Intelligent control method and apparatus for automatic feed production line

Abstract

The application discloses an intelligent control method and equipment for an automatic feeding production line, relates to the related technical field of forming material processing, and comprises the following steps: carrying out the automatic feeding production line of target plastic mold production in interaction, obtaining a plurality of feeding parameter spaces; collecting historical processing record data, determining the raw material utilization capacity of a preset control period; determining an initial feeding control parameter combination; determining a plurality of stability factors; generating a plurality of target feeding control parameters; and transmitting the plurality of control units of the plurality of feeding equipment to carry out feeding control of the automatic feeding production line. The application solves the technical problem that the feeding control of the existing plastic mold production cannot automatically adjust the feeding control parameters according to the raw material quantity of the hopper, thereby causing poor feeding control precision and stability and poor production efficiency, so that the abnormal situation cannot be reflected in time, and resources are wasted, and achieves the technical effect of improving the precision, stability and production efficiency of the feeding control.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of forming material processing technology, in particular to an intelligent control method and device for an automatic feeding production line. BACKGROUND

[0002] With the rapid development of the plastics industry, the feeding control method in the production process of plastic molds becomes increasingly important. Efficient feeding control not only ensures product quality, but also significantly improves production efficiency and resource utilization. The feeding control in the production of plastic molds relies on a vacuum suction machine to suck plastic particles from a storage area and transfer them to a dryer for drying. Then the dried plastic particles enter a hopper from the dryer. The hopper is transferred to the processing equipment at a certain feeding speed and feeding amount. The traditional feeding control cannot dynamically adjust the feeding parameters according to the plastic raw materials in the hopper to meet the equipment requirements, affecting the efficiency of mold production and the accuracy and stability of feeding control, resulting in large raw material feeding control errors and waste of plastic raw materials.

[0003] Therefore, in the current plastic mold production feeding control related technology, there is a technical problem that the feeding control parameters cannot be automatically adjusted according to the amount of raw materials in the hopper, which leads to poor feeding control accuracy, stability and poor production efficiency, and resources are wasted due to the inability to timely reflect abnormal conditions. SUMMARY

[0004] The present application provides an intelligent control method and device for an automatic feeding production line, which uses techniques such as constructing a feeding parameter space, cluster screening, and feeding stability analysis to solve the technical problem that the existing plastic mold production feeding control cannot automatically adjust the feeding control parameters according to the amount of raw materials in the hopper, which leads to poor feeding control accuracy, stability and poor production efficiency, and resources are wasted due to the inability to timely reflect abnormal conditions. The technical effect of improving the accuracy, stability and production efficiency of feeding control is achieved.

[0005] The application provides an intelligent control method for an automatic feeding production line, which comprises the following steps: interacting with the automatic feeding production line for target plastic mold production, obtaining a plurality of feeding parameter spaces of a plurality of feeding devices, wherein each feeding device is controlled by an independent control unit; collecting historical processing record data of the target plastic mold, performing raw material utilization analysis on the historical processing record data, and determining a raw material utilization tolerance of a preset control period; taking the raw material utilization tolerance as a hopper storage target, performing feeding parameter cluster screening on the plurality of feeding parameter spaces of the plurality of feeding devices, and determining an initial feeding control parameter combination, wherein the initial feeding control parameter combination comprises a plurality of initial feeding control parameters of the plurality of feeding devices; performing feeding stability analysis on the plurality of feeding devices respectively, and determining a plurality of stability factors; fine-tuning the plurality of initial feeding control parameters based on the plurality of stability factors, and generating a plurality of target feeding control parameters; and transmitting the plurality of target feeding control parameters to a plurality of control units of the plurality of feeding devices, and performing feeding control of the automatic feeding production line.

[0006] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: the plurality of feeding devices comprise a vacuum suction machine, a drying machine and a hopper.

[0007] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: traversing the historical processing record data to extract abnormal processing record data, and generating a plurality of abnormal processing record data, wherein the plurality of abnormal processing record data comprises an abnormal type, an abnormal duration and an abnormal raw material utilization amount; taking the abnormal type as an index, clustering the plurality of abnormal processing record data to obtain K abnormal processing record data sets, wherein each abnormal processing record data set corresponds to an abnormal type; classifying the K abnormal processing record data sets by using an abnormal duration division forest to obtain K division sets, each duration division set having a frequency identifier; screening the K division sets having the frequency identifier according to a preset frequency threshold to obtain K screened division sets; and performing centralized analysis on the abnormal raw material utilization amount of the abnormal processing record data in the K screened division sets to determine the raw material utilization tolerance.

[0008] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: determining K sets of screening abnormal unit raw material utilization amounts by using the abnormal raw material utilization amounts of the abnormal processing record data in the K sets of screening division sets and corresponding abnormal duration lengths; extracting a first set of screening abnormal unit utilization amounts from the K sets of screening abnormal unit raw material utilization amounts, and performing centralized analysis on the first set of screening abnormal unit raw material utilization amounts to determine a first target abnormal unit raw material utilization amount; performing centralized analysis on K-1 sets of screening abnormal unit raw material utilization amounts to determine K-1 target abnormal unit raw material utilization amounts; performing mean value calculation on the first target abnormal unit raw material utilization amount and the K-1 target abnormal unit raw material utilization amounts, and multiplying the calculation result by a time length of a preset control period to obtain the raw material utilization tolerance amount.

[0009] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: extracting a median as a starting point from the first set of screening abnormal unit raw material utilization amounts, comparing a data amount less than or equal to the starting point in the first set of screening abnormal unit raw material utilization amounts and a data amount greater than the starting point in the first set of screening abnormal unit raw material utilization amounts, and determining a first centralized direction according to a comparison result; moving the starting point to the first centralized direction by a first centralized step length to obtain a first centralized point, and determining whether a first centralized density of the first centralized point is greater than or equal to a starting point centralized density of the starting point; if yes, taking the first centralized point as a stage point, and updating the stage point as the starting point to continue moving iteration in the first centralized direction until a preset iteration number is met, taking a stage point corresponding to a maximum centralized density in the iteration process as a first target point, and taking a screening abnormal unit utilization amount corresponding to the first target point as a first target abnormal unit raw material utilization amount.

[0010] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: if yes, generating a first random value based on a random number generator; when the first random value is greater than or equal to a preset random value, accepting the first centralized point as the stage point; when the first random value is less than or equal to the preset random value, moving the starting point to a second centralized direction by a second centralized step length to obtain a second centralized point, and updating the starting point based on the second centralized point to determine a stage point, where the second centralized step length is greater than the first centralized step length, and the second centralized direction is an opposite direction of the first centralized direction.

[0011] In a possible implementation, the intelligent control method for the automatic feeding production line further performs the following processing: constructing a parameter cluster screening network layer; inputting the hopper storage target and the plurality of feeding parameter spaces into the parameter cluster screening network layer to perform parameter screening identification, and obtaining the initial feeding control parameter combination.

[0012] The application further provides an intelligent control system for an automatic feeding production line, comprising: a feeding parameter space obtaining module, which is used for obtaining a plurality of feeding parameter spaces of a plurality of feeding devices by interacting with an automatic feeding production line for target plastic mold production, wherein each feeding device is controlled by an independent control unit; a raw material utilization capacity determining module, which is used for collecting historical processing record data of the target plastic mold, performing raw material utilization analysis on the historical processing record data, and determining a raw material utilization capacity in a preset control period; an initial feeding control parameter combination determining module, which is used for taking the raw material utilization capacity as a hopper storage target, performing feeding parameter cluster screening on the plurality of feeding parameter spaces of the plurality of feeding devices, and determining an initial feeding control parameter combination, wherein the initial feeding control parameter combination comprises a plurality of initial feeding control parameters of the plurality of feeding devices; a feeding stability analysis module, which is used for performing feeding stability analysis on the plurality of feeding devices respectively, and determining a plurality of stability factors; a target feeding control parameter generating module, which is used for fine-tuning the plurality of initial feeding control parameters based on the plurality of stability factors, and generating a plurality of target feeding control parameters; and a production line feeding control module, which is used for transmitting the plurality of target feeding control parameters to a plurality of control units of the plurality of feeding devices, and performing feeding control on the automatic feeding production line.

[0013] The application further provides an electronic device, comprising:

[0014] a memory configured to store executable instructions;

[0015] a processor configured to execute the executable instructions stored in the memory to implement the intelligent control method for the automatic feeding production line.

[0016] The intelligent control method and device for the automatic feeding production line proposed in the application can interactively carry out the automatic feeding production line of the target plastic mold production, obtain a plurality of feeding parameter spaces, collect historical processing record data, determine the raw material utilization capacity of a preset control period, determine an initial feeding control parameter combination, determine a plurality of stability factors, generate a plurality of target feeding control parameters, and transmit the plurality of target feeding control parameters to a plurality of control units of a plurality of feeding devices to carry out feeding control of the automatic feeding production line. The technical problem that the feeding control of the existing plastic mold production cannot automatically adjust the feeding control parameters according to the hopper raw material quantity, thereby causing poor feeding control precision and stability and poor production efficiency, and causing the inability to timely reflect abnormal conditions and cause resource waste is solved, and the technical effects of improving the precision and stability of the feeding control and the production efficiency are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. The flowcharts are used to illustrate the operations performed by the system according to the embodiments of the present application in the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, according to the needs, various steps can be processed in reverse order or at the same time. At the same time, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0018] Figure 1 The intelligent control method flowchart for the automatic feeding production line provided by the embodiments of the present application is shown in the figure.

[0019] Figure 2 The intelligent control system structure diagram for the automatic feeding production line provided by the embodiments of the present application is shown in the figure.

[0020] Figure 3 The structure diagram of an electronic device provided by the embodiments of the present application is shown in the figure.

[0021] The figure mark explanation: feeding parameter space obtaining module 10, raw material utilization capacity determining module 20, initial feeding control parameter combination determining module 30, feeding stability analysis module 40, target feeding control parameter generating module 50, and production line feeding control module 60. DETAILED DESCRIPTION

[0022] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0023] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings, and the described embodiments should not be regarded as limitations to the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of the present application.

[0024] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The term "first\second" referred to only distinguishes similar objects, and does not represent a specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0025] The embodiments of the present application provide an intelligent control method for an automatic feeding production line, such as Figure 1 As shown in the figure, the method comprises:

[0026] In step S100, the automatic feeding production line for target plastic mold production is interacted to obtain a plurality of feeding parameter spaces of a plurality of feeding devices, wherein each feeding device is controlled by an independent control unit. By interacting with the automatic feeding production line, the current running state and parameter setting of each feeding device are obtained, that is, in the automatic production process of the plastic mold, the automatic feeding production line for target plastic mold production is first interacted, the automatic feeding production line is composed of a plurality of feeding devices, each feeding device is responsible for providing the required raw materials to the plastic mold in the production process. Specifically, the plurality of feeding devices are responsible for providing different types and different proportions of raw materials according to the needs of plastic mold production, and each feeding device is equipped with an independent control unit, which is responsible for parameter control of the operation of the device. By adjusting the parameters in the control unit, the running state of the feeding device can be controlled, such as feeding speed, feeding amount, etc. Each feeding device has its specific and corresponding feeding parameter space, which contains the parameter range that can be set by the device in different working states. For example, a feeding device may have a feeding speed range of 50 kg / min~100 kg / min and a raw material heating temperature range of 80℃~200℃. The plurality of feeding control parameters together constitute the feeding parameter space of the device, thereby ensuring more accurate, efficient and flexible feeding control in the plastic mold production process.

[0027] In a possible implementation, step S100 further includes that the plurality of feeding devices include a vacuum suction feeder, a dryer and a hopper. The vacuum suction feeder, also known as a vacuum feeding machine or a vacuum conveyor, is a dust-free closed pipe conveying equipment that uses vacuum suction to convey particulate and powder materials. It uses the pressure difference between vacuum and the environment to form gas flow in the pipe, which drives the movement of powder materials to complete the delivery of powder. Vacuum closed delivery can eliminate dust pollution, improve the working environment, reduce environmental and personnel pollution, improve cleanliness, occupy less space, complete powder delivery in narrow spaces, reduce labor intensity and improve work efficiency. The dryer is a mechanical equipment that uses heat energy to reduce the moisture content of materials. It is used for drying operations on objects and can be divided into two types: normal pressure dryer and vacuum dryer. It is used in wood making, ceramic production and other industries to remove moisture from materials to prevent deformation and cracking of products, and also facilitates transportation and storage. The hopper is a container used to store and transport materials. Its main function is to transport materials to downstream processing equipment through gravity, vibration or compression. The hopper is widely used in various production lines as an intermediate link for material transportation to ensure the continuity and stability of the production process.

[0028] In step S200, historical processing record data of the target plastic mold is collected, raw material utilization analysis is performed on the historical processing record data, and a raw material utilization tolerance of a preset control period is determined. The historical processing record data of the target plastic mold is collected, and raw material utilization analysis is performed on the historical processing record data. The historical processing record data may include the start and end time of each production, the type and quantity of raw materials used, the operating parameters of the production equipment (such as temperature, pressure, speed, etc.), and the product quality detection data produced, etc. Specifically, the raw material utilization analysis refers to analyzing the historical processing record data after collecting the data, focusing on the utilization of raw materials, such as calculating the raw material consumption rate, raw material utilization rate, and the amount of remaining or wasted raw materials in each production, while focusing on the differences in raw material utilization rate under different production conditions (such as different raw material ratios, different process parameters, etc.), so as to determine the raw material utilization tolerance of the preset control period. The raw material utilization tolerance is a range value, representing the maximum and minimum values of the expected raw material utilization rate within the preset control period. For example, assuming that the historical data shows that the average raw material utilization rate of the target plastic mold is 90%, and the standard deviation is 2%, in order to ensure the stability of production and the reasonable utilization of raw materials, the raw material utilization tolerance of the preset control period is set to 88% to 92%, indicating that if the raw material utilization rate is within this range within the preset control period, it is acceptable.

[0029] In one possible implementation, step S200 further includes step S210 of traversing the historical processing record data to extract abnormal processing records and generate a plurality of abnormal processing record data, wherein the plurality of abnormal processing record data includes an abnormal type, an abnormal duration, and an abnormal raw material utilization amount. The historical processing record is checked and analyzed one by one to identify those records that do not conform to the normal processing mode or exceed the preset threshold, and is arranged into a plurality of abnormal processing record data, each data item usually contains abnormal type, abnormal duration, and abnormal raw material utilization amount, etc. Specifically, the abnormal type describes the specific problems that occur in the processing process, such as equipment failure, operation error, raw material problem (such as substandard raw material quality, raw material supply interruption, etc.), process parameter abnormality (such as temperature, pressure, speed, etc. not within the set range), etc.; the abnormal duration refers to the time period from the beginning of the abnormality to the solution or automatic recovery of the normality; the abnormal raw material utilization amount refers to the amount of raw material that is wasted due to the influence of the processing process during the abnormality. By comparing the raw material utilization amount during normal processing and abnormal processing, the raw material utilization efficiency during abnormal processing can be determined, and corresponding measures can be taken for optimization.

[0030] Step S200 further includes step S220, clustering the plurality of abnormal processing record data indexed by the abnormal type, obtaining K abnormal processing record data sets, wherein each abnormal processing record data set corresponds to an abnormal type. Using a clustering algorithm (such as K-means, hierarchical clustering, DBSCAN, etc.), the plurality of abnormal processing record data is clustered based on the abnormal type, obtaining K abnormal processing record data sets, and the records in each set have the same abnormal type and may also have other similar characteristics (such as similar occurrence time, location or duration).

[0031] Step S200 further includes step S230, using K abnormal duration partition trees in the abnormal duration partition forest to classify the K abnormal processing record data sets, obtaining K partition sets, each duration partition set having a frequency identifier. For each abnormal processing record data set, a decision tree is constructed according to the abnormal duration partition, obtaining K different trees, and each tree performs duration partitioning for its corresponding abnormal type. Using these abnormal duration partition trees, the records in the original data set are assigned to different leaf nodes (i.e. different duration intervals). For each abnormal processing record data set, one or more duration intervals and the number of records in each interval (i.e. the frequency) are obtained, and finally K partition sets are obtained, each corresponding to an abnormal type and containing different duration intervals and their corresponding frequencies under this type, and the frequency identifier is attached, indicating how many records under this abnormal type fall into this duration interval.

[0032] Step S200 further includes step S240, screening the K partition sets with frequency identifiers according to a preset frequency threshold, obtaining K screened partition sets. The preset frequency threshold is used to determine which duration interval frequencies are significant, important or worthy of attention. For each partition set corresponding to an abnormal processing record data set, the frequency identifier of each duration interval is checked, and according to the preset frequency threshold, the duration intervals in each partition set are screened, and finally K screened partition sets are obtained, all having frequencies higher than the preset threshold.

[0033] Step S200 further includes step S250, centralized analysis of the abnormal raw material utilization amount of the abnormal processing record data in the K screened partition sets to determine the raw material utilization tolerance. For each screened partition set (i.e. each abnormal type), the average value of the abnormal raw material utilization amount is calculated, and based on historical data and quality requirements, a tolerance threshold of the raw material utilization amount is set. The abnormal raw material utilization amount in each screened partition set is compared with the set tolerance threshold to determine which abnormal processing record data has a raw material utilization amount that exceeds the acceptable range, and further determine the raw material utilization tolerance.

[0034] In a possible implementation, step S250 further includes step S251 of determining K sets of screening abnormal unit raw material utilization by dividing the abnormal raw material utilization of the abnormal processing record data in each of the K sets of screening division sets by the corresponding abnormal duration. The abnormal processing record data in each of the K sets of screening division sets is taken out, and for each abnormal processing record data, the abnormal raw material utilization is divided by the corresponding abnormal duration to obtain the raw material utilization per unit time (e.g., per minute, per hour, etc.). For example, if an abnormal processing record shows that the abnormal raw material utilization is 100 units and the abnormal duration is 2 hours, the unit raw material utilization is 100 units ÷ 2 hours = 50 units / hour. For each set of screening division (i.e., each abnormal type), all the calculated unit raw material utilizations are summarized to form a set of screening abnormal unit raw material utilizations, and K sets of screening abnormal unit raw material utilizations are determined.

[0035] Step S250 further includes step S252 of extracting a first set of screening abnormal unit utilizations from the K sets of screening abnormal unit raw material utilizations and performing centralized analysis on the first set of screening abnormal unit raw material utilizations to determine a first target abnormal unit raw material utilization. The first set of screening abnormal unit raw material utilizations is randomly extracted from the K sets of screening abnormal unit raw material utilizations. The data in the first set of screening abnormal unit utilizations is analyzed in detail, for example, by calculating the average, median, mode, etc., to understand the overall distribution and characteristics of the unit raw material utilization of the abnormal type, and finally determine the first target abnormal unit raw material utilization.

[0036] Step S250 further includes step S253 of performing centralized analysis on K-1 sets of screening abnormal unit raw material utilizations to determine K-1 target abnormal unit raw material utilizations. Similarly, the remaining sets of screening abnormal unit raw material utilizations are analyzed to determine K-1 target abnormal unit raw material utilizations. Step S250 further includes step S254 of performing average calculation on the first target abnormal unit raw material utilization and the K-1 target abnormal unit raw material utilizations, and multiplying the calculation result by the time length of the preset control period to obtain the raw material utilization tolerance.

[0037] In a possible implementation, step S253 further includes step S2531. A median is extracted from the first screened abnormal unit raw material utilization amount set as a starting point. The amount of data less than or equal to the starting point in the first screened abnormal unit raw material utilization amount set is compared with the amount of data greater than the starting point in the first screened abnormal unit raw material utilization amount set. The first concentration direction is determined according to the comparison result. The median of all data in the first screened abnormal unit raw material utilization amount set is calculated. The calculated median is set as the starting point of analysis. Then the entire data set is traversed. It is counted how many data in the first screened abnormal unit raw material utilization amount set are less than or equal to the starting point (i.e., the median). It is also counted how many data are greater than the starting point. According to the comparison result, the concentration direction of the data is determined. Specifically, if the amount of data less than or equal to the starting point is relatively large, that is, most of the data is concentrated on the left side of the median or is equal to the median, it is considered that the concentration direction of the data is low. If the amount of data greater than the starting point is relatively large, that is, most of the data is concentrated on the right side of the median, it is considered that the concentration direction of the data is high.

[0038] Step S253 further includes step S2532. The starting point is moved to the first concentration direction according to a first concentration step length. A first concentration point is obtained. It is determined whether the first concentration density of the first concentration point is greater than or equal to the starting point concentration density of the starting point. If yes, the first concentration point is taken as a stage point. The stage point is updated as the starting point. The iteration is continued according to the first concentration direction. The iteration process is continued until a preset iteration number is met. The stage point corresponding to the maximum concentration density in the iteration process is taken as a first target point. The screened abnormal unit utilization amount corresponding to the first target point is taken as a first target abnormal unit raw material utilization amount. The first concentration step length refers to the size of the screened abnormal unit raw material utilization amount moved at a time. According to the preset first concentration step length, the starting point is moved to the concentration direction by a certain amount. Thus, a new point, referred to as the first concentration point, is obtained. The concentration density of the point is calculated. The concentration density is used to measure the density of data points near the point. The concentration density of the first concentration point is compared with the starting point concentration density of the starting point. If the concentration density of the first concentration point is greater than or equal to the concentration density of the starting point, it is considered that the point is a more optimal point. If the concentration density of the first concentration point meets the condition (i.e., is greater than or equal to the concentration density of the starting point), the first concentration point is updated as a new starting point. The iteration is continued according to the first concentration direction. The iteration process is continued until a preset iteration number is met. The stage point corresponding to the maximum concentration density in the iteration process is taken as a first target point. The screened abnormal unit utilization amount corresponding to the first target point is taken as a first target abnormal unit raw material utilization amount.

[0039] In one possible implementation, step S253 further includes step S2533: if yes, generating a first random value based on a random number generator, and accepting the first concentration point as a stage point if the first random value is greater than or equal to a preset random value. A random value between 0 and 1 is generated by using a random number generator, and the preset random value is a preset threshold value, which is usually less than 1 (for example, 0.5). The random value is compared with the preset random value. If the first random value is greater than or equal to the preset random value, the current first concentration point is accepted as a stage point, which means that even if the concentration density of the first concentration point is not significantly improved (or is only slightly improved), the algorithm still has the opportunity to accept this new point as a stage point in the search process with the help of the random number.

[0040] Step S253 further includes step S2534: moving the starting point to a second concentration direction according to a second concentration step length to obtain a second concentration point, and updating and iterating the starting point based on the second concentration point to determine a stage point, wherein the second concentration step length is greater than the first concentration step length, and the second concentration direction is the opposite direction of the first concentration direction. The second concentration step length is a moving step length greater than the first concentration step length, and the second concentration direction is the opposite direction of the first concentration direction. In the search process, when the algorithm fails to find a better solution in a certain direction (i.e., the concentration density of the current point is not significantly improved), the algorithm can try to search in the opposite direction to find a possible better solution. When the first random value is less than or equal to the preset random value, the algorithm does not accept the current first concentration point as a stage point, but moves the starting point to the second concentration direction (i.e., the opposite direction of the first concentration direction) according to the second concentration step length to obtain a new point, which is referred to as a second concentration point. Then, the starting point is updated and iterated based on the second concentration point, i.e., the second concentration point is set as a new starting point, and the iteration process is restarted to finally determine a stage point.

[0041] Step S300, using the raw material utilization capacity, a wide hopper storage target is set for the plurality of feeding parameters of the plurality of feeding devices, and the initial feeding control parameter combination is determined, wherein the initial feeding control parameter combination includes a plurality of initial feeding control parameters of the plurality of feeding devices. According to the determined raw material utilization capacity, a target value of the hopper storage is set, for example, if the raw material utilization capacity is 88% to 92%, a hopper storage target close to the intermediate value but slightly conservative is set to ensure the stability of production and the rational utilization of raw materials, and then the feeding parameter space (the adjustable parameter range of the feeding speed, the feeding amount, the raw material ratio, the heating temperature, etc.) of each feeding device is analyzed. Based on the hopper storage target and the feeding parameter space of each feeding device, the feeding parameter cluster screening is performed. Specifically, by using algorithms and models, the control parameter settings of each device and their mutual influence and constraint relationship are comprehensively considered, and the initial feeding control parameter combination meeting the hopper storage target is found. The initial feeding control parameter combination includes a plurality of initial feeding control parameters of the plurality of feeding devices.

[0042] In a possible implementation, step S300 further includes step S310 of constructing a parameter cluster screening network layer. A screening model is constructed based on a neural network, a support vector machine, etc., including designing a plurality of hidden layers, different types of neurons, and specific activation functions, etc. The hopper storage target and the historical data of the feeding parameters are used as training data to input the screening model for training, to learn how to predict the performance of the hopper storage target according to the input parameter space, and to finally obtain the parameter cluster screening network layer through iterative optimization.

[0043] Step S300 further includes step S320 of inputting the hopper storage target and the plurality of feeding parameter spaces into the parameter cluster screening network layer for parameter screening and identification, to obtain the initial feeding control parameter combination. The hopper storage target and the plurality of feeding parameter spaces are provided as input to the parameter cluster screening network layer, and the parameter cluster screening network layer combines and evaluates these parameters, predicts the performance of the hopper storage target under each parameter combination, and according to the predicted performance, the network layer screens out the best initial feeding control parameter combination and outputs the screened best initial feeding control parameter combination, which is directly used to control the feeding process of the hopper, so as to optimize the hopper storage target.

[0044] Step S400, respectively, on the plurality of feeding equipment feeding stability analysis, determine a plurality of stability factors. Feeding equipment for feeding stability analysis refers to the analysis of its feeding control parameters, namely to evaluate the difference between the feeding control parameters and the actual operating parameters, the greater the difference, the worse the stability, and then determine the stability factor, the stability factor is an index for quantifying the evaluation of the stability of the feeding equipment, different feeding equipment may focus on different stability factors, for example, for a certain feeding equipment, the stability factor may include the volatility of the feeding speed, the error rate of the feeding amount, the accuracy of the raw material ratio, etc., which reflects the stability and reliability of the feeding equipment in the running process through quantitative analysis.

[0045] Step S500, based on the plurality of stability factors, fine-tune the plurality of initial feeding control parameters to generate a plurality of target feeding control parameters. The evaluation results of the stability factors are associated with the initial feeding control parameters, and it is found out which initial parameter settings have a direct impact on the stability factors, and the size and direction of the impact are determined. According to the evaluation results of the stability factors and the correlation between the initial parameters and the stability factors, a fine-tuning strategy is developed, such as adjusting the feeding speed, the feeding amount, the raw material ratio, etc. According to the fine-tuning strategy, the initial feeding control parameters are fine-tuned, and finally a plurality of target feeding control parameters are generated. These target parameters are the best parameter combination obtained through optimization under the premise of meeting the stability requirements.

[0046] Step S600, transmit the plurality of target feeding control parameters to the plurality of control units of the plurality of feeding equipment, and perform feeding control of the automatic feeding production line. After determining the plurality of target feeding control parameters that meet the stability requirements, these parameters are transmitted to the control unit of each feeding equipment through appropriate communication methods (such as Ethernet, wireless transmission, etc.). Each feeding equipment control unit will receive the target feeding control parameters sent by the system, wherein the control unit is usually a programmable logic controller (PLC), an industrial control computer (IPC), etc., which can parse and execute the received instructions. Specifically, after receiving the target feeding control parameters, the control unit sets these parameters as the current operating parameters of the equipment, including adjusting the feeding speed, the feeding amount, the raw material ratio, etc., to ensure that the equipment operates according to the predetermined requirements. Then the feeding equipment starts automatic feeding according to the new control parameters. During the automatic feeding process, the control unit monitors the running state of the equipment in real time and adjusts the parameters as needed to maintain the stability and accuracy of the feeding. If any abnormal or unstable situation is detected, the parameters are automatically adjusted or other measures are taken according to the preset rules and algorithms to ensure the smooth progress of production, further improving the efficiency and stability of the feeding control.

[0047] In the foregoing, reference is made to Figure 1The intelligent control method for the automatic feeding production line according to the embodiments of the present application is described in detail. Next, the intelligent control system for the automatic feeding production line according to the embodiments of the present application will be described with reference to Figure 2 The intelligent control system for the automatic feeding production line according to the embodiments of the present application is described.

[0048] The intelligent control system for the automatic feeding production line according to the embodiments of the present application is used to solve the technical problem that the existing feeding control for plastic mold production cannot automatically adjust the feeding control parameters according to the raw material amount of the hopper, thereby causing poor feeding control precision and stability and poor production efficiency, so that the abnormal situation cannot be timely reflected, causing resource waste. The intelligent control system for the automatic feeding production line achieves the technical effect of improving the precision, stability, and production efficiency of the feeding control. The intelligent control system for the automatic feeding production line includes: a feeding parameter space obtaining module 10, a raw material utilization capacity determining module 20, an initial feeding control parameter combination determining module 30, a feeding stability analyzing module 40, a target feeding control parameter generating module 50, and a production line feeding control module 60.

[0049] The feeding parameter space obtaining module 10 is used to interact with the automatic feeding production line for target plastic mold production, and obtain multiple feeding parameter spaces of multiple feeding devices, wherein each feeding device is controlled by an independent control unit.

[0050] The raw material utilization capacity determining module 20 is used to collect historical processing record data of the target plastic mold, analyze the raw material utilization of the historical processing record data, and determine the raw material utilization capacity of a preset control period.

[0051] The initial feeding control parameter combination determining module 30 is used to take the raw material utilization capacity as the hopper storage target, perform feeding parameter cluster screening on the multiple feeding parameter spaces of the multiple feeding devices, and determine an initial feeding control parameter combination, wherein the initial feeding control parameter combination includes multiple initial feeding control parameters of the multiple feeding devices.

[0052] The feeding stability analyzing module 40 is used to respectively analyze the feeding stability of the multiple feeding devices and determine multiple stability factors.

[0053] The target feeding control parameter generating module 50 is used to fine-tune the multiple initial feeding control parameters based on the multiple stability factors and generate multiple target feeding control parameters.

[0054] A production line feed control module 60 is configured to transmit the plurality of target feed control parameters to a plurality of control units of the plurality of feed devices for feed control of the automatic feed production line.

[0055] In the following, a detailed description of the specific configuration of the feed parameter space obtaining module 10 will be given. The feed parameter space obtaining module 10 can further comprise that the plurality of feed devices comprises a vacuum suction machine, a dryer and a hopper.

[0056] In the following, a detailed description of the specific configuration of the raw material utilization capacity determination module 20 will be given. The raw material utilization capacity determination module 20 can further comprise that the historical processing record data is traversed to extract abnormal processing record data, and a plurality of abnormal processing record data is generated, wherein the plurality of abnormal processing record data comprises an abnormal type, an abnormal duration and an abnormal raw material utilization amount; the plurality of abnormal processing record data is clustered with the abnormal type as an index to obtain K abnormal processing record data sets, wherein each abnormal processing record data set corresponds to an abnormal type; the K abnormal processing record data sets are classified by using an abnormal duration division forest to obtain K division sets, and each duration division set has a frequency identifier; the K division sets with the frequency identifier are filtered according to a preset frequency threshold to obtain K filtered division sets; and the abnormal raw material utilization amount of the abnormal processing record data in the K filtered division sets is centrally analyzed to determine the raw material utilization capacity.

[0057] In the following, a detailed description of the specific configuration of the raw material utilization capacity determination module 20 will be given. The raw material utilization capacity determination module 20 can further comprise that the abnormal raw material utilization amount of the abnormal processing record data in the K filtered division sets is divided by the corresponding abnormal duration to determine K filtered abnormal unit raw material utilization sets; a first filtered abnormal unit utilization set is extracted from the K filtered abnormal unit raw material utilization sets, and the first filtered abnormal unit raw material utilization set is centrally analyzed to determine a first target abnormal unit raw material utilization amount; K-1 filtered abnormal unit raw material utilization sets are centrally analyzed to determine K-1 target abnormal unit raw material utilization amounts; the first target abnormal unit raw material utilization amount and the K-1 target abnormal unit raw material utilization amounts are mean calculated, and the calculation result is multiplied by the time length of the preset control period to obtain the raw material utilization capacity.

[0058] Next, the specific configuration of the raw material utilization capacity determination module 20 will be described in detail. The raw material utilization capacity determination module 20 further comprises: extracting the median value in the first set of screened abnormal unit raw material utilization amounts as a starting point, comparing the data amount less than or equal to the starting point in the first set of screened abnormal unit raw material utilization amounts and the data amount greater than the starting point in the first set of screened abnormal unit raw material utilization amounts, determining the first concentration direction according to the comparison result; moving the starting point to the first concentration direction by a first concentration step length, obtaining a first concentration point, determining whether the first concentration density of the first concentration point is greater than or equal to the starting point concentration density of the starting point, if yes, taking the first concentration point as a stage point, and updating the stage point as the starting point, and continuing to move iteratively according to the first concentration direction, until the preset iteration number is met, taking the stage point corresponding to the maximum concentration density in the iteration process as the first target point, and taking the screened abnormal unit utilization amount corresponding to the first target point as the first target abnormal unit raw material utilization amount.

[0059] Next, the specific configuration of the raw material utilization capacity determination module 20 will be described in detail. The raw material utilization capacity determination module 20 further comprises: if yes, generating a first random value based on a random number generator, when the first random value is greater than or equal to a preset random value, then accepting the first concentration point as a stage point; when the first random value is less than or equal to a preset random value, moving the starting point to a second concentration direction by a second concentration step length, obtaining a second concentration point, and updating the starting point based on the second concentration point to determine the stage point, wherein the second concentration step length is greater than the first concentration step length, and the second concentration direction is the opposite direction of the first concentration direction.

[0060] Next, the specific configuration of the initial feed control parameter combination determination module 30 will be described in detail. The initial feed control parameter combination determination module 30 further comprises: constructing a parameter cluster screening network layer; inputting the hopper storage target and a plurality of feed parameter spaces into the parameter cluster screening network layer for parameter screening and identification, and obtaining the initial feed control parameter combination.

[0061] Figure 3 is a structural schematic diagram of an electronic device provided by an embodiment of the present application, which shows a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present application. Figure 3The electronic device shown is merely an example and should not impose any limitations on the functions and use ranges of the embodiments of the present application. The electronic device is in the form of a general computing device, and its components can include, but are not limited to, an input device 401, a processor 402, a memory 403, and an output device 404. The processor 402 can be one or more; the memory 403 can include a computer readable medium and at least one program product having a set of (at least one) program modules configured to perform the functions of the embodiments of the present application.

[0062] The memory 403 shown in the embodiments of the present application can employ any combination of one or more computer readable media; the computer readable storage media can be, but is not limited to, an infrared ray, a semiconductor system, a device, or an apparatus, or any combination of the above, for storing software programs, computer executable programs, and modules, such as the program instructions / modules corresponding to the intelligent control method for an automatic feeding production line in the embodiments of the present application. The processor 402 performs various functional applications and data processing of the computer device by running the software programs, instructions, and modules stored in the memory 403, i.e., implements the intelligent control method for an automatic feeding production line as described above.

[0063] The intelligent control system for an automatic feeding production line provided in the embodiments of the present application can execute the intelligent control method for an automatic feeding production line provided in any of the embodiments of the present application, and has the corresponding functional modules and beneficial effects of the execution method.

[0064] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for easy mutual differentiation, and do not limit the protection scope of the present application.

[0065] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent control method for an automated feeding production line, characterized in that, The method includes: An automated feeding production line that interactively produces target plastic molds obtains multiple feeding parameter spaces for multiple feeding devices, where each feeding device is controlled by an independent control unit; Collect historical processing record data of the target plastic mold, perform raw material utilization analysis on the historical processing record data, and determine the raw material utilization capacity of the preset control cycle; Using the wide capacity of the raw material utilization as the target for hopper storage, the feeding parameter cluster screening is performed on the multiple feeding parameter spaces of the multiple feeding devices to determine the initial feeding control parameter combination, wherein the initial feeding control parameter combination includes multiple initial feeding control parameters of multiple feeding devices; Feeding stability analysis was performed on the multiple feeding devices to determine multiple stability factors; Based on the multiple stability factors, the multiple initial feed control parameters are fine-tuned to generate multiple target feed control parameters; The multiple target feeding control parameters are transmitted to multiple control units of the multiple feeding devices to perform feeding control of the automatic feeding production line; The historical processing record data is traversed to extract abnormal processing records, generating multiple abnormal processing record data, wherein the multiple abnormal processing record data includes abnormal type, abnormal duration and abnormal raw material utilization. Using the aforementioned anomaly type as an index, the multiple anomaly processing record data are clustered to obtain K sets of anomaly processing record data, wherein each set of anomaly processing record data corresponds to an anomaly type; The K sets of abnormal processing records are classified by using K abnormal duration partitioning trees in the abnormal duration partitioning forest to obtain K partitioning sets, each duration partitioning set having a frequency identifier. K partition sets with frequency identifiers are filtered according to a preset frequency threshold to obtain K filtered partition sets; Based on the abnormal raw material utilization amount of abnormal processing record data in the K screening and partitioning sets, a centralized analysis is performed to determine the raw material utilization capacity. The abnormal raw material utilization rate of abnormal processing record data in the K screening and division sets is compared with the corresponding abnormal duration to determine the K sets of abnormal unit raw material utilization rates. Extract the first set of raw material utilization of the K sets of abnormal units, and perform a centralized analysis on the first set of raw material utilization of the abnormal units to determine the first target abnormal unit raw material utilization. A centralized analysis was conducted on the raw material utilization of K-1 screening abnormal units to determine the raw material utilization of K-1 target abnormal units; The average of the raw material utilization of the first target abnormal unit and the raw material utilization of K-1 target abnormal units is calculated, and the calculation result is multiplied by the time length of the preset control cycle to obtain the raw material utilization capacity.

2. The intelligent control method for an automated feeding production line as described in claim 1, characterized in that, The multiple feeding devices include a vacuum feeder, a dryer, and a hopper.

3. The intelligent control method for an automated feeding production line as described in claim 1, characterized in that, include: The median is extracted from the first set of raw material utilization of abnormal units as the starting point. The number of data in the first set of raw material utilization of abnormal units that is less than or equal to the starting point is compared with the number of data in the first set of raw material utilization of abnormal units that is greater than the starting point. The direction of the first set is determined based on the comparison result. According to the first concentration step size, the starting point is moved in the first concentration direction to obtain the first concentration point. It is determined whether the first concentration density of the first concentration point is greater than or equal to the starting concentration density of the starting point. If so, the first concentration point is taken as the stage point, and the stage point is updated as the starting point to continue to move and iterate in the first concentration direction until the preset number of iterations is met. The stage point corresponding to the maximum concentration density during the iteration process is taken as the first target point, and the utilization amount of the screening abnormal unit corresponding to the first target point is taken as the first target abnormal unit raw material utilization amount.

4. The intelligent control method for an automated feeding production line as described in claim 3, characterized in that, include: If so, a first random value is generated based on the random number generator. When the first random value is greater than or equal to a preset random value, the first set point is accepted as the stage point. When the first random value is less than or equal to a preset random value, the starting point is moved in the second concentration direction according to the second concentration step size to obtain the second concentration point, and the starting point is updated and iterated based on the second concentration point to determine the stage point, wherein the second concentration step size is greater than the first concentration step size, and the second concentration direction is the opposite direction of the first concentration direction.

5. The intelligent control method for an automated feeding production line as described in claim 1, characterized in that, include: Construct a parameter cluster filtering network layer; The target material in the hopper and multiple feeding parameter spaces are input into the parameter cluster filtering network layer for parameter filtering and identification to obtain the initial feeding control parameter combination.

6. An intelligent control system for an automated feeding production line, characterized in that, The system is used to implement the intelligent control method for an automated feeding production line as described in any one of claims 1-5, the system comprising: The feeding parameter space acquisition module is used to interactively operate an automated feeding production line for the production of target plastic molds, and to acquire multiple feeding parameter spaces of multiple feeding devices, wherein each feeding device is controlled by an independent control unit. The raw material utilization capacity determination module is used to collect historical processing record data of the target plastic mold, perform raw material utilization analysis on the historical processing record data, and determine the raw material utilization capacity of the preset control cycle. An initial feed control parameter combination determination module is used to perform feed parameter cluster screening on multiple feed parameter spaces of multiple feed devices with the raw material utilization capacity as the hopper storage target, and determine the initial feed control parameter combination, wherein the initial feed control parameter combination includes multiple initial feed control parameters of multiple feed devices. The feed stability analysis module is used to perform feed stability analysis on the multiple feed devices respectively and determine multiple stability factors. A target feed control parameter generation module, which fine-tunes the multiple initial feed control parameters based on the multiple stability factors to generate multiple target feed control parameters; The production line feeding control module is used to transmit the multiple target feeding control parameters to multiple control units of the multiple feeding devices to perform feeding control of the automatic feeding production line.

7. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the intelligent control method for an automated feeding production line as described in any one of claims 1-5.

Images