A method and system for digital control of a pultrusion apparatus

Abstract

The application provides a kind of pultrusion equipment digital control method and system, belong to plastic state material processing technical field, method includes: obtaining pultrusion process parameters;Obtain the pultrusion equipment of pultrusion equipment type;Train digital control model;Based on digital control model, according to pultrusion equipment type and pultrusion process parameters, digital control is carried out;In the process of digital control, real-time pultrusion operation image is obtained, and intelligent feedback is carried out according to pultrusion operation image.The application provides a kind of pultrusion equipment digital control method and system, trains digital control model, and carries out automatic intelligent digital control according to pultrusion process parameters and pultrusion equipment type.In addition, in the process of digital control, real-time pultrusion operation image is introduced, control strategy is adjusted according to pultrusion operation image, pultrusion process and product quality are optimized, digital control process is more comprehensive, further, product quality is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of plastic state material processing, in particular to a digital control method and system for pultrusion equipment. BACKGROUND

[0002] Pultrusion equipment is a device used for processing plastic and other plastic materials, such as: pultrusion equipment can be used to heat and cure fiber woven materials injected with liquid resin, then pull and draw out the product. Pultrusion equipment includes track type, hydraulic type and electric cylinder type, etc. The track type pultrusion equipment uses a track as a traction and transmission mechanism to extrude plastic state materials through an extruder and stretch and process them through the track. The track type pultrusion equipment is suitable for processing various plastic state materials, especially for stretching and processing larger size or more complex shaped products. The hydraulic type pultrusion equipment combines a hydraulic system and a traction mechanism to extrude plastic state materials through an extruder and stretch and process them through the traction mechanism. The electric cylinder type pultrusion equipment uses an electric cylinder as a driving device and combines a traction mechanism to provide power and control the stretching process.

[0003] Digital control of pultrusion equipment refers to monitoring, adjusting and optimizing the pultrusion process through digital technology and control systems to improve production efficiency, product quality and operational convenience. Digital control can also provide more data support and decision-making basis for production management and quality control, promoting traceability and continuous improvement of the production process.

[0004] The invention patent with application number CN201910994834.1 discloses a T-shaped pultrusion automatic straightening device and its control method, wherein the automatic straightening device comprises a rack, characterized in that it further comprises a clamping mechanism and a heat preservation box arranged in sequence; the clamping mechanism is fixedly installed on the rack to clamp the T-shaped pultrusion; the heat preservation box is fixedly installed on the rack and has first and second openings at both ends; the T-shaped pultrusion passes through the first and second openings, one end of the T-shaped pultrusion moves forward under the power action of the clamping mechanism, and the other end is hung with a counterweight. The above-mentioned invention is suitable for straightening T-shaped pultrusions and improves work efficiency.

[0005] However, the T-shaped pultrusion of the above-mentioned prior art does not monitor the state of the pultrusion material in real time during straightening, and cannot respond in time when the pultrusion process is abnormal, which makes the control process not comprehensive enough and reduces the production quality of the product.

[0006] Therefore, there is an urgent need for a pultrusion equipment digital control method and system to at least solve the above-mentioned problems. SUMMARY

[0007] One objective of this invention is to provide a digital control method and system for pultrusion equipment. This method trains a digital control model and performs automatic and intelligent digital control based on the parameters to be set and controlled during the pultrusion process and the type of pultrusion equipment. Simultaneously, real-time pultrusion operation images are introduced during the digital control process. Based on the actual situation and product quality information obtained from these images, the control strategy is adjusted to optimize the pultrusion process and product quality. This makes the digital control process more comprehensive and further improves product quality.

[0008] This invention provides a digital control method for pultrusion equipment, comprising:

[0009] Step 1: Obtain the pultrusion process parameters displayed on the preset digital display device on the pultrusion equipment;

[0010] Step 2: Obtain the types of pultrusion equipment. Pultrusion equipment types include: tracked, hydraulic, and electric cylinder types.

[0011] Step 3: Train the digital control model;

[0012] Step 4: Based on the digital control model, perform digital control according to the type of pultrusion equipment and pultrusion process parameters;

[0013] Step 5: During the digital control process, acquire pultrusion operation images in real time and provide intelligent feedback based on the pultrusion operation images.

[0014] Preferably, step 1: obtaining the pultrusion process parameters displayed on the preset digital display device on the pultrusion equipment, including:

[0015] Acquire the detection elements associated with the digital display device;

[0016] Obtain the detection channel of the detection element;

[0017] Based on the detection data from the detection channel, the pultrusion process parameters are determined.

[0018] Preferably, step 2: Obtain the types of pultrusion equipment, including:

[0019] Obtain historical operating parameters of pultrusion equipment for the preset pultrusion equipment type;

[0020] Based on the operating parameters, determine the operating characteristic group of the pultrusion equipment for each preset pultrusion equipment type;

[0021] By comparing the operation feature group with the real-time operation feature group of the pultrusion equipment, the preset pultrusion equipment type that matches the comparison is obtained as the pultrusion equipment type of the pultrusion equipment.

[0022] Preferably, step 3: training the digital control model, including:

[0023] Obtain the model training set, and split the model training set according to the type of device from which the training data in the model training set originates, to obtain the model sub-training set;

[0024] Based on the type of source device, control channel guidance labels are generated. At the same time, based on the model sub-training set, the data-driven model is trained.

[0025] Establish the association between control channel guidance labels and data-driven models;

[0026] Based on the model predictive control algorithm, a digital control model is generated according to the correlation relationship.

[0027] Preferably, the model training set is obtained, including:

[0028] Based on the industrial control record sharing platform, obtain pre-selected target data;

[0029] Determine the reference control information for the pre-selected target data;

[0030] Obtain target control information;

[0031] Calculate the similarity of control attributes based on the reference control information and the target control information;

[0032] Obtain verification data for the pre-selected target data;

[0033] Based on the verification data, determine the verification items and verification results, and obtain the true values ​​based on the verification items and verification results;

[0034] Based on the similarity of control attributes and the true values, the pre-selected target data that meet the selection criteria are selected, and the data are then aggregated to obtain the model training set.

[0035] Preferably, step 4: Based on the digital control model, digital control is performed according to the type of pultrusion equipment and pultrusion process parameters, including:

[0036] Determine the target control channel guide label based on the type of pultrusion equipment;

[0037] The parameter characteristics of the pultrusion process are input into the data-driven model associated with the target control channel guide tag for digital control.

[0038] Preferably, step 5: During the digital control process, real-time images of the pultrusion operation are acquired, and intelligent feedback is provided based on these images, including:

[0039] Acquire images of equipment in operation during digital control processes;

[0040] Analyze equipment operation images to obtain on-site thermal imaging temperature distribution;

[0041] Determine whether the on-site thermal imaging temperature distribution matches the target thermal imaging temperature distribution for the pultrusion operation.

[0042] If the conditions are met, determine the portion of the on-site thermal imaging temperature distribution that matches the target thermal imaging temperature distribution.

[0043] Based on the extraction relationship between the distribution area and the equipment operation image, the pultrusion operation image is determined;

[0044] Extract the material feature set of the pultrusion target from the pultrusion operation image;

[0045] Intelligent feedback is provided based on the material feature set.

[0046] Preferably, intelligent feedback is provided based on the material feature set, including:

[0047] Based on the feature type of the material feature in the material feature set, determine the feature representation of the material feature corresponding to the feature type. The feature types include: curing state, uniformity and flowability.

[0048] Based on the feature representation, determine the intelligent feedback action;

[0049] To obtain intelligent feedback and intervention between actions;

[0050] Based on the action intervention between intelligent feedback actions, the intelligent feedback actions are integrated to obtain a pre-feedback plan;

[0051] Conduct pre-tests of intelligent feedback based on the pre-feedback scheme, and obtain the pre-test results;

[0052] Evaluate the validity of the pre-test results and obtain the validity evaluation results;

[0053] The pre-feedback plan is adjusted based on the effectiveness evaluation results to obtain the target feedback plan;

[0054] Intelligent feedback is provided based on the target feedback plan.

[0055] Preferably, the pre-feedback plan is adjusted based on the effectiveness evaluation results to obtain the target feedback plan, including:

[0056] Determine whether the effectiveness assessment results meet the effectiveness assessment requirements;

[0057] If it does not meet the requirements, collect failure data for feedback failure analysis and obtain the feedback failure analysis results;

[0058] Based on the feedback failure analysis results, the integration of intelligent feedback actions is guided to obtain the target feedback scheme;

[0059] Among them, the fusion of intelligent feedback actions guided by feedback failure analysis results includes:

[0060] Based on the feedback failure analysis results, the adjustment items in the initial fusion scheme are determined;

[0061] Based on the adjustment items, determine the current fusion logic;

[0062] Based on the feedback failure analysis results and the current fusion logic, determine the adjustment logic;

[0063] Based on the adjustment logic and adjustment items, the replacement items are determined, and the adjustment items in the initial fusion scheme are replaced with the replacement items to obtain the transition scheme;

[0064] The transition schemes were integrated and pre-tested to determine the transition scheme with the highest effectiveness evaluation as the target feedback scheme.

[0065] This invention provides a digital control system for a pultrusion equipment, comprising:

[0066] The pultrusion process parameter acquisition subsystem is used to acquire the pultrusion process parameters displayed by the preset digital display device on the pultrusion equipment;

[0067] The pultrusion equipment type acquisition subsystem is used to acquire the types of pultrusion equipment, which include: tracked, hydraulic, and electric cylinder types.

[0068] The training subsystem is used to train digital control models;

[0069] The digital control subsystem is used for digital control based on a digital control model, according to the type of pultrusion equipment and pultrusion process parameters;

[0070] The intelligent feedback subsystem is used to acquire pultrusion operation images in real time during the digital control process and provide intelligent feedback based on the pultrusion operation images.

[0071] The beneficial effects of this invention are as follows:

[0072] This invention trains a digital control model and performs automatic and intelligent digital control based on the parameters that need to be set and controlled during the pultrusion process and the type of pultrusion equipment. At the same time, real-time pultrusion operation images are introduced during the digital control process. Based on the actual situation and product quality information obtained from the pultrusion operation images, the control strategy is adjusted to optimize the pultrusion process and product quality. The digital control process is more comprehensive, further improving product quality.

[0073] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0074] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0075] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0076] Figure 1 This is a schematic diagram of a digital control method for a pultrusion equipment according to an embodiment of the present invention;

[0077] Figure 2 This is a schematic diagram of a digital control system for a pultrusion equipment according to an embodiment of the present invention. Detailed Implementation

[0078] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0079] This invention provides a digital control method for pultrusion equipment, such as... Figure 1 As shown, it includes:

[0080] Step 1: Obtain the pultrusion process parameters displayed on the preset digital display device on the pultrusion equipment; wherein, the pultrusion equipment is: a processing equipment for plastic materials that is digitally controlled, such as: a device used to heat and cure woven materials injected with liquid resin and pultrude and pull out the product; the preset digital display device is: a device on the pultrusion equipment used to display and provide digital pultrusion process parameters, such as: a display screen or touch screen; the pultrusion process parameters are: parameters that need to be set and controlled during the pultrusion process, such as: extrusion speed, temperature profile, and traction force, etc.

[0081] Step 2: Obtain the types of pultrusion equipment. Pultrusion equipment types include: tracked, hydraulic, and electric cylinder types.

[0082] Step 3: Train the digital control model; wherein, the digital control model is: an AI model that replaces manual prediction and control of the equipment parameters of the pultrusion equipment in the pultrusion process based on process parameters;

[0083] Step 4: Based on the digital control model, digital control is carried out according to the type of pultrusion equipment and pultrusion process parameters; wherein, digital control is: using digital technology and control system to monitor, adjust and optimize the pultrusion process to realize automated and intelligent pultrusion operation;

[0084] Step 5: During the digital control process, real-time images of the pultrusion operation are acquired, and intelligent feedback is provided based on these images. The pultrusion operation images are captured by a camera mounted on the pultrusion equipment, showing the material in a plastic state during the pultrusion process. The intelligent feedback involves adjusting the control strategy based on the actual conditions captured by the pultrusion operation images and product quality information to optimize the pultrusion process and product quality.

[0085] The working principle and beneficial effects of the above technical solution are as follows:

[0086] This application trains a digital control model and performs automatic and intelligent digital control based on the parameters that need to be set and controlled during the pultrusion process and the type of pultrusion equipment. At the same time, real-time pultrusion operation images are introduced during the digital control process. Based on the actual situation and product quality information obtained from the pultrusion operation images, the control strategy is adjusted to optimize the pultrusion process and product quality. The digital control process is more comprehensive, further improving product quality.

[0087] In one embodiment, step 1: obtaining the pultrusion process parameters displayed by the preset digital display device on the pultrusion equipment, including:

[0088] The detection element associated with the digital display device is acquired; wherein, the detection element associated with the digital display device is a sensor that transmits detection data to the digital display device, such as a temperature sensor.

[0089] Obtain the detection channel of the detection element; wherein, the detection channel is: the data transmission channel of the detection element;

[0090] Based on the detection data from the detection channel, the pultrusion process parameters are determined.

[0091] The working principle and beneficial effects of the above technical solution are as follows:

[0092] This application obtains the detection channel of the detection element associated with the digital display device, and determines the pultrusion process parameters based on the detection data of the detection channel. The acquisition of the pultrusion process parameters is more appropriate.

[0093] In one embodiment, step 2: obtaining the type of pultrusion equipment, including:

[0094] Obtain historical operating parameters of pultrusion equipment for preset pultrusion equipment types; where the preset pultrusion equipment types are: tracked, hydraulic, and electric cylinder; historical operating parameters are the parameter settings of the pultrusion equipment used in past operation history;

[0095] Based on the operating parameters, determine the operating feature group of each preset pultrusion equipment type; wherein, the operating feature group is: the characteristic representation result of the operating parameters, such as: the operating feature representation vector;

[0096] By comparing the operational feature group with the real-time operational feature group of the pultrusion equipment, the preset pultrusion equipment type that matches the comparison is obtained as the pultrusion equipment type. Specifically, when comparing the operational feature group with the real-time operational feature group of the pultrusion equipment, vector matching is performed between the representation vector of the comparison operational feature group and the representation vector of the real-time operational feature group. If the angle between the matched vectors is within a preset angle range, then the comparison is considered to be successful.

[0097] The working principle and beneficial effects of the above technical solution are as follows:

[0098] This application introduces historical operating parameters of preset pultrusion equipment types, characterizes the operating parameters to obtain operating feature groups, compares the operating feature groups with real-time operating feature groups, and determines the preset pultrusion equipment type corresponding to the matching operating feature groups as the pultrusion equipment type, making the acquisition of pultrusion equipment type more accurate.

[0099] In one embodiment, step 3: training the digital control model includes:

[0100] Obtain the model training set, and split the model training set according to the source equipment type of the training data in the model training set to obtain the model sub-training set; wherein, the model training set includes: a collection of training data, which is: process records of manual control of pultrusion equipment obtained from the big data platform in the field of industrial control; the source equipment type is: the pultrusion equipment types recorded in the model training set;

[0101] Based on the type of source device, control channel guidance labels are generated. Simultaneously, based on the model sub-training set, a data-driven model is trained. The control channel guidance labels are labels generated based on the type of source device to guide the control channel. The data-driven model training based on the model sub-training set is implemented using deep learning technology.

[0102] Establish the association between control channel guidance labels and data-driven models;

[0103] Based on the model predictive control algorithm, a digital control model is generated according to the correlation relationship. The model predictive control algorithm is a control logic that determines the control channel based on real-time parameters and generates control commands based on the digital control model corresponding to the determined control channel to control the equipment.

[0104] The working principle and beneficial effects of the above technical solution are as follows:

[0105] Since directly acquired model training sets come from various types of equipment and may not be suitable for the pultrusion equipment requiring control, this paper proposes a method to obtain model sub-training sets by identifying the types of equipment from which the training data originates. Control channel guidance labels are generated based on these equipment types, and a data-driven model is trained for each sub-training set. Simultaneously, a correlation is established between the control channel guidance labels and the data-driven model. Finally, a global model predictive control algorithm is introduced to determine the control channel based on real-time parameters and trigger the data-driven model to generate corresponding control commands, making the generation process of the digital control model more rational.

[0106] In one embodiment, obtaining the model training set includes:

[0107] Based on the industrial control record sharing platform, pre-selected target data is obtained; wherein, the industrial control record sharing platform is a big data platform in the field of industrial control; the pre-selected target data is the process record of manual industrial digital control obtained by the industrial control record sharing platform;

[0108] Determine the reference control information for the pre-selected target data; wherein, the reference control information includes: the control equipment and control content of the pre-selected target data;

[0109] Obtain target control information; wherein, target control information includes: the required control equipment and control content;

[0110] Based on the reference control information and the target control information, the control attribute similarity is calculated; where the control attribute similarity is the degree of similarity between the reference control information and the target control information.

[0111] Obtain verification data for the pre-selected target data; wherein, the verification data is: the data source form filled in by the uploader of the pre-selected target data when uploading the corresponding pre-selected target data;

[0112] Based on the verification data, determine the verification items and results, and obtain the true value based on the verification items and results. The verification items are the fields filled in the data source form, such as recording the equipment model, factory, and time. The verification results are the results of filling in the verification items. When obtaining the true value based on the verification items and results, the verification data is verified, for example, by verifying the authenticity of the verification results against the enterprise qualification platform. The true value is calculated as follows: a true verification result is recorded as 1, and a false result as 0. The sum of the verification results is calculated, and then divided by the number of verification items to obtain the true value.

[0113] Pre-selected target data that meet the selection criteria are selected based on both control attribute similarity and true values, and this is then used to obtain the model training set. The selection criteria are: control attribute similarity greater than or equal to a manually preset control attribute similarity threshold, and true values ​​greater than or equal to a manually preset true value threshold.

[0114] The working principle and beneficial effects of the above technical solution are as follows:

[0115] This application introduces pre-selected target data from an industrial control record sharing platform, obtains reference control information, required control equipment, and control content from the pre-selected target data, and calculates control attribute similarity. It then obtains verification data from the pre-selected target data, parses the verification data to obtain verification items and results, calculates true values ​​based on the verification items and results, and aggregates pre-selected target data whose control attribute similarity and true values ​​both meet the screening criteria to generate a model training set, resulting in higher quality training data.

[0116] In one embodiment, step 4: Based on the digital control model, digital control is performed according to the type of pultrusion equipment and pultrusion process parameters, including:

[0117] Based on the type of pultrusion equipment, determine the target control channel guide label; wherein, the target control channel guide label is: a label subsequently used for determining the control channel, determined based on the type of pultrusion equipment;

[0118] The parameter characteristics of the pultrusion process parameters are input into the data-driven model associated with the target control channel guide label for digital control. The parameter characteristics include the parameter type and value of the pultrusion process parameters; the data-driven model associated with the target control channel guide label is the data-driven model corresponding to the control channel guide label that matches the target control channel guide label.

[0119] The working principle and beneficial effects of the above technical solution are as follows:

[0120] This application identifies the target control channel guide label of the pultrusion equipment, inputs the parameter characteristics of the pultrusion process parameters into the data-driven model associated with the target control channel guide label for digital control, improves data-driven efficiency, and makes the control process more precise.

[0121] In one embodiment, step 5: During the digital control process, real-time images of the pultrusion operation are acquired, and intelligent feedback is provided based on these images, including:

[0122] Acquire equipment operation images during the digital control process; wherein, the equipment operation images are: images of the work site where the pultrusion equipment is located;

[0123] Analyze equipment operation images to obtain on-site thermal imaging temperature distribution; wherein, on-site thermal imaging temperature distribution refers to the temperature distribution of the working area of ​​the pultrusion equipment obtained through thermal imaging technology;

[0124] Determine whether the on-site thermal imaging temperature distribution matches the target thermal imaging temperature distribution of the pultrusion operation; whereby the target thermal imaging temperature distribution refers to the thermal imaging temperature distribution of the pultrusion target (e.g., the pultruded pipe) in the pultrusion operation.

[0125] If the conditions are met, determine the portion of the on-site thermal imaging temperature distribution that matches the target thermal imaging temperature distribution; wherein, the portion of the distribution is: the distribution map in the on-site thermal imaging temperature distribution map that matches the target thermal imaging temperature distribution;

[0126] Based on the extraction relationship between the distribution portion and the equipment operation image, the pultrusion operation image is determined; wherein, the extraction relationship is: the distribution portion is based on where the thermal imaging temperature distribution is drawn in the equipment operation image;

[0127] Extract the material feature set of the pultrusion target from the pultrusion operation image; the material feature set includes multiple material features, such as material type, material temperature, and material flow rate;

[0128] Intelligent feedback is provided based on the material feature set.

[0129] The working principle and beneficial effects of the above technical solution are as follows:

[0130] During pultrusion, production anomalies may occur, such as low room temperature, failure to meet curing time requirements before subsequent operations, or collapse of uncured material. Therefore, digital control of the equipment operation images is introduced to acquire on-site thermal imaging temperature distribution. Based on the on-site thermal imaging temperature distribution and the target thermal imaging temperature distribution of the pultrusion operation, the image corresponding to the target pultruded material is determined. Furthermore, material feature sets are extracted from the pultrusion target in the pultrusion operation images, improving the efficiency of material feature extraction.

[0131] In one embodiment, intelligent feedback based on a set of material characteristics includes:

[0132] Based on the feature types of material features in the material feature set, the feature representation of the material feature corresponding to the feature type is determined. The feature types include: curing state, uniformity, and flowability. The feature representation is: a feature representation description vector based on the material features corresponding to the feature type.

[0133] Based on the feature representation, determine the intelligent feedback action; where the intelligent feedback action is: the feedback control signal, for example: if the feature representation is: insufficient flow, the intelligent feedback action is: increase the temperature of the heating device; when determining the intelligent feedback action, match the feature representation with the pre-selected feature representation in the preset standard feedback action determination library. If the match is successful, the standard feedback action corresponding to the matching pre-selected feature representation is taken as the intelligent feedback action.

[0134] Obtain action intervention between intelligent feedback actions; where action intervention is: action conflict between intelligent feedback actions;

[0135] Based on the action intervention between intelligent feedback actions, the intelligent feedback actions are integrated to obtain a pre-feedback scheme; wherein, the integrated intelligent feedback actions are: adjusting the intelligent feedback actions with action intervention, including: adjusting the execution time sequence and execution time interval of the intelligent feedback actions with action intervention;

[0136] Pre-testing of intelligent feedback is conducted based on the pre-feedback scheme, and the pre-test results are obtained. The pre-testing is: conducting a preliminary test of the intelligent feedback of the pre-feedback scheme to evaluate its effectiveness and feasibility. The pre-test results are: test results of whether the extrusion equipment provides timely feedback of control signals under simulated abnormal conditions.

[0137] Evaluate the effectiveness of the pre-test results and obtain the effectiveness evaluation results; the effectiveness evaluation results are: the pre-test results are evaluated and analyzed manually to determine the effectiveness and feasibility of the pre-feedback plan;

[0138] The pre-feedback scheme is adjusted based on the effectiveness evaluation results to obtain the target feedback scheme; wherein, the target feedback scheme is: the control scheme obtained after adjusting and optimizing the pre-feedback scheme based on the effectiveness evaluation results;

[0139] Intelligent feedback is provided based on the target feedback plan.

[0140] The working principle and beneficial effects of the above technical solution are as follows:

[0141] This application acquires the feature types of material features from a material feature set and determines the feature representation of each feature type. Based on the feature representation, it determines intelligent feedback actions, resulting in more accurate acquisition of these actions. It acquires the action interventions between intelligent feedback actions and, based on these interventions, fuses the intelligent feedback actions to obtain a pre-feedback scheme. The pre-feedback scheme is pre-tested using intelligent feedback, and the pre-test results are obtained. Based on the effectiveness evaluation results of the prediction results, the pre-feedback scheme is adjusted to obtain the target feedback scheme. Finally, intelligent feedback is performed based on the target feedback scheme, further improving the appropriateness of the feedback.

[0142] In one embodiment, adjusting the pre-feedback scheme based on the effectiveness evaluation results to obtain the target feedback scheme includes:

[0143] Determine whether the effectiveness assessment results meet the effectiveness assessment requirements; the effectiveness assessment requirements are: the products generated by the pre-test task meet the product quality requirements;

[0144] If not, failure data is collected for feedback failure analysis, and feedback failure analysis results are obtained; whereby, failure data refers to: assessment data generated by the pre-test task indicating that the product does not meet quality requirements; feedback failure analysis results refer to: feedback failure attribution results;

[0145] Based on the feedback failure analysis results, the integration of intelligent feedback actions is guided to obtain the target feedback scheme;

[0146] Among them, the fusion of intelligent feedback actions guided by feedback failure analysis results includes:

[0147] Based on the feedback failure analysis results, the adjustment items in the initial fusion scheme are determined; among them, the adjustment items are: the intelligent feedback actions that need to be adjusted;

[0148] Based on the adjustment items, determine the current fusion logic; where the current fusion logic is: the basis for formulating the adjustment items;

[0149] Based on the feedback failure analysis results and the current fusion logic, the adjustment logic is determined; among which, the adjustment logic is: the adjustment plan for the adjustment items;

[0150] Based on the adjustment logic and adjustment items, the replacement items are determined, and the adjustment items in the initial fusion scheme are replaced with the replacement items to obtain the transition scheme;

[0151] The transition schemes were integrated and pre-tested to determine the transition scheme with the highest effectiveness evaluation as the target feedback scheme.

[0152] The working principle and beneficial effects of the above technical solution are as follows:

[0153] This application performs feedback failure analysis when the effectiveness assessment result fails, obtains the feedback failure analysis result, guides the fusion of intelligent feedback actions based on the feedback failure analysis result, obtains the target feedback scheme, and determines adjustment items when guiding the fusion of intelligent feedback actions. The adjustment items are parsed to obtain the current fusion logic, and the adjustment logic is determined based on the feedback failure analysis result and the current fusion logic. After determining the adjustment logic, multiple replacement items are generated to sequentially replace the adjustment items in the initial fusion scheme to obtain multiple transitional schemes. These transitional schemes are then fused and pre-tested, and the transitional scheme with the highest effectiveness evaluation is determined as the target feedback scheme. The formulation of the target feedback scheme is more efficient and more reasonable.

[0154] This invention provides a digital control system for pultrusion equipment, such as... Figure 2 As shown, it includes:

[0155] Pultrusion process parameter acquisition subsystem 1 is used to acquire the pultrusion process parameters displayed by the preset digital display device on the pultrusion equipment;

[0156] Pultrusion Equipment Type Acquisition Subsystem 2 is used to acquire the types of pultrusion equipment, which include: tracked, hydraulic, and electric cylinder types.

[0157] Training subsystem 3 is used to train the digital control model;

[0158] Digital control subsystem 4 is used for digital control based on a digital control model, according to the type of pultrusion equipment and pultrusion process parameters;

[0159] The intelligent feedback subsystem 5 is used to acquire pultrusion operation images in real time during the digital control process and provide intelligent feedback based on the pultrusion operation images.

[0160] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A digital control method for pultrusion equipment, characterized in that, The application comprises the following steps: Step 1: Obtain the pultrusion process parameters displayed by the preset digital display device on the pultrusion equipment; Step 2: Obtain the pultrusion equipment type of the pultrusion equipment, which includes track type, hydraulic type and electric cylinder type; Step 3: Train the digital control model; Step 4: Based on the digital control model, perform digital control according to the pultrusion equipment type and the pultrusion process parameters; Step 5: In the process of digital control, real-time pultrusion operation images are obtained, and intelligent feedback is performed according to the pultrusion operation images, including: Obtain the equipment operation image of the digital control process; Analyze the equipment operation image to obtain the on-site thermal imaging temperature distribution; Determine whether the on-site thermal imaging temperature distribution conforms to the target thermal imaging temperature distribution of the pultrusion operation; If it conforms, determine the distribution part of the on-site thermal imaging temperature distribution that conforms to the target thermal imaging temperature distribution; Determine the pultrusion operation image according to the extraction relationship between the distribution part and the equipment operation image; Extract the material feature set of the pultrusion target in the pultrusion operation image; According to the feature type of the material feature in the material feature set, determine the feature representation of the material feature corresponding to the feature type, and the feature type includes solidification state, uniformity and flowability; Determine the intelligent feedback action according to the feature representation; Obtain the action intervention between the intelligent feedback actions; According to the action intervention between the intelligent feedback actions, fuse the intelligent feedback actions to obtain a pre-feedback scheme; Perform pre-test of intelligent feedback according to the pre-feedback scheme to obtain a pre-test result; Evaluate the effectiveness of the pre-test result to obtain an effectiveness evaluation result; Determine whether the effectiveness evaluation result meets the effectiveness evaluation requirement; If it does not meet, collect failure data for feedback failure analysis to obtain a feedback failure analysis result; Based on the feedback failure analysis result, guide the fusion of the intelligent feedback action to obtain a target feedback scheme; Wherein, based on the feedback failure analysis result, the fusion of the intelligent feedback action comprises: According to the feedback failure analysis result, determine the adjustment item in the initial fusion scheme; According to the adjustment item, determine the current fusion logic; According to the feedback failure analysis result and the current fusion logic, determine the adjustment logic; According to the adjustment logic and the adjustment item, determine the replacement item, and replace the adjustment item in the initial fusion scheme with the replacement item to obtain a transition scheme; Fuse based on the transition scheme and pre-test to determine the transition scheme with the highest effectiveness evaluation as the target feedback scheme; Perform intelligent feedback according to the target feedback scheme.

2. A method of digitally controlling a pultrusion apparatus as claimed in claim 1, wherein, Step 1: Obtain the pultrusion process parameters displayed by the preset digital display device on the pultrusion equipment, including: Obtain the detection element associated with the display of the digital display device; Obtain the detection channel of the detection element; According to the detection data of the detection channel, determine the pultrusion process parameters.

3. A method of digitally controlling a pultrusion apparatus as defined in claim 1, wherein, Step 2: Obtain the pultrusion equipment type of the pultrusion equipment, including: Obtain the operation parameters of the pultrusion equipment in the history of the preset pultrusion equipment type; According to the operation parameters, determine the operation feature group of the pultrusion equipment of each preset pultrusion equipment type; Compare the operation feature group with the real-time operation feature group of the pultrusion equipment to obtain the preset pultrusion equipment type that meets the comparison as the pultrusion equipment type of the pultrusion equipment.

4. A method of digitally controlling a pultrusion apparatus as defined in claim 1, wherein, Step 3: Train the digital control model, including: The model training set is obtained, the model training set is split according to the source device category of the training data in the model training set, and the model sub-training set is obtained; According to the source device category, the control channel guide label is generated, and at the same time, the data-driven model is trained according to the model sub-training set; An association relationship between the control channel guide label and the data-driven model is established; Based on the model predictive control algorithm, the digital control model is generated according to the association relationship.

5. A method of digitally controlling a pultrusion apparatus as defined in claim 4, wherein, Obtain the model training set, including: Based on the industrial control record sharing platform, obtain the preselected target data; Determine the reference control information of the preselected target data; Obtain the target control information; According to the reference control information and the target control information, calculate the control attribute similarity; Obtain the verification data of the preselected target data; According to the verification data, determine the verification item and the verification result, and obtain the true value based on the verification item and the verification result; According to the control attribute similarity and the true value, the preselected target data that meets the screening condition in the preselected target data is screened out, and the model training set is obtained by summarizing.

6. A method of digitally controlling a pultrusion apparatus as defined in claim 1, wherein, Step 4: Based on the digital control model, according to the pultrusion equipment category and the pultrusion process parameters, digital control is performed, including: According to the pultrusion equipment category, determine the target control channel guide label; Input the parameter characteristics of the pultrusion process parameters into the data-driven model associated with the target control channel guide label to perform digital control.

7. A digitally controlled system for a pultrusion apparatus, characterized by, It includes: The pultrusion process parameter acquisition subsystem is used to obtain the pultrusion process parameters displayed by the preset digital display device of the pultrusion equipment; The pultrusion equipment category acquisition subsystem is used to obtain the pultrusion equipment category of the pultrusion equipment, including: track type, hydraulic type and electric cylinder type; The training subsystem is used to train the digital control model; The digital control subsystem is used to perform digital control based on the digital control model according to the pultrusion equipment category and the pultrusion process parameters; The intelligent feedback subsystem is used to obtain the pultrusion operation image in real time during the digital control process, and perform intelligent feedback according to the pultrusion operation image; The intelligent feedback subsystem performs the following operations: Obtain the equipment operation image of the digital control process; Analyze the equipment operation image to obtain the field thermal imaging temperature distribution; Determine whether the field thermal imaging temperature distribution meets the target thermal imaging temperature distribution of the pultrusion operation; If it meets, determine the distribution part of the field thermal imaging temperature distribution that meets the target thermal imaging temperature distribution; According to the extraction relationship between the distribution part and the equipment operation image, determine the pultrusion operation image; Extract the material feature set of the pultrusion target in the pultrusion operation image; According to the feature type of the material feature in the material feature set, determine the feature representation of the material feature corresponding to the feature type, and the feature type includes: solidification state, uniformity and flowability; According to the feature representation, determine the intelligent feedback action; Obtain the action intervention between the intelligent feedback actions; According to the action intervention between the intelligent feedback actions, fuse the intelligent feedback actions to obtain the pre-feedback scheme; According to the pre-feedback scheme, perform the pre-test of intelligent feedback to obtain the pre-test result; Evaluate the effectiveness of the pre-test result to obtain the effectiveness evaluation result; Determine whether the effectiveness evaluation result meets the effectiveness evaluation requirement; If not, collect invalid data for feedback failure analysis, and obtain feedback failure analysis results; Based on the feedback failure analysis results, guide the fusion of intelligent feedback actions, and obtain the target feedback scheme; Among them, based on the feedback failure analysis results, guiding the fusion of intelligent feedback actions includes: According to the feedback failure analysis results, determine the adjustment item in the initial fusion scheme; According to the adjustment item, determine the current fusion logic; According to the feedback failure analysis results and the current fusion logic, determine the adjustment logic; According to the adjustment logic and the adjustment item, determine the replacement item, and replace the adjustment item in the initial fusion scheme with the replacement item to obtain a transition scheme; Based on the transition scheme, perform fusion and pre-test, and determine the transition scheme with the highest effectiveness evaluation as the target feedback scheme; According to the target feedback scheme, intelligent feedback is performed.

Images