Mold machining precision control method, device and equipment and storage medium

Abstract

The invention discloses a mold machining precision control method, device and equipment and a storage medium, and the method comprises the steps that for a plurality of to-be-machined preset feature faces of a to-be-machined part, tolerances and initial programming margins are obtained, and the tolerances and / or the initial programming margins of all the feature faces are not completely the same; in the process of machining any feature surface, the state deviation of the feature surface is periodically detected, when it is detected that the state deviation is larger than a deviation threshold value, a PID algorithm is called to compensate and update the current programming allowance, and the larger the value of the state deviation is, the larger the compensation amount for the current programming allowance is; and the current machining parameters used for machining the feature face are adjusted, and the feature face continues to be machined in combination with the compensated current programming allowance and the adjusted machining parameters. According to the mold machining precision control method and device, the equipment and the storage medium, the problem that the machining precision of an existing machining mode cannot be guaranteed is solved, and the technical effect of effectively improving the machining precision of the feature surface is achieved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of mold processing, and particularly to a method, device, equipment, and storage medium for controlling the machining accuracy of molds. Background Art

[0002] Mold processing refers to the process of manufacturing forming tools (such as stamping dies, injection molds, etc.) and blanking tools, which usually consists of an upper mold and a lower mold, and the material is formed by a press, and is widely used in fields such as electronic connectors and automotive parts.

[0003] In some technologies, during mold processing, it usually relies on manual experience, and a unified processing standard is adopted for all the feature surfaces of the parts to be processed. However, the above processing method cannot process according to the feature surfaces with different precision requirements on the parts to be processed, which easily leads to waste of resources in high-precision areas, and its processing accuracy cannot be guaranteed. Summary of the Invention

[0004] The purpose of the present invention is to provide at least one method, device, equipment, and storage medium for controlling the machining accuracy of molds, which can at least solve the technical problem that the above processing method cannot process according to the feature surfaces with different precision requirements on the parts to be processed, easily leads to waste of resources in high-precision areas, and its processing accuracy cannot be guaranteed, and can at least achieve the following: by performing different-precision processing on different feature surfaces to achieve precision grading control, and by using the PID algorithm to compensate and update the current programming allowance for dynamic compensation, and then continuing to process the feature surfaces with the adjusted processing parameters, the technical effect of effectively improving the machining accuracy of the feature surfaces is achieved.

[0005] To solve the above technical problems, at least one embodiment of the present application provides a method for controlling the machining accuracy of molds, including: For a plurality of preset feature surfaces to be machined on a part to be machined, obtain the tolerance and initial programming allowance corresponding to each of the feature surfaces, wherein the tolerances and / or initial programming allowances of the feature surfaces are not all the same; During the process of machining any one of the feature surfaces, periodically detect the state deviation of the feature surface. When it is detected that the state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming allowance, and the greater the value of the state deviation, the greater the compensation amount for the current programming allowance; Adjust the processing parameters currently used for machining the feature surface, and continue to machine the feature surface in combination with the compensated current programming allowance and the adjusted processing parameters; The step of, when it is detected that the state deviation is greater than the deviation threshold, calling the PID algorithm to compensate and update the current programming allowance includes: When it is detected that the state deviation is greater than the deviation threshold, based on the state deviation, calculate the compensation amount for the current programming margin through the PID algorithm; Superimpose the compensation amount on the current programming margin to obtain the compensated current programming margin.

[0006] At least one embodiment of the present application further provides a mold processing precision control device, including: An acquisition module, configured to acquire the tolerance and the initial programming margin corresponding to each of a plurality of preset feature surfaces to be machined for a component to be machined, wherein the tolerances and / or the initial programming margins of the respective feature surfaces are not completely the same; A compensation module, configured to periodically detect the state deviation of a feature surface during the machining of any one of the feature surfaces. When it is detected that the state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming margin. The greater the value of the state deviation, the greater the compensation amount for the current programming margin; A first adjustment module, configured to adjust the machining parameters currently used for machining the feature surface, and continue to machine the feature surface in combination with the compensated current programming margin and the adjusted machining parameters.

[0007] At least one embodiment of the present application further provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned mold processing precision control method.

[0008] At least one embodiment of the present application further provides a computer-readable storage medium, storing a computer program, and when the computer program is executed by a processor, the above-mentioned mold processing precision control method is implemented.

[0009] The mold processing accuracy control method, device, equipment, and storage medium provided by the embodiments of the present application, for multiple preset feature surfaces to be processed on a component to be processed, obtain the tolerance and initial programming allowance corresponding to each feature surface. During the process of processing any feature surface, the state deviation of the feature surface is periodically detected. When it is detected that the state deviation is greater than the deviation threshold, the PID algorithm is called to compensate and update the current programming allowance. Specifically, when it is detected that the state deviation is greater than the deviation threshold, based on the state deviation, the compensation amount for the current programming allowance is calculated through the PID algorithm; the compensation amount is superimposed on the current programming allowance to obtain the compensated current programming allowance. Then, the processing parameters currently used to process the feature surface are adjusted, and in combination with the compensated current programming allowance and the adjusted processing parameters, the feature surface is continuously processed. In this way, by performing processing with different precisions on different feature surfaces, precision hierarchical control is achieved. Moreover, by compensating and updating the current programming allowance through the PID algorithm for dynamic compensation, and then continuing to process the feature surface in combination with the adjusted processing parameters, the processing accuracy of the feature surface is effectively improved.

[0010] In some alternative embodiments, the method further includes: Construct a mapping table including different feature surfaces, as well as the feature identifiers and tolerances corresponding to each feature surface; Set the initial programming allowance corresponding to each of the feature surfaces, and construct a mapping table including different feature surfaces, as well as the initial programming allowances corresponding to each feature surface; The obtaining of the tolerance and initial programming allowance corresponding to each of the feature surfaces includes: Based on the feature identifiers predefined for the respective feature surfaces in the software model library, match the tolerance and initial programming allowance corresponding to each of the feature surfaces from the mapping table.

[0011] In this way, the tolerance and initial programming allowance corresponding to the feature surface can be quickly obtained, and the dependence on manual experience can be effectively reduced.

[0012] In some alternative embodiments, the state deviation of the feature surface is calculated through the following steps: During the process of processing the feature surface, the dimensional deviation, temperature deviation, and stress deviation of the feature surface are periodically obtained; Perform fusion calculation on the dimensional deviation, temperature deviation, and stress deviation obtained within the same period to obtain the state deviation.

[0013] In this way, through the fusion of data transmitted by sensors at different positions, multi-sensor data fusion is achieved, making the resulting state deviation more accurate.

[0014] In some alternative embodiments, the adjustment of the current machining parameters for machining the feature surface includes: Input the feature parameters of the currently machined part and the mold for machining the part into a pre - constructed machining prediction model for prediction, obtain the predicted machining parameters that match the feature parameters of the currently machined part and the mold for machining the part, and use the predicted machining parameters as the adjusted machining parameters.

[0015] In this way, by dynamically adjusting the machining parameters for different feature surfaces, the adjusted machining parameters are more suitable for the feature surface in the current machining process. Continuing to machine the feature surface based on the adjusted machining parameters can effectively improve the machining accuracy of the feature surface.

[0016] In some alternative embodiments, the feature parameters of the currently machined part include the material hardness of the part, the feature parameters of the mold for machining the part include the wear degree and vibration frequency of the cutting tool on the mold, and the machining parameters include the spindle speed and feed rate of the machining machine tool for controlling the cutting tool; The machining prediction model is obtained through the following steps: Construct an original model; Using the feature parameter samples of the part in multiple periods in the historical machining state, the feature parameter samples of the mold for machining the part, and the corresponding machining parameter samples obtained, with the qualification rate of the machined finished products corresponding to the samples in each period as a constraint, use the reinforcement learning algorithm to train the original model to obtain the machining prediction model.

[0017] In this way, by training the original model with sample data constrained by the qualification rate, the prediction accuracy of the obtained machining prediction model is improved.

[0018] In some alternative embodiments, the method further includes: When it is detected that the state deviation is less than or equal to the deviation threshold, directly adjust the current machining parameters for machining the feature surface, and continue to machine the feature surface based on the adjusted machining parameters.

[0019] In this way, continuing to machine the feature surface according to the adjusted machining parameters can effectively improve the machining accuracy of the feature surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] One or more embodiments are exemplarily illustrated by the pictures in the corresponding drawings. These exemplary illustrations do not limit the embodiments.

[0021] Figure 1It is a schematic flow chart of a method for controlling the machining accuracy of a mold provided by an embodiment of the present application; Figure 2 It is a schematic flow chart of a device for controlling the machining accuracy of a mold provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of an electronic device provided by another embodiment of the present application; Figure 4 It is a schematic flow chart of a method for controlling the machining accuracy of a mold provided by an embodiment of the present application; Figure 5 It is a schematic diagram of a UG shortcut key query interface provided by an embodiment of the present application; Figure 6 It is a schematic diagram of the relationship between a feature surface and a tolerance provided by another embodiment of the present application; Figure 7 It is a schematic diagram of the positions of a laser sensor and a temperature sensor provided by another embodiment of the present application. Detailed implementation manners

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the embodiments of the present application will be elaborated in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that in the embodiments of the present application, many technical details are presented for the readers to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation on the specific implementation manner of the present application. Each embodiment can be combined and cross-referenced with each other without conflict.

[0023] To facilitate the understanding of the embodiments of the present application, the relevant content regarding the method for controlling the machining accuracy of a mold is introduced here first.

[0024] Mold machining refers to the process of manufacturing forming tools (such as stamping dies, injection molds, etc.) and blanking tools, which usually consists of an upper mold and a lower mold, and the material is formed by a press, and is widely used in fields such as electronic connectors and automotive parts.

[0025] In some technologies, during mold machining, it usually relies on manual experience and adopts a unified machining standard for all the feature surfaces of the parts to be machined. However, the above machining method cannot machine according to the feature surfaces with different precision requirements on the parts to be machined, which is likely to cause waste of resources in high-precision areas, and its machining accuracy cannot be guaranteed.

[0026] In order to solve the technical problems that the existing processing methods cannot process according to the characteristic surfaces with different precision requirements on the mold, which easily leads to waste of resources in high-precision areas and the processing precision cannot be guaranteed, the present invention proposes a method for controlling the processing precision of the mold. The implementation details of the method for controlling the processing precision of the mold in this embodiment will be specifically described below. The following content is only the implementation details provided for convenient understanding and is not necessary for implementing this solution.

[0027] Embodiment 1: The method for controlling the processing precision of the mold in this embodiment can be applied to an electronic device with communication, computing, and data storage capabilities. Its specific process can be as Figure 1 shown and includes: Step 101, for multiple preset characteristic surfaces to be processed on the part to be processed, obtain the tolerance and initial programming allowance corresponding to each characteristic surface, where the tolerances and / or initial programming allowances of each characteristic surface are not completely the same.

[0028] Specifically, the part to be processed has multiple types of characteristic surfaces, such as clearance, parting surface, plastic position, exhaust, groove, hanging platform and other characteristic surfaces, as shown in Table 1 below. Each characteristic surface corresponds to a tolerance and an initial programming allowance. In addition, since the required processing precisions of each characteristic surface are not completely the same, the tolerances and / or initial programming allowances of each characteristic surface are not completely the same. Among them, the programming allowance refers to a certain amount added on the basis of the theoretical processing size during processing to ensure that the processed part meets the required size and quality requirements.

[0029] Table 1 Characteristic Surface - Tolerance Comparison Table

[0030] Step 102, during the process of processing any characteristic surface, periodically detect the state deviation of the characteristic surface. When the detected state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming allowance. The greater the value of the state deviation, the greater the compensation amount for the current programming allowance.

[0031] Specifically, during the process of processing the characteristic surface, the state of the characteristic surface is periodically monitored to obtain the state deviation. Among them, the state of the characteristic surface includes the dimension state, temperature state, and stress state, which can be monitored by installing sensors. In addition, the state deviation refers to the difference between the actual state value and the preset standard value. When the detected state deviation is greater than the deviation threshold, it indicates that the processing precision of the characteristic surface of the part is poor at this time, and it is necessary to call the PID algorithm to compensate and update the current programming allowance.

[0032] Specifically, when the detected state deviation is greater than the deviation threshold, calling the PID algorithm to compensate and update the current programming allowance includes: When the detected state deviation is greater than the deviation threshold, based on the state deviation, calculate the compensation amount for the current programming margin through the PID algorithm; Superimpose the compensation amount on the current programming margin to obtain the compensated current programming margin.

[0033] Specifically, when the detected state deviation is greater than the deviation threshold, it indicates that the machining accuracy of the feature surface of the component is poor at this time. Then, based on the state deviation, calculate the compensation amount for the current programming margin through the PID algorithm. Specifically, using the PID control algorithm, the relational expression of the control signal is: ; In the formula, is the proportional coefficient of the PID controller, is the integral coefficient of the PID controller, is the differential coefficient of the PID controller. Based on the state deviation, calculate the compensation amount for the current programming margin through the PID algorithm, superimpose the compensation amount on the current programming margin to obtain the compensated current programming margin. In this way, by compensating and updating the current programming margin for dynamic compensation, and then continuing to machine the feature surface in combination with the adjusted machining parameters, the machining accuracy of the feature surface can be effectively improved.

[0034] In some examples, the state deviation Δ>0.005mm. Through error calculation: e(t) = set value - measured value of the sensor (such as e(t) = 0.01mm), integral term calculation: = cumulative error value in the past 30 seconds (such as 0.15mm·s), differential term calculation: =(current deviation - deviation of the previous second) / 1s (such as 0.005mm / s).

[0035] Output the compensation amount: u(t) = 0.8 * 0.01 + 0.2 * 0.15 + 0.1 * 0.005 = 0.008 + 0.03 + 0.0005 = 0.0385mm. At this time, the compensation amount is 0.0385mm. Superimpose the compensation amount on the current programming margin, that is, adjust the program margin from -0.005mm to -0.005 - 0.0385 = -0.0435mm. It should be noted here that the "-" sign indicates that the tool moves in the direction close to the component to be machined.

[0036] In some examples, the coefficients can be set as: = 0.8 (compensate for tool wear), = 0.2 (eliminate steady-state error), = 0.1 (suppress sudden vibration). When the detected deviation e(t) = 0.01 mm, the system automatically calculates the compensation amount u(t) = 0.8 * 0.01 + 0.2 * (cumulative error integral) + 0.1 * (error change rate), and finally adjusts the margin from -0.005 mm to -0.008 mm.

[0037] Step 103, adjust the current machining parameters for machining the feature surface, and continue to machine the feature surface in combination with the compensated current programming margin and the adjusted machining parameters.

[0038] Specifically, the machining parameters refer to the parameters of the machining tool on the mold used to control the machining machine when machining the part to be machined. At this time, the machining parameters are not necessarily the optimal machining parameters for machining the feature surface. Therefore, it is necessary to adjust the current machining parameters for machining the feature surface. In this embodiment, after calling the PID algorithm to compensate and update the current programming margin, the compensated current programming margin is obtained, and the compensated current programming margin is combined with the adjusted machining parameters to continue machining the feature surface.

[0039] In this embodiment, for multiple preset feature surfaces to be machined on the part to be machined, the tolerance and initial programming margin corresponding to each feature surface are obtained. During the machining process of any feature surface, the state deviation of the feature surface is periodically detected. When the detected state deviation is greater than the deviation threshold, the PID algorithm is called to compensate and update the current programming margin. Specifically, when the detected state deviation is greater than the deviation threshold, based on the state deviation, the compensation amount for the current programming margin is calculated by the PID algorithm; the compensation amount is superimposed on the current programming margin to obtain the compensated current programming margin. Then, the current machining parameters for machining the feature surface are adjusted, and the feature surface is continuously machined in combination with the compensated current programming margin and the adjusted machining parameters. In this way, by machining different feature surfaces with different precisions, precision grading control is achieved. Moreover, by compensating and updating the current programming margin through the PID algorithm for dynamic compensation, and then continuing to machine the feature surface in combination with the adjusted machining parameters, the machining precision of the feature surface is effectively improved.

[0040] In some embodiments, the method further includes: Construct a mapping table including different feature surfaces, and the feature identifier and tolerance corresponding to each feature surface; Set the initial programming margin corresponding to each feature surface, and construct a mapping table including different feature surfaces and the initial programming margin corresponding to each feature surface; Obtain the tolerance and initial programming margin corresponding to each feature surface, including: Match the tolerance and initial programming margin corresponding to each feature surface from the mapping table based on the feature identifier predefined for each feature surface in the software model library.

[0041] Specifically, the feature identifiers include colors, letters, numbers, etc. Different feature surfaces can be identified using different colors, letters, and numbers. In some examples, distinct colors can be preferentially selected over letters and numbers to identify the feature surfaces. By establishing a mapping relationship among the feature surface - feature identifier - tolerance through a mapping table, it is possible to quickly match the feature surface corresponding to a color and the tolerance corresponding to the feature surface. That is, when machining a feature surface of a certain color, the corresponding feature surface and the corresponding tolerance can be quickly matched according to the color.

[0042] In this embodiment, each feature surface also corresponds to an initial programming allowance. Through a pre - constructed mapping table of feature surface - initial programming allowance, the initial programming allowance corresponding to each feature surface can be quickly obtained. Before the machining process, based on the feature identifiers predefined for each feature surface in the software model library, the tolerance and the initial programming allowance corresponding to each feature surface are matched from the above - mentioned pre - constructed mapping table. In this way, the tolerance and the initial programming allowance corresponding to the feature surface can be quickly obtained, effectively reducing the dependence on manual experience.

[0043] In some embodiments, the state deviation of the feature surface is calculated through the following steps: During the process of machining the feature surface, the size deviation, temperature deviation, and stress deviation of the feature surface are periodically obtained; The size deviation, temperature deviation, and stress deviation obtained within the same period are fused and calculated to obtain the state deviation.

[0044] Specifically, as Figure 7 shown, by installing a laser sensor on the spindle side of the machining tool - controlling machine tool to monitor the actual size, the size deviation can be obtained; by installing a temperature sensor outside the spindle of the machining tool - controlling machine tool to monitor the actual temperature, the temperature deviation can be obtained; by installing a force sensor on the fixture of the machining tool - controlling machine tool to monitor the actual clamping stress, the stress deviation can be obtained; where the deviation is the difference between the actual value and the preset standard value.

[0045] In this embodiment, after obtaining the size deviation, temperature deviation, and stress deviation of the feature surface, weight distribution is performed on the size deviation, temperature deviation, and stress deviation obtained within the same period. For example, the size deviation weight is 60% (accuracy ±0.005 mm), the temperature deviation weight is 30% (compensating for thermal expansion), and the stress deviation weight is 10% (detecting clamping stress). Then, the size deviation, temperature deviation, and stress deviation after weight distribution are fused and calculated to obtain the state deviation. The following is the fusion formula: ; The fusion value obtained by the above formula is the state deviation. When it is detected that the state deviation is greater than the deviation threshold, the PID algorithm is called to compensate and update the current programming margin. In this way, by fusing the data transmitted by sensors at different positions, multi-sensor data fusion is achieved, making the obtained state deviation more accurate.

[0046] In some embodiments, the machining parameters for machining the feature surface are adjusted, including: Input the feature parameters of the currently machined part and the mold for machining the part into a pre-constructed machining prediction model for prediction, obtain the predicted machining parameters that match the feature parameters of the currently machined part and the mold for machining the part, and use the predicted machining parameters as the adjusted machining parameters.

[0047] Specifically, the feature parameters of the currently machined part and the mold for machining the part include: the material hardness of the part to be machined, the wear degree and vibration frequency of the cutting tool on the mold. The machining prediction model is used to output machining parameters that are more suitable for the subsequent continuous machining of the feature surface in the current machining process. It can be understood that different feature surfaces correspond to different initial machining parameters, and the initial machining parameters can be understood as theoretical machining parameters or historical machining parameters. Input the feature parameters of the currently machined part and the mold for machining the part into the machining prediction model for prediction, obtain the predicted machining parameters that match the feature parameters of the currently machined part and the mold for machining the part. At this time, the predicted machining parameters are the machining parameters that are more suitable for the subsequent continuous machining of the feature surface in the current machining process. After obtaining the predicted machining parameters, use the predicted machining parameters as the adjusted machining parameters. In this way, by dynamically adjusting the machining parameters for different feature surfaces, the adjusted machining parameters are more suitable for the feature surface in the current machining process, and based on the adjusted machining parameters, the feature surface is continuously machined, which can effectively improve the machining accuracy of the feature surface.

[0048] In some embodiments, the feature parameters of the currently machined part include the material hardness of the part, and the feature parameters of the mold for machining the part include the wear degree and vibration frequency of the cutting tool on the mold. The machining parameters include the spindle speed and feed rate of the machining machine tool for controlling the cutting tool; The machining prediction model is obtained through the following steps: Construct an original model; Using the qualified rate of the machined products corresponding to the samples in each period as a constraint, for the obtained feature parameter samples of parts in multiple periods in the historical machining state, feature parameter samples of the molds for machining the parts, and the corresponding machining parameter samples, use the reinforcement learning algorithm to train the original model to obtain the machining prediction model.

[0049] Specifically, by collecting the characteristic parameter samples of components at multiple time periods in the historical processing state, the characteristic parameter samples of the molds used for processing the components, and the corresponding processing parameter samples as training sample data, a reinforcement learning algorithm is used to train the original model, thereby obtaining a processing prediction model. Among them, the historical training sample data needs to be constrained by the qualified rate of the processed products corresponding to the samples in each time period. That is, the characteristic parameters, the characteristic parameters of the molds, and the corresponding processing parameter samples corresponding to the processed products with a high qualified rate are used to train the original model using the reinforcement learning algorithm. In this way, by training the original model with the sample data constrained by the qualified rate, the prediction accuracy of the obtained processing prediction model is improved.

[0050] In some examples, the model training data: 1000 sets of historical processing records (including the wear degree and vibration frequency of the tool, the relationship between the material hardness of the component and the qualified rate); after obtaining the historical data, perform feature standardization on the data: the tool wear range is 0~0.5mm, and it is normalized to 0~1, the material hardness is HRC20~60, and it is normalized to 0~1. When making a prediction, input the current values: tool wear degree = 0.3mm (normalized 0.6), vibration frequency = 150Hz (normalized 0.5), material hardness = HRC50 (normalized 0.75). Calculate the feed rate through the following relational expression: Y = 0.8 * tool wear degree + 0.15 * material hardness + 0.05 * vibration frequency + b (bias term), where 0.8, 0.15, and 0.05 in the formula are all coefficients and can be set by oneself. For example, Y = 0.8 * 0.6 + 0.15 * 0.75 + 0.05 * 0.5 = 0.48 + 0.1125 + 0.025 = 0.6175. Then perform anti-normalization on the obtained result: Y = 0.6175, and the corresponding feed rate = 400 + (600 - 400) * 0.6175 = 523.5mm / min. It should be noted that the spindle speed can also be predicted according to the above steps.

[0051] In some examples, a training sample library is constructed, where the training sample library is used to store the characteristic parameter samples of components, the characteristic parameter samples of the molds used for processing the components, and the corresponding processing parameter samples; After adjusting the current processing parameters for machining the feature surface and continuing to machine the feature surface in combination with the compensated current programming allowance and the adjusted processing parameters, it further includes: Adding the characteristic parameters of the component corresponding to the predicted feature surface, the characteristic parameters of the mold used for processing the component, and the corresponding processing parameters to the training sample library to provide the samples required for the training of the original model.

[0052] In some embodiments, the method further includes: When the detected state deviation is less than or equal to the deviation threshold, directly adjust the current machining parameters for machining the feature surface, and continue machining the feature surface based on the adjusted machining parameters.

[0053] Specifically, when the detected state deviation is less than or equal to the deviation threshold, it indicates that the machining accuracy of the feature surface of the component is good at this time, so there is no need to perform dynamic compensation on the state deviation. Therefore, directly adjust the current machining parameters for machining the feature surface, and continue machining the feature surface based on the adjusted machining parameters. Among them, the steps for adjusting the machining parameters are the same as those taken when the state deviation is greater than the deviation threshold, which will not be elaborated here. In this way, continuing to machine the feature surface according to the adjusted machining parameters can effectively improve the machining accuracy of the feature surface.

[0054] In some examples, the initial machining parameters can be: red area (high-precision surface): spindle speed 8000 rpm, feed rate 500 mm / min, programmed allowance -0.01 mm; orange area (low-precision surface): spindle speed 3000 rpm, feed rate 1500 mm / min, programmed allowance 0.0 mm. Assume that the current is the red area of the high-precision surface. When the detected state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programmed allowance to obtain the compensated current programmed allowance. Replace the programmed allowance of the initial machining parameters with the compensated current programmed allowance, and update and replace the spindle speed and / or feed rate of the initial machining parameters with the adjusted machining parameters obtained through the machining prediction model. When the detected state deviation is less than or equal to the deviation threshold, directly update and replace the spindle speed and / or feed rate of the initial machining parameters with the adjusted machining parameters obtained through the machining prediction model, and the programmed allowance in the initial machining parameters remains unchanged.

[0055] Embodiment 2: As Figure 4 shown, this embodiment provides an exemplary content of Embodiment 1, that is, provides an exemplary process of the mold machining accuracy control method. The specific content includes: Step 1: Feature surface classification and dynamic coding - functional surface division: As shown in Table 1, the mold is divided into 10 types of feature surfaces (such as clearance, parting surface, plastic position, exhaust, groove, etc.), and dynamic tolerance rules are defined (such as when the temperature rises by 10 degrees Celsius, the cavity surface tolerance tightens by 0.001 mm). As Figure 6 shown, there is a corresponding relationship between the feature surface and the tolerance. For feature surfaces such as the sealing surface, a corresponding tolerance range (0~0.01) is set, and for feature surfaces such as the high-precision surface, a corresponding tolerance range (-0.01~+0.01) is set, etc.

[0056] UG Coloring Operation: Use UG / Open API to color the feature surface, and write the color ID into the model property library.

[0057] Step 2: UG Plug-in Development and Real-time Monitoring - Plug-in Function: As Figure 5 shown, set a shortcut key 1 for the color ID, shortcut key "Ctrl+Shift+C": after selecting the surface, it can be as Figure 5 shown in the right part to display tolerance, roughness and recommended tool (such as recommending a φ2mm carbide milling cutter in the red area).

[0058] Dynamic Compensation Logic: When the detected status deviation Δ>0.005mm, call the PID formula to adjust the allowance: PID Dynamic Compensation Control When the real-time status deviation of the feature surface exceeds the threshold: 1) Error calculation: e(t) = set value - measured value by the sensor (such as e(t)=0.01mm); 2) Integral term calculation: = Cumulative error value in the past 30 seconds (such as 0.15mm·s); 3) Differential term calculation: de(t) / dt = (current error - error of the previous second) / 1s (such as 0.005mm / s); 4) Output compensation amount: u(t) = 0.8*0.01 + 0.2*0.15 + 0.1*0.005 = 0.008 + 0.03 + 0.0005 = 0.0385mm; 5) Execute action: Adjust the program allowance from -0.005mm to -0.005 - 0.0385 = -0.0435mm.

[0059] Step 3: Machining Execution and AI Optimization - CAM Strategy Matching Such As: Red area (high-precision surface): Spindle speed 8000rpm, feed 500mm / min, program allowance -0.01mm; Orange area (low-precision surface): Spindle speed 3000rpm, feed 1500mm / min, allowance 0.0mm.

[0060] AI Parameter Prediction: Train the model based on historical data, input (HRC45), and output the optimal feed speed of 480mm / min.

[0061] 1) Model training data: 1000 groups of historical machining records (including the relationship between tool wear, material hardness, vibration frequency and qualification rate); 2) Feature standardization: Tool wear range 0~0.5mm, normalized to 0~1, material hardness HRC20~60, normalized to 0~1; 3) Prediction application: Input current values: tool wear = 0.3 mm (normalized to 0.6), material hardness = HRC50 (normalized to 0.75), vibration frequency = 150 Hz (normalized to 0.5); Calculation: Y = 0.8 * 0.6 + 0.15 * 0.75 + 0.05 * 0.5 = 0.48 + 0.1125 + 0.025 = 0.6175; Denormalization: Y = 0.6175, then the corresponding feed rate = 400 + (600 - 400) * 0.6175 = 523.5 mm / min.

[0062] Step 4: Closed-loop feedback and self-learning - CMM data feedback: The CMM inspection data automatically updates the tolerance mapping table. For example, if a certain guiding surface is qualified 5 times continuously, the tolerance is relaxed from ±0.01 mm to ±0.015 mm.

[0063] When machining multi-cavity molds, the main cavity (red) is machined first, and the spare cavity (yellow) is machined later, increasing the resource utilization rate by 20%.

[0064] In this way, it can break through the limitations of static color coding and form a technical closed-loop through dynamic compensation and AI prediction. Clearly defining the sensor deployment, data fusion formula, and PID parameter setting rules, it has strong feasibility. The machining efficiency is increased by 35%, and the new employee training cycle is shortened to 1 week. Table 2 below is the actual comparison table of the effects: Table 2 Actual comparison table of the effects

[0065] In this embodiment, there are the following innovation points: 1. Dynamic color coding system: Feature surface - color - tolerance dynamic mapping table (ID62 light green → high-precision surface, tolerance fluctuates with temperature by ±0.01 mm). Technical feature: The color ID is deeply integrated with the UG / CAM software, supporting real-time tolerance adjustment.

[0066] 2. Multi-sensor data fusion: Sensor deployment: Laser displacement sensor (on the spindle side), temperature sensor (on the spindle side), force sensor (on the fixture).

[0067] Fusion formula: ; Parameter meanings: The weight of the laser sensor is 60% (accuracy ±0.005 mm), the temperature sensor is 30% (compensating for thermal expansion), and the force sensor is 10% (detecting clamping stress); Application scenario: When X laser = +0.01 mm (out of tolerance), X temperature = +5 °C (thermal expansion 0.003 mm), X force = 200 N (stress deformation 0.001 mm), the fusion value X = 0.6 * 0.01 + 0.3 * 0.003 + 0.1 * 0.001 = 0.0069 mm, triggering the compensation mechanism.

[0068] 3. AI-assisted decision-making model: Input features: Tool wear degree, material hardness, real-time vibration frequency.

[0069] Predicted machining parameter Y = 0.8 * tool wear degree + 0.15 * material hardness + 0.05 * vibration frequency + b (bias term); Input example: Tool wear degree = 0.2 mm, material hardness = HRC45, vibration frequency = 120 Hz, calculated Y = 0.8 * 0.2 + 0.15 * 45 + 0.05 * 120 = 0.16 + 6.75 + 6 = 12.91 → Output feed rate 480 mm / min.

[0070] The effects that can be achieved in this embodiment include: Precision grading control: Through color coding (e.g., red → cavity surface, tolerance ±0.01 mm; colorless → clearance, tolerance ±0.1 mm), "precision machining in high-precision areas and rough machining in low-precision areas" is realized, and the machining efficiency is increased by 35%.

[0071] Dynamic tolerance compensation: Integrate a laser sensor to monitor dimensional deviations in real time, and automatically correct the program allowance through the PID control algorithm (e.g., the allowance is adjusted from -0.005 mm to -0.008 mm), and the qualified rate after compensation is increased from 82% to 98%.

[0072] Using the PID control algorithm, the control signal is calculated as: ; Parameter settings: = 0.8 (compensate for tool wear), = 0.2 (eliminate steady-state error), = 0.1 (suppress sudden vibrations).

[0073] Application example: When the laser sensor detects a dimensional deviation e(t) = 0.01 mm on the cavity surface, the system automatically calculates the compensation amount u(t) = 0.8 * 0.01 + 0.2 * (cumulative error integral) + 0.1 * (error change rate), and finally adjusts the allowance from -0.005 mm to -0.008 mm. Table 3 below is the experimental data table: Table 3 Experimental data table

[0074] AI Parameter Prediction: Train a deep learning model (processing prediction model) based on historical data to predict the optimal spindle speed and feed rate, reducing the programming time by 50%.

[0075] Embodiment 3: Another embodiment of the present application relates to a mold processing precision control device. The implementation details of the mold processing precision control device in this embodiment will be specifically described below. The following content is only the implementation details provided for convenience of understanding and is not necessary for implementing this solution. The schematic diagram of the mold processing precision control device in this embodiment can be as Figure 2 shown, including: An acquisition module 201, configured to acquire the tolerance and initial programming allowance corresponding to each of a plurality of preset feature surfaces to be machined for a part to be machined, wherein the tolerances and / or initial programming allowances of the respective feature surfaces are not completely the same; A compensation module 202, configured to periodically detect the state deviation of a feature surface during the machining of any feature surface. When the detected state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming allowance. The greater the value of the state deviation, the greater the compensation amount for the current programming allowance; A first adjustment module 203, configured to adjust the machining parameters currently used for machining the feature surface, and continue to machine the feature surface in combination with the compensated current programming allowance and the adjusted machining parameters.

[0076] In some embodiments, the compensation module 202 includes: A first acquisition unit, configured to periodically acquire the dimensional deviation, temperature deviation, and stress deviation of the feature surface during the machining of the feature surface; A fusion unit, configured to perform a fusion calculation on the dimensional deviation, temperature deviation, and stress deviation acquired within the same period to obtain the state deviation.

[0077] In some embodiments, the mold processing precision control device includes: A first construction module, configured to construct a mapping table including different feature surfaces, as well as the feature identifiers and tolerances corresponding to each feature surface; A second construction module, configured to set the initial programming allowance corresponding to each feature surface, and construct a mapping table including different feature surfaces, as well as the initial programming allowances corresponding to each feature surface; The acquisition module 201 is further configured to match the tolerance and initial programming allowance corresponding to each feature surface from the mapping table based on the feature identifier defined in the software model library in advance for each feature surface.

[0078] In some embodiments, the compensation module 202 further includes: A calculation unit, configured to calculate a compensation amount for the current programming margin based on a state deviation through a PID algorithm when it is detected that the state deviation is greater than a deviation threshold. An overlay unit, configured to overlay the compensation amount with the current programming margin to obtain the compensated current programming margin.

[0079] In some embodiments, the first adjustment module 203 is further configured to input the characteristic parameters of the currently processed part and the mold for processing the part into a pre-constructed processing prediction model for prediction, obtain predicted processing parameters that match the characteristic parameters of the currently processed part and the mold for processing the part, and use the predicted processing parameters as the adjusted processing parameters.

[0080] In some embodiments, the mold processing precision control device further includes: A model construction module, configured to construct an original model; and perform training on the original model by using a reinforcement learning algorithm with the qualification rate of the processed finished products corresponding to the samples in each time period as a constraint for the obtained characteristic parameter samples of parts in multiple time periods in the historical processing state, characteristic parameter samples of the mold for processing the parts, and corresponding processing parameter samples, to obtain a processing prediction model.

[0081] In some embodiments, the mold processing precision control device further includes: A second adjustment module, configured to directly adjust the current processing parameters for processing the feature surface when it is detected that the state deviation is less than or equal to the deviation threshold, and continue to process the feature surface based on the adjusted processing parameters.

[0082] It is worth mentioning that each module involved in this embodiment is a logical module. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or implemented by a combination of multiple physical units. In addition, to highlight the innovative part of this application, units not closely related to solving the technical problems proposed in this application are not introduced in this embodiment, but this does not mean that there are no other units in this embodiment.

[0083] Embodiment 4: Another embodiment of the present application relates to an electronic device, as Figure 3 shown, including: at least one processor 901; and a memory 902 communicatively connected to the at least one processor 901; wherein, the memory 902 stores instructions executable by the at least one processor 901, and the instructions are executed by the at least one processor 901 to enable the at least one processor 901 to execute the mold processing precision control method in the above embodiments.

[0084] Among them, the memory and the processor are connected in a bus manner. The bus can include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and the memory together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, etc., which are well known in the art, so they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a component or multiple components, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor is transmitted on the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor.

[0085] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory can be used to store the data used by the processor when executing operations.

[0086] Embodiment Five: Another embodiment of the present application relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the above method embodiments are implemented.

[0087] That is, those skilled in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions for causing a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0088] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present application, and in practical applications, various changes can be made in form and details without departing from the spirit and scope of the present application.

Claims

1. A method for controlling the machining accuracy of a mold, characterized in that, Including: For a plurality of preset feature surfaces to be machined on a workpiece to be machined, obtain the tolerance and initial programming allowance corresponding to each of the feature surfaces, wherein the tolerances and / or initial programming allowances of the feature surfaces are not all the same; During the machining process of any one of the feature surfaces, periodically detect the state deviation of the feature surface. When the detected state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming allowance. The greater the value of the state deviation, the greater the compensation amount for the current programming allowance; Adjust the machining parameters currently used to machine the feature surface, and continue to machine the feature surface in combination with the compensated current programming allowance and the adjusted machining parameters; The step of, when the detected state deviation is greater than the deviation threshold, calling the PID algorithm to compensate and update the current programming allowance, includes: When the detected state deviation is greater than the deviation threshold, calculate the compensation amount for the current programming allowance through the PID algorithm based on the state deviation; Superimpose the compensation amount and the current programming allowance to obtain the compensated current programming allowance.

2. The mold processing accuracy control method according to claim 1, characterized in that The method further includes: Construct a mapping table including different feature surfaces, and the feature identifiers and tolerances corresponding to each feature surface; Set the initial programming allowance corresponding to each of the feature surfaces, and construct a mapping table including different feature surfaces and the initial programming allowances corresponding to each feature surface; The step of obtaining the tolerance and initial programming allowance corresponding to each of the feature surfaces includes: Match the tolerance and initial programming allowance corresponding to each of the feature surfaces from the mapping table based on the feature identifiers predefined for the feature surfaces in the software model library.

3. The mold processing accuracy control method according to claim 1, wherein The state deviation of the feature surface is calculated through the following steps: During the machining process of the feature surface, periodically obtain the dimensional deviation, temperature deviation, and stress deviation of the feature surface; Perform fusion calculation on the dimensional deviation, temperature deviation, and stress deviation obtained in the same period to obtain the state deviation.

4. The mold processing precision control method according to claim 1, characterized in that The step of adjusting the machining parameters currently used to machine the feature surface includes: Input the feature parameters of the currently machined workpiece and the mold used to machine the workpiece into a pre-constructed machining prediction model for prediction, obtain the predicted machining parameters matching the feature parameters of the currently machined workpiece and the mold used to machine the workpiece, and use the predicted machining parameters as the adjusted machining parameters.

5. The method for controlling the machining accuracy of a mold according to claim 4, wherein The feature parameters of the currently machined workpiece include the material hardness of the workpiece, the feature parameters of the mold used to machine the workpiece include the wear degree and vibration frequency of the tool on the mold, and the machining parameters include the spindle speed and feed rate of the machining machine tool used to control the tool; The machining prediction model is obtained through the following steps: Construct an original model; Using the qualified rate of the machined products corresponding to the samples in each period as a constraint, train the original model with the feature parameter samples of the workpiece in multiple periods in the historical machining state, the feature parameter samples of the mold used to machine the workpiece, and the corresponding machining parameter samples by using a reinforcement learning algorithm to obtain the machining prediction model.

6. The mold processing precision control method according to claim 1, wherein The method further includes: When it is detected that the state deviation is less than or equal to the deviation threshold, directly adjust the current machining parameters for machining the feature surface, and continue to machine the feature surface based on the adjusted machining parameters.

7. A device for controlling the machining precision of a mold, characterized in that, It includes: An acquisition module, configured to acquire the tolerance and the initial programming allowance corresponding to each of a plurality of preset feature surfaces to be machined for a to-be-machined part, wherein the tolerances and / or the initial programming allowances of the feature surfaces are not completely the same; A compensation module, configured to periodically detect the state deviation of the feature surface during the machining of any one of the feature surfaces. When it is detected that the state deviation is greater than the deviation threshold, call the PID algorithm to compensate and update the current programming allowance. The greater the value of the state deviation, the greater the compensation amount for the current programming allowance; the compensation module includes: a calculation unit, configured to calculate the compensation amount for the current programming allowance based on the state deviation through the PID algorithm when it is detected that the state deviation is greater than the deviation threshold; a superposition unit, configured to superpose the compensation amount and the current programming allowance to obtain the compensated current programming allowance; A first adjustment module, configured to adjust the current machining parameters for machining the feature surface, and continue to machine the feature surface in combination with the compensated current programming allowance and the adjusted machining parameters.

8. An electronic device, characterized in that, It includes: At least one processor; And, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the mold machining accuracy control method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the mold machining accuracy control method according to any one of claims 1 to 6.

Images