Camshaft Synchronous Drive Method for Automatic Control of Textile Equipment

Abstract

The present application provides a camshaft synchronous drive method for the automatic control of textile equipment, which relates to the field of automatic control technology and includes: optimizing the camshaft speed based on the single-cycle target motion state of the follower and combining the first-level prediction model of the follower motion state to obtain the first target camshaft speed; when the first target camshaft speed does not belong to the rated camshaft speed range, optimizing the camshaft speed and the cam pressure angle by combining the second-level prediction model of the follower motion state to obtain the second target camshaft speed and the first target cam pressure angle; when the second target camshaft speed belongs to the rated camshaft speed range and the first target cam pressure angle belongs to the rated cam pressure angle range, sending the second target camshaft speed and the first target cam pressure angle to the user terminal. The present application solves the technical problem that the working parameters of the existing camshaft cannot be dynamically adjusted according to the changes in process requirements, and improves the operation stability and textile efficiency of textile equipment.

Description

Technical Field

[0001] This application relates to the field of automation control technology, and particularly to a camshaft synchronous drive method for the automation control of textile equipment. Background Art

[0002] In textile equipment, the camshaft controls processes such as fabric feeding, cutting, and stitching by converting the rotational motion of an electric motor into linear motion. The traditional camshaft drive process usually relies on the electric motor to directly drive the camshaft through a speed reducer, with fixed design and operating parameters. This makes the equipment lack flexibility and adaptability during the processing, and it is difficult to respond in real time to changes in the fabric motion state. At the same time, this design with fixed parameters results in a slow response speed of the equipment under different process requirements. In the case of load changes or inconsistent material properties, it cannot adaptively adjust the rotational speed and pressure angle, causing unstable motion of the follower, affecting the overall equipment performance, and resulting in problems such as unstable processing quality or low efficiency. Summary of the Invention

[0003] This application provides a camshaft synchronous drive method for the automation control of textile equipment, which solves the technical problem that the working parameters of the camshaft in existing textile equipment are fixed and lack the ability of dynamic adjustment, and it is difficult to make timely adjustments according to changes in process requirements. By adaptively adjusting the working parameters of the camshaft to precisely match the motion requirements of the follower in different stages, the technical effect of improving the operating stability and textile efficiency of the textile equipment is achieved.

[0004] In view of the above problems, this application provides a camshaft synchronous drive method for the automation control of textile equipment. The method includes: receiving the single-cycle target motion state of the follower, where the single-cycle target motion state of the follower includes process target speed timing information and return target speed timing information; activating the first-level prediction model of the follower motion state and the second-level prediction model of the follower motion state according to the cam model; optimizing the rotational speed of the camshaft based on the process target speed timing information and the return target speed timing information, in combination with the first-level prediction model of the follower motion state, to obtain the first target camshaft rotational speed; when the first target camshaft rotational speed does not belong to the rated rotational speed range of the camshaft, optimizing the rotational speed of the camshaft and the cam pressure angle based on the process target speed timing information and the return target speed timing information, in combination with the second-level prediction model of the follower motion state, to obtain the second target camshaft rotational speed and the first target cam pressure angle; when the second target camshaft rotational speed belongs to the rated rotational speed range of the camshaft, and the first target cam pressure angle belongs to the rated cam pressure angle range, sending the second target camshaft rotational speed and the first target cam pressure angle to the client.

[0005] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0006] Receive the single-cycle target motion state of the follower, where the single-cycle target motion state of the follower includes process target speed timing information and return target speed timing information. These information provide accurate target parameters for subsequent optimization, making the optimization process more precise. According to the cam model, enable the corresponding first-level prediction model of the follower motion state and the second-level prediction model of the follower motion state to evaluate the motion state of the follower. Based on the process target speed timing information and the return target speed timing information, combined with the first-level prediction model of the follower motion state, optimize the camshaft speed to obtain the first target camshaft speed; when the first target camshaft speed does not belong to the rated range of the camshaft speed, based on the process target speed timing information and the return target speed timing information, combined with the second-level prediction model of the follower motion state, optimize the camshaft speed and the cam pressure angle to obtain the second target camshaft speed and the first target cam pressure angle; when the second target camshaft speed belongs to the rated range of the camshaft speed and the first target cam pressure angle belongs to the rated range of the cam pressure angle, send the second target camshaft speed and the first target cam pressure angle to the client. By optimizing the camshaft speed and pressure angle through a multi-level prediction model and combining the rated range constraint, it is ensured that the most suitable operating parameters can be found under different loads and motion states, and problems such as equipment wear caused by overloading are avoided.

[0007] In summary, according to the motion requirements of the follower in different stages, this application combines the first-level prediction model of the follower motion state and the second-level prediction model of the follower motion state to perform hierarchical control optimization, find the optimal camshaft speed and pressure angle, realize the precise motion control of the follower, ensure that the follower runs according to the predetermined motion trajectory and speed, thereby achieving the technical effects of improving the stability and reliability of the entire textile equipment, ensuring the textile processing accuracy, and further improving the textile processing efficiency and fabric quality.

[0008] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. Description of the Drawings

[0009] Figure 1 It is a schematic flowchart of the camshaft synchronous drive method for textile equipment automation control provided by the embodiment of this application.

[0010] Figure 2Schematic diagram of the process for training the first-level prediction model and the second-level prediction model of the motion state of the follower in the camshaft synchronous drive method for textile equipment automation control provided by the embodiments of the present application.

[0011] Figure 3 Schematic diagram of the process for obtaining the first target camshaft speed in the camshaft synchronous drive method for textile equipment automation control provided by the embodiments of the present application. Detailed implementation manners

[0012] By providing a camshaft synchronous drive method for textile equipment automation control, the embodiments of the present application solve the technical problem that in existing textile equipment, the working parameters of the camshaft are fixed, lacking the ability of dynamic adjustment and being difficult to be adjusted in time according to the changes in process requirements. By adaptively adjusting the working parameters of the camshaft to precisely match the motion requirements of the follower in different stages, the technical effects of improving the operation stability and textile efficiency of the textile equipment are achieved.

[0013] As Figure 1 shown, the embodiments of the present application provide a camshaft synchronous drive method for textile equipment automation control, and the method includes:

[0014] Step S1: Receive the single-cycle target motion state of the follower, where the single-cycle target motion state of the follower includes process target speed time-series information and return target speed time-series information.

[0015] Specifically, the follower refers to a component that moves driven by the camshaft in textile equipment, including heald frames, reed seats, etc. Obtain the expected operating state of each follower's speed changing with time within a complete working cycle from the production process design document, that is, the single-cycle target motion state of the follower. The single-cycle target motion state of the follower includes process target speed time-series information and return target speed time-series information. Among them, the process target speed time-series information refers to the sequence of the speed changing with time when the follower moves forward (i.e., the "forward weaving" stage in the weaving process). The return target speed time-series information refers to the sequence of the speed changing with time when the follower moves backward (i.e., the "return weaving" stage in the weaving process).

[0016] By receiving the single-cycle target motion state of the follower, the motion requirements of the follower can be clarified, so as to determine the specific objectives of optimizing the working parameters of the camshaft and provide data reference for the subsequent optimization process.

[0017] Step S2: Activate the first-level prediction model and the second-level prediction model of the motion state of the follower according to the cam model.

[0018] Specifically, since camshafts of different models have different physical and motion characteristics, it is necessary to select and activate an appropriate prediction model according to the specific model of the cam. The prediction models include a primary prediction model for the motion state of the follower and a secondary prediction model for the motion state of the follower. These prediction models are pre-programmed or trained and can predict the motion performance of the follower corresponding to the camshaft under different working conditions based on the working parameters of the cam. Among them, the primary prediction model for the motion state of the follower is used to predict the motion state of the follower at different camshaft speeds. The primary prediction model for the motion state of the follower is a more refined prediction model for predicting the motion state of the follower at different camshaft speeds and different cam pressure angles.

[0019] Step S3: Based on the process target speed time series information and the return target speed time series information, combined with the primary prediction model for the motion state of the follower, optimize the camshaft speed to obtain the first target camshaft speed.

[0020] Specifically, taking the process target speed time series information and the return target speed time series information received in Step S1 as the optimization objectives, adjust the camshaft speed through an optimization algorithm. After each adjustment, use the primary prediction model for the motion state of the follower to predict the motion characteristics of the follower (process speed time series information and return speed time series information) at the current speed, and compare it with the optimization objectives (process target speed time series information and return target speed time series information). Based on the comparison results, continue to adjust and optimize the camshaft speed. Through continuous adjustment and optimization, find the ideal speed of the camshaft that matches the process target speed time series information and the return target speed time series information, that is, the first target camshaft speed.

[0021] Since the cam pressure angle and profile are design parameters and usually do not change, when performing optimization, first use the primary prediction model for the motion state of the follower to optimize the camshaft speed and find the camshaft speed that meets the motion requirements of the follower to achieve optimization at the lowest cost.

[0022] Step S4: When the first target camshaft speed does not belong to the rated range of the camshaft speed, based on the process target speed time series information and the return target speed time series information, combined with the secondary prediction model for the motion state of the follower, optimize the camshaft speed and the cam pressure angle to obtain the second target camshaft speed and the first target cam pressure angle.

[0023] Specifically, the rated speed range of the camshaft refers to the acceptable speed range of the camshaft determined by the design limitations of the equipment or technological requirements. If the speed exceeds this range, the equipment may not operate stably. For the first target camshaft speed obtained in step S3, check whether it is within the rated speed range of the camshaft. If the first target camshaft speed exceeds the rated speed range of the camshaft, it indicates that the preliminary optimization result does not meet the safety requirements of the equipment's mechanical structure and further optimization is needed.

[0024] At this time, using the same method as the optimization process in step S3, with the process target speed time series information and the return stroke target speed time series information as the optimization targets, the camshaft speed and the cam pressure angle are adjusted simultaneously through an algorithm, and the motion state of the follower is predicted using the secondary prediction model of the follower motion state. Through the joint optimization of the secondary prediction model, the second target camshaft speed and the first target cam pressure angle that match the process target speed time series information and the return stroke target speed time series information are obtained.

[0025] On the premise of ensuring the safe operation of the textile equipment, when optimizing only the camshaft speed cannot meet the motion requirements of the follower, by combining a more complex secondary prediction model, the speed and pressure angle of the camshaft are jointly optimized to further improve the motion accuracy and efficiency of the follower, making the motion of the follower more stable.

[0026] Step S5: When the second target camshaft speed belongs to the rated speed range of the camshaft and the first target cam pressure angle belongs to the rated pressure angle range of the cam, send the second target camshaft speed and the first target cam pressure angle to the user terminal.

[0027] Specifically, the rated pressure angle range of the cam is the pressure angle range allowed for the safe operation of the equipment. If the pressure angle is too large or too small, it will cause a decrease in system efficiency or an increase in mechanical wear, affecting the equipment life and product quality.

[0028] For the second target camshaft speed and the first target cam pressure angle obtained by optimization in step S4, respectively check whether they are within their respective safe and effective ranges. If the second target camshaft speed belongs to the rated speed range of the camshaft and the first target cam pressure angle belongs to the rated pressure angle range of the cam, this indicates that the joint optimization result can not only meet the motion requirements of the follower but also does not exceed the bearing limit of the mechanical structure of the textile equipment. At this time, send the joint optimization result, that is, the second target camshaft speed and the first target cam pressure angle, to the user's operation terminal or control system, so that the automatic control system adjusts the camshaft speed according to the second target camshaft speed and manually adjusts the cam pressure angle to the first target cam pressure angle to meet the motion requirements of the follower, thereby ensuring the stable operation of the textile equipment and improving the fabric quality.

[0029] Further, the method described in the embodiments of the present application further includes:

[0030] When the first target camshaft speed belongs to the rated camshaft speed range, send the first target camshaft speed to the user terminal.

[0031] Specifically, in step S3, if the first target camshaft speed is already within the rated camshaft speed range, it indicates that the first target camshaft speed at this time meets the safety requirements of the equipment mechanical structure. There is no need to perform subsequent optimization steps, and this first target camshaft speed can be directly sent to the user terminal for adjustment, so that the follower moves precisely according to the predetermined speed curve, improving production efficiency and fabric quality.

[0032] Further, the method described in the embodiments of the present application further includes:

[0033] When the second target camshaft speed does not belong to the rated camshaft speed range, and / or the first target cam pressure angle does not belong to the rated cam pressure angle range, based on the process target speed timing information and the return target speed timing information, and in combination with the secondary prediction model of the follower motion state, update the camshaft speed and the cam pressure angle to obtain the third target camshaft speed and the second target cam pressure angle; when the third target camshaft speed belongs to the rated camshaft speed range and the second target cam pressure angle belongs to the rated cam pressure angle range, send the third target camshaft speed and the second target cam pressure angle to the user terminal; otherwise, continue to update the third target camshaft speed and the second target cam pressure angle; when the preset number of times of directional optimization is reached and a convergent solution is still not generated, generate a camshaft adjustment instruction and send it to the user terminal.

[0034] Specifically, when any one of the second target camshaft speed and the first target cam pressure angle obtained in step S4 does not meet the corresponding rated range or both do not meet the corresponding rated range, it is necessary to update the camshaft speed and the cam pressure angle according to the process target speed timing information and the return target speed timing information of the follower, and in combination with the secondary prediction model of the follower motion state. According to the similar optimization process described above, further adjust the camshaft speed and the cam pressure angle to generate a new optimization result, that is, the third target camshaft speed and the second target cam pressure angle.

[0035] Determine the rated intervals for the third target camshaft speed and the second target cam pressure angle. When the third target camshaft speed belongs to the rated interval of the camshaft speed and the second target cam pressure angle belongs to the rated interval of the cam pressure angle, send the third target camshaft speed and the second target cam pressure angle to the client. When either the third target camshaft speed or the second target cam pressure angle does not meet the corresponding rated interval or both do not meet, repeat the above update optimization and determination process to find the optimization result where the camshaft speed belongs to the rated interval of the camshaft speed and the cam pressure angle belongs to the rated interval of the cam pressure angle, and send this optimization result to the client.

[0036] If the update continues until the number of optimization times reaches the preset number of directed optimizations and still no camshaft speed and cam pressure angle that meet the conditions are found, generate a camshaft adjustment instruction to indicate that the staff should manually intervene for manual adjustment to break through the current optimization bottleneck.

[0037] Further, as Figure 2 shown, activate the first-level prediction model for the follower motion state and the second-level prediction model for the follower motion state according to the cam model, including:

[0038] Collect the dataset for constructing the first-level prediction model of the follower motion state according to the cam model and the initial pressure angle. Among them, the dataset for constructing the first-level prediction model of the follower motion state includes the first camshaft speed time series record dataset, the first process identification speed time series dataset, and the first return identification speed time series dataset; use the first camshaft speed time series record dataset as the input, and use the first process identification speed time series dataset and the first return identification speed time series dataset as the supervision to train the first-level prediction model of the follower motion state; collect the dataset for constructing the second-level prediction model of the follower motion state according to the cam model. Among them, the dataset for constructing the second-level prediction model of the follower motion state includes the second camshaft speed time series record dataset, the pressure angle record dataset, the second process identification speed time series dataset, and the second return identification speed time series dataset; use the second camshaft speed time series record dataset and the pressure angle record dataset as the input, and use the second process identification speed time series dataset and the second return identification speed time series dataset as the supervision to train the second-level prediction model of the follower motion state.

[0039] Specifically, according to the aforementioned cam model and initial pressure angle, the operation data of the cam synchronous drive structure of the textile equipment under the same model and cam pressure angle are collected, including the rotational speed of the camshaft at different time points, as well as the speed change data of the follower during the forward stroke and the speed change data during the return stroke. These data are preprocessed and sorted and summarized in chronological order to generate the first camshaft rotational speed time series record data set, the first forward stroke identification speed time series data set, and the first return stroke identification speed time series data set. These data sets together constitute the data set for constructing the first-level prediction model of the follower motion state, which is used to train the first-level prediction model of the follower motion state.

[0040] Using a supervised learning algorithm, an initial model is constructed, such as a long short-term memory neural network. The first camshaft rotational speed time series record data set is used as the training data set, and the first forward stroke identification speed time series data set and the first return stroke identification speed time series data set are used as the supervised data sets. The initial model is supervised and trained. During the training process, the model learns how to predict the corresponding follower forward stroke speed and return stroke speed based on the camshaft rotational speed. By comparing the prediction results with the supervised data sets, the model continuously adjusts its internal parameters until convergence, so that the prediction results on the training data set approach the supervised data sets, and a trained first-level prediction model of the follower motion state is obtained. This model can predict the follower forward stroke speed and return stroke speed under a specific cam model and cam pressure angle based on the input camshaft rotational speed.

[0041] According to the camshaft model, the operation data of the cam synchronous drive structure of the textile equipment under this model are collected, including the rotational speed of the camshaft at different time points, different cam pressure angle data, as well as the speed change data of the follower during the forward stroke and the speed change data during the return stroke. The second camshaft rotational speed time series record data set, the pressure angle record data set, the second forward stroke identification speed time series data set, and the second return stroke identification speed time series data set are sorted and generated. Similar model training steps are adopted. Using the second camshaft rotational speed time series record data set and the pressure angle record data set as the input data sets, and the second forward stroke identification speed time series data set and the second return stroke identification speed time series data set as the supervised data sets, the second-level prediction model of the follower motion state is trained. This second-level prediction model of the follower motion state can predict the follower forward stroke speed and return stroke speed under the corresponding cam model based on the input camshaft rotational speed data and cam pressure angle data.

[0042] Through the above steps, a customized prediction model can be trained for each cam model to improve the accuracy and efficiency of follower motion state prediction, and provide an accurate and reliable judgment tool for subsequent working parameter optimization.

[0043] Further, using the first camshaft speed timing record data set as the input and the first process identification speed timing data set and the first return identification speed timing data set as the supervision, training the first-level prediction model of the follower motion state further includes:

[0044] Construct a first loss function for motion state prediction. Specifically, the first loss function for motion state prediction is used to count the proportion of the first speed deviation time series data that is greater than or equal to the speed deviation threshold in the first process prediction speed time series data and the first process identification speed time series data for any single training. Construct a second loss function for motion state prediction. Specifically, the second loss function for motion state prediction is used to count the proportion of the second speed deviation time series data that is greater than or equal to the speed deviation threshold in the first return prediction speed time series data and the first return identification speed time series data for any single training. Using the first camshaft speed timing record data set as the input and the first process identification speed timing data set and the first return identification speed timing data set as the supervision to train the long short-term memory neural network. When in continuous N trainings, at least more than f(0.9*N) times: both the first loss function for motion state prediction and the second loss function for motion state prediction are less than or equal to the first loss threshold, generate the first-level prediction model of the follower motion state, where f(0.9*N) represents rounding up 0.9*N.

[0045] Specifically, in the process of training the first-level prediction model of the follower motion state, the prediction results of the model are evaluated by constructing a loss function to further optimize the performance of the first-level prediction model of the follower motion state.

[0046] First, construct a first loss function for motion state prediction to measure the accuracy of the first process speed predicted by the model after each training. This function first calculates the difference between all the first process prediction speed time series data output by the model after each training and the corresponding actual first process identification speed time series data to obtain M first speed deviation time series data. Then, compare the M first speed deviation time series data with a preset speed deviation threshold, and count the proportion of the data volume in the M first speed deviation time series data that is greater than or equal to the speed deviation threshold, and use the calculation result as the first loss value. Among them, the speed deviation threshold is the upper limit of the speed difference determined according to the actual model prediction accuracy requirement. When the first speed deviation time series data is greater than or equal to this speed deviation threshold, it is considered that the deviation is too large and the prediction is inaccurate.

[0047] Use a similar method to construct a second loss function for motion state prediction. This second loss function measures the accuracy of the first return speed predicted by the model after each training by counting the proportion of the second speed deviation time series data that is greater than or equal to the speed deviation threshold in the first return prediction speed time series data and the first return identification speed time series data for any single training.

[0048] During the training process of the first-level prediction model for the follower motion state, the long short-term memory neural network (LSTM) is selected as the model architecture. Using the first camshaft speed time series record data set as the input and the first process identification speed time series data set and the first return identification speed time series data set as the supervision, the long short-term memory neural network is trained. After each training, the first and second loss functions defined above are used to evaluate the accuracy of the model prediction results. When in consecutive N trainings, in more than f(0.9*N) trainings, both the first and second loss functions are less than or equal to the first loss threshold, the model is considered to have converged, and the final first-level prediction model for the follower motion state is generated. Among them, f(0.9*N) represents rounding up 0.9*N, and the first loss threshold is the upper limit of the loss function value preset according to the actual model prediction accuracy requirements.

[0049] Through the custom loss function, the model can not only learn the law of the follower speed change but also ensure the stability and accuracy of the prediction. During the training process of the second-level prediction model for the follower motion state, a similar function construction logic is adopted to construct the corresponding loss function to evaluate the accuracy of the model training results.

[0050] Further, as Figure 3 shown, step S3 of the embodiment of the present application further includes:

[0051] Obtain a randomly assigned camshaft speed, process the randomly assigned camshaft speed through the first-level prediction model for the follower motion state to generate first process speed time series information and first return speed time series information; based on the first loss function for motion state prediction, calculate the first loss value between the first process speed time series information and the process target speed time series information; based on the second loss function for motion state prediction, calculate the second loss value between the first return speed time series information and the return target speed time series information; when both the first loss value and the second loss value are less than or equal to the second loss threshold, add the randomly assigned camshaft speed to the first target camshaft speed.

[0052] Specifically, the specific process of obtaining the first target camshaft speed by optimizing the camshaft speed based on the process target speed time series information and the return target speed time series information in combination with the first-level prediction model for the follower motion state includes:

[0053] First, a series of randomly generated camshaft speed values are generated. These randomly generated camshaft speed values are sequentially input into the previously trained first-level prediction model for the follower motion state to predict the corresponding first process speed time series information and first return speed time series information of the follower.

[0054] Use the motion state to predict the first loss function, calculate the deviation ratio between the predicted first process speed time series information and the process target speed time series information, which is denoted as the first loss value. Use the motion state to predict the second loss function, calculate the deviation ratio between the predicted first return speed time series information and the return target speed time series information, which is denoted as the second loss value.

[0055] Compare the calculated first loss value and second loss value with the set second loss threshold. When both the first loss value and the second loss value are less than or equal to the second loss threshold, it indicates that the motion state under the current camshaft speed assignment is close to or meets the target speed requirement. Among them, the second loss threshold is the upper limit of the difference between the predicted follower operating state corresponding to each randomly assigned camshaft speed and the follower target motion state. Add the randomly assigned camshaft speeds that meet the conditions to the first target camshaft speed set to provide parameter selection for subsequent optimization or actual application.

[0056] The above steps combine simulation optimization and prediction models. Through repeated iteration and evaluation, it gradually approaches the optimal or satisfactory camshaft speed, improving the operating efficiency and stability of the camshaft synchronous drive structure.

[0057] Further, when both the first loss value and the second loss value are less than or equal to the second loss threshold, adding the randomly assigned camshaft speed to the first target camshaft speed includes:

[0058] When the number of the first target camshaft speeds meets the preset number, output the first target camshaft speeds.

[0059] Specifically, continuously execute the process of randomly assigning camshaft speeds, prediction, calculating loss values, judgment, and collection. Add the randomly assigned camshaft speeds with the first loss value and the second loss value both less than or equal to the second loss threshold to the first target camshaft speeds until the number in the first target camshaft speeds meets the preset number, and then output the complete optimized speed set, that is, the first target camshaft speeds. The preset number is set according to actual application requirements and optimization goals.

[0060] Further, when the directional optimization is performed for a preset number of times and no convergent solution is still generated, generate a camshaft adjustment instruction and send it to the user terminal. After that, it further includes:

[0061] Receive the user feedback data of the user terminal for the camshaft adjustment instruction, and judge whether the user feedback data has the follower single-cycle updated target motion state; if so, perform the camshaft drive control process according to the follower single-cycle updated target motion state.

[0062] Specifically, since the automatic optimization cannot find a satisfactory solution, a camshaft adjustment instruction is generated at this time and sent to the user terminal. After receiving the camshaft adjustment instruction, the user terminal will make a manual adjustment according to the actual situation and send the adjusted result back as user feedback data. Analyze the received user feedback data to determine whether it contains the single-cycle updated target motion state of the follower, that is, the update of the expected motion state of the follower by the user according to the actual situation. If it contains, update the optimization target according to the single-cycle updated target motion state of the follower contained in the user feedback, and repeat the foregoing steps S1 to S5 for camshaft drive control.

[0063] By introducing the user feedback loop, it is ensured that in the case of not finding an automatic optimization solution, the accurate control of the follower can still be achieved through the participation of the user. It can effectively integrate automatic optimization and manual adjustment, and improve the adaptability and problem-solving ability for complex problems.

[0064] In summary, the camshaft synchronous drive method for textile equipment automation control provided by the embodiments of the present application has the following technical effects:

[0065] Receive the single-cycle target motion state of the follower, where the single-cycle target motion state of the follower includes process target speed timing information and return target speed timing information. These information provide accurate target parameters for subsequent optimization, making the optimization process more precise. According to the cam model, enable the corresponding first-level prediction model of the follower motion state and the second-level prediction model of the follower motion state to evaluate the motion state of the follower. Based on the process target speed timing information and the return target speed timing information, combine the first-level prediction model of the follower motion state to optimize the camshaft speed and obtain the first target camshaft speed; when the first target camshaft speed does not belong to the rated range of the camshaft speed, based on the process target speed timing information and the return target speed timing information, combine the second-level prediction model of the follower motion state to optimize the camshaft speed and the cam pressure angle, and obtain the second target camshaft speed and the first target cam pressure angle; when the second target camshaft speed belongs to the rated range of the camshaft speed and the first target cam pressure angle belongs to the rated range of the cam pressure angle, send the second target camshaft speed and the first target cam pressure angle to the user terminal. By optimizing the camshaft speed and pressure angle through a multi-level prediction model and combining the rated range constraint, it is ensured that the most suitable operating parameters can be found under different loads and motion states, and problems such as equipment wear caused by overloading are avoided.

[0066] Overall, in the embodiments of the present application, according to the motion requirements of the follower in different stages, hierarchical control optimization is carried out by combining the primary prediction model of the follower motion state and the secondary prediction model of the follower motion state to find the optimal camshaft rotation speed and pressure angle, realizing the precise motion control of the follower, ensuring that the follower runs according to the predetermined motion trajectory and speed, thereby achieving the technical effects of improving the stability and reliability of the entire textile equipment, ensuring the textile processing accuracy, and further improving the textile processing efficiency and fabric quality. The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A camshaft synchronous driving method for automatic control of textile equipment, characterized in that: include: Receiving a single-cycle target motion state of a driven member, wherein the single-cycle target motion state of the driven member includes a forward target speed timing information and a return target speed timing information; According to the cam model, the primary prediction model of the follower motion state and the secondary prediction model of the follower motion state are activated, where: According to the cam model and the initial pressure angle, a data set for constructing a primary prediction model of the motion state of the follower is collected, wherein the data set for constructing the primary prediction model of the motion state of the follower includes a first camshaft speed time series record data set, a first process mark speed time series data set, and a first return mark speed time series data set; Taking the first camshaft speed time series record data set as input, and taking the first process mark speed time series data set and the first return mark speed time series data set as supervision, training the first level prediction model of the follower motion state; According to the cam model, a data set for constructing a secondary prediction model of the motion state of the follower is collected, wherein the data set for constructing the secondary prediction model of the motion state of the follower includes a second camshaft speed time series record data set, a pressure angle record data set, a second process mark speed time series data set, and a second return mark speed time series data set; Taking the second camshaft speed time series record data set and the pressure angle record data set as input, and taking the second process mark speed time series data set and the second return mark speed time series data set as supervision, training the secondary prediction model of the motion state of the follower; Based on the forward target speed timing information and the return target speed timing information, the camshaft speed is optimized in combination with the first-level prediction model of the follower motion state to obtain a first target camshaft speed; When the first target camshaft speed does not belong to the rated camshaft speed range, optimizing the camshaft speed and the cam pressure angle based on the forward target speed timing information and the return target speed timing information and in combination with the secondary prediction model of the follower motion state, to obtain a second target camshaft speed and a first target cam pressure angle; When the second target camshaft speed belongs to the camshaft speed rated interval, and the first target cam pressure angle belongs to the cam pressure angle rated interval, the second target camshaft speed and the first target cam pressure angle are sent to the user end.

2. The method according to claim 1, characterized in that Also includes: When the first target camshaft speed belongs to the rated camshaft speed range, the first target camshaft speed is sent to the user end.

3. The method according to claim 1, characterized in that Also includes: When the second target camshaft speed does not belong to the rated camshaft speed interval, or / and the first target cam pressure angle does not belong to the rated cam pressure angle interval, based on the forward target speed timing information and the return target speed timing information, the camshaft speed and the cam pressure angle are updated in combination with the secondary prediction model of the follower motion state to obtain a third target camshaft speed and a second target cam pressure angle; When the third target camshaft speed belongs to the camshaft speed rated interval, and the second target cam pressure angle belongs to the cam pressure angle rated interval, sending the third target camshaft speed and the second target cam pressure angle to the user end; otherwise, continue to update the third target camshaft speed and the second target cam pressure angle; When a converged solution is not generated after the directional optimization is performed for a preset number of times, a camshaft adjustment instruction is generated and sent to the user end.

4. The method according to claim 3, characterized in that Taking the first camshaft speed time series record data set as input, taking the first process mark speed time series data set and the first return mark speed time series data set as supervision, training the first level prediction model of the follower motion state includes: Constructing a first loss function for motion state prediction, wherein the first loss function for motion state prediction is used to count the proportion of first speed deviation time series data of first process prediction speed time series data and first process identification speed time series data that is greater than or equal to a speed deviation threshold value in any training; Constructing a second loss function for motion state prediction, wherein the second loss function for motion state prediction is used to count the proportion of the second speed deviation time series data of the first return prediction speed time series data and the first return identification speed time series data that is greater than or equal to the speed deviation threshold value in any training; The first camshaft speed timing record data set is used as input, and the first process identification speed timing data set and the first return identification speed timing data set are used as supervision to train the long short-term memory neural network. When, in N consecutive trainings, at least more than f (0.9*N) times: the first loss function for motion state prediction and the second loss function for motion state prediction are both less than or equal to the first loss threshold, the first-level prediction model of the motion state of the follower is generated, wherein f (0.9*N) represents rounding up to 0.9*N.

5. The method according to claim 4, characterized in that Based on the forward target speed timing information and the return target speed timing information, the camshaft speed is optimized in combination with the primary prediction model of the follower motion state to obtain a first target camshaft speed, including: Obtaining a random value of a camshaft speed, processing the random value of the camshaft speed through the primary prediction model of the follower motion state, and generating first process speed timing information and first return speed timing information; Calculate a first loss value of the first process speed timing information and the process target speed timing information based on the motion state prediction first loss function; Predicting a second loss function based on the motion state, calculating a second loss value of the first return speed timing information and the return target speed timing information; When the first loss value and the second loss value are both less than or equal to a second loss threshold, the camshaft speed is randomly assigned and added to the first target camshaft speed.

6. The method according to claim 5, characterized in that When the first loss value and the second loss value are both less than or equal to a second loss threshold, randomly assigning a value to the camshaft speed to the first target camshaft speed includes: When the number of the first target camshaft speeds meets a preset number, the first target camshaft speed is output.

7. The method according to claim 3, characterized in that When the directional optimization fails to generate a converged solution after a preset number of times, a camshaft adjustment instruction is generated and sent to the user end, and then the following steps are further included: Receiving user feedback data of the camshaft adjustment instruction from the user end, and determining whether the user feedback data has a single-cycle update target motion state of the follower; If available, the camshaft drive control process is performed according to the single-cycle update target motion state of the driven member.

Images