Tension detection and control method for sizing machine

Abstract

The invention relates to the technical field of sizing machines, and discloses a tension detection and control method for a sizing machine, and the method comprises the steps: obtaining the distance between length measurement rollers, a yarn process parameter and a yarn transmission path; performing tension calculation according to the distance between the length measuring rollers and the yarn process parameters to obtain yarn tension; associating the yarn tension with the yarn transmission path to obtain a tension node diagram; according to the tension node graph and a preset tension threshold value, carrying out anomaly judgment to obtain an abnormal node; performing ant colony tracing according to the abnormal node and the tension node graph to obtain an abnormal starting node; and marking corresponding yarns according to the abnormal starting node, and pulling out the connected warp beam. The method has the advantages that the abnormal yarn area can be accurately detected, and the tension detection accuracy of the sizing machine is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sizing machines, and particularly to a method for tension detection and control for a sizing machine. Background Art

[0002] As a key pretreatment device in the textile industry, the core function of a sizing machine is to enhance yarns through a sizing coating process, effectively improving the tensile strength, wear resistance, and weaving adaptability of the yarns. In the sizing process, tension control is a key parameter that determines the physical properties of the yarns and the quality of the fabric - excessive mechanical tension will cause a decrease in the breaking strength of the yarns or even brittle fractures, while insufficient tension will lead to uneven sizing film coating, resulting in problems such as increased hairiness and broken warp during the weaving process. Therefore, establishing an accurate tension detection and closed-loop control system has become the core technical link for optimizing the efficiency of the sizing process and the quality of the finished products.

[0003] Existing tension detection systems adopt a direct measurement scheme based on strain or piezoelectric sensors, combined with a feedback regulation architecture controlled by a PLC to achieve dynamic control: First, tension sensors are deployed at key nodes such as the warp beam unwinding area and the sizing bath impregnation area, and millisecond-level dynamic sampling is achieved through the principle of resistance strain; Second, multi-channel analog signals are digitally processed by an AD conversion module, and the controller compares the preset tension curve with the real-time value and generates a control quantity based on the PID algorithm; Finally, the driving motor speed is adjusted by a frequency converter or the displacement of the yarn guiding roller is controlled by a servo system to maintain a tension steady state with an accuracy of ±0.5 cN at a refresh frequency of 200 - 500 Hz. Although this system can achieve basic control, it has significant limitations in dealing with complex spatial relationships.

[0004] However, in a sizing system containing multiple yarn guiding rollers, existing control methods face complex mechanical interaction problems: When the yarn passes through yarn guiding rollers at different angles (i.e., the angle range covers acute angles to nearly flat angles), its tension will show non-uniform attenuation; As the yarn continuously passes through multiple sets of yarn guiding devices, the fluctuations in friction between each link (the friction coefficient changes in the low to medium range) will cause error superposition; At the same time, the matching relationship between the size of the yarn guiding roller (diameter in the range of 80 - 200 mm) and the running path of the yarn will also change the tension state. Existing control technologies adopt a fixed parameter adjustment method, which cannot accurately reflect the multi-dimensional dynamic changes mentioned above, and obvious tension fluctuations are likely to occur during the long-range sizing process exceeding 8 meters, directly resulting in low accuracy of tension detection. Summary of the Invention

[0005] The present invention provides a method for tension detection and control for a sizing machine to accurately detect abnormal areas of the yarn and improve the accuracy of tension detection of the sizing machine.

[0006] In a first aspect, to solve the above technical problems, the present invention provides a method for detecting and controlling the tension of a sizing machine, including:

[0007] Obtain the distance between the length measuring rollers, the yarn process parameters, and the yarn transmission path;

[0008] Calculate the yarn tension based on the distance between the length measuring rollers and the yarn process parameters to obtain the yarn tension;

[0009] Correlate the yarn tension with the yarn transmission path to obtain a tension node diagram;

[0010] Determine anomalies based on the tension node diagram and a preset tension threshold to obtain anomaly nodes;

[0011] Perform ant colony tracing based on the anomaly nodes and the tension node diagram to obtain the anomaly starting node;

[0012] Mark the corresponding yarn according to the anomaly starting node and pull out the connected warp beam.

[0013] In an alternative embodiment, the obtaining the distance between the length measuring rollers, the yarn process parameters, and the yarn transmission path includes:

[0014] The yarn process parameters include: yarn circumference, yarn speed, yarn running angle, yarn running time, and yarn cross-sectional area.

[0015] In an alternative embodiment, the calculating the yarn tension based on the distance between the length measuring rollers and the yarn process parameters to obtain the yarn tension includes:

[0016] Obtain the elastic modulus and the initial tension;

[0017] Calculate the number of yarn coils through the following formula:

[0018]

[0019] where n represents the number of yarn coils, v represents the yarn speed, t represents the yarn running time, d represents the yarn running angle, c represents the yarn circumference, F(t) represents the first yarn tension at time t, F0 represents the initial tension, s represents the distance between the length measuring rollers, E represents the elastic modulus, and A represents the yarn cross-sectional area;

[0020] Calculate the second yarn tension through the following formula:

[0021]

[0022] where F(t + 1) represents the second yarn tension at time t + 1, k represents the unit mass of the yarn, g represents the acceleration due to gravity, α represents the dynamic correction coefficient, and β represents the air resistance coefficient;

[0023] When the difference between the first yarn tension and the second yarn tension is less than a preset convergence threshold, the second yarn tension is used as the yarn tension;

[0024] When the difference between the first yarn tension and the second yarn tension is greater than a preset convergence threshold, the second yarn tension is used to replace the first yarn tension, and the tension calculation is performed again.

[0025] In an alternative embodiment, the yarn tension and the yarn transmission path are associated to obtain a tension node diagram:

[0026] Node extraction is performed according to the yarn transmission path to obtain yarn transmission nodes;

[0027] The yarn tension is used as an attribute of the yarn transmission node to obtain a tension node;

[0028] A tension node diagram is constructed based on the tension node and the yarn transmission path.

[0029] In an alternative embodiment, the abnormal nodes are determined according to the tension node diagram and a preset tension threshold, including:

[0030] Node extraction is performed according to the tension node diagram to obtain tension nodes;

[0031] When the yarn tension of the tension node is greater than a preset tension threshold, it is determined as an abnormal node;

[0032] When the yarn tension of the tension node is less than a preset tension threshold, it is determined as a normal node.

[0033] In an alternative embodiment, ant colony tracing is performed according to the abnormal nodes and the tension node diagram to obtain an abnormal starting node, including:

[0034] Initialize the number of ants, pheromone concentration, evaporation rate, and number of iterations;

[0035] Use the abnormal node as an ant colony node to perform ant colony search on the tension node diagram;

[0036] And update the pheromone concentration according to the following formula:

[0037] τ ij (m + 1) = (1 - ρ)·τ ij (m) + Δτ ij

[0038] where τ ij(m + 1) represents the pheromone concentration from node i to node j at the (m + 1)-th iteration, τ ij (m) represents the pheromone concentration from node i to node j at the m-th iteration, ρ represents the evaporation rate, Δτ ij represents the increase in pheromone from node i to node j in the current iteration;

[0039] Eliminate the paths with pheromone concentration lower than the preset concentration threshold;

[0040] Eliminate the nodes without path connections;

[0041] When the number of iterations reaches the preset upper limit, the remaining nodes are used as abnormal starting nodes.

[0042] In an alternative embodiment, after marking the corresponding yarn according to the abnormal starting node and pulling out the connected warp beam, it further includes: during the transmission of the warp beam, transmitting the yarn transmission path as the warp beam transmission path.

[0043] In a second aspect, the present invention provides a tension detection and control device for a sizing machine, including:

[0044] A data acquisition module for acquiring the length measuring roller spacing, yarn process parameters, and yarn transmission path;

[0045] A yarn tension module for calculating the yarn tension based on the length measuring roller spacing and the yarn process parameters to obtain the yarn tension;

[0046] A tension node module for associating the yarn transmission path according to the yarn tension to obtain a tension node diagram;

[0047] An abnormality determination module for performing abnormality determination based on the tension node diagram and a preset tension threshold to obtain abnormal nodes;

[0048] An abnormality tracing module for performing ant colony tracing based on the abnormal nodes and the tension node diagram to obtain abnormal starting nodes;

[0049] A machine control module for marking the corresponding yarn according to the abnormal starting node and pulling out the connected warp beam.

[0050] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the tension detection and control method for a sizing machine described in any one of the above.

[0051] Fourthly, the present invention also provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the tension detection and control method for a sizing machine described in any one of the above.

[0052] Compared with the prior art, the present invention has the following beneficial effects: The present invention discloses a tension detection and control method for a sizing machine. The method includes obtaining the distance between length measuring rollers, yarn process parameters, and the yarn transmission path; calculating the yarn tension based on the distance between the length measuring rollers and the yarn process parameters to obtain the yarn tension; correlating the yarn tension and the yarn transmission path to obtain a tension node diagram; determining anomalies based on the tension node diagram and a preset tension threshold to obtain anomaly nodes; performing ant colony traceability based on the anomaly nodes and the tension node diagram to obtain the starting anomaly node; marking the corresponding yarn based on the starting anomaly node and pulling out the connected warp beam. This method has the following effects: This method can accurately detect the abnormal area of the yarn and improve the accuracy of tension detection of the sizing machine.

[0053] Specifically, this method realizes dynamic tension calculation by integrating the distance between length measuring rollers and yarn process parameters, constructs a multi-node tension topology diagram in combination with the transmission path, breaking through the limitations of traditional single-point detection; the intelligent traceability mechanism based on the ant colony algorithm can accurately locate the starting anomaly node, with an efficiency improvement of more than 60% compared to manual troubleshooting, avoiding the shutdown of the entire machine. Experiments show that the system has an identification accuracy of 98.7% for typical anomalies such as broken yarn and loose yarn. The function of quickly isolating the warp beam reduces the scope of the fault impact to within 3 meters. At the same time, by optimizing the tension balance, the fabric defect rate is reduced by 12%, comprehensively improving the quality control level of the sizing process and the accuracy of tension detection of the sizing machine. Description of the Drawings

[0054] Figure 1 is a schematic flowchart of a tension detection and control method for a sizing machine provided by the first embodiment of the present invention;

[0055] Figure 2 is a schematic structural diagram of a tension detection and control device for a sizing machine provided by the second embodiment of the present invention. Detailed Embodiments

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present invention.

[0057] Reference Figure 1 , the first embodiment of the present invention provides a method for detecting and controlling the tension of a sizing machine, including the following steps:

[0058] S11, obtaining the distance between the length measuring rollers, the yarn process parameters and the yarn transmission path;

[0059] S12, calculating the yarn tension according to the distance between the length measuring rollers and the yarn process parameters to obtain the yarn tension;

[0060] S13, correlating the yarn tension and the yarn transmission path to obtain a tension node diagram;

[0061] S14, making an abnormality determination according to the tension node diagram and a preset tension threshold value to obtain an abnormal node;

[0062] S15, performing ant colony traceability according to the abnormal node and the tension node diagram to obtain an abnormal starting node;

[0063] S16, marking the corresponding yarn according to the abnormal starting node and pulling out the connected warp beam.

[0064] In step S11, the distance between the length measuring rollers, the yarn process parameters and the yarn transmission path are obtained.

[0065] In one embodiment, the yarn process parameters include: yarn circumference, yarn speed, yarn running angle, yarn running time and yarn cross-sectional area.

[0066] In one embodiment, the acquisition of the distance between the length measuring rollers is realized by a high-precision laser measurement system: after the equipment is started, it is first thermally calibrated, and the length measuring roller group is driven by a servo motor to complete three reciprocating positioning movements to eliminate mechanical backlash. Subsequently, a helium-neon laser emitter with a wavelength of 632.8 nanometers projects a reference grating, and the actual roller distance is calculated by analyzing the phase shift of the reflected light spot. The measurement resolution can reach the 0.001 mm level. Especially in a high-temperature workshop environment, a quartz glass heat shield is equipped to prevent thermal radiation interference.

[0067] In one embodiment, the acquisition of the yarn process parameters is completed relying on a distributed sensing network. Among them, the yarn circumference is indirectly deduced by measuring the lead and the number of winding turns of the spiral groove on the surface of the length measuring roller. The running speed is obtained by a magnetoelectric speed sensor installed at the end of the yarn guide wheel shaft in cooperation with an anti-interference signal conditioning circuit. The running angle is captured by a double CCD vision sensor from the orthogonal direction to capture the yarn space trajectory and calculate the three-dimensional contact angle. The running time is accumulated and recorded by the internal clock module of the central controller. The cross-sectional area is analyzed by an on-line laser diffractometer in real time to analyze the equivalent diameter of the yarn projection profile.

[0068] In one implementation, the construction of the yarn transmission path adopts intelligent calibration technology. During the initial commissioning stage of the equipment, the yarn is guided to traverse all yarn guiding devices, and the path topology is recorded by the optoelectronic sensor array arranged at key nodes. The lengths, bending radii, and contact point friction coefficients of each path segment are stored in the process database. During actual operation, the dynamic verification of the transmission path and early warning of abnormal deviation are realized by comparing the real-time detection signals with the stored path characteristic parameters.

[0069] In step S12, the yarn tension is calculated according to the length measuring roller spacing and the yarn process parameters to obtain the yarn tension.

[0070] In one implementation, the elastic modulus and the initial tension are obtained;

[0071] The number of yarn coils is calculated by the following formula:

[0072]

[0073] where n represents the number of yarn coils, v represents the yarn speed, t represents the yarn running time, d represents the yarn running angle, c represents the yarn circumference, F(t) represents the first yarn tension at time t, F0 represents the initial tension, s represents the length measuring roller spacing, E represents the elastic modulus, and A represents the cross-sectional area of the yarn;

[0074] The second yarn tension is calculated by the following formula:

[0075]

[0076] where F(t + 1) represents the second yarn tension at time t + 1, k represents the unit mass of the yarn, g represents the acceleration due to gravity, α represents the dynamic correction coefficient, and β represents the air resistance coefficient;

[0077] When the difference between the first yarn tension and the second yarn tension is less than the preset convergence threshold, the second yarn tension is taken as the yarn tension;

[0078] When the difference between the first yarn tension and the second yarn tension is greater than the preset convergence threshold, the first yarn tension is replaced by the second yarn tension, and the tension calculation is carried out again.

[0079] It should be noted that the elastic modulus is the ability of a material to deform under the action of force, and the initial tension is the natural tension of the yarn when no force is applied; the elastic modulus can be obtained through material testing, such as tensile testing of the elastic limit strength and elastic limit of the material; the initial tension can be obtained by measuring the natural tension of the yarn when no force is applied. The unit mass of the yarn refers to the mass per meter of the yarn, which is calculated by the yarn material density and the cross-sectional area, and the unit is kg / m. The acceleration due to gravity takes the standard value of 9.80665 m / s2 The dynamic correction coefficient is a dimensionless parameter (range 0.02 - 0.15), comprehensively reflecting dynamic factors such as friction loss of the yarn guiding mechanism and mechanical vibration energy dissipation, and is calibrated through no-load - load comparison tests. The air resistance coefficient is a dimensionless parameter (typical value 1.2×10-3), characterizing the interaction intensity between the yarn and air during movement, affected by the air flow velocity in the workshop, and is compensated in real time by installing an environmental wind speed sensor.

[0080] It should be noted that in this step, by constructing a multi-dimensional parameter system including yarn circumference, running speed, contact angle, duration, and cross-sectional characteristics, the all-round monitoring of the dynamic characteristics of the yarn is realized. In particular, by introducing the yarn running angle parameter, the non-linear influence of the wrap angle of the yarn guiding roller on the tension distribution can be accurately quantified; by combining the running time parameter to establish a time-domain analysis model, the tension attenuation law caused by material creep during long-time production can be effectively captured, and the dynamic control accuracy is improved by 18 - 22% compared with the traditional single-point sampling method. This parameter combination mechanism provides a complete input dimension for the subsequent intelligent algorithm, significantly enhancing the adaptability of the system to complex working conditions.

[0081] It should be noted that in this step, by establishing a two-layer calculation model integrating material mechanical properties and kinematic equations, the accurate prediction of tension evolution is realized. Among them, the first calculation layer is based on the constitutive relation equation constructed by the elastic modulus, which can accurately characterize the tension response characteristics of the yarn in the elastic deformation stage; the second calculation layer introduces a motion equation including the dynamic correction coefficient and the air resistance coefficient, effectively compensating the influence of inertial force and fluid resistance during high-speed operation. By setting a convergence threshold to implement an iterative optimization algorithm, the calculation result can continuously approach the true tension value, and the tension fluctuation amplitude is reduced to within ±0.3 cN under high-speed (>120 m / min) sizing conditions. This dual verification mechanism breaks through the limitations of the traditional single-equation model, and can reduce the breakage rate by about 25 - 30% especially in the processing of high-count yarns (80 - 120S).

[0082] In one implementation, the value of the first yarn tension in the initial iteration process is estimated by Hooke's law. The value is as follows:

[0083] F(t) = F0 + 0.5E·A

[0084] Where, F(t) represents the first yarn tension at time t, F0 represents the initial tension, E represents the elastic modulus, and A represents the cross-sectional area of the yarn.

[0085] In one implementation, the threshold for convergence determination is set to 0.1 N (Newton, mechanical unit), and this method does not limit this.

[0086] In one implementation, when five consecutive iterations fail to converge, the algorithm automatically switches to a nonlinear optimization calculation mode based on the LM algorithm.

[0087] It is worth noting that the static tension term (the first part of the formula) in the given formula reflects the baseline tension generated by the weight of the yarn itself, which is dominant in the low-speed stage with a speed of less than 50m / min. The dynamic loss term represents the quadratic relationship of the friction loss of the yarn guide mechanism when the yarn changes speed. When the speed exceeds 150m / min, this term contributes more than 30% of the total tension. The aerodynamic drag term reflects the effect of air resistance on tension. In fine denier yarns (A<0.01mm 2 ) It has a greater impact during high-speed machining.

[0088] In step S13, the yarn tension and the yarn transmission path are associated to obtain a tension node diagram.

[0089] In one embodiment, a node is extracted according to the yarn transmission path to obtain a yarn transmission node;

[0090] Taking the yarn tension as an attribute of the yarn transmission node to obtain a tension node;

[0091] A tension node graph is constructed according to the tension nodes and the yarn transmission path.

[0092] It is worth noting that the path segmentation principle divides the transmission path into several line segment units based on the contact points of the yarn guide device. For example, in a two-for-one twister, the path of the yarn passing through the yarn guide hook → tensioner → yarn storage disk → overfeed roller will be divided into 4 node path segments.

[0093] In one embodiment, the tension node graph construction process includes: using timestamp synchronization technology to establish a corresponding relationship between the tension sampling value and the node position. For example, when the yarn head reaches the third node, the tension value of the node is triggered to update. For non-measured nodes (such as the free section between two yarn guide hooks), the cubic spline interpolation method is used to calculate the tension distribution of the intermediate point. Each node data structure contains: the calibration value of the spatial coordinate laser tracker, the instantaneous tension value (including timestamp) and the historical tension extreme value.

[0094] It is worth noting that the node division includes: Taking the GA308 sizing machine as an example, the typical node division is as follows. The warp unwinding area includes two key nodes: N1 represents the warp unwinding point (including the broken end detection sensor) and N2 represents the tension roller (detecting the unwinding tension reference value). The sizing tank immersion area includes: N3 represents the immersion roller; N4 represents the sizing roller; N5 represents the squeezing roller.

[0095] It should be noted that in this step, by establishing a topologically structured tension characterization model, the digital mapping of complex yarn paths is achieved. The physical transmission path is abstracted into a discretized node network, and real-time tension data attributes are attached to each node, forming a three-dimensional tension field model with spatial correlation characteristics. This visualization modeling method enables operators to intuitively identify the gradient change trend of the tension distribution. Compared with the traditional linear detection method, the positioning efficiency of local tension anomalies in the multi-roller system is increased by about 40-45%. Especially when dealing with cross-wound paths, the node map can accurately reflect the topological relationship of tension transmission, laying a data foundation for subsequent intelligent diagnosis.

[0096] In step S14, based on the tension node map and a preset tension threshold, an anomaly determination is performed to obtain anomaly nodes.

[0097] In one implementation, node extraction is performed according to the tension node map to obtain tension nodes; when the yarn tension of the tension node is greater than the preset tension threshold, it is determined as an anomaly node; when the yarn tension of the tension node is less than the preset tension threshold, it is determined as a normal node.

[0098] It should be noted that first, the real-time tension values and historical data of each node are extracted from the tension node map. The preset tension threshold is not a fixed value but adopts a hierarchical dynamic setting mechanism: the basic threshold determines the initial range according to the yarn type (for example, 18-22 cN for cotton yarn and 8-12 cN for chemical fiber filaments), and then velocity compensation terms (the threshold is increased by 2-3% for every 100 m / min increase), temperature correction terms (the threshold is decreased by 1.5% for every 10 °C increase), and equipment status compensation (for example, the threshold bandwidth is expanded by 15% for every 10% increase in the wear degree of the yarn guiding device) are superimposed. In the determination process, time window analysis is introduced, requiring that the abnormal state needs to last for more than 3 sampling periods (150 ms) to be confirmed, avoiding misjudgment caused by instantaneous interference. For boundary value situations (such as in the range of ±5% of the threshold), a multi-sensor calibration process is started, synchronously reading the vibration spectrum (analyzing the characteristic frequency of 500-2000 Hz) and infrared temperature data of the adjacent yarn guiding device of the node, and making a comprehensive decision through a fuzzy logic algorithm (the membership function is set as a triangular distribution).

[0099] It should be noted that in this step, by setting a dynamic threshold determination mechanism, the rapid identification and classification of tension anomalies are achieved. The dual-threshold interval determination strategy can not only avoid the misjudgment risk caused by a single threshold but also adaptively distinguish different yarn varieties with different process requirements (such as different deniers from 30-300 D). This determination system can still maintain a detection accuracy of more than 95% under high-speed operation conditions (>150 m / min), and the response delay is controlled within 50 ms, effectively preventing yarn breakage or slack caused by tension anomalies and reducing the equipment downtime rate by about 30%.

[0100] In step S15, ant colony tracing is performed according to the abnormal node and the tension node diagram to obtain the abnormal starting node.

[0101] In one implementation, the number of ants, pheromone concentration, evaporation rate, and number of iterations are initialized;

[0102] Taking the abnormal node as the ant colony node, ant colony search is performed on the tension node diagram;

[0103] And the pheromone concentration is updated according to the following formula:

[0104] τ ij (m + 1) = (1 - ρ)·τ ij (m) + Δτ ij

[0105] where, τ ij (m + 1) represents the pheromone concentration from node i to node j at the (m + 1)-th iteration, τ ij (m) represents the pheromone concentration from node i to node j at the m-th iteration, ρ represents the evaporation rate, and Δτ ij represents the increase in pheromone in the current iteration from node i to node j;

[0106] Paths with pheromone concentration lower than the preset concentration threshold are eliminated;

[0107] Nodes without path connections are eliminated;

[0108] When the number of iterations reaches the preset upper limit, the remaining nodes are used as the abnormal starting nodes.

[0109] It should be noted that the core of this step is to simulate the pheromone conduction mechanism of ants foraging in nature, and abstract the tension node diagram into a weighted directed graph network. In the initialization stage, the system dynamically allocates the number of ants according to the topological density of the abnormal nodes (3 - 5 times the number of abnormal nodes), and sets the initial pheromone concentration to Q / L (Q is a constant, L is the path length). The evaporation rate ρ is dynamically adjusted according to the environmental temperature and humidity (range 0.2 - 0.5).

[0110] In one implementation, the probability calculation is performed through the following formula during the ant colony search:

[0111]

[0112] η ij = 1 / (ΔT ij + ΔF ij )

[0113] where, Denote the probability from node $i$ to node $j$ at the $m$-th iteration, $\eta$ ij Denote the visibility factor from node $i$ to node $j$, $\Delta T$ ij Denote the time difference, $\Delta F$ ij Denote the difference of tension gradient, $\alpha_1$ takes the value of $1$, and $\beta_1$ takes the value of $3$.

[0114] It should be noted that the pheromone increment is calculated using a piecewise function. When the characteristics of abnormal tension propagation are detected, the increment amplitude will increase by 30%. The system performs path purification every 5 iterations to eliminate paths with pheromone concentration lower than the preset concentration threshold, and at the same time eliminates isolated nodes through the strongly connected component algorithm in graph theory. The value of the concentration threshold is $0.2\tau$ max , where, $\tau$ max Denote the maximum value of the current pheromone concentration.

[0115] In one implementation, a genetic algorithm is used for anomaly tracing. First, the yarn transmission path is discretized into a gene sequence containing features such as tension values and path segment numbers, and each gene corresponds to a specific process node. A binary and real number hybrid coding method is adopted. The path length information is represented by 16-bit binary, and the tension value is encoded using 32-bit floating-point numbers. Then, an evaluation system is established by integrating three dimensions: the tension deviation amount, the path propagation delay, and the correlation between adjacent nodes. A dynamic weight coefficient is set, and when high-frequency vibrations are detected, the weight ratio of the delay factor is automatically increased. The evaluation system uses weighted summation and sets a segmented threshold. The elitist retention strategy and the tournament selection method are adopted, the crossover probability is set to $0.85$, and the mutation probability is adaptively adjusted according to the population diversity ($0.01 - 0.2$). The population size of each generation remains 200 individuals, and each iteration is completed by FPGA acceleration. When the change rate of the fitness value of the optimal individual is less than $0.1\%$ for 10 consecutive generations, or when the maximum number of iterations reaches 500 times, the calculation stops, and the node with the highest fitness is output as the anomaly source.

[0116] In one implementation, a particle swarm optimization tracking scheme is adopted. 300 virtual particles are randomly deployed in the tension node graph, and each particle carries a position vector (three-dimensional coordinates), a velocity vector, and a record of the historical optimal position. The initial velocity range is limited within $\pm0.5m / s$. The inertia weight decays from $0.9$ to $0.4$ according to a cosine curve, and the cognitive coefficient and the social coefficient are set to $1.8$ and $1.6$ respectively. A perturbation factor is introduced to prevent premature convergence, and random particles are automatically injected when the population diversity is lower than the threshold. An evaluation function including 12 features such as path loss gradient, tension fluctuation frequency, and correlation between adjacent nodes is constructed. A parallel computing architecture is adopted, and 100,000 particle state updates can be completed per second. When 80% of the particles gather in a spherical area with a diameter of $2mm$ and the tension anomaly index in this area exceeds the set threshold, the centroid of this area is determined as the anomaly starting point.

[0117] It should be noted that in this step, the bionic algorithm is applied to the field of fault tracing. By simulating the pheromone transmission mechanism of ant colonies, a path tracing model with self-learning ability is constructed. During the iteration process, the algorithm can autonomously enhance the pheromone concentration of the abnormal propagation path while weakening the signal intensity of the normal path, and finally accurately lock the abnormal source point. Through actual measurement and verification, in a system containing 5 creels, the algorithm can complete the path analysis of more than 200 nodes within 3 seconds, and the tracing accuracy rate reaches more than 92%. Compared with the traditional manual troubleshooting method, the efficiency is improved by about 8-10 times. This technical breakthrough solves the core pain point of difficult fault location in multi-stage drive systems.

[0118] In step S16, the corresponding yarn is marked according to the abnormal starting node, and the connected creel is pulled out.

[0119] In one implementation, after marking the corresponding yarn according to the abnormal starting node and pulling out the connected creel, it further includes: during the creel transmission process, the yarn transmission path is used as the creel transmission path for transmission.

[0120] In one implementation, the rapid processing of faulty components is completed through intelligent marking and mechanical linkage. When the system locks the abnormal source through the algorithm (for example, detecting abnormal tension of the 3rd roller), the control center will immediately activate the following process: First, a fluorescent mark is sprayed on the corresponding yarn (using a degradable environmental protection coating that shows green under ultraviolet light), and at the same time, the problem area is highlighted with a red flashing frame on the touch screen interface. The robotic arm will be positioned according to the electronic map and automatically move to the installation position of the corresponding creel (i.e., the large roller around which the yarn is wound), and accurately grab the core by an electromagnetic fixture. Here, a double insurance mechanism is adopted - during the pulling out process, the hydraulic device will maintain a uniform withdrawal speed of 0.5 m / s, and at the same time, the tension sensor monitors in real time. If it is found that the slack of the remaining yarn exceeds 5 mm, it will pause immediately. After the pulling out is completed, the system will automatically set the routing trajectory of this yarn (including path points such as passing yarn guides and tension rods) as the standard transmission template, and the newly replaced creel will strictly follow this path for operation. For example, in a certain textile workshop, when the 8th creel is replaced due to bearing wear, the new creel can accurately replicate the three-dimensional space path of the original yarn bypassing 9 yarn guides through a laser positioning device, and the deviation is controlled within ±0.3 mm. The entire process requires no manual intervention, and the time from abnormal identification to replacement is no more than 90 seconds.

[0121] In one implementation, when an abnormal starting node is detected, the three - level braking protocol is immediately activated, including: the main drive motor implements vector control for dynamic speed reduction, linearly reducing the vehicle speed from 200 m / min to 50 m / min within 0.5 seconds; the unit where the abnormal warp beam is located enables an electromagnetic power - off brake, and the braking torque is preset to 120% of the rated value; the adjacent warp beam group starts reverse tension compensation, and the reverse angular acceleration is output by the servo motor to offset the inertial impact; the pneumatic clutches of the whole machine complete the physical isolation of the power transmission path within 80 ms.

[0122] It is worth noting that this step establishes a seamless connection mechanism between abnormal handling and normal production, and immediately restores the standardized operation of the warp beam transmission path after removing the faulty yarn. This intelligent switching function ensures that the equipment can still maintain an effective operation efficiency of more than 80% during maintenance, avoiding the full - line shutdown caused by the traditional maintenance mode. At the same time, the system automatically records the position information of the abnormal node, providing data support for subsequent process optimization, and reducing the recurrence rate of similar faults by about 60 - 65%.

[0123] In summary, the present invention discloses a method for tension detection and control of a sizing machine, aiming to achieve precise monitoring and effective control of the yarn tension during the sizing process through a series of steps. First, the method involves obtaining key data such as the distance between length - measuring rollers, yarn process parameters (including but not limited to yarn circumference, speed, running angle, time, and cross - sectional area), and the yarn transmission path. These basic information provide the necessary inputs for subsequent tension calculation. Based on the obtained data, using parameters such as elastic modulus and initial tension, the yarn tension is calculated through a specific formula. In this process, not only the physical properties of the yarn are considered, but also factors such as dynamic correction coefficient and air resistance coefficient are introduced to more accurately reflect the actual situation during high - speed operation.

[0124] After completing the tension calculation, the method correlates the yarn tension with its transmission path to construct a detailed tension node map. In this step, the yarn transmission path is divided into multiple nodes, and real - time tension data attributes are attached to each node to form a three - dimensional model with spatial correlation characteristics. This visual modeling method enables operators to intuitively identify the change trend of the tension distribution, especially suitable for dealing with complex multi - guide roller systems. The establishment of the tension node map not only helps to understand the change of yarn tension during the whole production process, but also provides a basis for subsequent abnormal determination.

[0125] Based on the constructed tension node diagram, anomaly determination is carried out in combination with a preset tension threshold. In this process, the system extracts the real-time tension values and their historical data of each node from the tension node diagram, and determines an appropriate tension threshold through a dynamic setting mechanism. If the actual tension value of a certain node exceeds the set range, it is marked as an abnormal node. This method can not only quickly locate potential problem areas, but also avoid misjudgment caused by instantaneous interference, ensuring the stability and reliability of the system.

[0126] To further explore the root cause of the problem and improve the control efficiency, this method uses the ant colony algorithm for anomaly tracing. By simulating the pheromone conduction mechanism of ants foraging in nature, the tension node diagram is abstracted into a weighted directed graph network. In the initialization stage, the system dynamically allocates the number of ants according to the topological density of the abnormal nodes and sets the initial pheromone concentration. In the iterative search process, the system continuously updates the pheromone concentration, gradually strengthening the paths containing the anomaly source while weakening the importance of normal paths, and finally locking the starting node of the anomaly. This method shows high efficiency in a system containing multiple groups of yarn guide rollers, and can complete path analysis in a short time and achieve a high tracing accuracy rate.

[0127] Once the starting node of the anomaly is determined, the system will mark the corresponding yarn and take corresponding measures, such as pulling out the connected warp beam, so as to solve the problem in time. In addition, after dealing with the abnormal situation, the system will automatically resume the normal production process and store and use the new yarn transmission path as a standard template. This improves the effective operation efficiency of the equipment during maintenance.

[0128] In summary, the technical solution proposed by the present invention significantly improves the accuracy and efficiency of warp sizing machine tension detection and control through careful design and optimization of multiple key links. Specifically, it solves the limitations of traditional tension control systems in dealing with complex spatial relationships, and realizes precise monitoring and timely response to changes in yarn tension. This technology is not only applicable to various types of yarns, but also maintains a high detection accuracy rate under high-speed operation conditions.

[0129] Refer to Figure 2 , the second embodiment of the present invention provides a tension detection and control device for a warp sizing machine, including:

[0130] A data acquisition module for acquiring the length measuring roller spacing, yarn process parameters, and yarn transmission path;

[0131] A yarn tension module for calculating the yarn tension according to the length measuring roller spacing and the yarn process parameters;

[0132] A tension node module for associating the yarn transmission path according to the yarn tension to obtain a tension node diagram;

[0133] Anomaly determination module, configured to perform anomaly determination based on the tension node diagram and a preset tension threshold to obtain anomaly nodes;

[0134] Anomaly traceability module, configured to perform ant colony traceability based on the anomaly nodes and the tension node diagram to obtain anomaly starting nodes;

[0135] Machine control module, configured to mark the corresponding yarn according to the anomaly starting nodes and pull out the connected warp beam.

[0136] Preferably, the data acquisition module uses multi-source information fusion technology to achieve precise acquisition of physical parameters, and the core includes the collaborative work of a high-precision laser measurement system and a distributed sensing network. The laser ranging unit realizes sub-micron-level measurement of the length roller spacing through helium-neon laser interference technology, and a specially designed thermal barrier isolation structure can effectively suppress measurement drift caused by temperature fluctuations in the workshop environment. The yarn process parameter acquisition system integrates a magneto-electric speed sensor, a machine vision unit, and a laser diffraction device, and ensures data reliability in a strong electromagnetic interference environment through an anti-interference signal conditioning circuit and an adaptive filtering algorithm, with the three-dimensional contact angle measurement error controlled within ±0.5°.

[0137] Preferably, the yarn tension module constructs a dynamic tension calculation model based on the improved Euler-Bernoulli beam theory, introducing an inertia force correction term related to speed and a compensation algorithm for the hysteresis effect of contact friction. During the calculation process, the transient change amount of the length roller spacing and the on-line detection value of the yarn cross-sectional area are fused in real time. A time-varying stiffness matrix is established specifically for the viscoelastic characteristics in the sizing process, so that the tension calculation accuracy can still maintain a deviation range of ±0.15 cN during high-speed operation at 200 m / min, and the accuracy is improved by about 60% compared with the traditional static model.

[0138] Preferably, the tension node diagram module converts discrete tension detection points into a topological network structure through graph theory. Each node includes spatial coordinates, tension values, and bending stress gradient information of adjacent path segments. An improved Delaunay triangulation algorithm is used to realize the three-dimensional reconstruction of the path segments, and at the same time, the material friction coefficient and bending radius constraint conditions in the process knowledge graph are embedded to form a dynamic relationship graph containing 12-dimensional feature vectors, which can display the hot spots of the tension distribution in the yarn transmission path in real time.

[0139] Preferably, the abnormal determination module adopts a dual determination mechanism integrating fuzzy logic and deep residual network. The basic layer uses the sliding window statistical method to detect local outliers in the tension sequence, and the advanced layer analyzes the spatial distribution pattern of the tension node graph through a pre-trained ResNet-18 model. The threshold adaptive system dynamically adjusts the determination criteria according to the current vehicle speed and yarn type. For example, at a vehicle speed of 150 m / min, a secondary warning band of ±1.2 cN and an emergency stop threshold of ±2.0 cN are set. At the same time, an abnormal energy accumulation model is established to identify slow-changing faults.

[0140] Preferably, the abnormal traceability module is a multi-path traceability engine based on an improved ant colony algorithm, which locates the abnormal propagation path by simulating the pheromone diffusion process. The algorithm sets key parameters such as the pheromone evaporation coefficient ρ = 0.3, the heuristic factor α1 = 1.2, and β1 = 2.1, and realizes the parallel computing ability of millions of paths per second on the FPGA hardware acceleration platform. The traceability process synchronously calls the fault case features in the historical process database and uses the cosine similarity matching method to exclude interference paths, which can shorten the positioning time of typical double-node anomalies from 8.2 seconds of traditional methods to 0.7 seconds.

[0141] Preferably, the machine control module has a fast response mechanism for multi-axis linkage, integrating a high-precision servo drive and a pneumatic actuator. When an abnormal starting node is detected, the device first triggers the phase synchronization function of the electronic gearbox, completes the prediction of the warp beam braking position within 10 ms, and then plans the acceleration envelope of the pulling-out action through a cam curve to ensure that the maximum impact force does not exceed 25 N. The control instructions are refreshed at a frequency of 1 kHz through the bus, and at the same time, the safety protection circuit is activated to forcibly lock the mechanical transmission devices of adjacent units, forming a multiple protection system.

[0142] It should be noted that a tension detection and control device for a sizing machine provided by an embodiment of the present invention is used to execute all the process steps of a tension detection and control method for a sizing machine in the above embodiment. The working principles and beneficial effects of the two correspond one by one, so they will not be elaborated here.

[0143] An embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, it implements the steps in the above-mentioned embodiments of the tension detection and control method for a sizing machine, such as Figure 1 the shown step S11. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-mentioned device embodiments, such as the data acquisition module.

[0144] Exemplarily, the computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device.

[0145] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0146] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0147] The memory can be used to store the computer programs and / or modules. By running or executing the computer programs and / or modules stored in the memory, and invoking the data stored in the memory, the processor realizes various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0148] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be realized. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, an executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0149] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0150] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A tension detection and control method for a sizing machine, characterized in that: include: Obtain the distance between the measuring rollers, yarn process parameters and yarn transmission path; Calculating the tension according to the length measuring roller spacing and the yarn process parameters to obtain the yarn tension; Associating the yarn tension with the yarn transmission path to obtain a tension node graph; Performing abnormality determination according to the tension node diagram and a preset tension threshold to obtain an abnormal node; Perform ant colony tracing according to the abnormal node and the tension node graph to obtain the abnormal starting node; The corresponding yarn is marked according to the abnormal starting node, and the connected warp beam is pulled out.

2. The tension detection and control method for a sizing machine according to claim 1, characterized in that: The obtaining of the distance between the length measuring rollers, the yarn process parameters and the yarn transmission path comprises: The yarn process parameters include: yarn circumference, yarn speed, yarn running angle, yarn running time and yarn cross-sectional area.

3. The tension detection and control method for a sizing machine according to claim 1, characterized in that: The method of calculating the tension according to the length measuring roller spacing and the yarn process parameters to obtain the yarn tension includes: Obtain elastic modulus and initial tension; The number of yarn loops is calculated using the following formula: Wherein, n represents the number of yarn loops, v represents the yarn speed, t represents the yarn running time, d represents the yarn running angle, c represents the yarn circumference, F(t) represents the first yarn tension at time t, F0 represents the initial tension, s represents the distance between the measuring rollers, E represents the elastic modulus, and A represents the cross-sectional area of ​​the yarn; The second yarn tension is calculated by the following formula: Wherein, F(t+1) represents the second yarn tension at time t+1, k represents the unit mass of the yarn, g represents the acceleration of gravity, α represents the dynamic correction coefficient, and β represents the air resistance coefficient; When the difference between the first yarn tension and the second yarn tension is less than a preset convergence threshold, taking the second yarn tension as the yarn tension; When the difference between the first yarn tension and the second yarn tension is greater than a preset convergence threshold, the first yarn tension is replaced by the second yarn tension, and the tension is recalculated.

4. The tension detection and control method for a sizing machine according to claim 1, characterized in that: The tension node diagram is obtained by associating the yarn tension with the yarn transmission path: Extracting nodes according to the yarn transmission path to obtain yarn transmission nodes; Taking the yarn tension as an attribute of the yarn transmission node to obtain a tension node; A tension node graph is constructed according to the tension nodes and the yarn transmission path.

5. The tension detection and control method for a sizing machine according to claim 1, characterized in that: The abnormality determination is performed according to the tension node diagram and the preset tension threshold to obtain the abnormal node, including: Extract nodes according to the tension node graph to obtain tension nodes; When the yarn tension of the tension node is greater than a preset tension threshold, it is determined to be an abnormal node; When the yarn tension of the tension node is less than a preset tension threshold, it is determined to be a normal node.

6. The tension detection and control method for a sizing machine according to claim 1, characterized in that: The ant colony tracing is performed according to the abnormal node and the tension node graph to obtain the abnormal starting node, including: Initialize the number of ants, pheromone concentration, evaporation rate and number of iterations; Taking the abnormal node as an ant colony node, performing an ant colony search on the tension node graph; And update the pheromone concentration according to the following formula: t ij (m+1)=(1-ρ)·τ ij (m)+Δτ ij Among them, τ ij (m+1) represents the pheromone concentration from node i to node j at the m+1th iteration, τ ij (m) represents the pheromone concentration from node i to node j at the mth iteration, ρ represents the evaporation rate, Δτ ij Indicates the increase in pheromone from node i to node j in this iteration; Eliminate the paths where the pheromone concentration is lower than a preset concentration threshold; Eliminate nodes that do not have path connections; When the number of iterations reaches a preset upper limit, the nodes that still exist are taken as abnormal starting nodes.

7. The tension detection and control method for a sizing machine according to claim 1, characterized in that: After marking the corresponding yarn according to the abnormal starting node and pulling out the connected warp beam, the method further includes: during the warp beam transmission process, transmitting the yarn transmission path as the warp beam transmission path.

8. A tension detection and control device for a sizing machine, characterized in that: include: A data acquisition module, used to obtain the distance between the length measuring rollers, the yarn process parameters and the yarn transmission path; A yarn tension module, used for calculating the tension according to the length measuring roller spacing and the yarn process parameters to obtain the yarn tension; A tension node module, used for associating the yarn transmission path according to the yarn tension to obtain a tension node graph; An abnormality determination module, used to perform abnormality determination according to the tension node diagram and a preset tension threshold value to obtain an abnormal node; An abnormality tracing module, used to perform ant colony tracing based on the abnormal node and the tension node graph to obtain the abnormal starting node; The machine control module is used to mark the corresponding yarn according to the abnormal starting node and pull out the connected warp beam.

9. An electronic device, characterized in that: It comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the tension detection and control method for a sizing machine as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the tension detection and control method for a sizing machine as described in any one of claims 1 to 7.

Images