. Summary of the Invention

[0005] To address the problems raised in the background art, the technical solution of this invention is as follows: A collaborative control method for a photovoltaic intelligent manufacturing equipment production line, the production line comprising multiple process units connected in series via physical buffer zones, the method establishing and mapping discrete material flows within the physical buffer zones into a virtual viscoelastic dynamic model with non-Newtonian fluid characteristics, the method comprising the following steps: The control unit of each process unit collects the current material inventory and the rate of change of the material inventory in the associated physical buffer area in real time; The control unit is based on a virtual viscoelastic dynamics model to calculate the virtual elastic restoring force that returns the material inventory to the preset equilibrium point, and the virtual viscous damping force that hinders the change of the material inventory state. In calculating the virtual viscous damping force, the control unit executes an asymmetric anisotropic damping generation strategy: real-time identification of the positive and negative polarities of the rate of change of material inventory to determine the immediate trend of material flow; when the trend is determined to be material accumulation, the first damping coefficient is called to calculate the overdamping force with strong inhibition characteristics; when the trend is determined to be material dispersion, the second damping coefficient with a value less than the first damping coefficient is called to calculate the underdamping force with weak inhibition characteristics. The control unit performs vector synthesis of the virtual elastic restoring force and the virtual viscous damping force obtained based on the asymmetric anisotropic damping generation strategy to generate a virtual adjustment correction amount, and then superimposes the virtual adjustment correction amount onto the basic transmission speed of the process unit to generate a dynamic speed command to drive the transmission mechanism. By employing an asymmetric anisotropic damping generation strategy, the process unit's response to upstream logistics fluctuations exhibits a nonlinear rheological characteristic that combines rigid suppression under accumulation trends with flexible following under dispersal trends.

[0006] Preferably, the calculation logic for the virtual adjustment correction in the virtual viscoelastic dynamics model follows a linear combination relationship including state deviation terms and state change rate terms, and satisfies the following dynamic equation: ,in, Characterizing the virtual adjustment correction amount, Characterizing the virtual elastic stiffness coefficient Characterizes the current material inventory Characterizes the preset equilibrium point, Characterizing the trend of flow The first or second damping coefficient can be dynamically switched. Characterizes the rate of change of material inventory; the equation defines the dynamic convergence trajectory of the system when it deviates from steady state.

[0007] Preferably, the method further includes establishing a virtual backpressure communication mechanism for reverse transmission of material flow along the production line: the control unit receives the virtual pressure value sent by the directly downstream process unit and introduces the virtual pressure value as a feedforward suppression variable into the generation logic of the dynamic speed command; the virtual pressure value is generated by the downstream process unit based on its own current material inventory and the virtual pressure value received from the next downstream process unit through weighted accumulation; through the virtual backpressure communication mechanism, the downstream material flow blockage state is reversely diffused to the upstream in the form of a scalar field, driving the upstream process unit to perform predictive deceleration before the physical buffer is full, and using the advance transmission of information flow to exchange for the redundancy of the physical buffer space.

[0008] Preferably, the step of calculating the virtual elastic restoring force includes implementing a nonlinear boundary hardening strategy: the control unit monitors the safety margin between the material inventory and the physical limit boundary of the physical buffer zone in real time; the control unit generates a nonlinear gain coefficient based on the safety margin, which is set to maintain a constant base value when the safety margin is greater than a preset threshold, and to increase nonlinearly and monotonically when the safety margin is less than the preset threshold and approaches zero; the control unit uses the nonlinear gain coefficient to weight and amplify the virtual elastic restoring force so as to generate an exponentially enhanced repulsive potential energy field when the material inventory approaches the physical limit boundary, so that the transmission mechanism can continuously reverse the speed without external hard limit triggering.

[0009] Preferably, the method further includes a baseline speed adaptive correction step: the control unit establishes a sliding time window and calculates the statistical average value of the virtual adjustment correction within the window; when the absolute value of the statistical average value exceeds a preset dead zone threshold, the control unit identifies a systematic speed mismatch and compensates for the base transmission speed based on the statistical average value, using the corrected base transmission speed as the updated baseline value for the generation of subsequent dynamic speed commands; through the correction step, the steady-state inventory deviation caused by equipment aging or environmental drift is eliminated, so that the virtual adjustment correction automatically converges to near zero under steady-state operation, restoring the dynamic range of the system's bidirectional adjustment.

[0010] Preferably, the method further includes performing an active resonance suppression step: the control unit monitors the sign-flipping frequency of the virtual adjustment correction in the time domain to estimate the real-time oscillation frequency of the cooperative control system; when the real-time oscillation frequency falls into a preset intrinsic resonance frequency band, the control unit applies a time-varying offset to the virtual elastic stiffness coefficient in the virtual viscoelastic dynamic model to dynamically change the natural frequency of the cooperative control system to disrupt the resonance phase synchronization condition; when a resonance trend is detected, the control unit temporarily increases the weight of the virtual viscous damping force to dissipate the resonant energy in the system.

[0011] Preferably, the method further includes an inertia adaptive gain scheduling step based on material inventory: the control unit maps the material inventory in real time to an inertia factor characterizing the current load inertia of the system; the control unit dynamically adjusts the virtual elastic stiffness coefficient and virtual damping coefficient in the virtual viscoelastic dynamic model in real time based on the inertia factor and according to a preset nonlinear mapping rule; the nonlinear mapping rule is set to increase the virtual elastic stiffness coefficient and virtual damping coefficient accordingly as the inertia factor increases, so as to maintain the natural frequency and damping ratio of the cooperative control system constant under different load conditions, and ensure the consistency of the system's dynamic response characteristics.

[0012] Preferably, the acquisition of the rate of change of material inventory in the state acquisition step includes: acquiring the difference in material inventory between two adjacent sampling periods; performing low-pass filtering on the difference in material inventory to filter out high-frequency noise interference, and generating a smoothed rate of change signal as the basis for calculating the virtual viscous damping force; the cutoff frequency of the low-pass filtering is set to be lower than the minimum effective response frequency of the cooperative control system to prevent the asymmetric anisotropic damping generation strategy from being mistakenly triggered due to sensor signal jitter.

[0013] Preferably, the step of generating dynamic speed commands further includes nonlinear saturation constraint processing: the control unit presets a speed constraint range based on the minimum dwell time or maximum physical cycle time limit of the process steps executed by the process unit; when the calculated dynamic speed command exceeds the speed constraint range, the control unit clamps the dynamic speed command at the boundary value of the speed constraint range, giving priority to ensuring that the single-machine process parameters do not exceed the limit.

[0014] Preferably, the baseline speed adaptive correction step also includes state freeze logic: the control unit monitors the operating status of the production line in real time; when the production line is detected to be in a non-steady-state condition of start-up, emergency stop or fault alarm, the control unit automatically suspends the baseline speed adaptive correction step to keep the current base transmission speed unchanged and prevent transient disturbances from being incorrectly integrated into the baseline value.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In photovoltaic intelligent manufacturing equipment, a virtual viscoelastic impedance model is constructed between adjacent process units. The material inventory and change rate of the downstream physical buffer zone are mapped in real time to the virtual elastic restoring force and virtual damping suppression force of the control loop. This mechanism changes the traditional discrete manufacturing rigid control method based on threshold switch. Based on the continuous fluid rheological characteristics of discrete logistics, the virtual damping component is calculated using the material inventory change rate. A mechanism for energy dissipation of logistics fluctuations is introduced at the control algorithm layer. When a transient disturbance occurs in the upstream transmission cycle, the damping mechanism generates nonlinear antagonistic stagnation according to the severity of the disturbance, actively absorbing and attenuating the oscillation energy. This allows the system state to smoothly return to the steady-state equilibrium point without mechanical emergency stop or intervention, relying on the convergence characteristics of the control law itself, thus eliminating the nonlinear cascade oscillation phenomenon common in long-chain production lines.

[0016] 2. Establish a virtual backpressure communication mechanism that transmits pressure in reverse along the material flow direction, so that the load pressure of downstream process units gradually permeates upstream in the form of a scalar field. Information flow and material flow are coupled in reverse, so that upstream process units no longer operate blindly. By receiving backpressure signals, they can perceive the downstream blockage trend in advance and introduce it as a feedforward suppression variable into the current speed generation logic. By using information lead time to exchange for physical space redundancy, upstream equipment is driven to perform predictive deceleration adjustment before the actual physical congestion arrives. What might have been a severe impact of local stacking is transformed into a small speed coordinated modulation of all equipment on the entire line, realizing distributed absorption and spatial attenuation of flow fluctuations.

[0017] 3. When calculating the virtual damping suppression component, an asymmetric anisotropic strategy is adopted. The system damping characteristics are dynamically reconstructed according to the trend of material inventory changes. When the buffer is detected to be filling, the system automatically adapts to a high damping coefficient to construct overdamped characteristics, providing strong virtual braking force to prioritize physical safety and prevent material collisions. When the buffer is detected to be emptying, the system automatically switches to a low damping coefficient to construct underdamped characteristics, allowing virtual elastic force to dominate and achieve rapid material replenishment response. The control parameter logic is automatically switched according to the flow polarity, resolving the inherent contradiction between the steady-state accuracy and dynamic response speed of traditional linear control within a single control loop. This ensures that the production line is not blocked while having a very high ability to eliminate starvation, maximizing the system's dynamic bandwidth. Attached Figure Description

[0018] Fig. 1 This is a flowchart illustrating the collaborative control of integrated asymmetric damping and virtual back pressure regulation in this invention. Fig. 2 The virtual force and stock change rate waveforms present the asymmetric anisotropic characteristics of this invention. Fig. 3 This is a control system architecture diagram of the multidimensional suppression and adaptive correction module of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] A collaborative control method for a photovoltaic intelligent manufacturing equipment production line operates on a distributed control architecture built upon an industrial fieldbus. This fieldbus, which can employ EtherCAT or Profinet communication protocols, comprises multiple process units connected in series according to the process flow. Adjacent process units are coupled through physical buffer zones. Each process unit is equipped with an independent control unit and sensor components for monitoring the material inlet and outlet status. This method establishes a mapping relationship between the discrete material flow within the physical buffer zone and a virtual viscoelastic dynamic model through the collaboration of various functional modules, achieving continuous collaborative adjustment of the production line's material flow. The control units of each process unit collect the current material inventory within the associated physical buffer zone in real time at a preset control cycle. and the rate of change in the inventory of this material. The control cycle can be set from 2ms to 10ms, depending on the rate of change of material inventory. The control unit acquires the material inventory difference between two adjacent sampling periods and applies a low-pass filter to the difference sequence. This low-pass filtering uses a first-order hysteresis filtering algorithm or a moving average filtering algorithm, with the cutoff frequency set below the minimum effective response frequency of the coordinated control system, for example, 5Hz to 10Hz, to filter out high-frequency noise components caused by sensor jitter and generate a smoothed rate of change signal. The control unit establishes a mapping between the physical buffer and a virtual viscoelastic dynamic model with non-Newtonian fluid characteristics, and calculates the virtual adjustment correction amount used to correct the transmission speed in real time based on this model. The calculation logic follows a linear combination of state deviation terms and state change rate terms, and satisfies the following dynamic equation: ,in, This represents the preset balance point, set at 50% of the physical cache capacity. The virtual elastic stiffness coefficient is used to generate the virtual elastic restoring force that brings the material inventory back to the equilibrium point. Characterizing the trend of flow The dynamically switching damping coefficient is used to generate a virtual viscous damping force that hinders changes in the material's inventory state.

[0021] When calculating the virtual viscous damping force, the control unit executes an asymmetric anisotropic damping generation strategy, and the control unit identifies the rate of change of material inventory in real time. The positive or negative polarity is used to determine the immediate trend of material flow. When detected When the value is greater than zero, indicating a trend of material accumulation, the control unit calls the first damping coefficient. Calculate the virtual viscous damping force; the first damping coefficient The numerical setting is used to construct the overdamped control characteristics to generate a virtual resistance force that inhibits material stacking when detected. When the value is less than zero, indicating a trend of material evacuation, the control unit activates the second damping coefficient. Calculate the virtual viscous damping force; the second damping coefficient The value is less than the first damping coefficient. This is used to construct underdamped control characteristics, allowing virtual elastic restoring force to dominate the regulation process; the control unit executes a virtual back pressure communication mechanism that transmits pressure in the reverse direction of the material flow along the production line, and receives virtual pressure values ​​sent by the direct downstream process unit via the industrial bus. This value is then used as a feedforward suppression variable in the dynamic speed command generation logic, and the local virtual pressure value is... The downstream process unit is determined based on its current material inventory. The received downstream virtual pressure values ​​are generated through the following weighted accumulation logic: ,in, This is the local load weighting coefficient. This is the pressure transmission attenuation coefficient, with a value ranging from 0 to 1; To determine the physical buffer's maximum capacity, this mechanism drives upstream process units to detect remote congestion and implement deceleration adjustments before the physical buffer reaches its limit. The control unit executes a nonlinear boundary hardening strategy to calculate virtual elastic restoring force, and monitors material inventory in real time. Safety margin between physical buffer limits and physical buffer boundaries Based on this safety margin The control unit uses hyperbolic or exponential functions to generate nonlinear gain coefficients. The nonlinear gain coefficient The setting rule is: when the safety margin When the value exceeds a preset threshold, the coefficient remains at a constant base value; when the safety margin... When the value is less than a preset threshold and approaches zero, the coefficient exhibits a non-linear monotonically increasing trend. The control unit utilizes this non-linear gain coefficient to weight and amplify the virtual elastic restoring force. .

[0022] The control unit performs a baseline speed adaptive correction step, sets a sliding time window with a length of 30 to 60 seconds, and calculates the virtual adjustment correction amount within this window. Statistical average ,when When the absolute value exceeds the preset dead zone threshold, the control unit adjusts the base transmission speed based on this statistical average. Compensation and correction can be performed, for example, by using a formula. ,in The corrected base transmission speed, calculated using the integral learning rate, serves as the updated baseline value for generating subsequent dynamic speed commands. This step also includes state freeze logic: when the production line is detected to be in a non-steady-state condition such as startup, emergency stop, or fault alarm, the execution of this correction step is paused, maintaining the current base transmission speed unchanged. The control unit executes a resonance active suppression step, monitoring the virtual adjustment correction amount. The sign-flipping frequency in the time domain is used to estimate the real-time oscillation frequency of the cooperative control system. When this real-time oscillation frequency falls into the preset intrinsic resonance frequency band, the control unit adjusts the virtual elastic stiffness coefficient in the virtual viscoelastic dynamic model. A time-varying offset is applied to change the natural frequency of the cooperative control system; simultaneously, the control unit temporarily increases the weight of the virtual viscous damping force, and executes an inertia adaptive gain scheduling step based on material inventory, mapping the material inventory in real time to an inertia factor characterizing the current load inertia of the system. The virtual elastic stiffness coefficient and virtual damping coefficient are dynamically adjusted in real time according to a preset nonlinear mapping rule, which is set as follows: with the inertia factor... The increase in virtual elastic stiffness coefficient and virtual damping coefficient will correspondingly increase the virtual elastic stiffness coefficient to maintain the natural frequency and damping ratio of the coordinated control system constant.

[0023] The control unit performs vector synthesis of the virtual elastic restoring force and the virtual viscous damping force obtained based on the asymmetric anisotropic damping generation strategy to generate a virtual adjustment correction. This correction is then added to the base transfer speed of the process unit. Generate dynamic speed commands to drive the transmission mechanism. The dynamic speed command undergoes nonlinear saturation constraint processing before output: the control unit presets a speed constraint range based on the minimum dwell time or maximum physical cycle time limit of the process steps executed by the process unit; when the calculated dynamic speed command exceeds this range, the control unit clamps its value at the boundary value.

[0024] Example 1: In a cascaded high-throughput photovoltaic cell production scenario including screen printing, high-temperature sintering, and automated inspection units, the collaborative control method of this invention is applied to solve the nonlinear cascaded oscillation problem caused by process cycle mismatch. The production line is designed with a single-line capacity of 7200 cells / hour. Adjacent process units are coupled through a physical buffer zone with a capacity of 500 cells. Under continuous operation, due to the reciprocating periodic characteristics of the upstream screen printing machine's squeegee movement, its output cycle exhibits periodic fluctuations with a frequency of approximately 0.5Hz, and the amplitude of this fluctuation reaches ±1 of the average cycle time. 5%, while the downstream detection unit is limited by the processing latency of the visual algorithm, and the receiving cycle has a random jitter of about 200ms. This dynamic mismatch between upstream and downstream cycles, under the traditional discrete threshold control mode, can easily cause the material inventory in the physical buffer to switch drastically between full load and empty state in a short period of time, causing frequent emergency stops and full-speed starts of the transmission mechanism, which in turn causes the microcrack rate of ultrathin silicon wafers (thickness less than 110μm) to rise to more than 0.5%. When the system faces the above conditions, the control unit of this invention collects the current material inventory in the associated physical buffer in real time. and the rate of change of material inventory The control unit uses the constructed virtual viscoelastic dynamics model to calculate the virtual adjustment correction amount in real time. In response to the periodic discharge surges generated by the upstream screen printing machine, namely When the system exhibits large positive fluctuations, its inherent asymmetric anisotropic damping generation strategy is triggered, and the control unit recognizes this. The accumulation trend automatically selects the first damping coefficient with a larger value. This overdamping characteristic immediately generates a strong virtual viscous damping force, simulating a high-viscosity fluid's stagnation effect at the control algorithm level. This suppresses the rapid rise in material inventory and prevents physical stacking caused by upstream surges directly transmitting to the downstream. When the upstream scraper returns, causing a temporary interruption in material discharge, i.e. When the voltage turns negative, the control unit quickly switches to the smaller second damping coefficient. This underdamped characteristic allows the system to be subjected to only weak damping, enabling a virtual elastic stiffness coefficient. The dominant elastic recovery force quickly drives the transmission mechanism to accelerate, and uses the material accumulated in the buffer area to quickly fill the downstream cycle gap, avoiding the risk of downstream equipment starvation shutdown.

[0025] In this process, the system not only achieves flexible adjustment at the single-machine level, but also realizes the coordination of multi-level units through a virtual backpressure communication mechanism. When the downstream detection unit is temporarily blocked due to re-inspection, the virtual pressure value generated by the control unit... The back pressure rises rapidly and is transmitted in reverse via the industrial bus to the control unit of the upstream sintering furnace. Upon receiving the back pressure signal, the upstream control unit directly adds it to the speed command generation logic as a feedforward suppression variable. Predictive deceleration is executed before the physical buffer reaches its full load threshold. This control strategy, based on the advance transmission of information flow, transforms the potential hard blockage at the downstream detection unit into a small speed attenuation distributed along the entire production line, avoiding localized avalanche effects. Furthermore, the system's resonance active suppression step monitors and adjusts the virtual adjustment amount in real time to address the inherent 0.5Hz periodic disturbance of the screen printing machine. The sign-flipping frequency falls into the intrinsic resonance band, which then affects the virtual elastic stiffness coefficient. By applying a time-varying offset, this active detuning operation disrupts the resonant phase condition, ensuring that the material flow within the physical buffer zone remains in a smooth laminar flow state throughout the entire production cycle without standing wave oscillations. Ultimately, while maintaining a high throughput of 7200 wafers / hour, the silicon wafer microcrack rate was reduced to below 0.05%, and the overall equipment efficiency (OEE) was improved by 3.5%. This result confirms that the present invention, through the construction of a virtual physical field, successfully transforms the rigid physical constraints in discrete manufacturing systems into flexible logical connections with adaptive capabilities, solving the industry problem of balancing efficiency and stability in high-throughput photovoltaic production lines.

[0026] Example 2: To verify the practical engineering effectiveness and performance superiority of the collaborative control method of the present invention under complex working conditions, a semi-physical simulation test platform was constructed, comprising three series process units (upstream printing machine, midstream sintering furnace, and downstream inspection machine). Based on the EtherCAT real-time industrial bus architecture, the controllers of each process unit adopted Beckhoff CX2040 high-performance embedded PCs, with a control cycle set to 2ms. The experiment simulated the complex disturbance environment in real photovoltaic production: the output cycle of the upstream printing machine was set to an average of 7200 pieces / hour, and a sinusoidal fluctuation with an amplitude of ±15% and a frequency of 0.5Hz was superimposed to simulate the periodic surge waves caused by the reciprocating motion of the scraper; the input cycle of the downstream inspection machine was superimposed with Gaussian white noise with a mean of 0 and a variance of 200ms to simulate the randomness of visual inspection processing delay. In addition, to reflect electromagnetic interference in the industrial environment, broadband noise with a signal-to-noise ratio of 25dB was mixed into the collected material inventory signal.

[0027] Three experimental schemes were designed for multi-dimensional comparative verification. Control group A (existing technology group) adopted the traditional high and low liquid level threshold switch control strategy, setting the high threshold at 80% and the low threshold at 20%. Control group B (partially missing control group) adopted the virtual viscoelastic dynamic model of this invention, but removed the asymmetric anisotropic damping generation strategy and the virtual back pressure communication mechanism, retaining only the linear virtual impedance control. The experimental group (the sample group of this invention) fully utilized all the core control logic of this invention, including the asymmetric damping and virtual back pressure mechanism. Each group ran continuously for 4 hours under the same input conditions, focusing on monitoring the inventory fluctuation range of the physical buffer area, the speed fluctuation rate of the transmission mechanism (speed standard deviation / average speed), and the peak acceleration. During the experiment, the material inventory after 5Hz low-pass filtering was collected in real time. and rate of change In the test group, when periodic surges from upstream were detected, causing... At that time, the system automatically calls the strong damping coefficient. Overshoot is suppressed by overdamping characteristics; when When switching to weak damping coefficient The experiment utilized underdamped characteristics for rapid material replenishment. Furthermore, when a simulated downstream detector experienced a random blockage lasting 10 seconds, the experimental group employed a virtual backpressure mechanism to ensure the upstream sintering furnace received the signal in advance, even before its local buffer was full. The signal is then used to perform pre-deceleration. Table 1 shows a comparison of the three schemes in terms of key performance indicators.

[0028] Table 1: Comparison Data Table

[0029] Data shows that, due to the discontinuity of the control mechanism, the control group A system fell into severe start-stop limit loop oscillations, resulting in a peak acceleration of up to 5.2g, far exceeding the safety threshold of silicon wafers. Although control group B introduced virtual impedance to achieve continuous adjustment, it still showed a response lag when dealing with asymmetric material output fluctuations and sudden downstream blockages due to the lack of directional damping differences and feedforward back pressure information, with a speed fluctuation rate of 12.3%. In contrast, the experimental group, with its asymmetric damping for directional suppression of surges and the back pressure mechanism for advanced prediction of blockages, stably converged the existing fluctuation range within a narrow band of 42% to 58%, with a speed fluctuation rate of only 4.8%, and the peak acceleration was suppressed within the safe range of 1.1g. This confirms that the present invention, through the deep synergy of virtual physical field, asymmetric damping, and back pressure communication, fundamentally reconstructs the dynamic characteristics of production line material flow, achieving high stability and low-damage transmission under complex disturbances.

[0030] Example 3: This example combines Figs. 1 to 3 The collaborative control method for photovoltaic intelligent manufacturing equipment production lines is explained, such as... Fig. 1As shown, the collaborative control method performs a physical buffer state acquisition step to monitor the material inventory and change rate in real time. The process is divided into parallel processing logics, which respectively execute an asymmetric anisotropic damping generation strategy that dynamically switches the damping coefficient according to the accumulation or evacuation trend, establish a virtual viscoelastic dynamic model that maps non-Newtonian fluid characteristics, and introduce a virtual back pressure communication mechanism to suppress downstream blockage pressure feedforward. The calculation results of the above branches are incorporated into the virtual adjustment correction synthesis step to complete the vector superposition of elastic restoring force and viscous damping force, and then enter the reference speed adaptive correction step to eliminate the steady-state deviation of the system. Finally, in the dynamic speed command generation step, the transmission mechanism is driven after nonlinear saturation constraint processing.

[0031] like Fig. 2 As shown, the horizontal axis represents time (seconds), the left vertical axis represents the dimensionless virtual force, and the right vertical axis represents the material inventory change rate Q (% / second). The graph shows that the material inventory change rate Q exhibits periodic sinusoidal fluctuations, and the associated virtual viscous damping force curve shows asymmetric anisotropy characteristics, that is, a large-amplitude inhibiting force is generated in the range where Q is positive, while a small-amplitude damping force is generated in the range where Q is negative; for example... Fig. 3 As shown, the architecture of the collaborative control system is centered on the process unit control unit. The system collects the inventory and rate of change signals of the material state in real time through sensor components. The control unit integrates a virtual viscoelastic dynamic model calculation module constrained by a nonlinear boundary hardening strategy, an asymmetric anisotropic damping generation strategy module, a resonance active suppression module responsible for frequency monitoring and detuning, and a reference speed adaptive correction module. At the same time, the control unit receives virtual back pressure communication feedforward suppression variables from downstream process units. All calculation results are summarized to the dynamic speed command generation module, and after vector synthesis and saturation constraint, the final control command is output to the transmission mechanism.

[0032] Example 4: In the development of a highly integrated photovoltaic intelligent manufacturing equipment production line control system, addressing the engineering calibration challenges of parameter setting in the virtual viscoelastic dynamics model, this invention proposes and implements a systematic parameter optimization and calibration procedure. This procedure aims to solve the problem of virtual elastic stiffness coefficient... With virtual damping coefficient In practical applications, parameter issues arise due to the lack of explicit physical mapping. To ensure the stability and response performance of the control model under different operating conditions, it is crucial to clarify the calibration object and influencing factors. The calibration object is the core parameters in the virtual viscoelastic dynamics model. and The main technical factors affecting parameter values ​​include: the effective capacity of the physical buffer, the maximum acceleration / deceleration capability of the transmission mechanism, the fluctuation frequency and amplitude of the upstream discharge cycle time, and the sensitivity of the downstream process unit to the continuity of material supply. The essence of the technical trade-off lies in: higher... It can improve the system's response speed to existing deviations, but too high a level can lead to system overshoot or even oscillation; a higher level... While this can enhance system stability and anti-disturbance capabilities, excessively high settings can increase system viscosity, leading to sluggish response. Based on the above analysis, a decision-making logic and operational process for parameter calibration are established. The first step is to perform offline basic parameter calibration. Under no-load conditions on the production line, the physical buffer is set to a half-full state. And apply a step-type speed disturbance signal, and adjust the speed response curve of the transmission mechanism by monitoring it. Record the data until the system's rise time meets the minimum requirement of the process cycle time and the overshoot is less than 5%. The value is used as the basic stiffness coefficient, and is maintained. Unchanged, gradually increasing Until the system's oscillation decay ratio reaches The attenuation standard is that the amplitude of the second peak is 25% of the amplitude of the first peak. The value is used as the basic damping coefficient.

[0033] The second step is to perform online adaptive calibration under operating conditions. During the production line's load operation, the inertia adaptive gain scheduling mechanism is activated, and the system collects material inventory data in real time. And mapped to a dimensionless inertia factor. Based on the principle of dynamic similarity, set and Proportional, that is To maintain the system's natural frequency Approximately constant under different loads; simultaneously set and Proportional, that is To maintain the system's damping ratio This constant mapping rule ensures that the system maintains consistent dynamic response characteristics across the entire operating range, from no-load to full-load. The third step involves fine-tuning the asymmetric damping characteristics, setting the first damping coefficient for each asymmetric anisotropic damping generation strategy. (Corresponding to the accumulation trend) and the second damping coefficient (Corresponding to evacuation trend), in actual production, when a material accumulation trend is detected ( When ), Set to 1.5 to 2.0 times the base damping coefficient to construct strong damping characteristics and prioritize suppressing overshoot and spillover risks; when a material evacuation trend is detected ( When ), The damping coefficient is set to 0.5 to 0.8 times the base damping coefficient to construct weak damping characteristics. The elastic restoring force is used to accelerate the feeding. Through long-term operation tests, the above ratio coefficient is fine-tuned until the comprehensive evaluation indicators of the system, including OEE, microcrack rate, and mean time between failures, reach the optimal level.

[0034] Example 5: To ensure the stability and control accuracy of the asymmetric damping generation strategy and virtual backpressure communication mechanism of this invention in actual industrial deployment, this invention constructs a standardized on-site pre-deployment calibration and debugging procedure. This aims to eliminate control model mismatches that may be caused by individual equipment differences, sensor installation errors, and fluctuations in the on-site electromagnetic environment through systematic parameter identification and dynamic optimization. Technicians need to perform on-site calibration of the effective capacity of the physical buffer zone. Under no-load conditions, the control unit drives the upstream equipment to output material in a step manner. Photoelectric sensors installed at the inlet and outlet of the buffer zone record the time difference of material passage, and the actual maximum capacity of the buffer zone is calculated and confirmed. This calibration value is written into the control algorithm as a benchmark for material inventory normalization calculation to ensure state deviation. The calculation accuracy is good for asymmetric damping coefficients. and The system was configured to perform dynamic response testing. After connecting a simulated load to the system, a sinusoidal disturbance signal with a frequency of 0.1Hz to 1.0Hz was artificially introduced as a virtual beat input. By monitoring the speed response curve of the transmission mechanism, adjustments were gradually made. Until the system's overshoot to positive disturbances is less than 5%.

[0035] The damping value recorded at this point serves as the basic damping coefficient under the accumulation trend; similarly, adjust... Until the system's response time to negative disturbances meets the minimum cycle time requirement of downstream equipment, the basic damping coefficient under the evacuation trend is determined. This debugging step ensures that the asymmetric damping mechanism can accurately match the physical inertia characteristics of the current production line. Finally, a stress test of the virtual backpressure communication link is conducted. Under full-load operation, a continuous blockage of the downstream unit is simulated, and the virtual pressure value received by the upstream unit is monitored. The trend of change and response time; if the response delay exceeds a preset threshold, such as 500ms, the pressure conduction attenuation coefficient needs to be adjusted. To optimize the reverse transmission efficiency of pressure signals and ensure the real-time performance of the entire line's coordinated control, the system completes the adaptation from the theoretical model to the engineering site through the execution of the above procedures, ensuring the reliability of the coordinated control strategy.

[0036] Example 6: To verify the stability of the cooperative control method of the present invention under extreme conditions and long-term operation, an offline physical parameter identification sub-routine was constructed based on the geometric and inertia characteristics of the physical buffer zone. While the production line was shut down for maintenance, a laser rangefinder was used to determine the effective transmission length of the physical buffer zone between each process unit. Precise measurements were performed, and the transmission mechanism was driven to run at a step speed under no-load conditions. Speed ​​response data was collected by a high-frequency encoder, and the equivalent moment of inertia of the system was identified using the least squares method. coefficient of friction with the base These physical parameters, as core constants, are written into the controller's underlying configuration file, providing a definite physical benchmark for the subsequent construction of the virtual dynamic model. For the setting of key control parameters in the asymmetric damping generation strategy, an online dynamic response optimization sub-procedure is executed. During the system's trial operation phase, variable frequency sinusoidal speed disturbance signals with frequencies ranging from 0.1Hz to 2.0Hz are injected into the control loop, and the control unit monitors the speed tracking error of the transmission mechanism in real time. The first damping coefficient is adjusted using an adaptive optimization algorithm. With the second damping coefficient The overshoot is determined until the overshoot under material accumulation (positive disturbance) conditions is less than 3%, and the settling time under material evacuation (negative disturbance) conditions is less than 20% of the minimum cycle time of the downstream equipment. and It is solidified into the optimal damping parameter set under the current operating conditions of the production line.

[0037] Furthermore, to address the real-time performance and stability of the virtual backpressure communication mechanism, a network latency and packet loss stress test sub-procedure was implemented. A network analyzer was used to simulate a high-load industrial bus communication environment, injecting random packet loss and latency jitter into the control network, and monitoring the virtual stress value received by the upstream control unit. To ensure integrity and real-time performance, if more than three data packets are lost within one minute or a single delay exceeds 10ms, the system automatically triggers degradation protection logic, temporarily switching to local independent control mode and recording anomaly logs for subsequent network optimization reference, ensuring the safe failure of collaborative control when the communication link is unstable. Finally, regarding the trigger threshold setting for the nonlinear boundary hardening strategy, a boundary safety margin verification sub-procedure is executed. Under full load operation, a complete downstream blockage condition is artificially created, causing the material inventory to approach the physical limit, and the nonlinear gain coefficient is recorded in real time. With safety margin The curve of change is adjusted, and the curvature parameter of the nonlinear function is adjusted until the virtual elastic restoring force can generate a reverse braking acceleration sufficient to overcome the maximum static friction force of the system when the material inventory reaches 5% before the physical limit, thereby achieving soft landing protection without the intervention of mechanical limit switches.

[0038] Example 7: The calibration of key control parameters for the virtual viscoelastic dynamics model adopts a closed-loop tuning procedure based on step response, with virtual elastic stiffness coefficients... The initial value is obtained by applying a step speed command to the unloaded conveyor and monitoring the material inventory. The response curve is determined, and the adjustment coefficient ensures that the response rise time matches the minimum physical cycle time of the downstream process unit while keeping the overshoot within a preset range; By gradually increasing the base damping coefficient until the oscillation decay ratio of the stock response curve meets the quarter decay standard, and based on the base damping value, the control unit adjusts the control system according to the material stock change rate. The sign direction will determine the first damping coefficient of the accumulation trend. Set to 1.5 to 2.0 times the base value to enhance overshoot suppression, and set the second damping coefficient for dissipation trend. Set to 0.5 to 0.8 times the base value to improve follow response speed.

[0039] The active suppression step for resonance involves extracting the oscillation frequency and using hysteresis-inclusive zero-crossing detection logic to process the virtual adjustment correction amount. For time-domain signals, a hysteresis interval threshold is set to 5% to 10% of the signal's full scale. A valid zero-crossing event is recorded when the signal amplitude crosses this interval and polarity reverses. This shields against electromagnetic interference or mechanical micro-vibration signals causing zero-point jitter. The control unit calculates the statistical frequency of valid zero-crossing events within a preset sliding time window. When the statistical frequency falls within the frequency band of 0.8 to 1.2 times the system's first-order natural frequency, determined through modal analysis or frequency sweep testing, an adjustment to the virtual elastic stiffness coefficient is triggered. Time-varying offset modulation alters the equivalent stiffness of the control loop, causing the closed-loop poles of the system to deviate from the mechanical resonance point; a virtual backpressure communication mechanism simulates pressure values. Generate based on current material inventory Relative physical cache limit The local pressure component is calculated by proportion, and a first-order linear weighting algorithm is used to combine the local pressure component with the received downstream virtual pressure value. Superposition, weighted algorithm pressure transmission attenuation coefficient Depending on the cascade length of the physical layout of the production line, the direct coupling coefficient between adjacent nodes is set to a value in the range of 0.8 to 0.95 to ensure that the blockage information is transmitted to at least three upstream process units along the reverse flow. The step-by-step attenuation characteristic is used to prevent small disturbances at the far end from being excessively amplified in the long chain transmission, causing a slowdown response across the entire line.

[0040] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A collaborative control method for a photovoltaic intelligent manufacturing equipment production line, characterized in that, The production line comprises multiple process units connected in series via physical buffers. The method establishes and maps the discrete material flows within the physical buffers into a virtual viscoelastic dynamic model exhibiting non-Newtonian fluid properties. The method includes the following steps: The control unit of each process unit collects the current material inventory and the rate of change of material inventory in the associated physical buffer area in real time; The control unit is based on a virtual viscoelastic dynamics model to calculate the virtual elastic restoring force that returns the material inventory to the preset equilibrium point, and the virtual viscous damping force that hinders the change of the material inventory state. In calculating the virtual viscous damping force, the control unit executes an asymmetric anisotropic damping generation strategy: real-time identification of the positive and negative polarities of the rate of change of material inventory to determine the immediate trend of material flow; when the trend is determined to be material accumulation, the first damping coefficient is called to calculate the overdamping force with strong inhibition characteristics; when the trend is determined to be material dispersion, the second damping coefficient with a value less than the first damping coefficient is called to calculate the underdamping force with weak inhibition characteristics. The control unit performs vector synthesis of the virtual elastic restoring force and the virtual viscous damping force obtained based on the asymmetric anisotropic damping generation strategy to generate a virtual adjustment correction amount, and then superimposes the virtual adjustment correction amount onto the basic transmission speed of the process unit to generate a dynamic speed command to drive the transmission mechanism. By employing an asymmetric anisotropic damping generation strategy, the process unit's response to upstream logistics fluctuations exhibits a nonlinear rheological characteristic that combines rigid suppression under accumulation trends with flexible following under dispersal trends.

2. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The calculation logic for the virtual adjustment correction in the virtual viscoelastic dynamic model follows a linear combination relationship including state deviation terms and state change rate terms, and satisfies the following dynamic equation: ,in, Characterizing the virtual adjustment correction amount, Characterizing the virtual elastic stiffness coefficient Characterizes the current material inventory Characterizes the preset equilibrium point, Characterizing the trend of flow The first or second damping coefficient can be dynamically switched. Characterizes the rate of change of material inventory; the equation defines the dynamic convergence trajectory of the system when it deviates from steady state.

3. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The method also includes establishing a virtual back pressure communication mechanism for reverse transmission of material flow along the production line: the control unit receives the virtual pressure value sent by the direct downstream process unit and introduces the virtual pressure value as a feedforward suppression variable into the generation logic of the dynamic speed command; the virtual pressure value is generated by the downstream process unit based on its own current material inventory and the virtual pressure value received from the next downstream unit through weighted accumulation; through the virtual back pressure communication mechanism, the downstream material flow blockage state is reversely diffused to the upstream in the form of a scalar field, driving the upstream process unit to perform predictive deceleration before the physical buffer is full, and using the advance transmission of information flow to exchange for the redundancy of the physical buffer space.

4. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The steps for calculating the virtual elastic restoring force include implementing a nonlinear boundary hardening strategy: the control unit monitors the safety margin between the material inventory and the physical limit boundary of the physical buffer zone in real time; the control unit generates a nonlinear gain coefficient based on the safety margin, which is set to maintain a constant base value when the safety margin is greater than a preset threshold, and to increase nonlinearly and monotonically when the safety margin is less than the preset threshold and approaches zero; the control unit uses the nonlinear gain coefficient to weight and amplify the virtual elastic restoring force to generate an exponentially enhanced repulsive potential energy field when the material inventory approaches the physical limit boundary, so that the transmission mechanism can continuously reverse the speed without external hard limit triggering.

5. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The method also includes a baseline speed adaptive correction step: the control unit sets up a sliding time window and calculates the statistical average value of the virtual adjustment correction within the window; when the absolute value of the statistical average value exceeds a preset dead zone threshold, the control unit identifies that there is a systematic speed mismatch, and compensates and corrects the base transmission speed based on the statistical average value, and uses the corrected base transmission speed as the updated baseline value for the generation of subsequent dynamic speed commands; By correcting the steps, the steady-state inventory deviation caused by equipment aging or environmental drift is eliminated, so that the virtual adjustment correction amount automatically converges to near zero under steady-state operation, restoring the dynamic range of the system's bidirectional adjustment.

6. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The method also includes performing an active resonance suppression step: the control unit monitors the sign-flipping frequency of the virtual adjustment correction in the time domain to estimate the real-time oscillation frequency of the cooperative control system; when the real-time oscillation frequency falls into the preset intrinsic resonance frequency band, the control unit applies a time-varying offset to the virtual elastic stiffness coefficient in the virtual viscoelastic dynamic model to dynamically change the natural frequency of the cooperative control system to disrupt the resonance phase synchronization condition; when a resonance trend is detected, the control unit temporarily increases the weight of the virtual viscous damping force to dissipate the resonant energy in the system.

7. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The method also includes an inertia adaptive gain scheduling step based on material inventory: the control unit maps the material inventory to an inertia factor that characterizes the current load inertia of the system in real time; the control unit dynamically adjusts the virtual elastic stiffness coefficient and virtual damping coefficient in the virtual viscoelastic dynamic model in real time based on the inertia factor and according to the preset nonlinear mapping rules. The nonlinear mapping rule is set to increase the virtual elastic stiffness coefficient and virtual damping coefficient accordingly with the increase of the inertia factor, so as to keep the natural frequency and damping ratio of the cooperative control system constant under different load conditions and ensure the consistency of the system's dynamic response characteristics.

8. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The acquisition of the rate of change of material inventory in the status acquisition step includes: acquiring the difference in material inventory between two adjacent sampling periods; performing low-pass filtering on the difference in material inventory to filter out high-frequency noise interference, and generating a smoothed rate of change signal as the basis for calculating the virtual viscous damping force; the cutoff frequency of the low-pass filtering is set to be lower than the minimum effective response frequency of the cooperative control system to prevent the asymmetric anisotropic damping generation strategy from being mistakenly triggered due to sensor signal jitter.

9. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 1, characterized in that, The process of generating dynamic speed commands also includes nonlinear saturation constraint processing: the control unit presets a speed constraint range based on the minimum dwell time or maximum physical cycle time limit of the process steps executed by the process unit; when the calculated dynamic speed command exceeds the speed constraint range, the control unit clamps the dynamic speed command at the boundary value of the speed constraint range to ensure that the single-machine process parameters do not exceed the limit.

10. The collaborative control method for a photovoltaic intelligent manufacturing equipment production line according to claim 5, characterized in that, The baseline speed adaptive correction step also includes state freeze logic: the control unit monitors the production line's operating status in real time; when the control unit detects that the production line is in a non-steady-state condition such as starting, emergency stop, or fault alarm, the control unit automatically suspends the baseline speed adaptive correction step to keep the current base transmission speed unchanged.

Images