A method and system for controlling the temperature of a package for an LED taping machine

Abstract

The application relates to the technical field of temperature control, in particular to a packaging temperature control method and system for an LED strapping machine, which comprises the following steps: collecting temperature data and current data of the LED strapping machine at multiple historical time nodes, obtaining an optimal solution of the temperature data of the LED strapping machine at a current time node by using a random hill climbing algorithm, and adjusting the temperature at the current time node based on the optimal solution; wherein the initial solution of the random hill climbing algorithm is as follows: obtaining the nearest neighbor temperature data of the temperature data at the multiple historical time nodes, calculating noise factors of the temperature data at the historical time nodes, removing the temperature data with noise factors greater than a set threshold as noise, and taking the mean value of the nearest neighbor temperature data after removing the noise as the initial solution. The application solves the problem that the accuracy is not high when the packaging temperature is controlled.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of temperature control. More particularly, the present application relates to a packaging temperature control method and system for an LED taping machine. BACKGROUND

[0002] The LED taping machine plays a key role in the production process of LED packaging, and the accurate control of its packaging temperature directly affects the welding quality, light efficiency stability and service life of the LED chip. The traditional packaging temperature control method usually relies on fixed temperature control parameters or simple regression prediction based on historical data, but in the case of temperature fluctuations, equipment aging and process changes in the production environment, the traditional method is difficult to dynamically adjust the temperature, resulting in unstable packaging quality and even causing welding defects.

[0003] At present, some intelligent temperature control methods introduce machine learning or optimization algorithms for temperature prediction and control to improve the stability of packaging. However, the existing methods may still have problems of insufficient prediction accuracy, slow response speed or excessive dependence on local optimal solutions under complex working conditions.

[0004] Therefore, in view of the optimization demand of the packaging temperature control of the LED taping machine, a packaging temperature control method and system based on an improved random hill climbing algorithm are proposed, which predicts the optimal output value before packaging temperature control through the collection and analysis of historical temperature data, and dynamically adjusts the temperature control strategy to improve the packaging stability and ensure the production quality. In the random hill climbing algorithm, the selection of the initial solution has an important influence on the convergence speed and final result of the algorithm. The initial solution is usually generated randomly to ensure the diversity and exploration ability of the algorithm. If the initial solution is near the optimal solution, the algorithm may converge quickly, but it may also fall into a local optimum, affecting the global search ability. Therefore, the reasonableness of the initial solution not only affects the convergence efficiency of the algorithm, but also determines the quality of the final solution.

[0005] Although some existing intelligent temperature control methods can achieve intelligent temperature control of the packaging temperature to some extent, when collecting temperature data, the LED taping machine usually has high-frequency working current, which may cause electromagnetic interference, and thus some noise data may be mixed into the temperature data. The existence of noise data greatly reduces the prediction accuracy of the initial solution of the random hill climbing algorithm for each temperature control, and thus the accuracy of the packaging temperature control is not high. SUMMARY

[0006] To solve the problem of low accuracy in packaging temperature control in the background art, the present application provides solutions in the following aspects.

[0007] In a first aspect, the present application provides a packaging temperature control method for an LED taping machine, comprising: collecting temperature data and current data of the LED taping machine at a plurality of historical time nodes, obtaining an optimal solution of the temperature data of the LED taping machine at a current time node by using a random hill climbing algorithm, and adjusting the temperature at the current time node based on the optimal solution; wherein the initial solution of the random hill climbing algorithm is: obtaining the most recent neighbor temperature data of the temperature data at the plurality of historical time nodes, calculating a noise factor of the temperature data at the first historical time node , , wherein, is the isolation degree of the temperature data at the first historical time node, is the descending rank of the temperature data at the first historical time node among all the historical time nodes, is the descending rank of the current data corresponding to the temperature data at the first historical time node among all the historical time nodes, is a preset hyperparameter, and is a normalization function, the isolation degree representing the abnormality degree of the temperature data; the most recent neighbor temperature data with a noise factor greater than a set threshold is regarded as noise and is removed, and the mean value of the noise-removed most recent neighbor temperature data is taken as the initial solution.

[0008] The above technical solution can effectively improve the accuracy of packaging temperature control by collecting the temperature data and current data of the LED taping machine at a plurality of historical time nodes, and obtaining an optimal solution of the temperature data of the LED taping machine at a current time node by using a random hill climbing algorithm.

[0009] Further, the isolation degree , , , is the distance between the packaging temperature at the first historical time node and the upper and lower limits of a box plot formed by the temperature data at the historical plurality of time nodes, is the variance of the temperature data at the historical plurality of time nodes, is a preset hyperparameter, and is a normalization function.

[0010] The aforementioned technical solution effectively identifies anomalous temperature data by calculating the deviation of historical temperature data from the overall data distribution. Using the distance between the upper and lower bounds of a box plot to measure the degree of data outlier makes the calculation of isolation level more accurate. Simultaneously, normalization processing combined with the overall data fluctuation improves the comparability of temperature data at different time points. By setting hyperparameters to avoid numerical instability in the calculation, the stability and robustness of the isolation level calculation are ensured, thereby improving the overall reliability in temperature data optimization and anomaly detection.

[0011] Furthermore, it also includes modifying the initial solution, specifically as follows:

[0012] A value is randomly selected from the set error range of the initial solution as the new solution. The new solution is compared with the objective function value of the initial solution, and the solution corresponding to the largest objective function value is taken as the corrected solution. The objective function value is calculated as follows: In the formula, For the first The objective function value of each solution. For the first The energy consumption of working for one minute for the temperature data corresponding to each solution. For the first The pass rate of the corresponding packaging temperature during operation for each solution. For the natural constant An exponential function with base 0.

[0013] The above technical solution refines the packaging temperature control by correcting the initial solution and selecting the solution with the largest objective function value. By randomly selecting new solutions within a set error range and evaluating the objective function, local optima are effectively avoided, improving the global applicability of the optimization results. The objective function comprehensively considers two key factors: energy consumption and yield rate. It reduces the impact of energy consumption on the objective function through exponential decay, ensuring that the optimized temperature setting reduces energy consumption while maintaining packaging quality. This solution can improve production efficiency, reduce energy consumption, and enhance the adaptability of packaging temperature control while meeting process requirements, making it more stable and efficient in actual production.

[0014] Furthermore, temperature sensors are used to collect temperature data of the LED taping machine at multiple historical time points, and current sensors are used to collect current data of the LED taping machine at multiple historical time points.

[0015] Furthermore, adjusting the temperature at the current time point specifically involves:

[0016] The optimal solution is taken as a target temperature, and a difference between a temperature at a current time node and the target temperature is detected in real time, and if the temperature at the current time node deviates from the target temperature, a heating or cooling device is automatically adjusted to adjust the temperature at the current time node to the target temperature.

[0017] Further, the plurality of historical time node temperature data are obtained by using a nearest neighbor algorithm. The plurality of historical time node temperature data are obtained by using a nearest neighbor algorithm. The plurality of historical time node temperature data are obtained by using a nearest neighbor algorithm.

[0018] Further, the packaging temperature data are further subjected to data cleaning, and the data cleaning comprises detecting and processing abnormal values in the packaging temperature data, and the abnormal values are removed by setting a temperature change threshold.

[0019] The above technical solution can effectively improve the accuracy and stability of temperature control by performing data cleaning on the packaging temperature data. By setting a temperature change threshold to detect and remove abnormal values, the influence of extreme temperature data on the calculation result can be avoided, and the influence of noise on the packaging temperature adjustment can be reduced, thereby improving the reliability of temperature prediction. The introduction of data cleaning enables subsequent temperature optimization and control to be based on more real and stable data for calculation, thereby improving the temperature balance of the packaging process, reducing the influence of temperature fluctuations on product quality, and improving production consistency and yield.

[0020] Further, the normalization function adopts maximum-minimum normalization.

[0021] The above technical solution adopts maximum-minimum normalization to process data, so that the packaging temperature data, noise factors, and target function values are kept within a unified numerical range, improving the stability and comparability of calculation. Through normalization, the influence of different data scales on the optimization process can be reduced, ensuring that each parameter has the same weight during calculation, improving the convergence speed and calculation efficiency of the algorithm. In addition, maximum-minimum normalization can maintain the distribution characteristics of the original data, making the optimization result more consistent with the actual packaging temperature control requirements, thereby improving the packaging quality and production stability.

[0022] Further, the threshold is set to 0.5.

[0023] In a second aspect, the present application provides a packaging temperature control system for an LED strapping machine, comprising a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the packaging temperature control method for the LED strapping machine is realized.

[0024] The present application has the following advantages:

[0025] The application obtains the optimal solution of the temperature data of the LED ribbon machine at the current time node by using the random hill climbing algorithm, and adjusts the temperature at the current time node based on the optimal solution, effectively improving the control accuracy of the packaging temperature of the LED ribbon machine. By removing noise from historical packaging temperature data and calculating the isolation degree of temperature, abnormal data can be identified and removed, thereby avoiding the interference of noise on temperature control. The implementation of the application can effectively improve the stability and production efficiency of the LED ribbon machine under different operating conditions, ensuring the accuracy of the packaging temperature, reducing energy consumption and improving production quality. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flow chart schematically showing a packaging temperature control method for an LED ribbon machine according to an embodiment of the application;

[0027] Figure 2 is a structural block diagram schematically showing a packaging temperature control system for an LED ribbon machine according to an embodiment of the application. DETAILED DESCRIPTION

[0028] A packaging temperature control method for an LED ribbon machine embodiment.

[0029] As Figure 1 shown, the flow chart of the packaging temperature control method for the LED ribbon machine according to the embodiment of the application includes the following steps:

[0030] S1: Collecting temperature data and current data of the LED ribbon machine at multiple historical time nodes.

[0031] In one embodiment, in order to accurately obtain the temperature change during the packaging process of the LED ribbon machine, a high-precision temperature sensor and a current sensor can be used to continuously monitor the temperature and current, ensuring the real-time and accuracy of data acquisition. The collection frequency of the temperature sensor and the current sensor is set to once per second, so that every subtle temperature fluctuation and current fluctuation during the packaging process can be recorded. This high-frequency data acquisition method not only provides more detailed historical temperature data and current data, but also enhances the perception ability of sudden temperature abnormalities and current abnormalities, which helps to identify abnormal conditions that may affect the packaging quality in advance, and provides accurate data support in the subsequent packaging temperature control process.

[0032] In addition, high-precision temperature sensors and current sensors are used to collect temperature and current, which can reduce the interference of environmental noise on temperature measurement results and improve the reliability of measurement data. For example, during the operation of the LED ribbon machine, it may be affected by external environmental temperature, air flow, mechanical vibration and other factors, resulting in fluctuations in the packaging temperature. Therefore, in the data collection, the self-calibration function of the sensor can be combined to ensure that high-precision temperature data and current data can still be obtained under complex working conditions. By collecting and storing temperature data and current data at multiple historical time nodes, a time series database of temperature and current can be constructed to provide reliable data support for subsequent packaging temperature optimization.

[0033] S2: obtaining the optimal solution of the temperature data of the LED ribbon machine at the current time node by using the random hill climbing algorithm.

[0034] In one embodiment, the process of obtaining the optimal solution of the temperature data of the LED ribbon machine at the current time node based on the initial solution of the random hill climbing algorithm is a known technology, and this scheme will not be described in detail. The initial solution of the random hill climbing algorithm is to obtain the nearest neighbor temperature data of the temperature data of the multiple historical time nodes, wherein the nearest neighbor temperature data of the temperature data of the multiple historical time nodes is obtained by using a nearest neighbor algorithm, the nearest neighbor algorithm is a known technology, and this scheme will not be described in detail, and the noise factor of the temperature data of the first historical time node is calculated , , wherein is the isolation degree of the temperature data of the first historical time node, is the descending rank of the temperature data of the first historical time node among all historical time nodes, is the descending rank of the current data corresponding to the temperature data of the first historical time node among all historical time nodes, is a preset hyperparameter, is a normalization function; the nearest neighbor temperature data with a noise factor greater than a set threshold is removed as noise, and the mean value of the nearest neighbor temperature data after removing the noise is taken as the initial solution.

[0035] By combining nearest neighbor information from historical temperature data, representative temperature data is selected as the initial solution, improving the stability and rationality of the algorithm. Calculating the isolation level of temperature data effectively measures the degree of data anomalies, and a noise factor is constructed by combining the ranking differences of temperature and current data to identify anomalous data from multiple perspectives. Normalization ensures the comparability of data with different dimensions, while preset adjustment parameters enhance the model's adaptability to different data distributions, thereby improving the optimization effect of the stochastic hill-climbing algorithm in temperature control, making the final temperature data more reliable and accurate.

[0036] In another embodiment, the initial solution is modified as follows:

[0037] A value is randomly selected from the set error range of the initial solution as the new solution. The new solution is compared with the objective function value of the initial solution, and the solution corresponding to the largest objective function value is taken as the corrected solution. The objective function value is calculated as follows: In the formula, For the first The objective function value of each solution. For the first The energy consumption of working for one minute for the temperature data corresponding to each solution. For the first The pass rate of the corresponding packaging temperature during operation for each solution. For the natural constant An exponential function with base 0.

[0038] By randomly selecting a new solution within the set error range of the initial solution and optimizing the solution by comparing it with the objective function value, it is ensured that both energy efficiency and a high yield rate are achieved during the packaging temperature control process. By combining the objective function with energy consumption and yield rate, the system's efficiency and stability can be balanced to find the optimal packaging temperature control strategy. Adjusting the objective function value using an exponential function allows for a more refined trade-off between energy consumption and yield rate, thereby improving packaging quality and energy efficiency while effectively reducing overall energy consumption.

[0039] In one embodiment, the calculation of the first The degree of isolation of temperature data at historical time points , In the formula, , The first The distance between the packaging temperature at a historical time point and the upper and lower bounds of the box plot, which is composed of the temperature data from multiple historical time points. The variance of temperature data at multiple historical time points. To preset hyperparameters, is a normalization function.

[0040] By calculating the deviation of historical temperature data from the overall data distribution, abnormal temperature points can be identified. The upper and lower boundaries of the box plot are used to measure the degree of data outliers, making the calculation of isolation more accurate, and the stability of abnormal detection is enhanced by considering the overall data fluctuation. The normalization process uses the maximum and minimum normalization to ensure that the calculation results of different time nodes remain within a unified range, improving the comparability and adaptability of the data. By setting the adjustment parameter, the numerical instability problem in the calculation process is further avoided, making the temperature data optimization and abnormal detection more reliable, thereby improving the precision and stability of temperature control.

[0041] In another embodiment, the isolation degree of the temperature data of the first historical time node is calculated as , , wherein , is the distance between the first historical time node and the upper and lower boundaries of the box plot formed by the temperature data of the historical multiple time nodes, is the median of the temperature data of the historical multiple time nodes, is a preset hyperparameter, is a normalization function. The normalization function uses maximum and minimum normalization.

[0042] By calculating the deviation of temperature data from the overall data distribution, abnormal data points can be effectively identified. The upper and lower boundaries of the box plot are used to measure the degree of data outliers, making the calculation of isolation more accurate, and the adaptability to uneven data distribution is improved by replacing the variance with the median of historical data. The normalization process uses the maximum and minimum normalization to ensure that the calculation results of different time nodes remain within a unified range, improving the comparability of the data. By setting the adjustment parameter, the numerical instability problem in the calculation process is avoided, making the temperature data optimization and abnormal detection more reliable, thereby improving the precision and stability of temperature control.

[0043] S3: Real-time adjustment of packaging temperature based on optimal solution.

[0044] In one embodiment, the optimal solution of the packaging temperature obtained by the optimization algorithm is used as the target temperature, and the difference between the current temperature and the target temperature is detected in real time. If the temperature deviates from the target value, the heating or cooling device is automatically adjusted to ensure that the temperature is stable within the optimal range. This process ensures accurate temperature control through real-time feedback and adjustment, thereby improving production efficiency and product quality.

[0045] The scheme of the present application improves the The nearest neighbor algorithm and noise elimination technique effectively improve the precision and stability of the package temperature control. By collecting package temperature data at multiple historical time nodes and combining noise factor analysis, abnormal temperature data can be accurately identified and eliminated, the initial solution is optimized, and the optimal solution is further obtained through correction, thereby realizing precise control of the package temperature. The scheme can adjust the temperature in real time, ensure the working stability of the LED taping machine, improve the production efficiency and product quality, and reduce the impact of energy consumption and abnormal temperature fluctuations.

[0046] An embodiment of a package temperature control system for an LED taping machine:

[0047] As shown in Figure 2 , a structural block diagram of a package temperature control system for an LED taping machine according to an embodiment of the present application, comprising a processor and a memory.

[0048] The present application also provides a package temperature control system for an LED taping machine. As shown in Figure 2 , the system comprises a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the package temperature control method for an LED taping machine according to the present application.

[0049] The package temperature control system for an LED taping machine also includes other components such as a communication interface, which are well known to those skilled in the art, and their settings and functions are known in the art, so they will not be described here.

[0050] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0051] In the description of this specification, "multiple" or "several" means at least two, such as two, three or more, unless otherwise explicitly specified.

[0052] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A method for controlling the temperature of a package for an LED taping machine, characterized by, The method comprises the following steps: Collecting temperature data and current data of the LED ribbon machine at multiple historical time nodes, obtaining an optimal solution of the temperature data of the LED ribbon machine at a current time node by using a random hill climbing algorithm, and adjusting the temperature at the current time node based on the optimal solution; The initial solution when using the random hill climbing algorithm is: obtaining temperature data of a plurality of historical time nodes , calculating a noise factor of the temperature data of the , wherein the isolation degree of the temperature data of the is a preset hyperparameter, is a normalization function, and the isolation degree represents an abnormality degree of the temperature data.​​​​​​​​ Discarding the nearest neighbor temperature data with a noise factor greater than a set threshold as noise, and taking the mean value of the noise-removed nearest neighbor temperature data as an initial solution; Further comprising, correcting the initial solution, specifically: A value is randomly selected from the set error range of the initial solution as a new solution, the new solution is compared with the objective function value of the initial solution, the solution corresponding to the maximum objective function value is taken as the corrected solution, and the calculation method of the objective function value is: , wherein, is the objective function value of the first solution, is the energy consumption of the first solution corresponding to the temperature data working for one minute, is the qualified rate of the first solution corresponding to the packaging temperature working, is an exponential function with the natural constant as the base.

2. The method for controlling the temperature of the package of claim 1, wherein, The degree of isolation , In the formula, , The first The distance between the packaging temperature at a historical time point and the upper and lower bounds of the box plot, which is composed of the temperature data from multiple historical time points. The variance of temperature data at multiple historical time points. To preset hyperparameters, This is the normalization function.

3. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. Collecting temperature data of the LED ribbon machine at multiple historical time nodes by using a temperature sensor, and collecting current data of the LED ribbon machine at multiple historical time nodes by using a current sensor.

4. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. The adjustment of the temperature at the current time node is specifically: Taking the optimal solution as a target temperature, and detecting the difference between the temperature at the current time node and the target temperature in real time, if the temperature at the current time node deviates from the target temperature, automatically adjusting the heating or cooling equipment to adjust the temperature at the current time node to the target temperature.

5. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. Utilizing The neighbor algorithm obtains temperature data of a plurality of historical time nodes The neighbor algorithm obtains temperature data of a plurality of historical time nodes 6. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. Further comprising, data cleaning of the temperature data and the current data, the data cleaning comprising detecting and processing abnormal values, the abnormal values being removed by setting a change threshold.

7. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. The normalization function adopts maximum-minimum normalization.

8. The method of claim 1, wherein the temperature of the package is controlled by a temperature control device. The set threshold is 0.

5.

9. A package temperature control system for an LED taping machine, comprising: The method comprises a memory and a processor, and the memory stores computer program instructions, when the computer program instructions are executed by the processor, the method for packaging temperature control of the LED ribbon machine in any one of claims 1-8 is realized.

Images