A monitoring method for a guide rail cutting device for automated material transport

Abstract

This invention provides a monitoring method for a guide rail cutting device for automated material transport, relating to the field of data processing technology. The method includes: acquiring real-time monitored material cutting data; when the material cutting data does not meet preset cutting requirements, acquiring a list of current parameter values ​​of a PID controller; optimizing the current parameter values ​​using a particle swarm optimization algorithm based on the material cutting data within a preset time period ending at the current time point, thereby acquiring a list of intermediate parameter values; acquiring the difference between each intermediate parameter value and its corresponding preset parameter threshold, and acquiring the average error value of all differences; if the average error value does not meet preset error conditions, determining an updated parameter value list based on the intermediate parameter values ​​and their corresponding preset parameter thresholds; and using the updated parameter value list to update the cutting signal issued by the PID controller, thereby reducing material waste caused by inaccurate cutting.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a monitoring method for a guide rail cutting device for automatic material transport. Background Technology

[0002] In automated production systems, material handling and precise cutting are core components ensuring production continuity and product quality. As a key execution unit, the guide rail cutting device's control precision directly impacts material size consistency and equipment operating efficiency. In cutting continuous materials, sensors detect the cutting position, prompting the control unit to send a cutting signal to the guide rail cutting device. The device receives the signal and performs the cutting operation, for example, by using an infrared sensor to emit a laser or an image sensor to detect the continuous material.

[0003] Existing control units typically use PID controllers to generate cutoff signals. PID control is a control system based on proportional, integral, and derivative parameters, achieving dynamic response by adjusting the proportional gain parameter Kp, integral gain parameter Ki, and derivative parameter Kd. However, in actual operating conditions, due to dynamic changes in material properties and equipment status, PID controllers with fixed parameters struggle to maintain optimal performance over long periods. In recent years, intelligent optimization algorithms (such as genetic algorithms) have been introduced into the field of PID parameter correction. Genetic algorithms search for the globally optimal parameter combination by simulating natural selection mechanisms, but they suffer from high computational complexity and low efficiency in optimizing PID parameters, leading to material waste. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a monitoring method for an automated material transport guide rail cutting device. The guide rail cutting device receives a cutting signal from a PID controller to cut the material. The method includes the following steps:

[0005] Obtain real-time monitoring data on material cutoffs;

[0006] When the material cutoff data does not meet the preset cutoff requirements, obtain the current parameter value list. Based on the material cutoff data within a preset time period with the current time point as the endpoint, use the particle swarm optimization algorithm to optimize the current parameter values, thereby obtaining the intermediate parameter value list. The intermediate parameter value list includes the intermediate parameter values ​​of m preset parameters of the PID controller, and the current parameter value list includes the current parameter values ​​of m preset parameters of the PID controller.

[0007] Obtain the difference between each intermediate parameter value and its corresponding preset parameter threshold, and obtain the average error value of all differences;

[0008] If the average error value meets the preset error condition, the cutoff signal issued by the PID controller is updated using the intermediate parameter value list;

[0009] If the average error value does not meet the preset error condition, an update parameter value list is determined based on the intermediate parameter value and its corresponding preset parameter threshold. The truncation signal issued by the PID controller is updated using the update parameter value list. The update parameter value list includes the update parameter values ​​of m preset parameters of the PID controller.

[0010] The threshold of the preset parameter is determined through the following steps:

[0011] Based on the material cutoff data from the preset initial time point to the specified time point, a genetic algorithm is used to obtain a list of target parameter values, and the target parameter values ​​are used as the thresholds of their corresponding preset parameters. The list of target parameters includes the target parameter values ​​of m preset parameters of the PID controller.

[0012] After a specified time point, for each additional preset time period, a genetic algorithm is used to optimize the target parameter value based on the material cutoff data from the preset initial time point to the end of the preset time period.

[0013] The present invention has at least the following beneficial effects: In summary, by acquiring real-time monitored material cutoff data, when the material cutoff data does not meet the preset cutoff requirements, a list of current parameter values ​​of the PID controller is obtained. Based on the material cutoff data within a preset time period ending at the current time point, a particle swarm optimization algorithm is used to optimize each current parameter value, thereby obtaining a list of intermediate parameter values. The difference between each intermediate parameter value and its corresponding preset parameter threshold is obtained, and the average error value of all differences is obtained. If the average error value does not meet the preset error condition, an updated parameter value list is determined based on the intermediate parameter value and its corresponding preset parameter threshold. The cutoff signal issued by the PID controller is updated using the updated parameter value list. The present invention uses material cutoff data within a preset time period ending at the current time point and a particle swarm optimization algorithm to quickly correct errors, improve the optimization rate, and avoid the technical problems of low optimization efficiency and failure to optimize to the latest time caused by using genetic algorithms. The calculation result of the genetic algorithm is used as the threshold, making the output signal of the updated PID controller more accurate and reducing material waste caused by inaccurate cutoff. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart illustrating a monitoring method for an automated material transport guide rail cutting device provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar tasks and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0018] This invention provides a monitoring method for a guide rail cutting device for automatic material transport, such as... Figure 1 As shown, the guide rail cutting device receives a cutting signal from a PID controller to cut off the material. The method includes the following steps:

[0019] S100, acquire real-time monitored material cutting data. Specifically, the material cutting data includes at least: the actual length of the cut material and the preset target length of the cut material. Specifically, the timestamp of the actual length of the cut material and the timestamp of the preset target length of the cut material correspond to each other. The actual length of the cut material is the length actually cut by the guide rail cutting device; the preset target length of the cut material is a pre-set target length.

[0020] S200, when the material cutoff data does not meet the preset cutoff requirements, obtain the current parameter value list, and use the particle swarm optimization algorithm to optimize the current parameter values ​​based on the material cutoff data within a preset time period with the current time point as the endpoint, thereby obtaining the intermediate parameter value list; the intermediate parameter value list includes the intermediate parameter values ​​of m preset parameters of the PID controller, and the current parameter value list includes the current parameter values ​​of m preset parameters of the PID controller.

[0021] Specifically, the material truncation data does not meet the preset truncation requirement if the absolute value of the difference between the actual length of the truncated material and the preset target length of the truncated material is not less than the preset truncation threshold. Further, the material truncation data does not meet the preset truncation requirement if the absolute value of the difference between the actual length of the truncated material and the preset target length of the truncated material is greater than the preset truncation threshold.

[0022] In another embodiment of the present invention: the material truncation data does not meet the preset truncation requirement when the ratio of the second value to the preset target length of the truncated material is greater than the preset specified truncation threshold, and the second value is the absolute value of the difference between the actual length of the truncated material and the preset target length of the truncated material.

[0023] Specifically, the current parameter value list A = {A1, A2, ..., Ai, ..., Am}, where Ai is the current parameter value of the i-th preset parameter, and the value of i ranges from 1 to m. The particle swarm optimization algorithm is used to optimize Ai to obtain the i-th intermediate parameter value Bi, thereby obtaining the intermediate parameter value list B = {B1, B2, ..., Bi, ..., Bm}.

[0024] Specifically, m = 3. Furthermore, the three preset parameters of the PID controller are: the proportional gain parameter corresponding to the proportional unit of the PID controller, the integral gain parameter corresponding to the integral unit of the PID controller, and the derivative gain parameter corresponding to the derivative unit of the PID controller.

[0025] S300, obtain the difference between each intermediate parameter value and its corresponding preset parameter threshold, and obtain the average error value of all difference values.

[0026] Specifically, the difference between the intermediate parameter value and its corresponding preset parameter threshold is the ratio of the first value to the intermediate parameter value, where the first value is the absolute value of the difference between the intermediate parameter value and its corresponding preset parameter threshold.

[0027] S400: If the average error value meets the preset error condition, update the cutoff signal issued by the PID controller using the intermediate parameter value list.

[0028] Specifically, the average error value meets a preset error condition: the average error value is less than a preset error threshold. This means that intermediate parameter values ​​are input into the PID controller to update the output signal. This invention uses the average difference between the intermediate parameter value and the corresponding preset parameter threshold for judgment, ensuring that the particle swarm optimization algorithm plays a positive role.

[0029] S500: If the average error value does not meet the preset error condition, an updated parameter value list is determined based on the intermediate parameter values ​​and their corresponding preset parameter thresholds. This updated parameter value list is used to update the cutoff signal issued by the PID controller. The updated parameter value list includes the updated parameter values ​​of m preset parameters of the PID controller. It can be understood that when the average error value does not meet the requirements, it indicates that the particle swarm optimization algorithm is not performing well. Therefore, in this case, an updated parameter value list is determined based on the intermediate parameter values ​​and their corresponding preset parameter thresholds. It can also be understood that the monitoring of material cutoff data is continuously performed in real time. When the preset cutoff requirement is not met, the particle swarm optimization algorithm is used to optimize the preset parameter values ​​of the PID controller based on the material cutoff data from the most recent preset time period.

[0030] Specifically, determining the list of updated parameter values ​​based on intermediate parameter values ​​and their corresponding preset parameter thresholds includes: using the weighted sum of the intermediate parameter values ​​and their corresponding preset parameter thresholds as the updated parameter values ​​to obtain the list of updated parameter values, wherein the sum of the first weight and the second weight is 1, the first weight is the weight of the intermediate parameter value, and the second weight is the weight of the preset parameter threshold corresponding to the intermediate parameter value.

[0031] In one embodiment of the present invention, if the difference between the average error value and the preset error threshold is greater than the preset maximum difference threshold, the first weight is set to 0 and the second weight is set to 1.

[0032] Furthermore, the first weight equals the second weight, which is equal to 1 / 2.

[0033] In summary, this invention acquires real-time material cutoff data. When the material cutoff data does not meet the preset cutoff requirements, it obtains a list of current parameter values ​​for the PID controller. Based on the material cutoff data within a preset time period ending at the current time point, it uses a particle swarm optimization algorithm to optimize the current parameter values, thereby obtaining a list of intermediate parameter values. It then obtains the difference between each intermediate parameter value and its corresponding preset parameter threshold, and calculates the average error value of all differences. If the average error value does not meet the preset error condition, it determines an updated parameter value list based on the intermediate parameter values ​​and their corresponding preset parameter thresholds. The updated parameter value list is then used to update the cutoff signal issued by the PID controller. This invention uses material cutoff data within a preset time period ending at the current time point and a particle swarm optimization algorithm to quickly correct errors, improving the optimization rate and avoiding the low optimization efficiency and failure to optimize to the latest time issues caused by using genetic algorithms. This reduces material waste caused by inaccurate cutoff.

[0034] The threshold of the preset parameter is determined through the following steps:

[0035] S001, based on the material cutoff data from the preset initial time point to the specified time point, a genetic algorithm is used to obtain a list of target parameter values, and the target parameter values ​​are used as the threshold values ​​of their corresponding preset parameters. The list of target parameters includes the target parameter values ​​of m preset parameters of the PID controller.

[0036] S002, after a specified time point, for each additional preset time period, a genetic algorithm is used to optimize the target parameter value based on the material cutoff data from the preset initial time point to the end of the preset time period.

[0037] In summary, a genetic algorithm is used to obtain a list of target parameter values ​​based on material cutoff data from a preset initial time point to a specified time point. These target parameter values ​​are then used as thresholds for their corresponding preset parameters. For each new preset time period after the specified time point, the genetic algorithm optimizes the target parameter values ​​based on the material cutoff data from the preset initial time point to the end of the preset time period. Since the genetic algorithm is a global optimization algorithm that uses the entire material cutoff data to obtain target parameter values, its computational complexity is high, and it may fail to calculate the latest time point. Therefore, the calculation results of the genetic algorithm are used as thresholds for reference, making the updated PID controller output signal more accurate and avoiding the problem of getting trapped in local optima when using only particle swarm optimization, thereby improving the accuracy of material cutoff.

[0038] Specifically, based on the material cutoff data within a preset time period ending at the current time point, the particle swarm optimization algorithm is used to optimize each current parameter value, thereby obtaining a list of intermediate parameter values, including:

[0039] S210, initialize the particle swarm, which includes a number of particles. The position vector of each particle includes m preset parameter values, and the velocity vector of each particle includes the velocity of each preset parameter value. The preset parameter values ​​are the values ​​corresponding to the preset parameters.

[0040] S220, Based on the particle's position vector, determine the predicted truncation data, which includes at least the predicted truncation length. In one embodiment of the present invention, a machine learning model is used to determine the predicted truncation data based on the particle's position vector.

[0041] S230, if the predicted truncation data meets the preset loss requirements, the m preset parameter values ​​are used as intermediate parameter values; otherwise, proceed to the next step; the first loss function is the mean square error of the predicted truncation length and the preset target length in the material truncation data.

[0042] Specifically, if the predicted truncation data meets the preset loss requirement, the difference between the predicted truncation length and the preset target length in the target truncation data is less than the preset prediction threshold.

[0043] S240, based on the material cutoff data within a preset time period with the current time point as the endpoint, update the particle position vector and particle velocity vector to obtain the updated particle position vector and particle velocity vector. The updated particle position vector includes: the updated m preset parameter values, and the updated particle velocity vector includes: the velocity of each updated preset parameter value.

[0044] S250, based on the updated position vector of the particle, determine the updated prediction truncation data, which includes at least the updated truncation length.

[0045] S260, if the updated prediction truncation data meets the preset loss requirements, the updated m preset parameter values ​​are used as intermediate parameter values; otherwise, execute S240.

[0046] Specifically, the updated predicted truncation data meets the preset loss requirement as follows: the difference between the updated truncation length and the preset target length in the target truncation data is less than the preset prediction threshold, or the number of updates is greater than the preset number threshold.

[0047] In summary, the particle swarm optimization algorithm is initialized, and the predicted truncation data is determined based on the particle position vectors. If the predicted truncation data meets the preset loss requirements, the corresponding m preset parameter values ​​are used as intermediate parameter values. Otherwise, the particle position vectors and particle velocity vectors are updated based on the material truncation data within a preset time period ending at the current time point, and the updated particle position vectors and updated particle velocity vectors are obtained. Based on the updated particle position vectors, the updated predicted truncation data is determined. If the updated predicted truncation data meets the preset loss requirements, the corresponding m updated preset parameter values ​​are used as intermediate parameter values; otherwise, the iteration is performed. This invention uses the particle swarm optimization algorithm to quickly obtain a list of intermediate parameter values ​​based on the material truncation data within a preset time period ending at the current time point.

[0048] Specifically, a list of target parameter values ​​is obtained using a genetic algorithm based on material cutoff data from a preset initial time point to a specified time point, including:

[0049] S010, Initialize the initial parameter value list set. The initial parameter value list set includes several initial parameter value lists, which include the initial parameter values ​​of m preset parameters of the PID controller.

[0050] S020: Based on the material cutoff data from the preset initial time point to the specified time point, obtain the fitness value of each initial parameter value list, and sort the initial parameter value list in ascending order according to its corresponding fitness value.

[0051] Specifically, the predicted genetic truncation length is determined based on an initial list of parameter values, and the fitness value is determined based on the error between the predicted genetic truncation length and the target truncation length. Furthermore, a machine learning model is used to obtain the predicted genetic truncation length based on the initial list of parameter values.

[0052] S030, Based on the sorted list of initial parameter values, select k initial parameter value lists. For example, select the first k initial parameter value lists from the sorted list of initial parameter values.

[0053] S040, perform simulated binary cross-interaction on the selected list of k initial parameter values ​​to obtain a cross-parameter value list, wherein the cross-parameter value list includes m cross-parameter values. Specifically, those skilled in the art will understand that any prior art method for simulating binary cross-interaction on an initial parameter value list falls within the scope of this invention, and will not be elaborated further here.

[0054] S050, randomly perturb the cross parameters in the list of k cross parameter values ​​to obtain a list of k new parameter values; and replace the k initial parameter value lists selected in the initial parameter value list set with the list of k new parameter values ​​to obtain a list of candidate parameter values, wherein the list of candidate parameter values ​​includes several candidate parameter value lists, and the candidate parameter value lists are either the initial parameter value lists or the new parameter value lists.

[0055] S060: Based on the material cutoff data from the preset initial time point to the specified time point, obtain the candidate fitness value of each candidate parameter value list set in the candidate parameter value list set. If the candidate fitness value meets the preset genetic iteration requirements, take the candidate parameter value list with the smallest candidate fitness value as the target parameter value list; otherwise, take the candidate parameter value list set as the initial parameter value list set and execute S030.

[0056] Specifically, a machine learning model is used to determine the predicted genetic truncation length corresponding to the candidate parameter value list, and the candidate fitness value of the candidate parameter value list set is determined based on the error between the predicted genetic truncation length and the target truncation length.

[0057] In summary, the initial parameter value list is initialized. Based on the material truncation data from a preset initial time point to a specified time point, the fitness value of each initial parameter value list is obtained, and the initial parameter value lists are sorted in ascending order according to their corresponding fitness values. Based on the sorted initial parameter value lists, k initial parameter value lists are selected. Simulated binary crossover is performed on the selected k initial parameter value lists to obtain a crossover parameter value list. The crossover parameters in the k crossover parameter value lists are randomly perturbed to obtain k new parameter value lists. The k new parameter value lists replace the selected k initial parameter value lists in the initial parameter value list set to obtain a candidate parameter value list set. Based on the material truncation data from the preset initial time point to the specified time point, the candidate fitness value of each candidate parameter value list set in the candidate parameter value list set is obtained. If the candidate fitness value meets the preset genetic iteration requirements, the candidate parameter value list with the smallest candidate fitness value is selected as the target parameter value list; otherwise, iteration is performed. This invention obtains the target parameter value list through global data using a genetic algorithm, making the target parameter value list more stable.

[0058] Furthermore, after a specified time point, for each additional preset time period, the material truncation data from the preset initial time point to the end of the preset time period is processed through steps S010 to S060 to re-obtain the target parameter value in order to optimize the target parameter value.

[0059] While specific embodiments of the invention have been described in detail by way of examples, those skilled in the art should understand that the examples are for illustrative purposes only and are not intended to limit the scope of the invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the invention.

Claims

1. A monitoring method of a rail cutting apparatus for automatically transporting a material, characterized by, The guide rail cutting device receives a cutting signal from a PID controller to cut off the material. The method includes the following steps: Obtain real-time monitoring data on material cutoffs; When the material cutoff data does not meet the preset cutoff requirements, obtain the current parameter value list. Based on the material cutoff data within a preset time period with the current time point as the endpoint, use the particle swarm optimization algorithm to optimize the current parameter values, thereby obtaining the intermediate parameter value list. The intermediate parameter value list includes the intermediate parameter values ​​of m preset parameters of the PID controller, and the current parameter value list includes the current parameter values ​​of m preset parameters of the PID controller. Based on material cutoff data within a preset time period ending at the current time point, the particle swarm optimization algorithm is used to optimize each current parameter value, thereby obtaining a list of intermediate parameter values, specifically including: S210, initialize the particle swarm, the particle swarm includes a number of particles, the position vector of the particles includes m preset parameter values, the velocity vector of the particles includes the velocity of each preset parameter value, the preset parameter value is the value corresponding to the preset parameter; S220, Based on the particle's position vector, determine the predicted truncation data, which includes at least the predicted truncation length; S230, if the predicted truncation data meets the preset loss requirement, the m preset parameter values ​​are used as intermediate parameter values; otherwise, proceed to the next step; the first loss function is the mean square error of the predicted truncation length and the preset target length in the material truncation data; where, if the predicted truncation data meets the preset loss requirement, the difference between the predicted truncation length and the preset target length in the target truncation data is less than the preset prediction threshold. S240, based on the material cutoff data within a preset time period with the current time point as the endpoint, update the position vector and velocity vector of the particles, and obtain the updated position vector and the updated velocity vector of the particles. The updated position vector of the particles includes: the updated m preset parameter values, and the updated velocity vector of the particles includes: the velocity of each preset parameter value after the update. S250, based on the updated position vector of the particle, determine the updated prediction truncation data, wherein the updated prediction truncation data includes at least the updated truncation length; S260, if the updated prediction truncation data meets the preset loss requirements, the updated m preset parameter values ​​are used as intermediate parameter values; otherwise, S240 is executed. Obtain the difference between each intermediate parameter value and its corresponding preset parameter threshold, and obtain the average error value of all differences; If the average error value meets the preset error condition, the cutoff signal issued by the PID controller is updated using the intermediate parameter value list; If the average error value does not meet the preset error condition, an updated parameter value list is determined based on the intermediate parameter values ​​and their corresponding preset parameter thresholds. This updated parameter value list is then used to update the cutoff signal issued by the PID controller. The updated parameter value list includes the updated parameter values ​​of m preset parameters of the PID controller. The thresholds of the preset parameters are determined through the following steps: Based on the material cutoff data from the preset initial time point to the specified time point, a genetic algorithm is used to obtain a list of target parameter values, and the target parameter values ​​are used as the thresholds of their corresponding preset parameters. The list of target parameters includes the target parameter values ​​of m preset parameters of the PID controller. After a specified time point, for each additional preset time period, a genetic algorithm is used to optimize the target parameter value based on the material cutoff data from the preset initial time point to the end of the preset time period.

2. The monitoring method of the automatic transport material guide rail cutting apparatus according to claim 1, characterized by, m is 3.

3. The monitoring method for the guide rail cutting device for automatic material transport according to claim 2, characterized in that, The three preset parameters of the PID controller are: the proportional gain parameter corresponding to the proportional unit of the PID controller, the integral gain parameter corresponding to the integral unit of the PID controller, and the derivative gain parameter corresponding to the derivative unit of the PID controller.

4. The monitoring method for the guide rail cutting device for automatic material transport according to claim 1, characterized in that, The average error value meets the preset error condition as follows: the average error value is less than the preset error threshold.

5. The monitoring method for the guide rail cutting device for automatic material transport according to claim 1, characterized in that, Determining the list of updated parameter values ​​based on intermediate parameter values ​​and their corresponding preset parameter thresholds specifically includes: taking the weighted sum of the intermediate parameter values ​​and their corresponding preset parameter thresholds as the updated parameter values, thereby obtaining the list of updated parameter values, wherein the sum of the first weight and the second weight is 1, the first weight is the weight of the intermediate parameter value, and the second weight is the weight of the preset parameter threshold corresponding to the intermediate parameter value.

6. The monitoring method for the guide rail cutting device for automatic material transport according to claim 5, characterized in that, The first weight is equal to the second weight.

7. The monitoring method for the guide rail cutting device for automatic material transport according to claim 1, characterized in that, The difference between the intermediate parameter value and the threshold of its corresponding preset parameter is the ratio of the first value to the intermediate parameter value, where the first value is the absolute value of the difference between the intermediate parameter value and the threshold of its corresponding preset parameter.

8. The monitoring method for the guide rail cutting device for automatic material transport according to claim 1, characterized in that, The material cutting data includes at least: the actual length of the cut material and the preset target length of the cut material.

9. The monitoring method for the guide rail cutting device for automatic material transport according to claim 8, characterized in that, The material truncation data does not meet the preset truncation requirements if the absolute value of the difference between the actual length of the truncated material and the preset target length of the truncated material is not less than the preset truncation threshold.

10. The monitoring method for the guide rail cutting device for automatic material transport according to claim 9, characterized in that, The material truncation data does not meet the preset truncation requirements when the absolute value of the difference between the actual length of the truncated material and the preset target length of the truncated material is greater than the preset truncation threshold.

Images