A loom control method, loom controller, medium and product

Abstract

This application provides a loom control method, a loom controller, a medium, and a product, relating to the field of edge computing technology. In this method, the loom controller first receives a global standard characteristic curve from a cloud server, and then generates an actual characteristic curve through measured data acquisition and nonlinear fitting in calibration mode. Next, when it receives target process parameters from the cloud, the loom controller first determines the corresponding expected performance index based on the global standard characteristic curve, and then uses the expected performance index as the target to inversely map the actual execution parameters into the actual characteristic curve. Finally, the loom controller uses the actual execution parameters to control the current loom production. This method alleviates the problem of inconsistent effects caused by the same parameters due to the difficulty in adapting uniform parameter settings to individual loom performance differences, and improves the consistency of the overall actual process execution effect of the loom group.

Description

Technical Field

[0001] This application relates to the field of edge computing technology, and in particular to a loom control method, a loom controller, a medium, and a product. Background Technology

[0002] In the textile industry, loom control systems commonly employ a distributed configuration of "one machine, one screen." To ensure a high degree of uniformity in the final quality of fabric products from the same batch, process specifications require all looms producing that type of fabric to adhere to a unified process standard. When adjusting the process parameters of multiple looms in the workshop, operators must travel back and forth between the physical locations of each loom, manually inputting or confirming information on the screen one by one, resulting in low efficiency.

[0003] To address the aforementioned efficiency issues, existing technologies have introduced a centralized management solution based on cloud servers. In this solution, the cloud server establishes a communication connection with all looms and operating terminals in the workshop. Each loom connects to the cloud via an IoT module that supports wired or wireless connections, collecting and uploading its own loom status information in real time. Operators can remotely and uniformly modify the process parameters of one or more looms through the screen of a terminal device, thereby enabling rapid batch distribution of parameters.

[0004] However, while existing unified control methods have achieved standardization in parameter values, in actual production, there is a phenomenon of different effects of the same parameters in a group of looms (especially when old looms are included). For example, the same 3.5 Bar air pressure setting may be powerful on a new loom but insufficient on an old loom with severely worn parts, resulting in inconsistent actual process execution effects. Summary of the Invention

[0005] This application provides a loom control method, loom controller, medium, and product to alleviate the technical problem that uniform parameter settings are difficult to adapt to the individual performance differences of looms, resulting in different effects of the same parameters, and improves the consistency of the actual process execution effect of the entire loom group.

[0006] In a first aspect, this application provides a loom control method applied to a loom controller, the loom controller being configured on the loom and communicatively connected to a cloud server. The method includes: receiving a global standard characteristic curve issued by the cloud server, the global standard characteristic curve representing a first correspondence between the process parameters and weaving performance indicators of the loom within the entire process range under a preset standard loom state; controlling the loom to enter a calibration mode, driving the loom to operate under multiple gradient process parameters covering a preset working range under the calibration mode, and collecting the measured performance indicators corresponding to each gradient process parameter; performing nonlinear fitting based on multiple sets of gradient process parameters and corresponding measured performance indicators to obtain an actual characteristic curve, the actual characteristic curve representing a second correspondence between the process parameters and weaving performance indicators of the loom under the current mechanical state; responding to the target process parameters uniformly issued by the cloud server to multiple looms, determining the expected performance indicators corresponding to the target process parameters based on the global standard characteristic curve; using the expected performance indicators as target values, performing reverse mapping calculation on the actual characteristic curve to obtain the actual execution parameters that enable the current loom to achieve the expected performance indicators; and using the actual execution parameters to control the current loom to perform weaving production.

[0007] By adopting the above technical solution, the loom controller first establishes a theoretical benchmark correspondence between process parameters and weaving performance indicators under a preset standard state using a global standard characteristic curve. Then, the loom controller generates an actual characteristic curve through measured data acquisition and nonlinear fitting in calibration mode. This actual characteristic curve quantifies and represents the individual performance response differences of the current loom due to aging or mechanical wear. Next, upon receiving a unified command from the cloud, the loom controller uses the global standard characteristic curve as an index to convert the target process parameters into expected performance indicators representing the final weaving quality. Using these expected performance indicators as target values, it performs inverse calculations on the actual characteristic curve to derive the actual execution parameters that are suitable for the current machine state. Finally, the loom controller uses the actual execution parameters to control the current loom for weaving production. This method alleviates the phenomenon of inconsistent effects of the same parameters caused by differences in the state of loom equipment without requiring manual intervention on a machine-by-machine basis, and improves the consistency of the overall actual process execution effect of the loom group.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the current loom is controlled to perform weaving production using actual execution parameters, specifically including: driving the current loom to operate according to the actual execution parameters; during the operation of the current loom, acquiring the current operating performance index at a preset sampling frequency, and calculating the instantaneous deviation rate between the current operating performance index and the expected performance index; when the instantaneous deviation rate exceeds the deviation fluctuation range, generating a compensation value for the actual execution parameters based on the instantaneous deviation rate; superimposing the compensation value with the actual execution parameters to obtain the correction parameters, and using the correction parameters to control the current loom to continue weaving production.

[0009] By adopting the above technical solution, the loom controller first monitors the current operating performance indicators in real time and calculates the instantaneous deviation rate. Then, when the instantaneous deviation rate exceeds the fluctuation range, a compensation value is generated and added to the actual execution parameters. This method realizes dynamic adjustment of parameters, suppresses random disturbances in the production process in real time, and ensures that the loom performance indicators remain stable near the expected target under dynamic conditions, thereby stabilizing the process execution effect of a single loom.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of controlling the current loom to continue weaving production using the correction parameters, the method further includes: generating a historical deviation sequence based on the instantaneous deviation rate and the corresponding time, and calculating the trend characteristics of the historical deviation sequence within a preset time monitoring window; when the trend characteristics meet a preset unidirectional drift condition, determining the current working point of the target process parameter on the actual characteristic curve, wherein the unidirectional drift condition indicates that the mechanical state of the current loom has shifted; and fitting and correcting a preset neighborhood centered on the current working point on the actual characteristic curve based on the data within the preset time monitoring window.

[0011] By adopting the above technical solution, the loom controller identifies the unidirectional drift of the loom's mechanical state through historical deviation sequences, and specifically corrects the corresponding neighborhood of the actual characteristic curve to ensure that the actual characteristic curve always adapts to the current state of the loom and guarantees the accuracy of parameter mapping.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, based on data within a preset time monitoring window, a preset neighborhood centered on the current operating point on the actual feature curve is fitted and corrected. Specifically, this includes: extracting correction parameters and corresponding current operating performance indicators located within the preset time monitoring window to obtain a correction sample set; assigning a first confidence weight to the correction sample set and assigning a second confidence weight to the original data of the actual feature curve within the preset neighborhood; performing a weighted regression operation using the correction sample set and the original data based on the first and second confidence weights to update the parameter mapping relationship within the preset neighborhood; and generating the corrected actual feature curve based on the updated parameter mapping relationship.

[0013] By adopting the above technical solution, the loom controller first assigns differentiated weights to old and new data based on their temporal proximity (the first confidence weight is positively correlated with the proximity of the data acquisition time). Then, a weighted regression algorithm is used to dynamically update the parameter mapping relationship in the neighborhood of the current working point during loom operation. This method takes into account both the timeliness and fundamentality of the data, improving the accuracy of curve correction.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the expected performance index is used as the target value, and a reverse mapping calculation is performed on the actual characteristic curve to obtain the actual execution parameters that enable the current loom to achieve the expected performance index. Specifically, this includes: finding the theoretical process parameters corresponding to the expected performance index on the actual characteristic curve; obtaining the hardware safety boundary threshold of the current loom, which is determined based on the loom's service life and historical maintenance records; and determining the boundary value of the hardware safety boundary threshold as the actual execution parameter when the theoretical process parameters exceed the range of the hardware safety boundary threshold.

[0015] By adopting the above technical solution, the loom controller verifies the theoretical process parameters obtained by reverse mapping by combining the loom hardware safety boundary threshold, thus preventing the parameters from exceeding hardware limits. This method balances the safe operation of the loom with the expected process execution effect.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, a nonlinear fitting is performed based on multiple sets of gradient process parameters and corresponding measured performance indicators to obtain the actual characteristic curve. Specifically, this includes: calculating the standard rate of change of the global standard characteristic curve at different process parameter points, where the standard rate of change represents the change in weaving performance indicators caused by the adjustment of a unit process parameter under a preset standard loom state; calculating the performance decay ratio of the measured performance indicators relative to the corresponding values ​​in the global standard characteristic curve under each gradient process parameter; correcting the standard rate of change using the performance decay ratio to obtain the actual rate of change of the loom across the entire process range; and fitting the gradient process parameters and corresponding measured performance indicators based on the actual rate of change to obtain the actual characteristic curve.

[0017] By adopting the above technical solution, the loom controller first uses measured data to proportionally correct the standard rate of change of the global standard characteristic curve. Then, based on the corrected rate, it constructs an actual characteristic curve that conforms to physical laws. This method uses prior theoretical data to compensate for the sparseness of field sampling points, enabling the construction of a high-precision global parameter correspondence with only a small number of samples.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, with the expected performance index as the target value, a reverse mapping calculation is performed on the actual characteristic curve to obtain the actual execution parameters that enable the current loom to achieve the expected performance index. Specifically, this includes: finding the theoretical process parameters corresponding to the expected performance index on the actual characteristic curve; determining the current rate of change of the actual characteristic curve at the theoretical process parameters based on the actual rate of change; when the current rate of change is lower than a preset minimum efficiency threshold, determining the critical position on the actual characteristic curve where the actual rate of change equals the preset minimum efficiency threshold, and determining the process parameters corresponding to the critical position as the actual execution parameters; when the current rate of change is greater than or equal to the preset minimum efficiency threshold, determining the theoretical process parameters as the actual execution parameters.

[0019] By adopting the above technical solution, the loom controller first calculates the rate of change of the actual characteristic curve at the current operating point. Then, when this rate falls below the minimum efficiency threshold, it is determined that the actual execution parameters will no longer be increased. This method alleviates the ineffective energy consumption after the loom enters the performance saturation zone, achieving an optimal balance between energy efficiency and equipment wear while ensuring that the process meets standards.

[0020] In a second aspect, this application provides a loom controller, including one or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors calling the computer instructions to cause the loom controller to perform the methods described in the first aspect and any possible implementation thereof.

[0021] Thirdly, this application provides a computer-readable storage medium including computer instructions that, when executed on a loom controller, cause the loom controller to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, this application provides a computer program product, including a computer program / instruction that, when run on a loom controller, causes the loom controller to perform the method described in the first aspect and any possible implementation thereof.

[0023] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By adopting the technical means of receiving the cloud-based global standard feature curve, fitting the individual actual feature curve through gradient calibration, and obtaining the actual execution parameters adapted to the loom's own state by back mapping the expected performance index, the problem of different effects of the same parameters caused by the inability of unified process parameters to adapt to the individual performance differences of looms is effectively alleviated. This achieves the consistency of the actual process execution effect of the entire loom group, while taking into account both the efficiency of centralized management and the adaptability of individual looms.

[0024] 2. By employing the technique of analyzing the trend of historical deviation sequences to identify unidirectional drift of the loom's mechanical state and fitting and correcting the corresponding neighborhood of the actual characteristic curve, the problem of mismatch between the actual characteristic curve and the current state and parameter mapping distortion caused by the deviation of the loom's mechanical state is effectively alleviated. This achieves continuous adaptation between the actual characteristic curve and the real-time state of the loom, ensuring the long-term stability of the process execution effect.

[0025] 3. By adopting the technical means of comparing the actual rate of change of theoretical process parameters with the preset minimum efficiency threshold to determine the final actual execution parameters, the problem of inefficient loom operation and unstable process execution caused by the low operating efficiency corresponding to theoretical process parameters is effectively alleviated. In this way, the loom operating efficiency and process performance indicators are balanced, and the rationality of weaving operation and process execution is improved. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating a loom control method in an embodiment of this application; Figure 2 This is another flowchart illustrating the loom control method in this application embodiment; Figure 3 This is a schematic diagram of the physical device structure of a loom controller in an embodiment of this application. Detailed Implementation

[0027] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0029] The following describes the process of the method provided in this implementation. Please refer to [link / reference]. Figure 1 This is a flowchart illustrating a loom control method in an embodiment of this application.

[0030] 101. Receive the global standard characteristic curve sent by the cloud server. The global standard characteristic curve represents the first correspondence between the process parameters and weaving performance indicators of the loom in the preset standard loom state within the entire process range.

[0031] A cloud server refers to a server cluster deployed on cloud infrastructure, possessing powerful computing and storage capabilities, used for centralized management and scheduling of process parameter settings for multiple looms; a global standard characteristic curve refers to a parameter mapping relationship established based on loom operating data under ideal or standard conditions, which fully describes the functional relationship between process parameters and weaving performance indicators across the entire process parameter range; a preset standard loom state refers to the ideal mechanical state of the loom at the factory or after standard maintenance, in which the performance of each component meets design specifications; loom process parameters refer to adjustable parameters affecting the weaving process, including but not limited to spindle speed, warp tension, weft tension, and air pressure setpoints; weaving performance indicators refer to quantitative indicators reflecting weaving quality and efficiency, including but not limited to fabric density, warp breakage rate, weft stop rate, and production efficiency; the full process range refers to the complete value range of process parameters during normal loom production operation; the first correspondence relationship refers to the functional mapping relationship between process parameters as input variables and weaving performance indicators as output variables under the preset standard loom state.

[0032] Specifically, the loom controller receives global standard characteristic curve data through a communication connection established with the cloud server. This global standard characteristic curve is a parameter-performance mapping relationship constructed by the cloud server based on historical operating data of a large number of looms under standard conditions, using machine learning algorithms or mathematical modeling methods. The data received by the loom controller contains predicted values ​​of weaving performance indicators corresponding to the complete range of process parameter values, forming one or more continuous characteristic curves. These curves reflect the influence of different process parameter settings on the weaving results under ideal mechanical conditions, providing a standard reference benchmark for subsequent parameter adjustments and performance predictions. The loom controller stores the received global standard characteristic curve data in local memory and establishes a corresponding data index structure to support rapid parameter lookup and mapping calculations.

[0033] In some embodiments, the process of receiving the global standard feature curve can be implemented as follows: The loom controller first establishes a secure communication connection with the cloud server, verifies the server's identity and data transmission permissions, then receives the global standard feature curve data packet pushed by the cloud server, performs integrity verification and format parsing on the received data, and finally stores the parsed feature curve data in the loom controller's local database and creates an index. It is understood that other methods can also be used to implement the global standard feature curve receiving process, which are not limited here.

[0034] 102. Control the loom to enter the calibration mode. In the calibration mode, drive the loom to run under multiple gradient process parameters covering the preset working range, and collect the measured performance indicators corresponding to each gradient process parameter.

[0035] Calibration mode refers to the operating mode set by the loom controller for parameter testing and data acquisition. In this mode, the loom automatically adjusts the process parameters and records the operating data according to the predetermined test program. The preset working range refers to the effective range of process parameters determined according to the current loom type and production needs. This range covers the parameter area that may be used during normal production of the loom. Multiple gradient process parameters refer to process parameter test points distributed at certain intervals or gradients within the preset working range. These parameter points uniformly cover the entire working range to obtain sufficient sampling data. The measured performance index refers to the weaving performance data measured by the loom under specific process parameters during actual operation, which is obtained in real time through various sensors and detection equipment.

[0036] Specifically, the loom controller first switches the loom's operating state to calibration mode. In this mode, normal production tasks are paused and an automatic testing program is initiated. To avoid waste fabric or affecting fabric continuity during the calibration process, the calibration mode is preferably triggered during the gap period when the warp beam is changed, the downtime maintenance period after fabric laying, or the stage of trial weaving using waste yarn. The loom controller adjusts the loom's process parameter settings one by one according to a preset parameter gradient sequence. The loom operates for a preset stabilization time under each parameter setting to eliminate the transient effects of parameter switching. During the stable operation of each gradient process parameter, the loom controller collects weaving performance index data in real time through an integrated sensor system, including but not limited to measuring the actual spindle speed through a speed sensor, measuring warp and weft yarn tension through a tension sensor, counting the number of warp breaks and weft stops through a photoelectric sensor, and detecting fabric quality parameters through an image recognition system. The loom controller filters and statistically analyzes the collected raw data, calculates the average performance index value corresponding to each gradient process parameter, and records the parameter-index data pairs to form a calibration dataset. Throughout the calibration process, the loom controller monitors the loom's operating status in real time to ensure the safety of the testing process and the accuracy of data acquisition.

[0037] In some embodiments, the data acquisition process in calibration mode can be implemented in several ways: Optionally, the loom controller first calculates the distribution points of gradient process parameters according to a preset working range, divides the range of process parameters into multiple test points at equal intervals, then sequentially sets the loom to run under the parameters of each test point, and waits for the loom to stabilize before starting to collect measured performance index data. The acquisition process lasts for a preset time to obtain sufficient statistical samples, and finally, the collected data is averaged and outlier removed. Optionally, the loom controller first sets initial sparse test points within a preset working range for coarse acquisition, then analyzes the gradient and nonlinear characteristics of the initial data, increases the sampling density in areas of rapid change and sets more test points, and reduces the sampling density in areas of gradual change, obtaining a higher quality calibration dataset through iterative optimization of the sampling point distribution. It is understood that other methods can also be used to implement loom control and data acquisition in calibration mode, which are not limited here.

[0038] 103. Based on multiple sets of gradient process parameters and corresponding measured performance indicators, nonlinear fitting is performed to obtain the actual characteristic curve. The actual characteristic curve represents the second correspondence between process parameters and weaving performance indicators of the loom under the current mechanical state.

[0039] Nonlinear fitting refers to a numerical calculation method that uses nonlinear mathematical functions to fit discrete data points to curves, and can describe complex nonlinear mapping relationships between variables. The actual characteristic curve refers to the function curve between process parameters and weaving performance indicators obtained by fitting the actual operating data of the current loom, reflecting the true performance response characteristics of the loom under the current mechanical state. The current mechanical state refers to the actual mechanical wear, component aging status, and overall performance level of the loom at the time of data collection. The second correspondence refers to the actual functional mapping relationship between process parameters as independent variables and weaving performance indicators as dependent variables under the current mechanical state of the loom, which is different from the first correspondence under the ideal state.

[0040] Specifically, the loom controller takes multiple sets of gradient process parameters and corresponding measured performance index data collected in calibration mode as input, and selects an appropriate nonlinear function form for curve fitting calculation. The loom controller first preprocesses the collected data, including data cleaning, outlier detection, and smoothing filtering. Then, it selects nonlinear function forms such as polynomial functions, exponential functions, logarithmic functions, or piecewise functions as the fitting basis. During the fitting process, the loom controller uses optimization algorithms such as least squares method, gradient descent method, or genetic algorithm to determine the optimal function parameters through iterative calculation, minimizing the sum of squared errors between the fitted curve and the measured data points. The accuracy indicators used by the loom controller to evaluate the fitting results include the coefficient of determination, root mean square error, and maximum absolute error, ensuring that the actual characteristic curve accurately reflects the mapping relationship between the current loom's process parameters and performance indexes. After fitting, the loom controller stores the mathematical expression and parameters of the actual characteristic curve in a local database, providing an accurate mathematical basis for subsequent parameter mapping calculations.

[0041] In some embodiments, nonlinear fitting can be used to obtain the actual characteristic curve in various ways: Optionally, the loom controller first analyzes the distribution characteristics and trends of gradient process parameters and measured performance index data, selects multiple candidate nonlinear function forms including quadratic polynomials, cubic spline functions, and Gaussian process functions, then performs parameter estimation and fitting calculations for each function form, evaluates the fitting accuracy and generalization ability of each function through cross-validation, and finally selects the function form with the best overall performance as the actual characteristic curve; Optionally, the loom controller adopts a piecewise fitting strategy, dividing the preset working interval into multiple sub-intervals, selecting the most suitable local nonlinear function for fitting within each sub-interval, ensuring the continuity of function values ​​and derivatives at the boundaries of the sub-intervals, and then combining the fitting results of each sub-interval to form a complete piecewise actual characteristic curve. This method can better capture the local characteristics of the process parameter-performance index relationship in different regions. It is understood that other methods can also be used to achieve nonlinear fitting and actual characteristic curve construction based on measured data, which are not limited here.

[0042] 104. In response to the target process parameters uniformly issued by the cloud server to multiple looms, the expected performance indicators corresponding to the target process parameters are determined based on the global standard characteristic curve.

[0043] The unified distribution of cloud server data to multiple looms refers to the cloud server simultaneously sending the same process parameter setting instructions to all relevant looms in the workshop to achieve batch parameter adjustment; the target process parameter refers to the standardized process parameter setting value determined by the cloud server based on the production plan and quality requirements; the expected performance index refers to the theoretical weaving performance value corresponding to the target process parameter under ideal standard conditions.

[0044] Specifically, after receiving the target process parameters uniformly distributed by the cloud server, the loom controller immediately initiates the parameter mapping calculation program. The loom controller uses the received target process parameters as input values ​​and performs lookup and interpolation calculations in the locally stored global standard characteristic curve. When the target process parameter exactly corresponds to a known data point on the global standard characteristic curve, the loom controller extracts the corresponding weaving performance index value as the expected performance index. When the target process parameter lies between known data points, the loom controller uses numerical methods such as linear interpolation, spline interpolation, or Lagrange interpolation to calculate the accurate expected performance index value based on the position and value of adjacent data points. The loom controller also needs to consider the case of multi-dimensional process parameters. In actual weaving, a single performance index is often affected by the coupling of multiple parameters such as spindle speed and weft tension. Therefore, the global standard characteristic curve is represented as a feature hypersurface in multi-dimensional space. When performing parameter mapping, the loom controller is actually solving for parameter solutions in a multi-dimensional vector space. The two-dimensional curve descriptions in the above steps are only simplified examples for ease of understanding and should not be considered as a limitation on dimensions.

[0045] In some embodiments, the mapping from target process parameters to expected performance indicators can be achieved in multiple ways: Optionally, the loom controller first parses the target process parameter data packet sent by the cloud server, extracts the specific values ​​and parameter type identifiers of each process parameter, then establishes a multi-dimensional index in the global standard feature curve database to find the corresponding performance indicator range, calculates the accurate expected performance indicator value using bilinear interpolation or cubic spline interpolation methods, and finally verifies the rationality of the calculation results to ensure that the values ​​are within the normal range; Optionally, the loom controller adopts a segmented search strategy, compares the target process parameter with the parameter range of the global standard feature curve to locate the segment to which it belongs, selects the nearest neighbor data point within the determined segment for local fitting calculation, and obtains a more stable expected performance indicator by combining the performance indicator values ​​of multiple neighboring points through a weighted average method, while recording the confidence level of the interpolation calculation for subsequent parameter adjustment reference.

[0046] 105. Using the expected performance index as the target value, perform reverse mapping calculation on the actual characteristic curve to obtain the actual execution parameters that enable the current loom to achieve the expected performance index.

[0047] Reverse mapping calculation refers to the inverse function calculation process of deriving the corresponding process parameters from the weaving performance index, which is the opposite of forward mapping; the actual execution parameters refer to the process parameter settings obtained through reverse mapping calculation, which enable the current loom to achieve the expected performance index under its actual mechanical state.

[0048] Specifically, the loom controller uses the expected performance index as a known output value and performs inverse calculations on the actual characteristic curve to determine the corresponding process parameter input values. Since the actual characteristic curve describes the functional relationship between process parameters and weaving performance index under the current loom state, the loom controller needs to solve for the inverse function or use numerical methods. When the actual characteristic curve is an explicit function with an analytical inverse function, the loom controller substitutes the expected performance index value to calculate the corresponding process parameters. When the actual characteristic curve is a complex nonlinear function or a piecewise function, the loom controller uses numerical algorithms such as binary search, Newton's iteration method, or the golden section method to find the process parameter solution that makes the weaving performance index equal to the expected value through successive approximation. In the case of multiple parameters, the loom controller uses multidimensional optimization algorithms such as genetic algorithms, particle swarm optimization, or gradient descent to search for the optimal parameter combination in the multidimensional parameter space. The loom controller also needs to handle the possibility of multiple solutions. When an expected performance index corresponds to multiple process parameter solutions, the loom controller selects the most suitable solution as the actual execution parameter based on criteria such as minimum energy consumption, minimum equipment wear, or minimum parameter adjustment range.

[0049] It should be noted that, in the embodiments of this application, "obtaining the actual execution parameters that enable the current loom to achieve the expected performance index" should be interpreted broadly in logic. That is, this step aims to find a parameter solution that can achieve or is closest to the expected performance index. When limited by the physical limits of the loom, safety boundaries, or energy efficiency ratio, the calculated actual execution parameters may not be able to completely reproduce the theoretical expected performance index numerically. However, at the control logic level, these parameters are considered to be the optimal solution achievable under the current state, and the mapping objective of this step is considered to have been completed.

[0050] In some embodiments, when the loom controller finds the theoretical process parameters corresponding to the expected performance index on the actual characteristic curve, it first searches for process parameter points within the domain of the actual characteristic curve that make the function output value equal to the expected performance index through function value comparison or numerical search algorithms. The loom controller uses an interval search method to gradually narrow down the parameter range, and selects the parameter point with the smallest deviation as the theoretical process parameter by calculating the deviation between the performance index value corresponding to each candidate parameter point and the expected value. When the actual characteristic curve has flat sections or oscillations in certain areas, the loom controller uses multi-point sampling and averaging to improve search accuracy and avoid inappropriate parameter selection due to numerical errors. In the process of obtaining the hardware safety boundary threshold of the current loom, the loom controller first extracts the service life information from the loom's equipment file database, including the production date, cumulative operating time, and load history, and then queries the historical maintenance record database to obtain the equipment's maintenance frequency, replacement part types, and failure mode statistics. Based on the equipment aging curve and reliability analysis methods, combined with the safety operation specifications provided by the loom manufacturer, the loom controller calculates the hardware load capacity boundary values ​​at the current point in time, including the safety thresholds of key parameters such as maximum speed limit, maximum tension limit, and maximum air pressure limit. The loom controller also considers the impact of seasonal factors, ambient temperature and humidity, and the current health status of the equipment on the safety boundary, dynamically adjusting the hardware safety boundary threshold to ensure absolute safety of equipment operation. When theoretical process parameters exceed the range of the hardware safety boundary threshold, the loom controller first determines the specific type and magnitude of the exceeded parameter, and then determines the boundary value of the corresponding hardware safety boundary threshold as the actual execution parameter. During the parameter limiting process, the loom controller records the difference between the original theoretical parameter value and the actual execution parameter value, calculates the resulting performance index deviation, and generates corresponding warning information. At this time, although the actual execution parameter may not be able to fully reproduce the expected performance index under the global standard due to the hardware safety boundary, in the definition of this application, the process of determining the boundary value within the hardware allowable range as the actual execution parameter is regarded as the calculation step of "obtaining the expected performance of the current loom as much as possible" under the current constrained conditions. When the loom controller performs the operation of "determining the boundary value as the actual execution parameter", although the actual physical performance corresponding to the parameter may be lower than the expected value under the global standard curve, in the control method of this application, the boundary value has been identified as the effective execution target that the current loom can achieve under safety constraints. This approach avoids the risk of equipment damage caused by pursuing theoretical performance targets and maintains the integrity of the control logic.

[0051] 106. Use actual execution parameters to control the current loom for weaving production.

[0052] Controlling the current loom refers to the loom controller setting operating parameters and managing the production process of the specific loom equipment it currently manages; weaving production refers to the actual textile production process carried out by the loom according to the set process parameters.

[0053] Specifically, the loom controller sends the calculated actual execution parameters one by one to each actuator and control unit of the loom. The loom controller sends the spindle speed parameter to the frequency converter via digital or analog signals to control the spindle motor's operating speed; it sends the tension parameter to the tension controller to adjust the tension of the warp and weft yarns; it sends the air pressure parameter to the pneumatic system to adjust the air pressure output of the air-jet loom; and it sends the temperature parameter to the heating system to control the ambient temperature of the textile environment. The loom controller employs a gradual adjustment strategy during parameter distribution to avoid sudden parameter changes impacting the loom's operation. By setting a reasonable rate of parameter change, it ensures a smooth transition of the loom to the new operating state.

[0054] The loom control method described in this application involves the loom controller first receiving a global standard characteristic curve from a cloud server to establish a first correspondence between process parameters and weaving performance indicators under a preset standard loom state. Then, the loom controller controls the loom to enter calibration mode, collecting gradient process parameters and corresponding measured performance indicators covering a preset working range to obtain the actual operating data of the current loom and identify individual performance differences caused by its mechanical state. Next, the loom controller performs nonlinear fitting based on multiple sets of measured data to obtain an actual characteristic curve, quantifying a second correspondence between process parameters and weaving performance indicators under the current mechanical state, and clarifying the individual performance response characteristics of the loom. Finally, the loom controller responds to the target process parameters uniformly distributed from the cloud, determines the expected performance indicators through the global standard characteristic curve, and obtains the actual execution parameters for production control by reverse mapping from the actual characteristic curve, ensuring that the process parameters are adapted to the current mechanical state of the loom. This method alleviates the problem of inconsistent effects caused by the same parameters due to the difficulty in adapting uniform parameter settings to individual loom performance differences, and improves the consistency of the overall actual process execution effect of the loom group.

[0055] The following provides a more detailed description of the process of the method provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the loom control method in this application.

[0056] 201. Receive the global standard characteristic curve sent by the cloud server. The global standard characteristic curve represents the first correspondence between the process parameters and weaving performance indicators of the loom under the preset standard loom state across the entire process range. (Refer to the execution process of step 101) 202. Control the loom to enter calibration mode. In calibration mode, drive the loom to run under multiple gradient process parameters covering a preset working range, and collect the measured performance indicators corresponding to each gradient process parameter. (Refer to the execution process of step 102) 203. Calculate the standard rate of change of the global standard characteristic curve at different process parameter points. The standard rate of change represents the change in weaving performance index caused by the adjustment of a unit process parameter under the preset standard loom condition.

[0057] Different process parameter points refer to the discrete positions of process parameter values ​​distributed on the global standard characteristic curve; the standard rate of change represents the change range of weaving performance index corresponding to the unit change of the process parameter under the preset standard loom condition, and its mathematical meaning is the derivative or slope of the global standard characteristic curve at a specific process parameter point; the adjustment of a unit process parameter refers to the parameter change when the process parameter value increases or decreases by one standard unit; the change in weaving performance index refers to the corresponding change range of the weaving performance value caused by the adjustment of the process parameter.

[0058] Specifically, the loom controller first performs mathematical analysis on the global standard characteristic curve, using numerical differentiation to calculate the derivative values ​​of the curve at each process parameter point. The loom controller selects an appropriate differential step size and calculates the rate of change between adjacent data points using forward difference, backward difference, or central difference formulas. When the global standard characteristic curve is in the form of an analytical function, the loom controller directly performs differentiation on the function expression to obtain the derivative function, and then substitutes each process parameter point into the derivative function to calculate the corresponding standard rate of change. When the global standard characteristic curve is in the form of discrete data points, the loom controller uses the finite difference method to calculate the slope between each data point and its adjacent points as an approximate value of the standard rate of change for that point. The loom controller also needs to handle the calculation of the rate of change at curve boundary points, using one-sided difference or extrapolation methods to ensure that each process parameter point within the global range has a corresponding standard rate of change value. During the calculation process, the loom controller smooths abnormal rates of change, using moving average or low-pass filtering methods to eliminate numerical noise and ensure the continuity and rationality of the standard rate of change.

[0059] In some embodiments, the standard rate of change can be calculated as follows: The loom controller first divides the global standard characteristic curve into multiple equally spaced calculation intervals according to the range of process parameters. In each interval, representative process parameter points are selected as calculation nodes. Then, the standard rate of change at each node is numerically calculated using the second-order accurate central difference formula. Finally, the change rate values ​​between nodes are filled in using spline interpolation to construct a complete rate of change distribution. Optionally, the loom controller first analyzes the local curvature features of the global standard characteristic curve to identify areas of drastic change. In areas with large curvature, a smaller differential step size is used to improve calculation accuracy, while in areas with small and gentle curvature, a larger differential step size is used to improve calculation efficiency. Then, the calculation results of each area are combined to obtain a standard rate of change distribution with optimized accuracy.

[0060] 204. Calculate the performance degradation ratio of the measured performance index relative to the corresponding value in the global standard characteristic curve under each gradient process parameter.

[0061] The performance degradation ratio is used to indicate the degree of deviation of the measured performance index from the theoretical standard value. It is usually expressed as a percentage, representing the degree of degradation or improvement of the current loom performance relative to the standard state.

[0062] Specifically, the loom controller processes each gradient process parameter and its corresponding measured performance index data collected in calibration mode one by one. First, the loom controller searches for a standard performance index value that perfectly matches the current gradient process parameter in the global standard characteristic curve. When a perfectly matching parameter point cannot be found, the loom controller uses linear interpolation or spline interpolation methods to calculate the theoretical performance index value of the gradient process parameter on the global standard characteristic curve. Then, the loom controller calculates the difference between the measured performance index and the theoretical performance index value. The performance degradation ratio is obtained by dividing the difference by the theoretical performance index value and multiplying by a percentage factor. When the measured performance index is lower than the theoretical value, the performance degradation ratio is positive, indicating a performance decrease; when the measured performance index is higher than the theoretical value, the performance degradation ratio is negative, indicating that the performance exceeds the standard. The loom controller performs a reasonableness check on the calculated performance degradation ratio, eliminating obviously abnormal values ​​and smoothing the degradation ratios of adjacent data points to ensure the continuity and stability of the degradation ratio sequence. The loom controller establishes a mapping relationship between each gradient process parameter point and its corresponding performance degradation ratio, providing accurate degradation reference data for subsequent standard change rate correction.

[0063] In some embodiments, the performance degradation ratio can be calculated in the following way: the loom controller uses a weighted average method to process the measured performance indicators of multiple measurements. First, the mean and standard deviation of the performance indicators of multiple sampling points under each gradient process parameter are calculated. Then, weights are assigned to each sampling point based on data quality to perform a weighted average calculation to obtain a more stable representative value of the measured performance indicators. Finally, the performance degradation ratio is calculated by comparing it with the corresponding theoretical value in the global standard characteristic curve.

[0064] 205. The standard rate of change is corrected by the performance decay ratio to obtain the actual rate of change of the loom across the entire process range.

[0065] The actual rate of change is used to represent the true rate of change of process parameters after taking into account the individual performance differences of the current loom, and reflects the actual strength of the impact of process parameter adjustments on performance indicators under the current mechanical state of the loom.

[0066] Specifically, the loom controller first establishes a correction relationship model between the performance degradation ratio and the standard rate of change, mapping the degradation information to the adjustment coefficient of the rate of change through mathematical calculations. The loom controller adopts a multiplicative correction model: Actual rate of change = Standard rate of change × (1 - Performance degradation ratio). This model assumes that performance degradation affects parameter sensitivity proportionally. When the degradation ratio is positive, the actual rate of change is less than the standard rate of change, reflecting a decrease in the loom's response capability. The above-mentioned correction algorithm based on the degradation ratio, rather than a complex global resampling fitting, is used because the loom controller, as an edge-side device, has limited computing resources. This algorithm utilizes the physical law that mechanical wear usually exhibits linear or quasi-linear response degradation, enabling rapid reconstruction of the global curve with very few sampling points, reducing the computational load at the edge and ensuring real-time control. The loom controller can also adopt an additive correction model: Actual rate of change = Standard rate of change - Degradation correction term, where the degradation correction term is obtained by fitting the difference between measured data and theoretical data. After calculating the actual rate of change for each gradient process parameter point, the loom controller uses interpolation or fitting methods to extend the discrete rate values ​​to a continuous function across the entire process range, ensuring that the corresponding actual rate of change can be obtained at any process parameter value. The loom controller constructs a continuous actual rate of change function using methods such as cubic spline interpolation, polynomial fitting, or radial basis function interpolation, and evaluates the interpolation accuracy through cross-validation, selecting the optimal interpolation method to ensure the accuracy and smoothness of the actual rate of change across the entire process range.

[0067] In some embodiments, the standard rate of change can be corrected in the following way: the loom controller adopts a piecewise linear correction strategy, dividing the entire process range into multiple sub-intervals according to the characteristics of the performance decay ratio. In each sub-interval, a suitable correction model and parameters are selected, and the correction coefficients of each sub-interval are determined by least squares fitting. Then, at the boundaries of the sub-intervals, continuity constraints are used to ensure the global smoothness of the actual rate of change function. This method can better adapt to the non-uniform distribution characteristics of performance decay in different working intervals of the loom.

[0068] 206. Fit the gradient process parameters to the corresponding measured performance indicators based on the actual rate of change to obtain the actual characteristic curve.

[0069] The actual characteristic curve represents a continuous function curve that describes the complete mapping relationship between the process parameters and weaving performance indicators of the loom under the current mechanical state.

[0070] Specifically, the loom controller first uses discrete data points formed by gradient process parameters and corresponding measured performance indicators as basic constraints, and simultaneously introduces the actual rate of change as a derivative constraint into the fitting calculation process. The loom controller employs a spline interpolation method with derivative constraints, determining the coefficients of the spline function by solving a system of linear equations. This ensures that the fitted curve not only passes through all measured data points but also satisfies that the derivative value at each data point equals the corresponding actual rate of change. The loom controller establishes a constrained optimization problem: minimizing the fitting error objective function J=Σ[f(x)] i )-y i ] 2 +λΣ[f'(x i )-s i ] 2 , where x i For gradient process parameters, y i For actual measured performance indicators, s i Let λ represent the actual rate of change, and λ be the derivative constraint weighting coefficient. The loom controller uses the Lagrange multiplier method or sequential quadratic programming algorithm to solve this constrained optimization problem, obtaining the optimal fitting function that conforms to both the measured data and the rate of change constraint. During the fitting process, the loom controller also considers the physical rationality constraints of the function, ensuring that the monotonicity and concavity of the actual characteristic curve conform to the physical laws of the weaving process throughout the entire process range, avoiding unreasonable oscillations or singular behaviors.

[0071] 207. In response to the target process parameters uniformly issued by the cloud server to multiple looms, determine the expected performance indicators corresponding to the target process parameters based on the global standard characteristic curve. (Refer to the execution process of step 104) 208. Using the expected performance index as the target value, perform reverse mapping calculation on the actual characteristic curve to obtain the actual execution parameters that enable the current loom to achieve the expected performance index.

[0072] This step includes: finding the theoretical process parameters corresponding to the expected performance indicators on the actual characteristic curve; determining the current rate of change of the actual characteristic curve at the theoretical process parameters based on the actual rate of change; when the current rate of change is lower than the preset minimum efficiency threshold, determining the critical position on the actual characteristic curve where the actual rate of change equals the preset minimum efficiency threshold, and determining the process parameters corresponding to the critical position as the actual execution parameters; when the current rate of change is greater than or equal to the preset minimum efficiency threshold, determining the theoretical process parameters as the actual execution parameters.

[0073] The theoretical process parameters represent the process parameter solutions on the actual characteristic curve that make the weaving performance index equal to the expected value; the current rate of change represents the first derivative value of the actual characteristic curve at the theoretical process parameters; the preset minimum efficiency threshold represents the lower limit value of the rate of change set by the loom controller, used to determine whether the loom's operating efficiency is within a reasonable range; the critical position represents the process parameter point on the actual characteristic curve where the rate of change equals the minimum efficiency threshold; and the actual execution parameters represent the process parameter settings finally determined by the loom controller for controlling the loom's operation.

[0074] Specifically, the loom controller first searches for process parameter solutions within the domain of the actual characteristic curve that make the function value equal to the expected performance index. The loom controller employs numerical search algorithms such as binary search, Newton-Raphson iteration, or the golden section method to find the theoretical process parameter values ​​that meet the conditions through a step-by-step approximation approach. When the actual characteristic curve is a monotonic function, the loom controller uses a binary search method: setting an initial search interval [a, b], calculating the function value f(c) at the midpoint c = (a + b) / 2; if f(c) > the expected performance index, then setting b = c; otherwise, setting a = c, repeating the iteration until |f(c) - expected performance index| < tolerance. When the actual characteristic curve is a non-monotonic function, the loom controller adopts a global search strategy, dividing the domain into multiple sub-intervals for separate solutions, and prioritizing the solution with the lowest energy consumption or mechanical loss as the theoretical process parameter, which serves as the benchmark for subsequent performance evaluation. After obtaining the theoretical process parameters, the loom controller determines the current rate of change by differentiating the actual characteristic curve at that point or by searching pre-stored actual rate of change data. The loom controller compares the current rate of change with a preset minimum efficiency threshold. If the current rate of change is lower than the threshold, it indicates that the parameter adjustment efficiency of the loom near that operating point is too low, and the marginal benefit of further increasing the process parameters diminishes. The loom controller searches along the actual characteristic curve in the direction of parameter reduction to find the critical position where the actual rate of change equals the preset minimum efficiency threshold, and determines the process parameters corresponding to this position as the actual execution parameters. When the current rate of change is greater than or equal to the preset minimum efficiency threshold, it indicates that the loom has good parameter response efficiency at that operating point, and the loom controller determines the theoretical process parameters as the actual execution parameters.

[0075] In this situation, continuing to increase process parameters to fully achieve the expected performance indicators will lead to an imbalance between energy consumption and revenue (i.e., entering the performance saturation zone). Therefore, the operation of "obtaining the expected performance indicators for the current loom" in this step is specified as: finding the optimal parameter solution corresponding to the expected performance indicators under the premise of satisfying the preset energy efficiency ratio constraint. That is, when the efficiency is lower than the threshold, this critical position is identified as the acceptable actual execution parameters corresponding to the expected performance intention under the current mechanical state, thereby solving the inefficient operation problem that may be caused by simple mathematical inverse calculation.

[0076] In some embodiments, the reverse mapping calculation and parameter determination process can be implemented in multiple ways: Optionally, the loom controller first uses the cubic spline interpolation method to reconstruct the actual characteristic curve to obtain the analytical expression and derivative function, and then uses Newton's iteration method to solve the equation f(x) = expected performance index, with the iterative formula being x n+1 =x n -[f(x) n ) - Expected performance metrics] / f'(x nThe process involves iterating through multiple initial values ​​to ensure all possible solutions are found, and finally, an engineering feasibility evaluation is performed on each solution to select the optimal theoretical process parameters. Alternatively, the loom controller can employ a piecewise linear approximation method to linearize the actual characteristic curve near key points. By using linear interpolation, the approximate values ​​of the theoretical process parameters are quickly estimated. Then, a high-precision numerical method is used for local optimization near the approximate values. This method significantly improves the solution efficiency while ensuring computational accuracy, and is particularly suitable for the rapid parameter adjustment requirements in real-time control scenarios.

[0077] 209. Drive the current loom to operate according to the actual execution parameters; Specifically, the loom controller decomposes the calculated actual execution parameters into specific process control quantities, including spindle speed setpoints, warp tension setpoints, weft tension setpoints, air pressure setpoints, and temperature setpoints. These parameter values ​​are then converted into corresponding control signal formats by a digital signal processing unit. The loom controller sends speed parameters to the frequency converter to control the spindle motor's operating speed via fieldbus, Ethernet communication, or analog signal output; tension parameters to the tension controller to adjust the tension of the warp and weft insertion systems; air pressure parameters to the pneumatic control valve to adjust the airflow pressure of the air-jet loom; and temperature parameters to the heating controller to adjust the temperature conditions of the weaving environment. The loom controller employs a gradual adjustment strategy during parameter distribution, setting reasonable limits on parameter change rates to avoid sudden parameter changes impacting the loom's mechanical system. A maximum change rate is set to ensure that each actuator can smoothly respond to parameter adjustment commands. The loom controller monitors the parameter response status of each actuator in real time, confirming through feedback signals that the actual execution parameters have been correctly set.

[0078] 210. During the current operation of the loom, the current operating performance index is obtained at a preset sampling frequency, and the instantaneous deviation rate between the current operating performance index and the expected performance index is calculated. The preset sampling frequency represents the reciprocal of the data acquisition time interval preset by the loom controller, used to control the time resolution of performance monitoring. The current operating performance index represents the actual weaving performance data collected by the sensor system in real-time operation of the loom. The expected performance index represents the theoretical performance target value obtained by the loom controller from the global standard characteristic curve based on the target process parameters. The instantaneous deviation rate represents the real-time relative error percentage of the current operating performance index relative to the expected performance index, used to quantify the instantaneous difference between the actual operating effect of the loom and the expected target.

[0079] Specifically, after the loom begins operating according to the actual parameters, the loom controller immediately initiates a performance monitoring program, periodically triggering data acquisition tasks based on a preset sampling frequency and time interval. The loom controller acquires key performance parameters in real time during the weaving process through an integrated multi-sensor system. This includes detecting the actual spindle speed using photoelectric sensors, measuring the actual tension values ​​of warp and weft yarns using tension sensors, monitoring the actual air pressure output of the pneumatic system using pressure sensors, counting the number of warp breaks and weft stops per unit time using a yarn break detector, measuring fabric density and quality parameters using a fabric detection system, and monitoring the actual energy consumption level of the loom using a power sensor. The loom controller performs real-time filtering on the acquired raw sensor data, using a moving average filter to eliminate high-frequency noise and a median filter to remove pulse interference, ensuring the accuracy and stability of the current operating performance data. The loom controller calculates the instantaneous deviation rate using a standardized relative error formula: Instantaneous Deviation Rate = (Current Operating Performance Index - Expected Performance Index) / Expected Performance Index × 100%. This formula converts absolute error into a relative error percentage, facilitating the unified evaluation and comparison of performance indicators with different dimensions. The loom controller performs sign judgment and numerical range check on the calculated instantaneous deviation rate. A positive value indicates that the actual performance exceeds expectations, while a negative value indicates that the actual performance is lower than expected.

[0080] 211. When the instantaneous deviation rate exceeds the deviation fluctuation range, generate a compensation value for the actual execution parameters based on the instantaneous deviation rate; The deviation fluctuation range represents the normal fluctuation range of the instantaneous deviation rate preset by the loom controller. This range takes into account random disturbances and measurement errors during the weaving process. The compensation value represents the incremental adjustment of the actual execution parameters to correct the current operating deviation.

[0081] Specifically, the loom controller first compares the calculated instantaneous deviation rate e(t) with a preset deviation fluctuation range, typically set between ±3% and ±5%, based on the precision requirements of the weaving process and the normal fluctuation characteristics of the equipment. When the absolute value of the instantaneous deviation rate e(t) exceeds the upper limit of the fluctuation range, the loom controller initiates a parameter compensation calculation program, using a proportional-integral-derivative control algorithm to generate a compensation value u(t). The loom controller establishes a PID controller mathematical model: u(t) = K p ×e(t)+K i ×∫e(t)dt+K d ×(de(t) / dt), where K p K is the proportional gain coefficient. i K is the integral gain coefficient. dThe three coefficients, representing the differential gain coefficients, are determined through system identification or experimental adjustment based on the loom's dynamic characteristics and control performance requirements. When calculating the proportional term, the loom controller multiplies the instantaneous deviation rate by the proportional gain. The proportional term reflects the instantaneous correction strength of the current deviation; an excessively large proportional gain can lead to system oscillation, while an excessively small gain can result in slow adjustment speed. When calculating the integral term, the loom controller performs numerical integration on the historical deviation rate. The integral term is used to eliminate the system's steady-state error and prevent the accumulation of long-term small deviations into significant performance degradation. When calculating the differential term, the loom controller performs numerical differentiation on the trend of deviation rate changes. The differential term is used to predict the direction of deviation development and apply reverse adjustment in advance, improving the system's dynamic response performance and stability. The loom controller also considers the physical constraints and safety limitations of the compensation value, imposing amplitude limits on the calculated compensation value to ensure that the adjusted parameters after compensation do not exceed the loom hardware's tolerance and the reasonable range of the process.

[0082] 212. The compensation value is superimposed with the actual execution parameters to obtain the correction parameters. The correction parameters are then used to control the current loom to continue weaving production.

[0083] The correction parameter refers to the command value of the loom controller after closed-loop adjustment, which is finally output to the hardware actuator at the bottom of the loom. It includes the basic setting based on the standard curve and the dynamic adjustment based on real-time feedback.

[0084] Specifically, after calculating the compensation value for the current deviation using a PID algorithm or other control algorithms, the loom controller reads the currently effective actual execution parameter (as a reference value) and performs an algebraic addition operation between the calculated compensation value and the actual execution parameter. For example, if the actual execution parameter is "spindle speed 500 rpm" and the calculated compensation value is "+5 rpm", the loom controller obtains a correction parameter of "505 rpm" through superposition. After obtaining the correction parameter, the loom controller first performs a safety check to ensure that it does not exceed the physical limits of the loom hardware (such as maximum air pressure, maximum speed, etc.). After the check passes, the loom controller converts the correction parameter into a corresponding low-level electrical signal (such as a PWM signal, analog voltage signal, or bus communication command) and sends it to the current loom's servo drive, air valve controller, or tension regulator, etc., via the I / O interface. The actuator responds to the correction parameter by adjusting its own amplitude or frequency of movement, thereby changing the current operating state of the loom to counteract external interference or system fluctuations and ensure that the performance indicators in the weaving production process return to the expected target.

[0085] In some embodiments, after controlling the current loom to continue weaving production using correction parameters, the loom controller can identify mechanical state drift by monitoring historical deviation trends online and perform local adaptive correction of the actual characteristic curve based on a weighted regression algorithm. Specifically, while controlling production using correction parameters, the loom controller continuously records the instantaneous deviation rate and corresponding time at each moment, generating a historical deviation sequence in memory. The loom controller sets a preset time monitoring window (e.g., the most recent 10 minutes or the most recent 1000 weaving cycles) and calculates the changing trend characteristics of the historical deviation sequence within this window, such as the rate of movement of the mean deviation or the slope of the deviation. The loom controller determines whether the changing trend characteristics meet a preset unidirectional drift condition, which is used to distinguish between random noise and systematic offset. If the condition is met (e.g., the deviation continues to increase in the positive direction), it is determined that the mechanical state of the current loom (e.g., spring fatigue, nozzle wear) has shifted.

[0086] At this point, the loom controller determines the current target process parameters, identifies the corresponding current operating point on the original actual characteristic curve, and defines a preset neighborhood centered on this point. To correct the curve within this neighborhood to match the current mechanical state, the loom controller first extracts all correction parameters and their resulting current operating performance indicators within a preset time monitoring window, constructing a correction sample set. The loom controller employs weighted processing logic: assigning a first confidence weight to the data in the correction sample set, which is positively correlated with the proximity of the data acquisition time to the current moment, i.e., newer data has a larger weight; simultaneously, assigning a second confidence weight to the original data (raw data) of the actual characteristic curve within the preset neighborhood as a basic constraint to maintain the curve shape. Subsequently, based on the first and second confidence weights, the loom controller performs a weighted regression operation (such as weighted least squares) using the correction sample set and the raw data to solve for the optimal fitting parameters, thereby updating the parameter mapping relationship within the preset neighborhood. Finally, the loom controller uses the updated mapping relationship to replace the corresponding segments in the original curve, generating the corrected actual characteristic curve for subsequent parameter reverse mapping calculations, thereby achieving the self-evolution and calibration of the mapping relationship.

[0087] The loom control method described in this application involves the loom controller first receiving a global standard characteristic curve and collecting measured performance indicators of gradient process parameters. It calculates the standard change rate and performance degradation ratio, corrects the actual change rate, and then fits the actual characteristic curve. Prior theoretical data is used to compensate for the sparseness of on-site sampling points, constructing a high-precision global parameter correspondence. Next, the loom controller responds to the target process parameters from the cloud to determine the expected performance indicators. It then combines the actual change rate with a preset minimum efficiency threshold for inverse mapping to obtain the actual execution parameters, avoiding ineffective energy consumption in the performance saturation zone and balancing energy efficiency and equipment wear. Finally, the loom controller drives the loom to operate according to the actual execution parameters, monitors the operating performance at a preset sampling frequency, and calculates the instantaneous deviation rate. When the deviation exceeds the range, a compensation value is generated and superimposed to obtain the correction parameters, achieving dynamic parameter adjustment and real-time suppression of random disturbances in the production process. This method alleviates the problems of parameter mapping distortion caused by loom mechanical state deviation and operational instability caused by inefficient process parameters, ensuring the continuous stability of the loom's process execution effect.

[0088] The method provided in the above embodiments can be executed by a loom controller. The loom controller in the embodiments of this invention is described below from a hardware processing perspective; please refer to [link to relevant documentation]. Figure 3 This is a schematic diagram of the physical device structure of a loom controller in an embodiment of this application.

[0089] It should be noted that, Figure 3 The structure of the loom controller shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0090] like Figure 3 As shown, the loom controller includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on a program stored in Read-Only Memory (ROM) 302 or a program loaded from storage section 308 into Random Access Memory (RAM) 303, such as performing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0091] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0092] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs the various functions defined in the present invention.

[0093] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0094] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0095] Specifically, the loom controller in this embodiment includes a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the loom control method provided in the above embodiment.

[0096] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the loom controller described in the above embodiments; or it may exist independently and not assembled into the loom controller. The storage medium carries one or more computer programs that, when executed by a processor of the loom controller, cause the loom controller to implement the loom control method provided in the above embodiments.

[0097] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0098] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0099] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0100] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A loom control method characterized by, The application is applied to a loom controller configured on a loom and connected with a cloud server, and the method comprises: receiving a global standard characteristic curve issued by the cloud server, the global standard characteristic curve representing a first corresponding relationship between process parameters and weaving performance indicators of the loom in a global process range under a preset standard loom state; controlling the loom to enter a calibration mode, driving the loom to run under a plurality of gradient process parameters covering a preset working interval in the calibration mode, and collecting measured performance indicators corresponding to each gradient process parameter; based on a plurality of sets of gradient process parameters and corresponding measured performance indicators, performing nonlinear fitting to obtain an actual characteristic curve representing a second corresponding relationship between the process parameters and the weaving performance indicators of the loom under a current mechanical state; in response to target process parameters issued by the cloud server to a plurality of looms, determining an expected performance indicator corresponding to the target process parameters based on the global standard characteristic curve; taking the expected performance indicator as a target value, performing reverse mapping calculation on the actual characteristic curve to obtain actual execution parameters for the current loom to reach the expected performance indicator; controlling the current loom to perform weaving production by using the actual execution parameters.

2. The method of claim 1, wherein, controlling the current loom to perform weaving production by using the actual execution parameters, specifically comprising: driving the current loom to run according to the actual execution parameters; in the process of running the current loom, acquiring a current running performance indicator at a preset sampling frequency, and calculating an instantaneous deviation rate between the current running performance indicator and the expected performance indicator; when the instantaneous deviation rate exceeds a deviation fluctuation range, generating a compensation value for the actual execution parameters based on the instantaneous deviation rate; superimposing the compensation value and the actual execution parameters to obtain a modified parameter, and using the modified parameter to control the current loom to continue weaving production.

3. The method of claim 2, wherein, After the step of controlling the current loom to continue weaving production by using the modified parameter, the method further comprises: based on the instantaneous deviation rate and the corresponding time, generating a historical deviation sequence, and calculating a change trend feature of the historical deviation sequence within a preset time monitoring window; when the change trend feature meets a preset one-way drift condition, determining a current working point on the actual characteristic curve corresponding to the target process parameter, the one-way drift condition indicating that the mechanical state of the current loom has deviated; based on data within the preset time monitoring window, fitting and correcting a preset neighborhood of the actual characteristic curve centered on the current working point.

4. The method of claim 3, wherein, based on data within the preset time monitoring window, fitting and correcting a preset neighborhood of the actual characteristic curve centered on the current working point, specifically comprising: extracting the modified parameters and corresponding current running performance indicators within the preset time monitoring window to obtain a modified sample set; assigning a first confidence weight to the modified sample set and a second confidence weight to original data in the preset neighborhood of the actual characteristic curve; Based on the first confidence weight and the second confidence weight, a weighted regression operation is performed on the original data and the modified sample set to update a parameter mapping relationship in the preset neighborhood; Based on the updated parameter mapping relationship, a corrected actual characteristic curve is generated.

5. The method of claim 1, wherein, With the expected performance indicator as a target value, a reverse mapping calculation is performed on the actual characteristic curve to obtain an actual execution parameter that enables the current loom to achieve the expected performance indicator, specifically including: A theoretical process parameter corresponding to the expected performance indicator is found on the actual characteristic curve; A hardware safety boundary threshold of the current loom is obtained, and the hardware safety boundary threshold is determined based on the service life and historical maintenance records of the loom; When the theoretical process parameter exceeds the range of the hardware safety boundary threshold, a boundary value of the hardware safety boundary threshold is determined as the actual execution parameter.

6. The method of claim 1, wherein, Based on a plurality of sets of gradient process parameters and corresponding measured performance indicators, a nonlinear fitting is performed to obtain an actual characteristic curve, specifically including: A standard change rate of the global standard characteristic curve at different process parameter points is calculated, and the standard change rate represents the change amount of the weaving performance indicator caused by the adjustment of the unit process parameter in the preset standard loom state; A performance attenuation ratio of the measured performance indicator relative to the corresponding value in the global standard characteristic curve is calculated under each gradient process parameter; The actual change rate of the loom in the full process range is obtained by correcting the standard change rate using the performance attenuation ratio; Based on the actual change rate, the gradient process parameters and the corresponding measured performance indicators are fitted to obtain the actual characteristic curve.

7. The method of claim 6, wherein, With the expected performance indicator as a target value, a reverse mapping calculation is performed on the actual characteristic curve to obtain an actual execution parameter that enables the current loom to achieve the expected performance indicator, specifically including: A theoretical process parameter corresponding to the expected performance indicator is found on the actual characteristic curve; Based on the actual change rate, a current change rate of the actual characteristic curve at the theoretical process parameter is determined; When the current change rate is lower than a preset minimum performance threshold, a critical position on the actual characteristic curve where the actual change rate is equal to the preset minimum performance threshold is determined, and a process parameter corresponding to the critical position is determined as the actual execution parameter; When the current change rate is greater than or equal to the preset minimum performance threshold, the theoretical process parameter is determined as the actual execution parameter.

8. A loom controller characterized by, One or more processors and a memory; The memory is coupled to the one or more processors, and the memory is configured to store computer program code including computer instructions, and the one or more processors are configured to invoke the computer instructions to enable the loom controller to perform the method of any one of claims 1-7.

9. A computer-readable storage medium comprising computer instructions, wherein, When the computer instructions are run on the loom controller, the loom controller is enabled to perform the method of any one of claims 1-7.

10. A computer program product comprising computer programs / instructions, characterized in that, When the computer program / instructions are run on the loom controller, the loom controller performs the method as described in any one of claims 1-7.

Images