Forming defect detection method and system for PE septic tank
By acquiring and comparing the forming process parameters and model data of PE septic tanks, a deviation vector matrix is generated, defects are predicted using the mechanical performance correlation module, and directional detection is performed, the accuracy and speed problems in traditional detection methods are solved, and high efficiency is achieved , accurate defect detection effect.
Patent Information
- Application Number
- CN202411375823.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2024-12-31
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The traditional PE septic tank molding defect detection method has low control accuracy and slow response speed, which is difficult to meet the efficient, accurate and safe needs of modern industrial production.
Through the control terminal of the interactive PE septic tank primary molding machine, we obtain the molding process control monitoring parameters and reference parameters, generate a deviation vector matrix, use the mechanical performance correlation module to process data to predict mechanical performance defects, and compare and identify the concept model and monitoring model. Dimension defects are combined with the prediction results to conduct directional detection to achieve efficient and accurate defect detection.
It realizes efficient and accurate defect detection, solves the problems of low control accuracy and slow response speed, and improves the accuracy and speed of detection.
Smart Images
Figure CN119223357A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of defect detection, and in particular to a molding defect detection method and system for a PE septic tank. Background Art
[0002] With the rapid development of industrial production and transportation, PE septic tanks have been widely used in the field of sewage treatment. However, with the continuous expansion of production scale and the complexity of process flow, the molding process of PE septic tanks and its defect detection are also facing increasing challenges. Traditional PE septic tank molding defect detection methods mainly rely on simple physical tests, which have problems such as low control accuracy and slow response speed, and are difficult to meet the high efficiency, precision and safety requirements of modern industrial production. Summary of the invention
[0003] The present application provides a molding defect detection method and system for PE septic tanks, which are used to solve the technical problems of low control accuracy and slow response speed in the prior art.
[0004] In view of the above problems, the present application provides a molding defect detection method and system for PE septic tanks.
[0005] In a first aspect of the present application, a molding defect detection method for a PE septic tank is provided, the method comprising:
[0006] Interactive PE septic tank one-time forming machine control terminal obtains one-time forming process control monitoring parameters, one-time forming process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model; performs deviation comparison on the one-time forming process control monitoring parameters and the one-time forming process control benchmark parameters to generate a deviation vector matrix; processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result; performs directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result; adds the size defect area identification result and the mechanical property defect identification result into the forming defect detection result.
[0007] A second aspect of the present application provides a molding defect detection system for a PE septic tank, the system comprising:
[0008] A control terminal interaction unit, wherein the control terminal interaction unit interacts with the PE septic tank one-time molding machine control terminal to obtain one-time molding process control monitoring parameters, one-time molding process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model; a first deviation comparison unit, wherein the first deviation comparison unit performs deviation comparison on the one-time molding process control monitoring parameters and the one-time molding process control benchmark parameters to generate a deviation vector matrix; a mechanical property defect prediction result acquisition unit, wherein the mechanical property defect prediction result acquisition unit processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; a second deviation comparison unit, wherein the second deviation comparison unit performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result; an directional detection unit, wherein the directional detection unit performs directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result; a molding defect detection result acquisition unit, wherein the molding defect detection result acquisition unit adds the size defect area identification result and the mechanical property defect identification result into the molding defect detection result.
[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0010] The present application interacts with the PE septic tank one-time forming machine control terminal to obtain one-time forming process control monitoring parameters, one-time forming process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model; performs deviation comparison on one-time forming process control monitoring parameters and one-time forming process control benchmark parameters to generate a deviation vector matrix; processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result; performs directional detection based on the mechanical property defect prediction result to obtain a mechanical property defect identification result; adds the size defect area identification result and the mechanical property defect identification result to the forming defect detection result. The present invention solves the technical problems of low control accuracy and slow response speed in the prior art, obtains process parameters and model data, performs deviation comparison to generate a deviation vector matrix, uses a mechanical property association module to process data to predict mechanical property defects, compares the model to identify size defects, and combines the prediction results to detect mechanical property defects in a directional manner, thereby achieving the technical effect of efficient and accurate defect detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0012] Figure 1 A schematic flow chart of a molding defect detection method for a PE septic tank provided in an embodiment of the present application;
[0013] Figure 2 A schematic diagram of the structure of a molding defect detection system for a PE septic tank provided in an embodiment of the present application.
[0014] Explanation of the accompanying drawings: control terminal interaction unit 11, first deviation comparison unit 12, mechanical property defect prediction result acquisition unit 13, second deviation comparison unit 14, orientation detection unit 15, molding defect detection result acquisition unit 16. DETAILED DESCRIPTION
[0015] The present application provides a molding defect detection method and system for PE septic tanks, aiming to solve the technical problems of low control accuracy and slow response speed in the prior art. By acquiring process parameters and model data, deviation comparison is performed to generate a deviation vector matrix, and a mechanical property association module is used to process data to predict mechanical property defects. The model is compared to identify dimensional defects, and mechanical property defects are detected in a targeted manner based on the prediction results, thereby achieving the technical effect of efficient and accurate defect detection.
[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0017] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0018] Embodiment 1
[0019] like Figure 1 As shown, the present application provides a molding defect detection method for a PE septic tank, the method comprising:
[0020] Step S100: Interact with the PE septic tank one-step forming machine control terminal to obtain one-step forming process control monitoring parameters, one-step forming process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model.
[0021] In an embodiment of the present application, by interacting with the control terminal of the PE septic tank one-time molding machine, the one-time molding process control monitoring parameters, the one-time molding process control benchmark parameters, the PE septic tank conceptual model and the PE septic tank monitoring model are obtained. The one-time molding process control monitoring parameters are process data monitored and recorded in real time by the control terminal of the PE septic tank one-time molding machine. Including but not limited to temperature, pressure, speed, etc. The one-time molding process control benchmark parameters are preset ideal or standard process parameters used to guide the operation of the molding machine. The PE septic tank conceptual model is a design blueprint of the PE septic tank, which describes in detail the geometric shape, size, structure and other characteristics of the product. The PE septic tank monitoring model is a monitoring model generated by the PE septic tank one-time molding machine control terminal using three-dimensional scanning technology or other measurement means to collect real-time geometric data of the PE septic tank actually produced.
[0022] Step S200: performing deviation comparison between the primary molding process control monitoring parameters and the primary molding process control reference parameters to generate a deviation vector matrix.
[0023] In an embodiment of the present application, when performing a deviation comparison between a primary molding process control monitoring parameter and a primary molding process control reference parameter, the actual value of each monitoring parameter is subtracted from its corresponding reference value to obtain a deviation value. The deviation values of all monitoring parameters are organized into a matrix according to the time sequence and parameter type. For example, for all monitoring data within a molding cycle, a two-dimensional matrix is generated, in which rows represent time points and columns represent different monitoring parameters. Through the above process, a deviation vector matrix is generated.
[0024] Step S300: Processing the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result.
[0025] In the embodiment of the present application, the mechanical property association module is responsible for associating the deviation of the process parameters with the mechanical property defects of the PE septic tank, and predicting the possible mechanical property defects based on the deviation vector matrix. The deviation vector matrix is input as input data to the mechanical property association module, and based on the results of the model matching, the possible mechanical property defects are predicted to generate the mechanical property defect prediction results. The mechanical property defect prediction results include information such as the type, location, and severity of the defect.
[0026] Step S400: performing deviation comparison between the PE septic tank conceptual model and the PE septic tank monitoring model to obtain size defect area identification results.
[0027] In the embodiment of the present application, when the PE septic tank conceptual model and the PE septic tank monitoring model are compared for deviation, the dimensions in the conceptual model and the monitoring model, such as length, width, height, etc., are compared item by item. The deviation value of each dimension is calculated, and the magnitude and direction of the deviation are recorded. According to the preset deviation threshold, the dimension deviation exceeding the threshold is identified, and these deviations indicate potential dimension defects. The defective areas are marked, and the dimension defect area identification result is generated.
[0028] Step S500: performing directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result.
[0029] In the embodiment of the present application, when directional detection is performed based on the prediction results of mechanical property defects, the area that needs to be focused on is determined based on the prediction results of mechanical property defects. The prediction results provide specific defect locations and types to guide the positioning of the detection equipment and the setting of detection parameters. An appropriate detection method is selected according to the defect type. For example, for predicted material strength defects, non-destructive testing technologies such as ultrasonic testing and X-ray testing are used; for surface defects, visual inspection or laser scanning technology is used. Subsequently, the target area is inspected in detail by controlling the detection equipment, and the detection equipment obtains high-resolution detection data, including physical properties and geometric shapes, according to pre-set parameters and methods.
[0030] Next, the inspection data is analyzed in detail using machine learning and image processing techniques. For image data, such as X-ray or ultrasonic images, deep learning algorithms such as convolutional neural networks are used for feature extraction and defect identification. Convolutional neural networks are pre-trained based on historical data, where the historical data includes normal and defective images, each with a corresponding label indicating whether it has a defect and the type and location of the defect. For non-image data, such as physical parameters such as stress distribution and density, anomaly detection methods are used for analysis. Specifically, z-score analysis is used to identify abnormal data points. The z-score measures the degree to which a data point deviates from the mean, by calculating the difference between each data point and the mean and normalizing it to standard deviation units. The z-score of each inspection data point is calculated, and the abnormal points are identified based on the preset threshold. These abnormal points represent possible mechanical property defects. These abnormal points are compared with the mechanical property defect prediction results to confirm the existence of the defect and its specific location and nature. For example, if the z-score of a data point exceeds the threshold and is consistent with the defect location in the prediction result, the mechanical property defect is confirmed to exist at that location, and its specific properties, such as cracks, holes, etc., are recorded.
[0031] Finally, the confirmed mechanical property defect identification results are recorded, including the location and type of the defects, to obtain the mechanical property defect identification results.
[0032] Step S600: adding the size defect area identification result and the mechanical property defect identification result to the molding defect detection result.
[0033] In the embodiment of the present application, the size defect area identification results and the mechanical property defect identification results are summarized. According to the nature and characteristics of the defects, the size defects and mechanical property defects are classified, such as size deviation, cracks, pores, etc. A unique code or identifier is assigned to each defect. The size defect and mechanical property defect information after the summary and classification code is added to the molding defect detection result.
[0034] Furthermore, in the method provided in the embodiment of the application, the step of configuring the primary molding process control reference parameters includes:
[0035] Obtain a PE septic tank mold model; input the PE septic tank mold model into a molding process parameter identification table, and output molding process control parameters; perform molding health sample backtracking based on the PE septic tank mold model and the molding process control parameters to obtain a molding process record control parameter set; perform overall density sorting on the molding process record control parameter set to obtain the one-time molding process control benchmark parameters.
[0036] In an embodiment of the present application, the mold model of the PE septic tank is obtained by scanning the QR code or barcode on the mold, or directly retrieving the mold number from the database. Then the obtained PE septic tank mold model is input into the molding process parameter identification table. The molding process parameter identification table is a preset database containing standard process parameters corresponding to different mold models. By searching the identification table, the molding process control parameters corresponding to the mold model are obtained. These parameters include specific process control data such as temperature, pressure, and time.
[0037] Then, the molding healthy sample is traced back according to the PE septic tank mold model and the output molding process control parameters. The historical production data is retrieved from the historical database to find successful molding records similar to the current mold model and process parameters. These records provide reference data for successful production under similar conditions. Collect these successful molding records to obtain a set of molding process record control parameters.
[0038] Next, the molding process record control parameter set is subjected to overall density sorting, and the primary molding process control benchmark parameters are obtained through weight distribution calculation, deviation modulus calculation, weighted synthesis and density calculation.
[0039] Furthermore, in the method provided in the embodiment of the application, the overall density sorting of the molding process record control parameter set is performed to obtain the primary molding process control reference parameter, including:
[0040] Traverse the molding process control parameters for weight distribution to generate parameter attribute weight distribution results; obtain the first molding process record control parameter of the molding process record control parameter set; based on the first molding process record control parameter, traverse the molding process record control parameter set for deviation modulus value to obtain a deviation modulus value array set; traverse the deviation modulus value array set for normalization to obtain a deviation modulus value normalized array set; according to the parameter attribute weight distribution result, add the deviation modulus value normalized array set to generate a first distribution distance set; calculate the mean of the first distribution distance set, and set the inverse of the mean as the first molding process record control parameter distribution density; add the first molding process record control parameter distribution density into the control parameter distribution density set; extract the maximum value of the control parameter distribution density set and record the control parameter, and set it as the one-time molding process control reference parameter.
[0041] In an embodiment of the present application, all process parameters are extracted from a molding process record control parameter set. The variance of each parameter in a successful molding sample is calculated. In order to convert the variance into a weight, a method of reverse normalized variance is used. Specifically, the inverse of the variance of each parameter is calculated to reflect its importance. The smaller the variance, the larger the inverse, indicating that the importance of the parameter is higher. In order to make the sum of the weights 1, the calculated weights are normalized. The normalized weight values are stored to generate parameter attribute weight distribution results. Next, a molding process record control parameter is randomly extracted from the molding process record control parameter set as the first molding process record control parameter. The first molding process record control parameter is used as a reference value, each record in the molding process record control parameter set is traversed, the actual value is extracted, each actual value is compared with the reference value, and the deviation modulus value is calculated. The deviation modulus value represents the absolute difference between the actual value and the reference value. All the calculated deviation modulus values are stored in an array in order to form a deviation modulus value array. For each deviation modulus value array, the minimum and maximum values of the array are calculated, and these values are used for normalization. When normalizing, the normalized value is obtained by dividing the difference between the maximum value and the minimum value by the difference between the deviation modulus value and the minimum value. The values in each deviation modulus value array are traversed, and each deviation modulus value is converted to a normalized value using the aforementioned calculation process. Finally, a deviation modulus value normalized array set is obtained.
[0042] For each normalized array of deviation modulus values, perform weighted summation according to the corresponding parameter weights to generate a first distribution distance set. Calculate the mean of the first distribution distance set, and set the reciprocal of the mean as the distribution density of the first molding process record control parameter. The mean reflects the average level of all deviation values, and the reciprocal is used to indicate that the more concentrated the parameter value, the higher the distribution density. Store the distribution density of the first molding process record control parameter in a set, repeat the above steps, perform the same processing on each molding process record control parameter, and finally generate a set containing all parameter distribution densities. Finally, extract the record control parameter with the largest distribution density from the control parameter distribution density set and set it as the primary molding process control benchmark parameter.
[0043] Furthermore, step S300 in the method provided in the application embodiment also includes:
[0044] Taking the PE septic tank mold model as a constraint, the deviation vector matrix record data set and the defect mechanical property identification array data set are collected; and the mechanical property association loss function is constructed:
[0045]
[0046] Among them, LOSS represents the output value of the loss function, y ji Characterizes the i-th predicted defect mechanical property type of the j-th group of training data, Y j0 A defective mechanical property identification array representing the j-th group of training data, N representing the total number of defective mechanical property attributes of the defective mechanical property identification array of the j-th group of training data, and M representing the number of training groups of the analysis loss function; according to the mechanical property association loss function, combined with the deviation vector matrix record data set and the defective mechanical property identification array data set, the mechanical property association module is configured.
[0047] In the embodiment of the present application, the PE septic tank mold model is used as a constraint, and deviation vector matrix record data sets are collected from sensors and monitoring systems in the production process. These data sets contain the deviations between various process parameters and benchmark parameters in the actual production process. At the same time, defect mechanical property identification array data sets are collected, which contain mechanical property defect information corresponding to each production, such as strength, toughness, cracks and other defect characteristics. In order to measure the difference between predicted defects and actual defects, a mechanical property correlation loss function is constructed.
[0048] The deviation vector matrix record data set is input into the mechanical property association module, and the predicted mechanical property defect value is calculated through forward propagation. According to the loss function formula, the difference between the predicted value and the actual value is calculated to obtain the loss value. The gradient of the loss value relative to the model parameters is calculated using the back propagation algorithm. Based on these gradients, the model parameters are updated using an optimization algorithm such as the gradient descent method to minimize the loss function value. The forward propagation, loss calculation, and back propagation processes are repeatedly performed to iteratively update the model parameters until the loss function value converges or the preset stop condition is reached.
[0049] Through the above steps, the configuration of the mechanical property association module is completed by using the deviation vector matrix record data set and the defective mechanical property identification array data set according to the mechanical property association loss function.
[0050] Furthermore, step S400 in the method provided in the application embodiment also includes:
[0051] A first positioning axis is identified for the PE septic tank conceptual model, and a second positioning axis is identified for the PE septic tank monitoring model; based on the first positioning axis and the second positioning axis, the PE septic tank conceptual model and the PE septic tank monitoring model are overlapped to obtain a non-intersecting area, which is added to the size defect area identification result.
[0052] In an embodiment of the present application, CAD software or other design tools are used to identify a first positioning axis for the conceptual model of the PE septic tank. The first positioning axis is a key reference line in the model, such as a central axis or a symmetry axis. In the present application, the longitudinal center line of the conceptual model is selected as the first positioning axis. Similarly, the longitudinal center line of the PE septic tank monitoring model is selected as the second positioning axis.
[0053] According to the identified first positioning axis and second positioning axis, the PE septic tank conceptual model and the monitoring model are aligned. The alignment process is performed using a three-dimensional transformation matrix, including translation, rotation, and scaling operations, to ensure that the two models are aligned in the same coordinate system. For example, by translating and rotating the second positioning axis of the monitoring model, it is made to coincide with the first positioning axis of the conceptual model. The aligned conceptual model and the monitoring model are overlapped, and the difference between the two models is calculated. Through point cloud comparison or voxel comparison technology, the non-intersection areas between the two models are identified. These areas are the parts that do not conform to the design in actual production. The identified non-intersection areas are added to the size defect area identification results. The size defect area identification results contain information such as the specific location, size, and shape of each defect area.
[0054] Furthermore, in the method provided in the embodiment of the application, a first positioning axis is identified for the PE septic tank conceptual model, and a second positioning axis is identified for the PE septic tank monitoring model, including:
[0055] Mark a reference positioning axis for the PE septic tank mold; position the PE septic tank conceptual model into the mold to obtain the first positioning axis that coincides with the reference positioning axis; position the PE septic tank monitoring model into the mold to obtain the second positioning axis that coincides with the reference positioning axis.
[0056] In an embodiment of the present application, a computer-aided design software, such as AutoCAD or SolidWorks, is used to identify a reference positioning axis in the design file of the PE septic tank mold. The reference positioning axis is the geometric center axis or symmetry axis of the mold, which serves as a reference line for model alignment. The PE septic tank conceptual model is imported into the design software. The position and direction of the conceptual model are adjusted so that it coincides with the reference positioning axis of the mold. When the center line or symmetry axis of the conceptual model coincides with the reference positioning axis, the coincident axis is identified as the first positioning axis.
[0057] Similarly, the PE septic tank monitoring model is positioned in the mold, and the position and direction of the monitoring model are adjusted to coincide with the reference positioning axis of the mold. When the center line or symmetry axis of the monitoring model coincides with the reference positioning axis, the coincident axis is marked as the second positioning axis.
[0058] Furthermore, in the method provided in the embodiment of the application, the PE septic tank conceptual model and the PE septic tank monitoring model are overlapped according to the first positioning axis and the second positioning axis to obtain a non-intersection area, and the non-intersection area is added to the size defect area identification result, including:
[0059] Extract the vertical distance set of the PE septic tank monitoring model and the PE septic tank conceptual model in the non-intersecting area; delete the coordinate points whose vertical distance set is less than or equal to the consistent distance threshold to obtain a reconstructed non-intersecting coordinate set; perform hierarchical clustering analysis on the reconstructed non-intersecting coordinate set based on the neighborhood distance threshold to obtain a reconstructed non-intersecting area set; delete the areas whose reconstructed non-intersecting area set is less than or equal to the area threshold to obtain the size defect area identification result.
[0060] In an embodiment of the present application, the ICP algorithm is used to align the PE septic tank monitoring model point cloud with the conceptual model point cloud. In the aligned point cloud, the vertical distance of each point between the monitoring model and the conceptual model is calculated. All calculated vertical distance values are stored as coordinate points to form a vertical distance set. The vertical distance set is compared with a consistent distance threshold, and the coordinate points that are less than or equal to the consistent distance threshold are deleted to obtain a reconstructed non-intersection coordinate set. The consistent distance threshold is pre-set by technical experts based on production requirements and design tolerances.
[0061] The neighborhood distance threshold is a neighborhood distance threshold that is pre-set as needed, such as 10 mm, which is used to determine the neighboring relationship in cluster analysis. When performing hierarchical cluster analysis on the reconstructed non-intersecting coordinate set based on the neighborhood distance threshold, the Euclidean distance between each pair of points in the reconstructed non-intersecting coordinate set is calculated to generate a distance matrix. A clustering tree is constructed using an agglomerative hierarchical clustering algorithm, such as a single chain, complete link or average link method. According to the clustering results, each cluster is regarded as a reconstructed non-intersecting region. Through this process, a set of reconstructed non-intersecting regions is obtained.
[0062] Use point cloud processing tools to calculate the area of each reconstructed non-intersecting region. For example, use triangulation to calculate the surface area of each cluster region. Delete regions with an area less than or equal to the area threshold. Keep significant defect regions. The area threshold is pre-set, such as 100 square millimeters, to filter out smaller defect regions. After the above steps, the remaining set of reconstructed non-intersecting regions is the size defect region, and these regions are output as the size defect region identification results.
[0063] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:
[0064] The present application interacts with the PE septic tank one-time forming machine control terminal to obtain one-time forming process control monitoring parameters, one-time forming process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model; performs deviation comparison on one-time forming process control monitoring parameters and one-time forming process control benchmark parameters to generate a deviation vector matrix; processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result; performs directional detection based on the mechanical property defect prediction result to obtain a mechanical property defect identification result; adds the size defect area identification result and the mechanical property defect identification result to the forming defect detection result. The present invention solves the technical problems of low control accuracy and slow response speed in the prior art, obtains process parameters and model data, performs deviation comparison to generate a deviation vector matrix, uses a mechanical property association module to process data to predict mechanical property defects, compares the model to identify size defects, and combines the prediction results to detect mechanical property defects in a directional manner, thereby achieving the technical effect of efficient and accurate defect detection.
[0065] Embodiment 2
[0066] Based on the same inventive concept as the molding defect detection method for PE septic tank in the aforementioned embodiment, Figure 2 As shown, the present application provides a molding defect detection system for PE septic tanks, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:
[0067] The control terminal interaction unit 11 interacts with the PE septic tank primary molding machine control terminal to obtain the primary molding process control monitoring parameters, the primary molding process control benchmark parameters, the PE septic tank conceptual model and the PE septic tank monitoring model;
[0068] A first deviation comparison unit 12, wherein the first deviation comparison unit 12 performs deviation comparison on the primary molding process control monitoring parameter and the primary molding process control reference parameter to generate a deviation vector matrix;
[0069] A mechanical property defect prediction result acquisition unit 13, wherein the mechanical property defect prediction result acquisition unit 13 processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result;
[0070] A second deviation comparison unit 14, wherein the second deviation comparison unit 14 performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result;
[0071] A directional detection unit 15, wherein the directional detection unit 15 performs directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result;
[0072] The molding defect detection result acquisition unit 16 adds the size defect area identification result and the mechanical property defect identification result into the molding defect detection result.
[0073] Furthermore, the system is also used to implement the following functions:
[0074] Obtain a PE septic tank mold model; input the PE septic tank mold model into a molding process parameter identification table, and output molding process control parameters; perform molding health sample backtracking based on the PE septic tank mold model and the molding process control parameters to obtain a molding process record control parameter set; perform overall density sorting on the molding process record control parameter set to obtain the one-time molding process control benchmark parameters.
[0075] Furthermore, the system is also used to implement the following functions:
[0076] Traverse the molding process control parameters for weight distribution to generate parameter attribute weight distribution results; obtain the first molding process record control parameter of the molding process record control parameter set; based on the first molding process record control parameter, traverse the molding process record control parameter set for deviation modulus value to obtain a deviation modulus value array set; traverse the deviation modulus value array set for normalization to obtain a deviation modulus value normalized array set; according to the parameter attribute weight distribution result, add the deviation modulus value normalized array set to generate a first distribution distance set; calculate the mean of the first distribution distance set, and set the inverse of the mean as the first molding process record control parameter distribution density; add the first molding process record control parameter distribution density into the control parameter distribution density set; extract the maximum value of the control parameter distribution density set and record the control parameter, and set it as the one-time molding process control reference parameter.
[0077] Furthermore, the system is also used to implement the following functions:
[0078] Taking the PE septic tank mold model as a constraint, the deviation vector matrix record data set and the defect mechanical property identification array data set are collected; and the mechanical property association loss function is constructed:
[0079]
[0080] Among them, LOSS represents the output value of the loss function, y ji Characterizes the i-th predicted defect mechanical property type of the j-th group of training data, Y j0 A defective mechanical property identification array representing the j-th group of training data, N representing the total number of defective mechanical property attributes of the defective mechanical property identification array of the j-th group of training data, and M representing the number of training groups of the analysis loss function; according to the mechanical property association loss function, combined with the deviation vector matrix record data set and the defective mechanical property identification array data set, the mechanical property association module is configured.
[0081] Furthermore, the system is also used to implement the following functions:
[0082] A first positioning axis is identified for the PE septic tank conceptual model, and a second positioning axis is identified for the PE septic tank monitoring model; based on the first positioning axis and the second positioning axis, the PE septic tank conceptual model and the PE septic tank monitoring model are overlapped to obtain a non-intersecting area, which is added to the size defect area identification result.
[0083] Furthermore, the system is also used to implement the following functions:
[0084] Mark a reference positioning axis for the PE septic tank mold; position the PE septic tank conceptual model in the mold to obtain the first positioning axis that coincides with the reference positioning axis; position the PE septic tank monitoring model in the mold to obtain the second positioning axis that coincides with the reference positioning axis.
[0085] Furthermore, the system is also used to implement the following functions:
[0086] Extract the vertical distance set of the PE septic tank monitoring model and the PE septic tank conceptual model in the non-intersecting area; delete the coordinate points whose vertical distance set is less than or equal to the consistent distance threshold to obtain a reconstructed non-intersecting coordinate set; perform hierarchical clustering analysis on the reconstructed non-intersecting coordinate set based on the neighborhood distance threshold to obtain a reconstructed non-intersecting area set; delete the areas whose reconstructed non-intersecting area set is less than or equal to the area threshold to obtain the size defect area identification result.
[0087] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0088] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0089] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A molding defect detection method for a PE septic tank, characterized in that: include: Interactive PE septic tank one-step molding machine control terminal, obtain one-step molding process control monitoring parameters, one-step molding process control benchmark parameters, PE septic tank conceptual model and PE septic tank monitoring model; Performing deviation comparison between the primary molding process control monitoring parameters and the primary molding process control reference parameters to generate a deviation vector matrix; Processing the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; Performing deviation comparison between the PE septic tank conceptual model and the PE septic tank monitoring model to obtain size defect area identification results; Performing directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result; The size defect area identification result and the mechanical property defect identification result are added to the molding defect detection result.
2. The molding defect detection method for PE septic tank according to claim 1, characterized in that: The one-step molding process control reference parameter configuration step comprises: Get the PE septic tank mold model; Input the PE septic tank mold model into the molding process parameter identification table, and output the molding process control parameters; Perform molding health sample backtracking according to the PE septic tank mold model and the molding process control parameters to obtain a molding process record control parameter set; The molding process record control parameter set is subjected to overall density sorting to obtain the primary molding process control reference parameter.
3. The molding defect detection method for PE septic tank according to claim 2, characterized in that: The molding process record control parameter set is subjected to overall density sorting to obtain the primary molding process control reference parameter, including: Traverse the molding process control parameters to perform weight distribution and generate parameter attribute weight distribution results; Obtaining a first molding process record control parameter of the molding process record control parameter set; Based on the first molding process record control parameter, traverse the molding process record control parameter set to perform deviation module value, and obtain a deviation module value array set; Traversing the deviation modulus value array set to perform normalization processing to obtain a deviation modulus value normalized array set; According to the parameter attribute weight distribution result, the deviation modulus value normalized array set is summed to generate a first distribution distance set; Calculating the mean of the first distribution distance set, and setting the inverse of the mean as the distribution density of the control parameter of the first molding process record; Adding the control parameter distribution density recorded in the first molding process into the control parameter distribution density set; The maximum value of the control parameter distribution density set is extracted and recorded as the control parameter, which is set as the primary molding process control reference parameter.
4. The molding defect detection method for PE septic tank according to claim 1, characterized in that: The deviation vector matrix is processed by a mechanical property association module to generate a mechanical property defect prediction result, including: Taking the PE septic tank mold model as a constraint, the deviation vector matrix record data set and the defect mechanical property identification array data set are collected; Construct the mechanical property correlation loss function: Among them, LOSS represents the output value of the loss function, y ji Characterizes the i-th predicted defect mechanical property type of the j-th group of training data, Y j0 represents the defect mechanical property identification array of the j-th group of training data, N represents the total number of defect mechanical property attributes of the defect mechanical property identification array of the j-th group of training data, and M represents the number of training groups for the analysis loss function; The mechanical property association module is configured according to the mechanical property association loss function in combination with the deviation vector matrix record data set and the defective mechanical property identification array data set.
5. The molding defect detection method for PE septic tank according to claim 1, characterized in that: The PE septic tank conceptual model and the PE septic tank monitoring model are compared for deviations to obtain size defect area identification results, including: Identifying a first positioning axis for the PE septic tank conceptual model and identifying a second positioning axis for the PE septic tank monitoring model; According to the first positioning axis and the second positioning axis, the PE septic tank conceptual model and the PE septic tank monitoring model are overlapped to obtain non-intersecting areas, which are added to the size defect area identification results.
6. The molding defect detection method for PE septic tank according to claim 5, characterized in that: Identifying a first positioning axis for the PE septic tank conceptual model and identifying a second positioning axis for the PE septic tank monitoring model include: Mark the reference positioning axis for the PE septic tank mold; Positioning the PE septic tank conceptual model into a mold to obtain the first positioning axis that coincides with the reference positioning axis; The PE septic tank monitoring model is placed into the mold for positioning, and the second positioning axis that coincides with the reference positioning axis is obtained.
7. The molding defect detection method for PE septic tank according to claim 5, characterized in that: According to the first positioning axis and the second positioning axis, the PE septic tank conceptual model and the PE septic tank monitoring model are overlapped to obtain a non-intersection area, and added to the size defect area identification result, including: Extracting a vertical distance set between the PE septic tank monitoring model and the PE septic tank conceptual model in the non-intersection area; Deleting coordinate points whose vertical distance set is less than or equal to a consistent distance threshold to obtain a reconstructed non-intersection coordinate set; Performing a hierarchical clustering analysis on the reconstructed non-intersecting coordinate set based on a neighborhood distance threshold to obtain a reconstructed non-intersecting region set; The areas in the reconstructed non-intersection area set that are smaller than or equal to the area threshold are deleted to obtain the size defect area identification result.
8. Molding defect detection system for PE septic tank, characterized in that: The system comprises: A control terminal interaction unit, wherein the control terminal interaction unit interacts with the PE septic tank one-step forming machine control terminal to obtain one-step forming process control monitoring parameters, one-step forming process control benchmark parameters, a PE septic tank conceptual model, and a PE septic tank monitoring model; A first deviation comparison unit, which performs deviation comparison on the primary molding process control monitoring parameter and the primary molding process control reference parameter to generate a deviation vector matrix; A mechanical property defect prediction result acquisition unit, wherein the mechanical property defect prediction result acquisition unit processes the deviation vector matrix through a mechanical property association module to generate a mechanical property defect prediction result; A second deviation comparison unit, wherein the second deviation comparison unit performs deviation comparison on the PE septic tank conceptual model and the PE septic tank monitoring model to obtain a size defect area identification result; A directional detection unit, wherein the directional detection unit performs directional detection according to the mechanical property defect prediction result to obtain a mechanical property defect identification result; A molding defect detection result acquisition unit is used to add the size defect area identification result and the mechanical property defect identification result into the molding defect detection result.