Demoulding device and demoulding method for plastic mould
By collecting and classifying the actual parameters in the plastic mold release process in real time, and intelligently adjusting them in combination with preset parameters, the problem of inaccurate parameter control in traditional mold release methods is solved, and product quality and production efficiency are improved.
Patent Information
- Application Number
- CN202411882392.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-19
- Publication Date
- 2025-03-11
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Traditional plastic mold demolding methods are difficult to accurately control demolding parameters and are susceptible to human factors, resulting in unstable product quality and low production efficiency.
By obtaining the process accuracy requirements information of the plastic mold, preset demolding parameters of the prefabricated demolding device, and collecting actual demolding parameters during the demolding cycle for classification processing and analysis, and real-time monitoring and regulation of demolding parameters are realized to realize an automated and intelligent demolding process.
It effectively reduces the defect rate of plastic molds during the mold release process, improves product quality and production efficiency, and realizes real-time monitoring and intelligent regulation of the demolding parameter status.
Smart Images
Figure CN119589907A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of mold production, in particular to a demoulding device and a demoulding method for a plastic mold. Background Art
[0002] With the widespread application of plastic products in various fields, the requirements for the production efficiency and product quality of plastic molds are also increasing. The traditional plastic mold demolding method mainly relies on experience and manual operation, and there are problems such as demolding parameters are difficult to accurately control, the demolding process is easily interfered by human factors, and the product quality is unstable. For example, in the traditional injection molding process, the demolding parameters are difficult to accurately control, and defects such as deformation of plastic products, surface scratches, and incomplete demolding are prone to occur, affecting product quality and production efficiency. In addition, due to the complexity of the demolding process and the differences between different plastic materials and product structures, it is difficult for traditional methods to achieve real-time monitoring and intelligent control of demolding parameters, and it is impossible to promptly detect and solve abnormal situations in the demolding process, resulting in difficulty in stable control of product quality and difficulty in improving production efficiency.
[0003] In recent years, with the rapid development of automation technology, sensor technology, artificial intelligence technology and other fields, new technical means have been provided for the demoulding process of plastic molds. For example, by installing sensors to monitor various parameters in the demoulding process in real time, such as mold temperature, pressure, ejection speed, etc., real-time data of the demoulding process can be obtained and used to analyze the influence of demoulding parameters on product quality. At the same time, by optimizing and controlling the demoulding parameters using artificial intelligence algorithms, the automation and intelligence of the demoulding process can be realized, and the demoulding efficiency and product quality can be improved. However, the existing plastic mold demoulding methods and devices still have deficiencies in the intelligent control of demoulding parameters, lack of real-time monitoring and analysis of the demoulding parameter status, and it is difficult to adjust the demoulding strategy in real time according to the state of the demoulding parameters, resulting in the demoulding efficiency and product quality. It is difficult to further improve. Therefore, there is an urgent need for a plastic mold demoulding method and device that can realize intelligent control of demoulding parameters to meet the growing production needs. Summary of the invention
[0004] The invention overcomes the shortcomings of the prior art and provides a demoulding device and a demoulding method for a plastic mold.
[0005] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is: The invention discloses a demoulding method for a plastic mold, comprising the following steps: Acquire the process accuracy requirement information of the plastic mold, prefabricate the preset demoulding parameters of the demoulding device at each demoulding time node according to the process accuracy requirement information, and obtain various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; During each demoulding cycle, actual demoulding parameters of the demoulding device are collected at several preset time nodes, and the collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period; Acquire various preset demoulding parameter subsets of the demoulding device within a preset time period according to various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; Analyze the demoulding parameter status of the demoulding device according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and obtain the status of each demoulding parameter of the demoulding device within the preset time period; If all demoulding parameters of the demoulding device are in normal state within the preset time period, the demoulding parameters of the demoulding device will not be adjusted; if there are demoulding parameters of the demoulding device marked as drifting state within the preset time period, the demoulding parameters of the demoulding device will be adjusted.
[0006] Preferably, the collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, specifically: Constructing a classification graph framework, initializing an iteration endpoint, and importing the collected actual demoulding parameters into the iteration endpoint of the classification graph framework; Continuously iteratively bifurcates the iterative endpoints to continuously iteratively bifurcate the actual demoulding parameters in the iterative endpoints to form a plurality of bifurcated nodes; Determine whether there are repeated actual demoulding parameters in each bifurcation node; when there are repeated actual demoulding parameters in at least one bifurcation node, continue to perform iterative bifurcation operation; When there are no repeated actual demoulding parameters in each bifurcation node, the iterative bifurcation operation is stopped to obtain a classification map of the actual demoulding parameters.
[0007] Preferably, the collected demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, and the following steps are also included: Perform wavelet analysis on the actual demoulding parameters of each bifurcation node in the classification spectrum diagram to obtain the characteristic vector of the actual demoulding parameters in each bifurcation node; The Euclidean distance between the actual demoulding parameters in each bifurcation node is calculated according to the eigenvector of the actual demoulding parameters in each bifurcation node; Compare and analyze the Euclidean distances between the actual demoulding parameters in each bifurcation node, merge the bifurcation nodes corresponding to the Euclidean distances less than the preset distance value, and merge to obtain several node blocks; Calculate the average value of the actual demoulding parameters in each node block, and define the calculated average value as the block center of the node block; Calculate the Euclidean distance between the actual demoulding parameter in each node block and the center of the block, and determine whether there is an actual demoulding parameter in each node block whose Euclidean distance between the actual demoulding parameter and the center of the block is greater than a preset value; If so, the actual demoulding parameter in the node block is marked as noise data, and the actual demoulding parameter marked as noise data is erased; After erasing all actual demoulding parameters marked as noise data, the status of each node block in the classification atlas is updated; After the status is updated, the actual demoulding parameters stored in each node block in the classification atlas are sorted based on the acquisition timestamp to obtain various actual demoulding parameter sets of the demoulding device within a preset time period.
[0008] Preferably, the demoulding parameter status of the demoulding device is analyzed according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and the status of each demoulding parameter of the demoulding device within the preset time period is obtained, specifically: A dynamic time warping algorithm is introduced to calculate the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period based on the dynamic time warping algorithm; Comparing the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period with a preset similarity threshold; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is greater than a preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a normal state; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is not greater than the preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a drift state.
[0009] The following steps are also included: Acquire a demoulding parameter of the demoulding device that is marked as a drift state within a preset time period, and define the parameter as a drift demoulding parameter; Calculating the parameter difference between the actual demoulding parameter and the preset demoulding parameter corresponding to each time node of the drift demoulding parameter within the preset time period; The calculated parameter differences are averaged to obtain the average drift amplitude of the drift demoulding parameter within a preset time period; Comparing the average drift amplitude of the drift demoulding parameter within a preset time period with a preset amplitude; If the drift amplitude average value of the drift demoulding parameter within the preset time period is greater than the preset amplitude, after the plastic mold of the demoulding cycle is demoulded, the plastic mold is calibrated as a plastic mold that needs to be defect-detected; If the drift amplitude average value of the drift demolding parameter within the preset time period is not greater than the preset amplitude, after the plastic mold of the demolding cycle is demolded, the plastic mold is calibrated as a plastic mold that does not require defect detection.
[0010] Preferably, if the demoulding device has demoulding parameters marked as drifting within a preset time period, the demoulding parameters of the demoulding device are regulated, specifically: Prefabricate the control measures corresponding to the average values of various drift amplitudes of various demoulding parameters in the demoulding process within a preset time period in advance; Construct a knowledge graph and import various pre-made regulatory measures into the knowledge graph; Obtaining a demoulding parameter of the demoulding device that is marked as drifting in a preset time period, which is defined as a drift demoulding parameter; and obtaining an average value of the drift amplitude of the drift demoulding parameter in the preset time period; Importing the average value of the drift amplitude of the drift demoulding parameter within a preset time period into the knowledge graph for matching, and obtaining corresponding control measures; The control measures obtained by matching are transmitted to the control terminal of the demoulding device, so that the demoulding parameters of the demoulding device are regulated based on the control measures obtained by matching, so as to reduce the defective rate of the plastic mold during the demoulding process.
[0011] The present invention also discloses a demoulding device for a plastic mold, which is applied to any one of the demoulding methods for a plastic mold, wherein the demoulding device comprises an injection upper mold, an injection lower mold, a lower mold frame and an ejection mechanism; The injection upper mold and the injection lower mold can be interlocked to form an injection cavity; the injection lower mold is installed on the lower mold frame; The ejection mechanism comprises a fixing frame, which is mounted on one side of the lower injection mold, on which a pneumatic cylinder is fixedly mounted, an output end of the pneumatic cylinder is cooperatively connected with a push rod, a push block is fixedly connected to the end of the push rod, and a plurality of ejection rods are arranged on the push rod; The ejector rod can be extended into the injection cavity by the drive of the pneumatic cylinder to push out the plastic mold after injection molding.
[0012] The present invention solves the technical defects existing in the background technology, and the present invention has the following beneficial effects: the actual demoulding parameters collected are classified and processed, and various actual demoulding parameter sets of the demoulding device within a preset time period are obtained; various preset demoulding parameter subsets of the demoulding device within a preset time period are obtained according to various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; the demoulding parameter state of the demoulding device is analyzed according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and the state of each demoulding parameter of the demoulding device within a preset time period is obtained; if each demoulding parameter of the demoulding device within the preset time period is in a normal state, the demoulding parameters of the demoulding device are not regulated; if there are demoulding parameters marked as drifting states in the demoulding device within the preset time period, the demoulding parameters of the demoulding device are regulated. The present invention can effectively reduce the defective rate of plastic molds in the demoulding process, and improve product quality and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.
[0014] Figure 1 The overall method flow chart of this demoulding method is as follows; Figure 2 This is a schematic diagram of the overall structure of the demoulding device; Figure 3 This is a schematic diagram of the first internal structure of the demoulding device; Figure 4 This is a schematic diagram of the second internal structure of the demoulding device; In the figure: 8011, injection upper mold; 8022, injection lower mold; 8033, lower mold frame; 8044, fixed frame; 8055, pneumatic cylinder; 8066, push block; 8077, ejector rod; 8088, push rod. DETAILED DESCRIPTION
[0015] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
[0016] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.
[0017] like Figure 1 As shown, the present invention discloses a demoulding method for a plastic mold, comprising the following steps: S102: Acquire process accuracy requirement information of the plastic mold, prefabricate preset demoulding parameters of the demoulding device at each demoulding time node according to the process accuracy requirement information, and obtain various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; S104: During each demoulding cycle, actual demoulding parameters of the demoulding device are collected at a number of preset time nodes, and the collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period; S106: obtaining various preset demoulding parameter subsets of the demoulding device within a preset time period according to various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; S108: analyzing the demoulding parameter states of the demoulding device according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and obtaining the states of various demoulding parameters of the demoulding device within the preset time period; S110: If all demoulding parameters of the demoulding device are in normal state within the preset time period, the demoulding parameters of the demoulding device are not adjusted; if there are demoulding parameters of the demoulding device marked as drifting state within the preset time period, the demoulding parameters of the demoulding device are adjusted.
[0018] Among them, demoulding parameters include but are not limited to ejection speed, ejection distance, mold temperature, demoulding force, holding pressure, and cooling time.
[0019] Preferably, the collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, specifically: Constructing a classification graph framework, initializing an iteration endpoint, and importing the collected actual demoulding parameters into the iteration endpoint of the classification graph framework; Continuously iteratively bifurcates the iterative endpoints to continuously iteratively bifurcate the actual demoulding parameters in the iterative endpoints to form a plurality of bifurcated nodes; Determine whether there are repeated actual demoulding parameters in each bifurcation node; when there are repeated actual demoulding parameters in at least one bifurcation node, continue to perform iterative bifurcation operation; When there are no repeated actual demoulding parameters in each bifurcation node, the iterative bifurcation operation is stopped to obtain a classification map of the actual demoulding parameters.
[0020] It should be noted that the classification graph framework is first constructed, which is the basic architecture of the entire classification process. This framework is like a tree-structured blueprint for organizing and classifying demoulding parameters. Initializing the iteration endpoint is to determine the starting point of this classification process. Importing the actual demoulding parameters into the iteration endpoint is like placing the elements to be classified at the starting position of the classification system to prepare for subsequent classification operations. Continuously iterating and bifurcating the iteration endpoint is the core operation of this classification method. Each iteration bifurcation will further subdivide the parameters in the current node according to certain characteristics of the actual demoulding parameters (such as value range, relationship between parameters, etc.). For example, for the ejection speed parameter, if the initial iteration endpoint contains all the collected ejection speed values, then in the iterative bifurcation process, it can be divided into different bifurcation nodes according to the speed range (such as low speed, medium speed, and high speed). This process is repeated continuously, so that the parameters in each bifurcation node gradually have more similar characteristics. After each iterative bifurcation, it is determined whether there are repeated actual demoulding parameters in each bifurcation node. If there are duplications, it means that the current bifurcation is not detailed enough and it is necessary to continue the iterative bifurcation operation. The iterative bifurcation operation is stopped only when there are no repeated actual demoulding parameters in each bifurcation node. This process ensures that the parameters in each bifurcation node are unique, so as to obtain a reasonable and accurate classification spectrum of actual demoulding parameters. This spectrum can clearly show the distribution of different types of demoulding parameters within a preset time period, covering various parameters that have an important impact on the demoulding process, such as ejection speed and ejection distance.
[0021] Through the above-mentioned classification and processing method of actual demoulding parameters, various demoulding parameters of the demoulding device within a preset time period can be effectively classified and sorted. Demolding parameters of different types and characteristics can be clearly distinguished to form an organized classification atlas. This helps to gain an in-depth understanding of the distribution law of various parameters in the demoulding process, and provides an accurate data basis for subsequent state analysis, performance optimization, and fault diagnosis of the demoulding device. For example, based on the parameter distribution in the classification atlas, the relationship between different types of demoulding parameters can be analyzed to find out the key parameter combinations that may affect the demoulding effect, and then the operating parameters of the demoulding device can be adjusted in a targeted manner to improve the success rate of demoulding and product quality, while also helping to improve the efficiency and stability of the demoulding process.
[0022] Preferably, the collected demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, and the following steps are also included: Perform wavelet analysis on the actual demoulding parameters of each bifurcation node in the classification spectrum diagram to obtain the characteristic vector of the actual demoulding parameters in each bifurcation node; The Euclidean distance between the actual demoulding parameters in each bifurcation node is calculated according to the eigenvector of the actual demoulding parameters in each bifurcation node; Compare and analyze the Euclidean distances between the actual demoulding parameters in each bifurcation node, merge the bifurcation nodes corresponding to the Euclidean distances less than the preset distance value, and merge to obtain several node blocks; Calculate the average value of the actual demoulding parameters in each node block, and define the calculated average value as the block center of the node block; Calculate the Euclidean distance between the actual demoulding parameter in each node block and the center of the block, and determine whether there is an actual demoulding parameter in each node block whose Euclidean distance between the actual demoulding parameter and the center of the block is greater than a preset value; If so, the actual demoulding parameter in the node block is marked as noise data, and the actual demoulding parameter marked as noise data is erased; After erasing all actual demoulding parameters marked as noise data, the status of each node block in the classification atlas is updated; After the status is updated, the actual demoulding parameters stored in each node block in the classification atlas are sorted based on the acquisition timestamp to obtain various actual demoulding parameter sets of the demoulding device within a preset time period.
[0023] It should be noted that the actual demolding parameters of each bifurcation node in the classification spectrum are processed by wavelet analysis. Wavelet analysis is a time-frequency analysis method that can decompose signals into components of different scales and frequencies. Here, by processing the demolding parameters through wavelet analysis, the characteristic information of the actual demolding parameters in each bifurcation node at different scales can be mined, thereby obtaining the characteristic vectors of the actual demolding parameters in each bifurcation node. These characteristic vectors can describe the characteristics of the demolding parameters more comprehensively and meticulously, such as the local variation characteristics and periodic characteristics of the demolding parameters. The Euclidean distance between the actual demolding parameters in each bifurcation node is calculated according to the characteristic vector. The Euclidean distance is a commonly used distance measurement method that can measure the spatial distance between two vectors. By calculating this distance, the similarity between the actual demolding parameters in different bifurcation nodes can be quantified. Then, the Euclidean distance between the actual demolding parameters in each bifurcation node is compared and analyzed, and the bifurcation nodes corresponding to the Euclidean distance less than the preset distance value are fused. The purpose of this step is to merge bifurcation nodes with similar characteristics together to form several node blocks. This can further simplify the classification map, reduce unnecessary classification levels, and also gather similar demolding parameters together for subsequent analysis and processing. The average value of the actual demolding parameters in each node block is calculated and defined as the block center of the node block. This block center can be regarded as a representative value of the actual demolding parameters in the node block. Then the Euclidean distance value between the actual demolding parameters in each node block and its block center is calculated, and by comparing with the preset value, it is determined whether there is an actual demolding parameter in each node block whose Euclidean distance value between its block center is greater than the preset value. If such a parameter exists, it means that the difference between the parameter and other parameters in the block is large, which may be caused by measurement errors, abnormal conditions, etc., so it is calibrated as noise data, and the actual demolding parameters marked as noise data are erased. This helps to improve the quality of the data and remove abnormal values that may affect the analysis results. After erasing the noise data, the status of each node block in the classification map is updated. This step ensures that the classification map can reflect the actual situation after data cleaning. Finally, the actual demoulding parameters stored in each node block in the classification atlas are sorted based on the acquisition timestamp. The acquisition timestamp records the acquisition time sequence of the demoulding parameters. By sorting the actual demoulding parameters in this order, various actual demoulding parameter sets of the demoulding device within the preset time period can be obtained. The actual demoulding parameter set obtained in this way has not only been processed by classification, fusion, denoising, etc., but also arranged in chronological order, which can better reflect the dynamic changes of the demoulding parameters of the demoulding device within the preset time period.
[0024] Through the above series of steps, the demoulding parameters of the demoulding device within a preset time period can be deeply processed. First, wavelet analysis and Euclidean distance related operations realize the feature mining, similarity analysis and reasonable fusion of bifurcation nodes of demoulding parameters, simplify the classification structure and aggregate similar parameters. Secondly, outliers are eliminated through block center calculation and noise data calibration, thereby improving data quality. Finally, the timestamp-based sorting processing enables the actual demoulding parameter set to accurately reflect the changes of parameters over time during the demoulding process, which helps to more accurately analyze the working status of the demoulding device, optimize the demoulding process, improve the demoulding quality, and accurately diagnose and predict problems that may occur during the demoulding process.
[0025] Preferably, the demoulding parameter status of the demoulding device is analyzed according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and the status of each demoulding parameter of the demoulding device within the preset time period is obtained, specifically: A dynamic time warping algorithm is introduced to calculate the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period based on the dynamic time warping algorithm; Comparing the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period with a preset similarity threshold; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is greater than a preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a normal state; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is not greater than the preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a drift state.
[0026] It should be noted that the similarity between the calculated actual demoulding parameter set and the corresponding preset demoulding parameter subset is compared with the preset similarity threshold. This preset similarity threshold is a pre-set standard, which defines the boundary of whether the demoulding parameter state is normal. The setting of this threshold can be based on a large amount of experimental data, previous production experience or theoretical calculation. For example, if in previous successful demoulding cases, the similarity between the actual parameters and the preset parameters during normal demoulding is generally within a certain range, then the lower limit of this range can be used as the preset similarity threshold. If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is greater than the preset similarity threshold, it means that the actual demoulding parameters are relatively close to the preset parameters, and the demoulding process is basically carried out as expected, so the corresponding demoulding parameter state in the demoulding device is marked as a normal state. On the contrary, if the similarity is not greater than the preset similarity threshold, it means that there is a large difference between the actual demoulding parameters and the preset parameters. At this time, the corresponding demoulding parameter state in the demoulding device is marked as a drift state.
[0027] The following steps are also included: Acquire a demoulding parameter of the demoulding device that is marked as a drift state within a preset time period, and define the parameter as a drift demoulding parameter; Calculating the parameter difference between the actual demoulding parameter and the preset demoulding parameter corresponding to each time node of the drift demoulding parameter within the preset time period; The calculated parameter differences are averaged to obtain the average drift amplitude of the drift demoulding parameter within a preset time period; Comparing the average drift amplitude of the drift demoulding parameter within a preset time period with a preset amplitude; If the drift amplitude average value of the drift demoulding parameter within the preset time period is greater than the preset amplitude, after the plastic mold of the demoulding cycle is demoulded, the plastic mold is calibrated as a plastic mold that needs to be defect-detected; If the drift amplitude average value of the drift demolding parameter within the preset time period is not greater than the preset amplitude, after the plastic mold of the demolding cycle is demolded, the plastic mold is calibrated as a plastic mold that does not require defect detection.
[0028] It should be noted that the demoulding parameters marked as drifting states of the demoulding device within the preset time period are obtained and defined as drift demoulding parameters. This step is based on the previous analysis results of the demoulding parameter state to clarify which parameters are abnormal (drifting state). Then the parameter difference between the actual demoulding parameters and the preset demoulding parameters corresponding to each time node of the drift demoulding parameters within the preset time period is calculated. This calculation process is to quantify the degree to which the actual demoulding parameters deviate from the preset demoulding parameters at each time node. For example, if the ejection speed is a drift demoulding parameter, then the difference between the actual ejection speed and the preset ejection speed at each time node is calculated, and these differences can reflect the fluctuation of the ejection speed during the demoulding process. The calculated differences of each parameter are averaged to obtain the average drift amplitude of the drift demoulding parameter within the preset time period. This average value combines the parameter drift within the entire preset time period and can more comprehensively describe the degree of deviation of the drift demoulding parameter. Then the average drift amplitude is compared with the preset amplitude. The preset amplitude is a pre-set limit, which is determined based on product quality requirements, previous experience or theoretical analysis. For example, if the preset amplitude is set to a certain value, when the drift amplitude average value exceeds this value, it means that the drift is more serious. If the drift amplitude average value of the drift demolding parameter within the preset time period is greater than the preset amplitude, after the plastic mold of the demolding cycle is demolded, the plastic mold is calibrated as a plastic mold that requires defect detection. This is because a larger drift amplitude may affect the molding quality of the plastic mold, and further testing is required to determine whether there are defects. Conversely, if the drift amplitude average value is not greater than the preset amplitude, after the plastic mold of the demolding cycle is demolded, the plastic mold is calibrated as a plastic mold that does not require defect detection, indicating that although there is a certain parameter drift, this drift is within an acceptable range and is unlikely to have a significant impact on the quality of the plastic mold.
[0029] By further analyzing the drift demoulding parameters, calculating the average drift amplitude and comparing it with the preset amplitude, it is possible to accurately judge whether the plastic mold needs to be inspected for defects after demoulding. This method comprehensively considers the overall drift of the demoulding parameters within the preset time period, avoiding the decision of whether to detect defects in the plastic mold based only on a single time node or a simple status mark. It helps to more accurately screen out plastic molds that may have quality problems, improves the efficiency of quality inspection, reduces unnecessary inspection work, and also ensures product quality, ensuring that only plastic molds that may have defects are inspected for defects, effectively saving production costs, and thus optimizing the quality control link in the entire plastic mold production process.
[0030] Preferably, if the demoulding device has demoulding parameters marked as drifting within a preset time period, the demoulding parameters of the demoulding device are regulated, specifically: Prefabricate the control measures corresponding to the average values of various drift amplitudes of various demoulding parameters in the demoulding process within a preset time period in advance; Construct a knowledge graph and import various pre-made regulatory measures into the knowledge graph; Obtaining a demoulding parameter of the demoulding device that is marked as drifting in a preset time period, which is defined as a drift demoulding parameter; and obtaining an average value of the drift amplitude of the drift demoulding parameter in the preset time period; Importing the average value of the drift amplitude of the drift demoulding parameter within a preset time period into the knowledge graph for matching, and obtaining corresponding control measures; The control measures obtained by matching are transmitted to the control terminal of the demoulding device, so that the demoulding parameters of the demoulding device are regulated based on the control measures obtained by matching, so as to reduce the defective rate of the plastic mold during the demoulding process.
[0031] It should be noted that before the actual demoulding operation, according to experience, theoretical analysis or previous experimental data, corresponding solutions are formulated for the drift conditions of different degrees (average drift amplitude) that may occur in different demoulding parameters. For example, for the demoulding parameter of ejection speed, if its drift amplitude average value is small, the corresponding control measure is to fine-tune the air pressure of the ejection cylinder. Constructing a knowledge graph is a way to represent knowledge in a graphical structure, which can effectively organize and manage information. Various control measures prefabricated in advance are imported into this knowledge graph, making the knowledge graph a knowledge base containing the corresponding relationship between the drift of demoulding parameters and the control measures. In this knowledge graph, a connection is established between each drift amplitude average range (for different demoulding parameters) and the corresponding control measures, which facilitates subsequent query and matching to obtain the demoulding parameters (defined as drift demoulding parameters) marked as drifting states in the demoulding device within a preset time period, and obtain the average drift amplitude within the preset time period. This step is based on the previous analysis results of the demoulding parameter state, and clarifies the demoulding parameters that need to be regulated and the severity of their drift. For example, if the mold temperature is marked as a drift demolding parameter, and the average value of its drift amplitude is calculated to be a specific value, this value will be used as the basis for subsequent matching of control measures in the knowledge graph. The average value of the drift amplitude of the drift demolding parameter within a preset time period is imported into the knowledge graph for matching to obtain the corresponding control measures. Through this matching method, suitable control measures for the current drift situation can be found quickly and accurately. The control measures obtained by matching are then transmitted to the control terminal of the demolding device, so that the demolding parameters of the demolding device are regulated based on these control measures. For example, if the matched control measure is to adjust the value of the holding pressure, the control terminal will receive this instruction and adjust the holding pressure accordingly to reduce the defective rate of the plastic mold during the demolding process.
[0032] By prefabricating the control measures corresponding to the demoulding parameter drift and constructing a knowledge graph, it is possible to quickly obtain and implement corresponding control measures based on the drift demoulding parameters and their drift amplitude average values that occur during the actual demoulding process. This method realizes the automation and intelligence of demoulding parameter control and improves the response speed to abnormal situations during the demoulding process. By adjusting the demoulding parameters in a timely and accurate manner, the defective rate of plastic molds during the demoulding process can be effectively reduced, product quality and production efficiency can be improved, and problems such as product defects and production delays caused by abnormal demoulding parameters can be reduced.
[0033] In addition, in actual application, the method may further include the following steps: The characteristic images of the plastic molds that are judged as defective are obtained from the historical inspection results, and the fuzzy clustering algorithm is introduced. The characteristic images are clustered based on the fuzzy clustering algorithm to cluster the characteristic images with the same defect category and defect position into the same subset to generate the clustering results. According to the clustering result, a feature image is randomly extracted from each subset, and feature transformation processing is performed on the extracted image to obtain a feature texture matrix of each feature image; Constructing a multi-branch tree, and importing the feature texture matrix of each feature image into each tree node of the multi-branch tree for storage, to obtain a multi-branch tree storage model; and regularly updating the multi-branch tree storage model; Acquire a mold image of a plastic mold to be defect-detected, divide the mold image into a plurality of sub-region images, perform feature transformation processing on each sub-region image, and acquire a feature texture matrix of each sub-region image; Importing the feature texture matrix of each sub-region image into the multi-tree storage model, calculating the Pearson correlation coefficient value between the feature texture matrix of each sub-region image and the feature texture matrix of each tree node in the multi-tree storage model, and obtaining a plurality of Pearson correlation coefficient values; Compare each Pearson correlation coefficient value with a preset coefficient threshold; if at least one Pearson correlation coefficient value is greater than the preset coefficient threshold, the plastic mold that needs to be defectively inspected is determined to be a defective product; If the values of each Pearson correlation coefficient are not greater than the preset coefficient threshold, the plastic mold that needs to be inspected for defects is determined to be a good product.
[0034] It should be noted that the characteristic images of the plastic molds that are judged as defective are obtained in the historical inspection results, and then the fuzzy clustering algorithm is introduced to cluster these characteristic images. The purpose of this step is to classify and organize the historical defective characteristic images so that images with similar defect features (category and position) can be grouped together, thereby providing a more organized data structure for subsequent processing and generating clustering results. Feature transformation processing is an operation that represents the texture features of an image in the form of a matrix. This feature texture matrix can describe the texture features of the image in a quantitative way. For example, it contains a mathematical representation of information such as the direction, frequency, and roughness of the image texture, which helps to compare and match images in subsequent steps. Construct a multi-branch tree, and import the feature texture matrix of each feature image into each tree node of the multi-branch tree for storage, so as to obtain a multi-branch tree storage model. A multi-branch tree is a tree-shaped data structure, and each node can have multiple child nodes. Storing the feature texture matrix in the nodes of the multi-branch tree can easily organize and manage these feature data. The multi-tree storage model is updated regularly to ensure that the storage model can reflect the latest feature information of defective plastic molds. For example, with changes in the production process or improvements in detection technology, newly discovered defective features can be promptly incorporated into the storage model. The Pearson correlation coefficient is a statistical indicator that measures the linear correlation between two variables (here are two feature texture matrices). By calculating this coefficient value, the similarity between the sub-region image to be detected and the historical defective feature image can be quantified to obtain several Pearson correlation coefficient values. If there is at least one Pearson correlation coefficient value greater than the preset coefficient threshold, it means that the plastic mold to be detected has a high similarity with the historical defective feature image in a certain sub-region, then the defect detection plastic mold will be judged as a defective product; if each Pearson correlation coefficient value is not greater than the preset coefficient threshold, it means that the plastic mold to be detected is very different from the historical defective feature image, and it is judged as a good product. By clustering the characteristic images of historical defective plastic molds using a fuzzy clustering algorithm, constructing a multi-tree storage model, and comparing the sub-region image of the plastic mold to be inspected with the characteristic texture matrix in the storage model based on the Pearson correlation coefficient, the quality of the plastic mold that needs to be inspected can be judged efficiently and accurately. The defect feature information in the historical data is fully utilized, and through reasonable data organization (clustering and multi-tree storage) and effective comparison methods (Pearson correlation coefficient), the accuracy and efficiency of defect detection are improved, the possibility of misjudgment is reduced, and it is helpful to timely screen out defective products in the plastic mold production process, so as to further optimize the demoulding parameters according to the defect characteristics of the screened defective products.
[0035] like Figures 2 to 4As shown, the present invention also discloses a demoulding device for a plastic mold, which is applied to any one of the demoulding methods for a plastic mold, wherein the demoulding device comprises an injection upper mold 8011, an injection lower mold 8022, a lower mold frame 8033 and an ejection mechanism; The injection upper mold and the injection lower mold can be interlocked to form an injection cavity; the injection lower mold is installed on the lower mold frame; The ejection mechanism includes a fixing frame 8044, which is installed on one side of the lower injection mold, and a pneumatic cylinder 8055 is fixedly installed on the fixing frame. The output end of the pneumatic cylinder is matched with a push rod 8088, and the end of the push rod is fixedly connected to a push block 8066, and a plurality of ejection rods 8077 are arranged on the push rod; The ejector rod can be extended into the injection cavity by the drive of the pneumatic cylinder to push out the plastic mold after injection molding.
[0036] It should be noted that, in the initial state, the injection mold and the injection mold are interlocked to form an injection cavity. At this time, the ejection mechanism is not in operation, and the push rod and the ejector rod do not extend into the injection cavity. The plastic raw material is injected into the injection cavity for injection molding. During the injection molding process, the injection mold and the injection mold are kept tightly interlocked to ensure that the plastic can be formed in the set shape.
[0037] After the injection molding is completed, the demoulding operation is required. At this time, the pneumatic cylinder in the ejection mechanism is started. After receiving the start signal, the pneumatic cylinder starts working, and its output end pushes the push rod forward. When the push rod moves forward, it drives the push block at the end to move forward together, and at the same time, several ejector rods on the push rod also move into the injection cavity with the push rod. The ejector rod extends into the injection cavity. Since the ejector rod contacts the completed plastic mold, the completed plastic mold will be pushed out of the lower injection mold as the ejector rod continues to move forward, thereby completing the demoulding operation.
[0038] The pneumatic cylinder is used as the power source, and the pneumatic system has the characteristics of fast response speed. Once the demoulding procedure is started, the pneumatic cylinder can quickly push the push rod and the ejector rod to move. Compared with the traditional manual demoulding or some complex mechanical demoulding methods, it can greatly shorten the demoulding time and improve production efficiency. The setting of multiple ejector rods can simultaneously apply thrust to multiple parts of the injection mold, so that the plastic mold can be more evenly and quickly separated from the injection mold, avoiding the mold jamming phenomenon that may occur due to single-point demoulding. The structural design of the ejection mechanism is relatively simple, mainly composed of a fixed frame, a pneumatic cylinder, a push rod, a push block and an ejector rod. The connection relationship between these components is clear, and it is easy to manufacture and install. The fixed frame is installed on one side of the injection mold, which will not take up too much space. The entire demoulding device can be easily integrated into the existing plastic mold production equipment without large-scale transformation of the overall layout of the equipment.
[0039] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A method for demoulding a plastic mold, characterized in that: The following steps are involved: Acquire the process accuracy requirement information of the plastic mold, prefabricate the preset demoulding parameters of the demoulding device at each demoulding time node according to the process accuracy requirement information, and obtain various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; During each demoulding cycle, actual demoulding parameters of the demoulding device are collected at several preset time nodes, and the collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period; Acquire various preset demoulding parameter subsets of the demoulding device within a preset time period according to various preset demoulding parameter sets of the demoulding device in the entire demoulding cycle; Analyze the demoulding parameter status of the demoulding device according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and obtain the status of each demoulding parameter of the demoulding device within the preset time period; If all demoulding parameters of the demoulding device are in normal state within the preset time period, the demoulding parameters of the demoulding device are not adjusted; If the demoulding device has demoulding parameters marked as drifting within a preset time period, the demoulding parameters of the demoulding device are adjusted.
2. A method for demoulding a plastic mold according to claim 1, characterized in that: The collected actual demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, specifically: Constructing a classification graph framework, initializing an iteration endpoint, and importing the collected actual demoulding parameters into the iteration endpoint of the classification graph framework; Continuously iteratively bifurcates the iterative endpoints to continuously iteratively bifurcate the actual demoulding parameters in the iterative endpoints to form a plurality of bifurcated nodes; Determine whether there are repeated actual demoulding parameters in each bifurcation node; when there are repeated actual demoulding parameters in at least one bifurcation node, continue to perform iterative bifurcation operation; When there are no repeated actual demoulding parameters in each bifurcation node, the iterative bifurcation operation is stopped to obtain a classification map of the actual demoulding parameters.
3. A plastic mold demoulding method according to claim 2, characterized in that: The collected demoulding parameters are classified and processed to obtain various actual demoulding parameter sets of the demoulding device within a preset time period, and the following steps are also included: Perform wavelet analysis on the actual demoulding parameters of each bifurcation node in the classification spectrum diagram to obtain the characteristic vector of the actual demoulding parameters in each bifurcation node; The Euclidean distance between the actual demoulding parameters in each bifurcation node is calculated according to the eigenvector of the actual demoulding parameters in each bifurcation node; Compare and analyze the Euclidean distances between the actual demoulding parameters in each bifurcation node, merge the bifurcation nodes corresponding to the Euclidean distances less than the preset distance value, and merge to obtain several node blocks; Calculate the average value of the actual demoulding parameters in each node block, and define the calculated average value as the block center of the node block; Calculate the Euclidean distance between the actual demoulding parameter in each node block and the center of the block, and determine whether there is an actual demoulding parameter in each node block whose Euclidean distance between the actual demoulding parameter and the center of the block is greater than a preset value; If so, the actual demoulding parameter in the node block is marked as noise data, and the actual demoulding parameter marked as noise data is erased; After erasing all actual demoulding parameters marked as noise data, the status of each node block in the classification atlas is updated; After the status is updated, the actual demoulding parameters stored in each node block in the classification atlas are sorted based on the acquisition timestamp to obtain various actual demoulding parameter sets of the demoulding device within a preset time period.
4. A plastic mold demoulding method according to claim 1, characterized in that: The demoulding parameter status of the demoulding device is analyzed according to various actual demoulding parameter sets and preset demoulding parameter subsets of the demoulding device within a preset time period, and the status of each demoulding parameter of the demoulding device within the preset time period is obtained, specifically: A dynamic time warping algorithm is introduced to calculate the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period based on the dynamic time warping algorithm; Comparing the similarity between various actual demoulding parameter sets and corresponding preset demoulding parameter subsets within a preset time period with a preset similarity threshold; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is greater than a preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a normal state; If the similarity between a certain actual demoulding parameter set and the corresponding preset demoulding parameter subset is not greater than the preset similarity threshold, the corresponding demoulding parameter state in the demoulding device is marked as a drift state.
5. A plastic mold demoulding method according to claim 4, characterized in that: The following steps are also included: Acquire a demoulding parameter of the demoulding device that is marked as a drift state within a preset time period, and define the parameter as a drift demoulding parameter; Calculating the parameter difference between the actual demoulding parameter and the preset demoulding parameter corresponding to each time node of the drift demoulding parameter within the preset time period; The calculated parameter differences are averaged to obtain the average drift amplitude of the drift demoulding parameter within a preset time period; Comparing the average drift amplitude of the drift demoulding parameter within a preset time period with a preset amplitude; If the drift amplitude average value of the drift demoulding parameter within the preset time period is greater than the preset amplitude, after the plastic mold of the demoulding cycle is demoulded, the plastic mold is calibrated as a plastic mold that needs to be defect-detected; If the drift amplitude average value of the drift demolding parameter within the preset time period is not greater than the preset amplitude, after the plastic mold of the demolding cycle is demolded, the plastic mold is calibrated as a plastic mold that does not require defect detection.
6. A plastic mold demoulding method according to claim 1, characterized in that: If the demoulding device has demoulding parameters marked as drifting within a preset time period, the demoulding parameters of the demoulding device are regulated, specifically: Prefabricate the control measures corresponding to the average values of various drift amplitudes of various demoulding parameters in the demoulding process within a preset time period in advance; Construct a knowledge graph and import various pre-made regulatory measures into the knowledge graph; Obtaining a demoulding parameter of the demoulding device that is marked as drifting in a preset time period, which is defined as a drift demoulding parameter; and obtaining an average value of the drift amplitude of the drift demoulding parameter in the preset time period; Importing the average value of the drift amplitude of the drift demoulding parameter within a preset time period into the knowledge graph for matching, and obtaining corresponding control measures; The control measures obtained by matching are transmitted to the control terminal of the demoulding device, so that the demoulding parameters of the demoulding device are regulated based on the control measures obtained by matching, so as to reduce the defective rate of the plastic mold during the demoulding process.
7. A demoulding device for a plastic mold, applied to a demoulding method for a plastic mold according to any one of claims 1 to 6, characterized in that: The demoulding device comprises an injection upper mold, an injection lower mold, a lower mold frame and an ejection mechanism; The injection upper mold and the injection lower mold can be interlocked to form an injection cavity; the injection lower mold is installed on the lower mold frame; The ejection mechanism comprises a fixing frame, which is mounted on one side of the lower injection mold, on which a pneumatic cylinder is fixedly mounted, an output end of the pneumatic cylinder is cooperatively connected with a push rod, a push block is fixedly connected to the end of the push rod, and a plurality of ejection rods are arranged on the push rod; The ejector rod can be extended into the injection cavity by the drive of the pneumatic cylinder to push out the plastic mold after injection molding.