A method for controlling the winding tension of a spunbond geotextile based on real-time winding diameter estimation

Abstract

The present application belongs to the technical field of PID control application, and particularly relates to a kind of based on real-time estimation of winding diameter spinning geotextile winding tension control method.The present application first calibrates the parameters of spun-bond geotextile material and winding process parameters, real-time acquisition of winding roller, traction roller angular velocity and real-time tension;Real-time estimation of winding diameter is carried out by using double-channel fusion strategy, combined with speed ratio method and layer-by-layer cumulative winding diameter model considering radial compressibility, and the fusion weight is adaptively updated by extended Kalman filter, to output high-precision fusion winding diameter estimation value;Based on winding diameter, the target winding tension is calculated, and the compound closed-loop control is constituted by inertia feedforward torque compensation and gain scheduling PID feedback, to adjust the electromagnetic torque of winding motor in real time.The present application effectively improves the estimation accuracy of winding diameter, suppresses the tension fluctuation under acceleration and deceleration conditions, ensures the flatness of finished fabric and the quality of winding package, and adapts to the high-speed production demand of wide width and large winding package.

Description

Technical Field

[0001] This invention belongs to the field of PID control application technology, and in particular relates to a method for controlling the winding tension of spunbond geotextile based on real-time prediction of roll diameter. Background Technology

[0002] Spunbond geotextiles are core geosynthetic materials with filtration, isolation, and reinforcement functions, widely used in engineering fields such as water conservancy embankments, road infrastructure, and ecological restoration. As the final and core process in the production of spunbond geotextiles, the stability of the winding tension directly determines the flatness of the fabric surface, the neatness of the roll ends, and the consistency of the product's mechanical properties. It is a key process link to ensure finished product quality and improve production efficiency. With the development of geotextile production towards wider widths, larger rolls, and higher speeds, higher requirements are placed on the precision and dynamic stability of winding tension control.

[0003] Existing tension control methods for spunbond geotextile winding mostly employ conventional PID feedback control schemes. The roll diameter calculation is generally based on the angular velocity ratio of the traction roller and the winding roller. While this method is simple in principle, it has weak anti-interference capabilities and is easily affected by roller slippage and speed detection noise, resulting in limited accuracy in roll diameter estimation. Furthermore, existing roll diameter models often assume a constant material thickness, failing to consider the radial compressibility of spunbond geotextile: as the number of winding layers increases, the inner layer material experiences radial compressive stress due to the accumulated tension of the outer layers, leading to a continuous decrease in the actual effective thickness. This causes the roll diameter estimation deviation to increase with the roll diameter, resulting in inaccurate tension settings. In addition, most control schemes do not provide feedforward compensation for the dynamic increase in rotational inertia with the roll diameter during winding. Relying solely on feedback adjustment results in response lag, easily leading to significant tension fluctuations under acceleration and deceleration conditions. This can cause quality defects such as fabric wrinkling and loose rolls, making it difficult to meet the demands of high-quality production. Summary of the Invention

[0004] To address the technical problems existing in the background art, this invention proposes a method for controlling the winding tension of spunbond geotextiles based on real-time roll diameter estimation.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] S1. Determine the material parameters and winding process parameters of the spun clay geotextile. The material parameters include the core diameter. Initial inherent thickness of geotextile Radial compressibility coefficient and characteristic compressive stress The winding process parameters include the initial winding tension. and maximum target volume ;

[0007] S2. Real-time data acquisition: Real-time acquisition of the winding roller angular velocity during the winding process. angular velocity of traction roller and the real-time tension of the geotextile in the winding section ;

[0008] S3. Perform dual-channel fusion estimation of roll diameter, and use a dual-channel fusion strategy to estimate the current take-up roll diameter in real time. Specifically, it includes:

[0009] S31, based on the diameter of the traction roller and the angular velocity of the traction roller With the angular velocity of the take-up roll The speed ratio is used to calculate the estimated roll diameter for the first channel: ;

[0010] S32. Establish a layer-by-layer cumulative roll diameter model considering the radial compressibility of spun clay geotextile, based on real-time tension. And the historical tension of each layer, calculate the effective incorporation thickness of each layer layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. , where i is the level index;

[0011] S33. Based on the extended Kalman filter framework, with the roll diameter and roll diameter change rate as state variables, the prediction results of S31 and S32 are introduced into the filter as two independent observations. The estimation error covariance of each channel is calculated in real time through covariance recursion, and the fusion weights of the two channels are adaptively updated to output the fused roll diameter prediction. ;

[0012] S4. Based on the estimated fusion roll diameter The current target winding tension is calculated in real time using a preset tension-wound diameter decrease curve. ;

[0013] S5. Based on the estimated roll diameter Feedforward calculation of the torque compensation required due to changes in moment of inertia, while simultaneously using the target tension. With measured tension deviation As a feedback input, the PID controller generates a feedback torque compensation amount, and the feedforward torque and feedback torque are combined and output to the winding motor driver to adjust the electromagnetic torque of the winding motor in real time, thereby realizing the composite closed-loop control of the winding tension.

[0014] Preferably, the formula for calculating the effective incorporation thickness of the i-th layer in step S32 is: ,in, The radial compressive stress currently borne by the i-th layer of material is generated by the cumulative tension of each outer layer; radial compressive stress Based on the theory of elastic thin-layer accumulation, when the total number of involved layers is n: Where W is the winding width, Let J be the winding tension when the j-th layer is wound in. The wrap angle corresponding to the j-th layer, This is the estimated volume diameter when the j-th layer is involved.

[0015] Preferably, the calculation of the effective incorporation thickness of each layer is performed layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. The specific implementation is as follows: .

[0016] Preferably, in step S33, the state equation of the extended Kalman filter framework is expressed in terms of state variables: Establish, Where F is the state transition matrix and the rate of change of volume diameter is... The thickness is determined by the current winding speed v(t) and the effective thickness of the geotextile. ,in, , The effective thickness of the single layer being involved at the current moment is defined by the state equation, which embeds the material compressibility characteristics into the state transition process, so that the filter pre-estimation stage already includes physical constraint information.

[0017] Preferably, in step S33, the estimated results of S31 and S32 are introduced into the filter as two independent observations. The estimation error covariance of each channel is calculated in real time through covariance recursion, the fusion weights of the two channels are adaptively updated, and the fused volume diameter estimate is output. Specifically, it includes:

[0018] The predicted roll diameter results of the first channel and the predicted roll diameter results of the second channel are treated as two independent observations, and corresponding observation equations are constructed respectively. The observation matrices of the two channels only extract the roll diameter component in the state vector, and their respective observation noise and initial observation noise covariance are configured.

[0019] According to the state transition equation, the prior state estimate at the current time is recursively calculated from the posterior state estimate at the previous time step, and the prior estimate error covariance matrix at the current time step is calculated simultaneously.

[0020] The observations from two channels are introduced sequentially to perform two Kalman update operations. First, the observations from the first channel are introduced to update the prior state estimate for the first time, and the innovation, innovation covariance, and Kalman gain of the first channel are calculated to obtain the state estimate and error covariance matrix after the first update. Then, the observations from the second channel are introduced to update the state and covariance results after the first update for the second time, and the innovation, innovation covariance, and Kalman gain of the second channel are calculated to finally obtain the posterior state estimate and posterior error covariance matrix at the current time.

[0021] Based on the innovation covariance of the two channels, extract adaptive fusion weights: ,in, The fusion weight for the first channel, For the fusion weight of the second channel, The trace of the matrix; the estimated fused volume diameter is obtained by weighting according to the fusion weights. .

[0022] As a preferred option, based on the estimated fusion roll diameter The current target winding tension is calculated in real time using a preset tension-wound diameter decrease curve. ,include:

[0023] The obtained estimated fusion volume Substitute the preset tension-roll diameter decrease curve to calculate the target tension value at the current moment: ,in, This is the tension attenuation coefficient, with a value ranging from 0.3 to 0.8; The pressure uniformity compensation coefficient ranges from 0.05 to 0.2; the target tension value is limited by upper and lower limits to obtain the final output target winding tension value at time t.

[0024] Preferably, the specific calculation method for the feedforward torque compensation amount in step S5 is as follows:

[0025] Based on the estimated roll diameter Real-time calculation of the equivalent moment of inertia referred to the motor shaft And feedforward to calculate torque compensation: ,in, The angular acceleration of the take-up roller. Transmission efficiency is a measure of energy loss in transmission mechanisms such as gearboxes. The reduction ratio is the ratio of the motor speed to the winding roller speed. The equivalent moment of inertia, referred to the total system moment of inertia on the motor shaft, varies in real time with the roll diameter. The calculation method is as follows: ,in, The moment of inertia of the motor rotor. The moment of inertia of the core. This represents the cumulative mass of the geotextile currently rolled up.

[0026] Preferably, in step S5, the parameters of the PID controller are based on the current roll diameter. Adaptive adjustment of gain scheduling is performed, specifically including: scaling factor Updated to: Integral coefficient Updated to: Differential coefficients Updated to: ,in, These are the initial PID parameters. All of these are gain scheduling coefficients.

[0027] Compared with existing technologies, the advantages and positive effects of this invention are as follows: It employs a dual-channel fusion roll diameter estimation scheme, combining the speed ratio method with a layer-by-layer cumulative model considering the radial compressibility of spunbond geotextiles. Through extended Kalman filtering and adaptive allocation of fusion weights, it effectively overcomes roll diameter estimation deviations caused by roller slippage, detection noise, and material compression, significantly improving estimation accuracy. Simultaneously, it constructs a feedforward-feedback composite closed-loop control architecture. Based on real-time roll diameter feedforward compensation of rotational inertia dynamic disturbances, coupled with gain-scheduled PID that adaptively adjusts with roll diameter, it combines fast dynamic response with high steady-state control accuracy. This effectively suppresses tension fluctuations under acceleration and deceleration conditions, ensuring fabric flatness and roll end face uniformity, adapting to the high-speed production needs of wide-width, large-roll packages. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart illustrating a method for controlling the winding tension of spunbond geotextiles based on real-time roll diameter prediction. Detailed Implementation

[0030] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0031] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways than those described herein, and therefore the invention is not limited to the specific embodiments disclosed in the following specification.

[0032] In the example of spunbond geotextile production, the winding process is a crucial final step in ensuring finished product quality. During winding, the geotextile is wound onto the core at a constant or varying linear speed. As the roll diameter gradually increases, the system's moment of inertia continuously grows, and the inner layer material undergoes physical compression due to the radial compressive stress generated by the accumulated tension of the outer layer, resulting in a continuous decrease in the actual effective winding thickness of each layer. Traditional roll diameter estimation methods are based solely on the angular velocity ratio between the traction roller and the winding roller, assuming a constant material thickness and ignoring the roll diameter deviation caused by the radial compressibility characteristics, thus leading to deviations of the tension setpoint from the process target. Furthermore, control schemes relying solely on PID feedback do not provide feedforward compensation for dynamic changes in moment of inertia, resulting in response lag under acceleration and deceleration conditions, easily leading to significant tension fluctuations and causing quality defects such as fabric wrinkling and loose rolls. To address the aforementioned technical problems, this embodiment provides a method for controlling the winding tension of spunbond geotextile based on real-time roll diameter prediction. By integrating roll diameter prediction with feedforward-feedback composite closed-loop control through a dual-channel fusion mechanism, high-precision and stable control of the winding tension is achieved. The specific implementation process is as follows: Figure 1 As shown.

[0033] First, the material parameters and winding process parameters of the spun clay geotextile are calibrated. The material parameters include the core diameter. Initial inherent thickness of geotextile Radial compressibility coefficient and characteristic compressive stress The winding process parameters include the initial winding tension. and maximum target volume .

[0034] Specifically, the core diameter refers to the physical outer diameter of the hollow paper tube or metal mandrel on the take-up roller, which is input by the operator through the human-machine interface during production line changeover. In this embodiment, the typical configuration is 76mm to 152mm, preferably 152mm (corresponding to a 6-inch standard paper tube). The core diameter serves as the initial reference value for the system's roll diameter calculation, and all subsequent roll diameter recursive calculations are accumulated layer by layer from this starting point. The initial inherent thickness of the geotextile refers to the nominal single-layer physical thickness of the spunbond geotextile under no external compression, provided by the material supplier or measured using a laboratory micrometer under zero pressure. In this embodiment, for a typical 150g / m² polypropylene spunbond geotextile, its initial inherent thickness is calibrated to be 1.8mm to 2.5mm, preferably 2.0mm. This parameter serves as the reference value for calculating the effective roll thickness of each layer in the subsequent layer-by-layer cumulative roll diameter model. The radial compressibility coefficient characterizes the nonlinear compressive deformation physical properties of spunbond geotextile under radial compressive stress, defined as the relative compression per unit thickness under unit compressive stress. The calibration method for this parameter is as follows: In a laboratory environment, geotextile samples cut to standard dimensions are placed between the upper and lower plates of a universal testing machine. Increasing normal pressures from 0 to the maximum expected working compressive stress are applied at a constant loading rate, and the sample thickness variation data at each pressure level are recorded simultaneously. The slope characteristic value of the pressure-compressibility relationship curve is obtained through polynomial fitting or piecewise linear regression. In this embodiment, for polypropylene spunbond geotextile, the characteristic compressive stress refers to the reference stress threshold used to define the segmented characteristics of material compression behavior in the layer-by-layer cumulative roll diameter model. Physically, it represents the stress value at the inflection point where the radial compressive stress between geotextile layers transitions from the elastically recoverable compressible region to the plastically non-recoverable compacted region. This parameter is also determined using the pressure-thickness relationship curve of the universal testing machine, i.e., finding the stress value corresponding to the inflection point where the slope changes significantly on the curve. In this embodiment, for polypropylene spunbond geotextile... The initial winding tension refers to the target tension value set at the beginning of the winding process, which is usually determined based on the geotextile's width, weight, and production line speed, among other process conditions. The maximum target roll diameter refers to the upper limit of the maximum winding diameter allowed for a single roll of product, which is usually determined by the physical space constraints of downstream storage, handling, and slitting equipment. In this embodiment, a typical configuration is 1200mm to 1500mm, preferably 1200mm.

[0035] It is understandable that all calibration parameters in the above steps are written digitally into the non-volatile memory of the control system, serving as static reference data for subsequent roll diameter estimation, tension calculation, and control algorithms during runtime calculations.

[0036] Next, real-time data acquisition is performed, specifically by collecting the angular velocity of the winding roller during the winding process. angular velocity of traction roller and the real-time tension of the geotextile in the winding section Specifically, the acquisition of the winding roll angular velocity is achieved through a high-precision incremental rotary encoder installed at the end of the winding roll's rotating shaft. The incremental rotary encoder has a resolution of 2048 to 10240 pulses per revolution (preferably 4096 pulses / revolution in this embodiment). The quadrature A-phase and B-phase pulse signals output by the encoder are connected to the high-speed counter hardware interface of the control system. The processor performs real-time decoding processing on the encoder pulse signals using the M / T frequency measurement method or the variable M / T method to calculate the instantaneous angular velocity value of the winding roll (unit: rad / s). To suppress quantization noise and motor ripple interference in the encoder signal, after acquiring a single angular velocity sample value, the processor inputs it to a second-order low-pass digital filter with a cutoff frequency configured to 50Hz to 100Hz for smoothing. The filtered angular velocity data serves as the effective input parameter for the roll diameter prediction calculation in step S3. The acquisition method for the traction roll angular velocity is the same as that for the winding roll angular velocity, achieved through an incremental rotary encoder independently installed at the end of the traction roll shaft. The configuration parameters (resolution, filter cutoff frequency, etc.) of the traction roller encoder are consistent with those of the take-up roller encoder. The traction roller is located upstream of the take-up roller, and its function is to provide a stable linear velocity reference for the geotextile and isolate the influence of tension fluctuations in the preceding section on the take-up section. The real-time tension of the geotextile in the take-up section is detected in real time by a tension sensor. This tension sensor is a strain gauge force sensor, installed at the support bearing seat of the tension detection floating roller located between the take-up roller and the traction roller, or mounted on the guide roller bearing seat base plate using a directly embedded pressure sensing structure. The tension sensor converts the normal force applied by the geotextile to the detection roller into a proportional analog voltage signal. This analog voltage signal is digitized by a high-precision analog-to-digital converter circuit (resolution not less than 16 bits, sampling rate not less than 1kHz) and then transmitted to the processor. The processor performs median filtering or moving average filtering on the digital tension signal to filter out burr interference caused by mechanical vibration and electrical noise. The output filtered tension value is the real-time tension value of the geotextile in the take-up section.

[0037] It is understood that the acquisition of the three parameters—the angular velocity of the take-up roller, the angular velocity of the traction roller, and the real-time tension—is performed cyclically with a unified control cycle. In this embodiment, the control cycle is set to 2ms to 10ms (preferably 5ms), that is, a complete data refresh is performed every 5 milliseconds to ensure that the roll diameter estimation and tension control algorithms have sufficient dynamic response capabilities.

[0038] Then, considering that in the winding process of spun clay geotextile, the real-time accurate prediction of the current roll diameter is the basic prerequisite for tension control. Traditional roll diameter calculation generally adopts a single speed ratio method, that is, the current roll diameter is estimated based on the product of the ratio of the angular velocity of the traction roller and the winding roller and the known diameter of the traction roller. This method is simple in principle and requires little calculation, but it has the following prominent problems: First, the physical premise of the speed ratio method is that there is no relative slippage between the geotextile and the roller surface. However, in actual production, factors such as long-term wear of the traction roller coating layer, changes in the surface friction coefficient of the geotextile, and inertial lag under high-speed conditions can all introduce roller surface slippage, causing the angular velocity speed ratio to deviate from the actual roll diameter ratio, resulting in a systematic estimation error. Second, the speed ratio method implicitly assumes that the material thickness is constant and does not consider the radial compressibility characteristics of spunbond geotextile. As the number of winding layers increases, the inner layer material is subjected to radial compressive stress generated by the accumulated tension of the outer layer, and the fiber mesh pores are continuously compressed. The actual effective winding thickness decreases layer by layer. If the roll diameter is still accumulated with a constant initial thickness, the calculated roll diameter will be systematically higher than the actual physical roll diameter. Moreover, this deviation will continue to accumulate and amplify as the roll diameter increases, eventually causing the tension set value to deviate from the process target, resulting in overpressure of the inner layer or relaxation of the outer layer.

[0039] To address the dual shortcomings of the traditional speed ratio method—weak anti-interference capability and neglect of the radial compressibility of materials—a dual-channel fusion roll diameter prediction strategy based on an extended Kalman filter framework is proposed. Specifically, this step retains the speed ratio method as the first channel for roll diameter prediction, leveraging its advantages of good real-time performance and computational simplicity. Simultaneously, a second channel is added, establishing a layer-by-layer cumulative roll diameter model considering the radial compressibility of spunbond geotextiles. By calculating the effective roll thickness of each layer under actual radial compressive stress and accumulating the results, the nonlinear compressive physical characteristics of the material are directly embedded into the roll diameter calculation process. Based on this, the prediction results from the two channels are introduced into the extended Kalman filter as independent observations. A covariance recursive mechanism is used to evaluate the estimation error of each channel in real time, adaptively and dynamically allocating fusion weights—when the prediction accuracy of a channel decreases due to roller slippage or parameter calibration deviation, its information covariance increases, and its fusion weight automatically decreases, causing the filter to rely more heavily on the other, more reliable channel.

[0040] Specifically, first, based on the diameter of the traction roller and the angular velocity of the traction roller With the angular velocity of the take-up roll The speed ratio is used to calculate the estimated roll diameter for the first channel: A layer-by-layer cumulative roll diameter model considering the radial compressibility of spunbond geotextile was established, based on real-time tension. And the historical tension of each layer, calculate the effective incorporation thickness of each layer layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. , where i is the level index.

[0041] Furthermore, the physical basis of the layer-by-layer cumulative roll diameter model is as follows: Spunbond geotextile is not a rigid material; its fiber web structure undergoes pore compression deformation under radial compressive stress, resulting in an actual effective roll thickness that is less than the initial inherent thickness. As the number of roll layers increases, the inner geotextile layer bears radial compressive stress generated by the cumulative tension of all the outer roll layers. The compressive stress value increases with the number of layers, causing the effective roll thickness of each layer to show a decreasing physical trend. If this compression effect is ignored in the roll diameter calculation and each layer thickness is assumed to be the initial inherent thickness, the calculated theoretical roll diameter will be systematically higher than the actual physical roll diameter, and this deviation will continue to accumulate and amplify as the roll diameter increases. The formula for calculating the effective roll thickness of the i-th layer is: ,in, The radial compressive stress currently borne by the i-th layer of material is generated by the cumulative tension of each outer layer; radial compressive stress Based on the theory of elastic thin-layer accumulation, when the total number of involved layers is n: Where W is the winding width, Let J be the winding tension when the j-th layer is wound in. The wrap angle corresponding to the j-th layer, This is the estimated roll diameter for the j-th layer. It can be understood that the calculation of the effective roll thickness for each layer is described layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. The specific implementation is as follows: During system operation, whenever a new layer of geotextile is wound in, the processor first obtains the current real-time tension value as the winding tension of that layer and records it. Simultaneously, it records the system's estimated winding diameter at the current moment as a reference winding diameter for that layer. Subsequently, the processor iterates through all wound layers and recalculates the updated radial compressive stress of each layer according to the aforementioned radial compressive stress formula. Since the winding tension of a new layer affects the radial compressive stress of all inner layers, the effective winding thickness of all inner layers must be recalculated after each new layer is wound in. The processor sequentially substitutes the updated radial compressive stress into the effective winding thickness calculation formula to obtain the updated effective thickness of each layer. Finally, the processor performs an accumulation and summation operation on the effective winding thickness of all layers and outputs the estimated winding diameter of the second channel at the current moment.

[0042] Understandably, this layer-by-layer cumulative model directly embeds the radial compression physical properties of the material into the roll diameter calculation process, without relying on the roller surface angular velocity signal, and is therefore unaffected by roller surface slippage and speed detection noise. However, the accuracy of this model is limited by the measurement accuracy of the roll tension of each layer and the calibration accuracy of the material compression parameters, and when the total number of roll layers is large (usually exceeding several hundred layers), the computational load of layer-by-layer recalculation will increase. Therefore, this invention uses an extended Kalman filter framework in the next step to adaptively fuse the prediction results of the first and second channels to obtain a fused roll diameter prediction that combines real-time performance and accuracy. Specifically, based on the extended Kalman filter framework, with roll diameter and roll diameter change rate as state variables, the prediction results of the first and second channels are introduced into the filter as two independent observations. The estimation error covariance of each channel is calculated in real time through covariance recursion, the fusion weights of the two channels are adaptively updated, and the fused roll diameter prediction is output.

[0043] Furthermore, this step employs the Extended Kalman Filter (EKF) framework to perform probabilistically optimal fusion of the dual-channel roll diameter prediction results. The core reason for choosing Kalman filtering as the fusion framework is that, based on the minimum mean square error criterion, Kalman filtering can automatically and dynamically adjust the fusion weights according to the estimation error covariance of each information source, even in the presence of process and observation noise, ensuring that the final fusion result is statistically optimal and unbiased. The reason for introducing Extended Kalman filtering instead of standard linear Kalman filtering is that there is a nonlinear relationship between the roll diameter change rate and the state variables (involving the material compression model), requiring local linearization through the Jacobian matrix. Specifically, the state equation of the Extended Kalman Filter framework uses the following state variables: Establish, Where F is the state transition matrix and the rate of change of volume diameter is... The thickness is determined by the current winding speed v(t) and the effective thickness of the geotextile. ,in, , Given the effective thickness of the single layer being entrained at the current moment, the state equation embeds the material compressibility characteristics into the state transition process, ensuring that the filter pre-estimation stage already includes physical constraint information. The predicted roll diameter results from the first and second channels are treated as two independent observations, and corresponding observation equations are constructed for each. The observation matrices for both channels extract only the roll diameter component from the state vector, and their respective observation noise and initial observation noise covariance are configured. Based on the state transition equation, the prior state estimate for the current moment is recursively calculated from the posterior state estimate of the previous moment, while simultaneously recursively calculating the prior estimation error covariance matrix for the current moment. Two Kalman update operations are performed sequentially using observations from two channels. First, the observations from the first channel are used to update the prior state estimate, calculating the innovation, innovation covariance, and Kalman gain for the first channel to obtain the updated state estimate and error covariance matrix. Then, the observations from the second channel are used to update the state and covariance results, calculating the innovation, innovation covariance, and Kalman gain for the second channel to obtain the current posterior state estimate and posterior error covariance matrix. Finally, adaptive fusion weights are extracted based on the innovation covariance of each channel. ,in, The fusion weight for the first channel, For the fusion weight of the second channel, The trace of the matrix; the estimated fused volume diameter is obtained by weighting according to the fusion weights. .

[0044] It is understandable that the serial measurement update method used in this embodiment introduces the observations from the two channels into the same filter for a two-step update. Mathematically, this is equivalent to merging the two observations into a joint observation vector for a one-time update (joint observation update). However, the serial method has better numerical stability and computational flexibility. In actual engineering implementation, this dual-channel fused roll diameter estimate serves as the roll diameter input reference for tension calculation and feedforward torque compensation.

[0045] Then based on the estimated fusion volume diameter The current target winding tension is calculated in real time using a preset tension-wound diameter decrease curve. Specifically, the estimated fusion roll diameter will be obtained. Substitute the preset tension-roll diameter decrease curve to calculate the target tension value at the current moment: ,in, The tension attenuation coefficient ranges from 0.3 to 0.8, and in this embodiment, it is preferably 0.5 for polypropylene spunbond geotextile. The pressure uniformity compensation coefficient ranges from 0.05 to 0.2, and is preferably 0.12 in this embodiment; the target tension value is limited at the upper and lower limits to obtain the final output target winding tension value at time t.

[0046] The final execution process involves feedforward calculation of the torque compensation required due to changes in moment of inertia based on the roll diameter prediction. Simultaneously, the deviation between the target tension and the measured tension is used as feedback input. A PID controller generates the feedback torque compensation, which is then combined with the feedback torque and output to the take-up motor driver. This allows for real-time adjustment of the take-up motor's electromagnetic torque, achieving composite closed-loop control of the take-up tension. Furthermore, the core design idea of ​​this step lies in the fact that as the roll diameter continuously increases during take-up, the equivalent moment of inertia referred to the motor shaft exhibits a monotonically increasing trend. Under acceleration and deceleration conditions, the rate of change of moment of inertia will generate additional inertial torque disturbances. These disturbances are difficult to eliminate in time using only the PID feedback loop, leading to significant transient tension fluctuations. Therefore, this invention adds a feedforward compensation channel outside the feedback control loop. Based on the real-time roll diameter prediction, it actively calculates and compensates for the inertial torque component caused by changes in moment of inertia, achieving dual-channel coordinated control of feedforward and feedback.

[0047] Specifically, the calculation method for the feedforward torque compensation is based on the estimated roll diameter. Real-time calculation of the equivalent moment of inertia referred to the motor shaft And feedforward to calculate torque compensation: ,in, The angular acceleration of the take-up roller. Transmission efficiency is a measure of energy loss in transmission mechanisms such as gearboxes. The reduction ratio is the ratio of the motor speed to the winding roller speed. The equivalent moment of inertia, referred to the total system moment of inertia on the motor shaft, varies in real time with the roll diameter. The calculation method is as follows: ,in, The moment of inertia of the motor rotor. The moment of inertia of the core. This represents the cumulative mass of the geotextile currently rolled up.

[0048] Furthermore, the parameters of the PID controller are adaptively adjusted through gain scheduling based on the current roll diameter. The core physical basis of this adaptive gain scheduling is that as the roll diameter increases, the equivalent moment of inertia of the system increases, and the time constant and gain characteristics of the controlled object change. If the PID parameters remain unchanged, the control performance will gradually deteriorate. This embodiment establishes a functional relationship between the PID parameters and the roll diameter through a gain scheduling strategy, enabling the controller parameters to automatically adjust with the roll diameter to maintain consistent control quality. The proportional gain coefficient is used in this process. Updated to: Integral coefficient Updated to: Differential coefficients Updated to: ,in, These are the initial PID parameters. All of these are gain scheduling coefficients. , , The optimal value is determined using a genetic algorithm. Finally, the deviation between the target tension and the measured tension is used as feedback input, and a PID controller generates the feedback torque compensation. The feedforward torque and feedback torque are combined and output to the take-up motor driver. After receiving the torque command, the motor driver converts it into a corresponding electromagnetic torque output through an internal current closed-loop control algorithm, driving the take-up roller to run with the corresponding torque, thereby achieving precise control of the tension of the geotextile in the take-up section.

[0049] Understandably, the feedforward channel handles the compensation for known and predictable disturbance components caused by changes in rotational inertia. Its response speed is limited only by the system's computation cycle, eliminating the need for passive adjustment after deviations occur. The feedback channel (PID controller) is responsible for handling residual tension deviations caused by model parameter deviations and external unknown disturbances (such as fluctuations in material thickness and changes in roller friction coefficient), ensuring steady-state accuracy through a closed-loop self-correction mechanism. The synergistic effect of feedforward and feedback enables the system to possess both fast dynamic response performance and accurate steady-state tracking performance, effectively solving the technical problem of large tension fluctuations caused by response lag under acceleration and deceleration conditions in traditional pure feedback control schemes.

[0050] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for controlling the winding tension of spunbond geotextile based on real-time roll diameter prediction, characterized in that, Includes the following steps: S1. Determine the material parameters and winding process parameters of the spun clay geotextile. The material parameters include the core diameter. Initial inherent thickness of geotextile Radial compressibility coefficient and characteristic compressive stress The winding process parameters include the initial winding tension. and maximum target volume ; S2. Real-time data acquisition: Real-time acquisition of the winding roller angular velocity during the winding process. angular velocity of traction roller and the real-time tension of the geotextile in the winding section ; S3. Perform dual-channel fusion estimation of roll diameter, and use a dual-channel fusion strategy to estimate the current take-up roll diameter in real time. Specifically, it includes: S31, based on the diameter of the traction roller and the angular velocity of the traction roller With the angular velocity of the take-up roll The speed ratio is used to calculate the estimated roll diameter for the first channel: ; S32. Establish a layer-by-layer cumulative roll diameter model considering the radial compressibility of spun clay geotextile, based on real-time tension. And the historical tension of each layer, calculate the effective incorporation thickness of each layer layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. , where i is the level index; S33. Based on the extended Kalman filter framework, with the roll diameter and roll diameter change rate as state variables, the prediction results of S31 and S32 are introduced into the filter as two independent observations. The estimation error covariance of each channel is calculated in real time through covariance recursion, and the fusion weights of the two channels are adaptively updated to output the fused roll diameter prediction. ; S4. Based on the estimated fusion roll diameter The current target winding tension is calculated in real time using a preset tension-wound diameter decrease curve. ; S5. Based on the estimated roll diameter Feedforward calculation of the torque compensation required due to changes in moment of inertia, while simultaneously using the target tension. With measured tension deviation As a feedback input, the PID controller generates a feedback torque compensation amount, and the feedforward torque and feedback torque are combined and output to the winding motor driver to adjust the electromagnetic torque of the winding motor in real time, thereby realizing the composite closed-loop control of the winding tension.

2. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 1, characterized in that, The formula for calculating the effective in-th layer thickness in step S32 is as follows: ,in, The radial compressive stress currently borne by the i-th layer of material is generated by the cumulative tension of each outer layer; radial compressive stress Based on the theory of elastic thin-layer accumulation, when the total number of involved layers is n: Where W is the winding width, Let J be the winding tension when the j-th layer is wound in. The wrap angle corresponding to the j-th layer, This is the estimated volume diameter when the j-th layer is involved.

3. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 2, characterized in that, The effective incorporation thickness of each layer is calculated layer by layer. And sum them up to obtain the estimated roll diameter of the second channel. The specific implementation is as follows: .

4. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 1, characterized in that, In step S33, the state equation of the extended Kalman filter framework is expressed in terms of state variables: Establish, Where F is the state transition matrix and the rate of change of volume diameter is... The thickness is determined by the current winding speed v(t) and the effective thickness of the geotextile. ,in, , The effective thickness of the single layer being involved at the current moment is defined by the state equation, which embeds the material compressibility characteristics into the state transition process, so that the filter pre-estimation stage already includes physical constraint information.

5. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 4, characterized in that, In step S33, the prediction results of S31 and S32 are introduced into the filter as two independent observations. The estimation error covariance of each channel is calculated in real time through covariance recursion, and the fusion weights of the two channels are adaptively updated to output the fused convolution diameter prediction. Specifically, it includes: The predicted roll diameter results of the first channel and the predicted roll diameter results of the second channel are treated as two independent observations, and corresponding observation equations are constructed respectively. The observation matrices of the two channels only extract the roll diameter component in the state vector, and their respective observation noise and initial observation noise covariance are configured. According to the state transition equation, the prior state estimate at the current time is recursively calculated from the posterior state estimate at the previous time step, and the prior estimate error covariance matrix at the current time step is calculated simultaneously. The observations from two channels are introduced sequentially to perform two Kalman update operations. First, the observations from the first channel are introduced to update the prior state estimate for the first time, and the innovation, innovation covariance, and Kalman gain of the first channel are calculated to obtain the state estimate and error covariance matrix after the first update. Then, the observations from the second channel are introduced to update the state and covariance results after the first update for the second time, and the innovation, innovation covariance, and Kalman gain of the second channel are calculated to finally obtain the posterior state estimate and posterior error covariance matrix at the current time. Based on the innovation covariance of the two channels, extract adaptive fusion weights: ,in, The fusion weight for the first channel, For the fusion weight of the second channel, The trace of the matrix; the estimated fused volume diameter is obtained by weighting according to the fusion weights. .

6. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 1, characterized in that, Based on the estimated fusion volume diameter The current target winding tension is calculated in real time using a preset tension-wound diameter decrease curve. ,include: The obtained estimated fusion volume Substitute the preset tension-roll diameter decrease curve to calculate the target tension value at the current moment: ,in, This is the tension attenuation coefficient, with a value ranging from 0.3 to 0.8; The pressure uniformity compensation coefficient ranges from 0.05 to 0.2; the target tension value is limited by upper and lower limits to obtain the final output target winding tension value at time t.

7. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 1, characterized in that, The specific calculation method for the feedforward torque compensation in step S5 is as follows: Based on the estimated roll diameter Real-time calculation of the equivalent moment of inertia referred to the motor shaft And feedforward to calculate torque compensation: ,in, The angular acceleration of the take-up roller. Transmission efficiency is a measure of energy loss in transmission mechanisms such as gearboxes. The reduction ratio is the ratio of the motor speed to the winding roller speed. The equivalent moment of inertia, referred to the total system moment of inertia on the motor shaft, varies in real time with the roll diameter. The calculation method is as follows: ,in, The moment of inertia of the motor rotor. The moment of inertia of the core. This represents the cumulative mass of the geotextile currently rolled up.

8. The method for controlling the winding tension of spunbond geotextile based on real-time roll diameter estimation according to claim 1, characterized in that, In step S5, the parameters of the PID controller are determined based on the current roll diameter. Adaptive adjustment of gain scheduling is performed, specifically including: scaling factor Updated to: Integral coefficient Updated to: Differential coefficients Updated to: ,in, These are the initial PID parameters. All of these are gain scheduling coefficients.

Images