Batch and staged detection interactive machine parameter optimization system and method
By employing a batch and phased testing method and optimizing machine parameters using processors and databases, the problems of low efficiency in parameter adjustment and long testing time during frequent line changes are solved, achieving efficient parameter adjustment and testing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-25
- Publication Date
- 2026-03-27
AI Technical Summary
When existing industrial production lines frequently change lines, the efficiency of adjusting machine parameters is low and the testing time is lengthy. The lack of phased testing results in low production efficiency.
A batch and phased testing method is adopted. An interactive machine parameter optimization system is used to record parameter categories and test results through processors and databases, and the test is carried out in stages to recommend the best parameter combination.
While shortening the inspection time, it improves the efficiency of parameter adjustment, making it suitable for production lines that frequently change lines, reducing unnecessary inspection actions, and improving production efficiency.
Smart Images
Figure CN121742364A_ABST
Abstract
Description
Technical Field
[0001] This application relates to a parameter optimization system and method for a machine tool, and more particularly to an interactive machine tool parameter optimization system and method for batch and staged testing. Background Technology
[0002] Modern industrial production lines are dominated by a small-batch, high-variety production model, which necessitates frequent line changes. Frequent line changes require constant adjustments to machine operating parameters. However, current production lines often rely on production line personnel to adjust parameters based on rules of thumb (such as trial and error). This method is inefficient and limited by the experience and intuition of the production line personnel, resulting in complex and time-consuming machine parameter adjustments.
[0003] In addition, during the parameter adjustment process, the products produced by the machine based on various parameters need to be tested. However, with the increase in testing items and the increasing complexity of the process, a long measurement waiting time is required, resulting in longer and longer quality inspection time.
[0004] Current common machine testing methods mostly involve comprehensive product quality inspection, lacking the concept of phased testing. For example, pre-planned experimental methods (such as the Taguchi method and exhaustive search) primarily involve setting all experimental parameter combinations at once before the experiment begins and then testing all these combinations in one go. Furthermore, even interactive experimental methods (such as Bayesian optimization) dynamically adjust subsequent experimental plans based on the experimental process, but only recommend one set of parameters for each round of testing. In other words, current machine testing methods, lacking phased testing and batch-specific parameter recommendations, are highly unfavorable for industrial production processes that emphasize production efficiency and require rapid adjustments to optimal parameters, and are also detrimental to production lines that frequently change lines. Summary of the Invention
[0005] The main objective of this application is to provide an interactive machine parameter optimization system and method for batch and phased testing. By using phased testing, it further saves testing time in a tedious and lengthy testing process. Furthermore, by incorporating a batch parameter recommendation mode, the system can reduce the number of rounds required for recommendation, thereby effectively shortening the time to find the optimal parameters.
[0006] In one embodiment, the interactive machine parameter optimization system of this application includes:
[0007] A database that records multiple parameter categories; and
[0008] A processor, coupled to the database, executes the following program:
[0009] (a) Obtain a batch of recommended groups, a testing stage, and a preset stop condition;
[0010] (b) Based on the multiple parameter categories and the number of recommended groups in the batch, generate at least one parameter combination;
[0011] (c) Based on at least one product produced by each set of the parameter combination, perform at least one stage of testing on the at least one product to generate a stage test result corresponding to each stage test;
[0012] (d) Determine whether the number of stages of the current phase of detection has reached the required number of detection stages;
[0013] (e) When the number of stages reaches the number of detection stages, determine whether the preset stop condition has been met;
[0014] (f) When it is determined that the preset stopping condition has been met, an optimal parameter combination is generated based on all the detection results of this stage obtained by the procedure (c);
[0015] (g) If it is determined that the preset stop condition has not been met, repeat procedure (b) to (f);
[0016] The procedure (c) includes: when the number of stages has not reached the number of testing stages, obtaining the parameter combinations corresponding to the testing results of the stage that meet the standards of the first stage, and conducting the next stage testing on at least one product produced based on the parameter combinations.
[0017] In one embodiment, the interactive machine parameter optimization method of this application includes:
[0018] a) Retrieve multiple parameter categories from a database;
[0019] b) A processor sets a batch recommended number of groups, a number of detection stages, and a preset stop condition;
[0020] c) The processor generates at least one set of parameter combinations based on the multiple parameter categories and the recommended number of groups in the batch;
[0021] d) The processor performs at least one stage of testing on at least one product produced based on each combination of parameters, so as to generate a one-stage test result corresponding to each stage test;
[0022] e) The processor determines whether the current stage of the current detection has reached the required number of detection stages;
[0023] f) When the number of stages reaches the detection stage number, determine whether the preset stop condition has been met;
[0024] g) When the preset stopping condition is met, the processor generates an optimized parameter combination based on all the detection results obtained in step d) for this stage; and
[0025] h) If the preset stopping condition is not met, repeat steps c) to g);
[0026] Step d) includes: when the number of stages has not reached the number of testing stages, obtaining the parameter combinations corresponding to the testing results of the stage that meet the standards of the first stage, and conducting the next stage testing on at least one product produced based on the parameter combinations.
[0027] Compared to related technologies, this application uses a batch-recommended parameter approach combined with phased testing to obtain optimal parameters in the shortest time, making it suitable for production lines that require frequent line changes and highly value parameter adjustment efficiency. Attached Figure Description
[0028] Figure 1 This is an embodiment of the block diagram of the optimization system of this application;
[0029] Figure 2 An embodiment of the flowchart of the optimization method of this application;
[0030] Figure 3 This is a schematic diagram of the input interface for this application;
[0031] Figure 4 This is a first embodiment of a schematic diagram of the recommended parameters page of this application;
[0032] Figure 5 This is a first embodiment of a schematic diagram of the test result reporting page of this application;
[0033] Figure 6 This is a second embodiment of a schematic diagram of the test result reporting page of this application;
[0034] Figure 7 The third embodiment is a schematic diagram of the test result reporting page of this application;
[0035] Figure 8 The fourth embodiment is a schematic diagram of the test result reporting page of this application;
[0036] Figure 9 This is a second embodiment of a schematic diagram of the recommended parameters page of this application.
[0037] Explanation of reference numerals in the attached figures
[0038] 1: Optimize the system
[0039] 11: Processor
[0040] 12: Database
[0041] 121: Parameter Category
[0042] 122: Adjustment Factor
[0043] 123: Testing Items
[0044] 124: Quality Standards
[0045] 13: Communication Unit
[0046] 14: Human-computer interface
[0047] 2: Machine
[0048] S20~S28: Optimization Steps Detailed Implementation
[0049] This application discloses an interactive machine parameter optimization system and method for batch and phased testing. It can recommend multiple parameter combinations for experiments in batches, thereby shortening the experimental time and dividing the testing items into multiple testing phases to avoid unnecessary testing actions. Furthermore, this application discloses an optimization system and method capable of interactive experiments. It recommends parameter combinations for the next round of experiment (i.e., round n) based on all test results up to the present (i.e., round 1 to round n-1), or by utilizing all historical testing data (test results) to recommend parameter combinations for the next round of experiment, thereby improving the adaptability of the recommended parameter combinations.
[0050] Specifically, conventional testing methods for equipment typically involve pre-testing, where all experimental parameter combinations are set before the experiment begins, followed by a full run of testing with these combinations and the generation of results. In contrast, the optimization system and method of this application involve interactive testing, creating a sense of interaction between the system and the testing equipment. In this application, the system first generates a batch of recommended parameter combinations. The testing equipment then performs the first round of testing based on these recommended combinations and returns the results to the system. The system then recommends parameter combinations for the next round of testing based on the results of the first round.
[0051] Please refer to the first one. Figure 1 This is an embodiment of the block diagram of the optimization system of this application. For example... Figure 1 As shown, the batch and phased testing interactive machine parameter optimization system (hereinafter referred to as optimization system 1) of this application includes at least a processor 11 and a database 12, wherein the processor 11 is coupled to the database 12.
[0052] In one embodiment, the processor 11 is a Central Processing Unit (CPU), a Microcontroller Unit (MCU), a Programmable Logic Controller (PLC), a System-on-Chip (SoC), or a Field-Programmable Gate Array (FPGA), but is not limited thereto. The database 12 is a Double Data Rate (DDR) memory, Flash memory, Random Access Memory (RAM), Read Only Memory (ROM), Hard Disk Drive (HDD), Solid State Drive (SSD), or a combination of the above, but is not limited thereto. The database 12 records computer-executable program code (not shown in the figure). When the optimization system 1 starts and the processor 11 reads and executes the computer-executable program code, the batch and phased interactive machine parameter optimization method of this application (hereinafter referred to as the optimization method) can be implemented.
[0053] The optimized system 1 of this application may also have a communication unit 13 coupled to the processor 11. For example... Figure 1 As shown, the communication unit 13 of the optimization system 1 is connected to the external machine 2 via wired or wireless means. One objective of this application is to automatically recommend suitable parameter combinations based on the product corresponding to the machine 2 after line change when the machine 2 is changing production lines. Thus, after the machine 2 performs necessary checks on the parameter combinations recommended by the optimization system 1, it can determine the optimal parameter combination to be used for producing this product.
[0054] like Figure 1 As shown, database 12 records multiple parameter categories 121, one or more adjustment factors 122 for each parameter category 121, one or more test items 123 included in one or more test stages, and quality standards 124 for each test item 123, etc. The data recorded in database 12 belongs to the problems that the optimization system 1 of this application needs to predefine (e.g., automatically defined according to the product specifications), and is recorded in database 12 after definition. In another embodiment, the data in database 12 can also be manually entered into the optimization system 1 and stored in database 12 by the user when they want to use the optimization system 1.
[0055] In one embodiment, the optimization system 1 records multiple parameter categories 121 and adjustment factors 122 (e.g., minimum adjustment amount, adjustment unit, upper limit value, and lower limit value) corresponding to each parameter category 121, based on one or more products that the machine 2 can produce. When the optimization system 1 is triggered to recommend one or more sets of parameter combinations, the processor 11 can generate recommended values for each parameter category 121 based on these adjustment factors 122. Thus, the machine 2 can use one or more sets of parameter combinations recommended by the optimization system 1 to conduct experiments and ultimately find the optimal parameter combination.
[0056] The parameter category 121 and adjustment factor 122 are shown in the table below, but are not limited thereto.
[0057]
[0058] In the table above, parameter category 121 is listed as twelve examples (i.e., parameter 1 to parameter 12), but this is only an example. Different products may have different numbers and contents of parameter categories 121, and are not limited to the table above. Taking an injection molding machine as an example, the parameter category 121 may include, for example, injection speed, injection pressure, holding speed, and holding pressure.
[0059] After machine 2 produces the corresponding product (i.e., the product to be tested) according to one or more parameter combinations recommended by optimization system 1, machine 2 can test these products based on the test items 123 and the quality standards 124 of each test item 123. Finally, optimization system 1 can recommend one or more parameter combinations for the next round based on the test results of these products. In one embodiment, the quality standards 124 may include the unit, test value specification, optimization target, specification requirements, and acceptance criteria of the test items 123, but are not limited thereto.
[0060] The testing items 123 and quality standards 124 are shown in the table below, but are not limited thereto.
[0061]
[0062] In the table above, four inspection items 123 are used as examples (including appearance inspection 1, appearance inspection 2, detail inspection 3 and detail inspection 4), but this is only an example. Different products have different numbers and contents of inspection items 123, and are not limited to the table above.
[0063] like Figure 1As shown, the optimization system 1 of this application also has a human-machine interface 14 coupled to the processor 11. In one embodiment, the optimization system 1 receives the number of recommended batch groups, the number of detection stages, and the preset stop condition input by the user from the outside through the human-machine interface 14. As mentioned above, the optimization system 1 of this application shortens the experimental time through batch recommendation and staged detection. In one embodiment, the number of recommended batch groups represents the number of parameter combinations that the optimization system 1 needs to recommend simultaneously in each round (e.g., 3 groups, 5 groups, etc.); the number of detection stages represents how many stages the machine 2 needs to divide into when detecting products (e.g., in the table above, multiple detection items 123 can be divided into two detection stages (including appearance detection stage and detail detection stage)); the preset stop condition represents how long the optimization system 1 should repeat batch recommendation (i.e., the number of rounds).
[0064] In another embodiment, the optimization system 1 can also store the recommended batch number, detection stage number and preset stop condition in the database 12 through preset settings, so that the processor 11 can read and use them, without limitation.
[0065] When machine 2 needs to change production lines and requires the optimization system 1 to recommend parameter combinations, the user first operates the optimization system 1 through the human-machine interface 14, selecting the product to be tested within the optimization system 1. Then, the optimization system 1 automatically retrieves the pre-stored corresponding template and obtains data from the database 12, including multiple parameter categories 121, multiple adjustment factors 122, multiple test items 123, and the quality standards 124 for each test item 123.
[0066] The optimization system 1 also receives (or reads from the database 12) the number of recommended batch groups (e.g., 3 groups), the number of testing stages (e.g., 2 stages), and the preset stopping condition (e.g., 6 rounds) required by the user for this test through the human-machine interface 14. After obtaining the above data, the processor 11 of the optimization system 1 generates one or more parameter combinations based on each parameter category 121 and the number of recommended batch groups, and the number of these one or more parameter combinations is the same as the number of recommended batch groups input by the user. For example, if the number of recommended batch groups is K, the processor 11 will generate K sets of parameter combinations at once, where each set of parameter combinations contains the recommended values of all parameter categories 121. It is worth mentioning that the processor 11 generates one or more parameter combinations based on one or more adjustment factors 122 of each parameter category 121 and the number of recommended batch groups, that is, it calculates and generates the recommended values of each parameter category 121, so the recommended values of each parameter category 121 in each set of parameter combinations are not repeated. The number of recommended batch groups K is a positive integer, and K≥1.
[0067] After generating the K sets of parameter combinations, machine 2 can automatically obtain the K sets of parameter combinations, or the user can input the K sets of parameter combinations into machine 2. Thus, machine 2 can produce one or more products based on the K sets of parameter combinations, with each product corresponding to a specific set of parameter combinations. Then, machine 2 or other testing equipment (not shown in the figure) will test the one or more products respectively, and generate corresponding test results.
[0068] It is worth mentioning that, when inspecting the product, based on the set number of inspection stages, machine 2 will only perform one or more inspection items in the first stage of inspection and feed the corresponding inspection results back to optimization system 1. Only when a product passes all the inspection items in the first stage will machine 2 proceed to the next stage of inspection. This achieves the technical effect of improving inspection efficiency through staged inspection.
[0069] In one embodiment, the optimization system 1 can also obtain the number of repeated tests M required by the user through the human-machine interface 14. The number of repeated tests M represents the number of times the user wants each parameter combination to be tested. For example, if the recommended batch size K is 3 and the number of repeated tests M is 2, it means that machine 2 needs to produce 2 corresponding products for each parameter combination recommended by the optimization system 1 in this round. Therefore, machine 2 needs to produce a total of 6 products in this round, and perform a phased testing procedure on each of these 6 products.
[0070] After machine 2 performs all inspection items 123 of the i-th stage inspection (where 1≤i≤L, and L is the number of inspection stages) on K products produced according to K sets of parameter combinations, the optimization system 1 can obtain the inspection results of these products in the i-th stage inspection. In this application, the optimization system 1 obtains the inspection results of each product (i.e., each set of parameter combinations) in the i-th stage inspection, and only when it is determined that a product (i.e., a set of parameter combinations) has passed all inspection items 123 of the i-th stage inspection will the feedback mechanism for the next stage inspection (i.e., the i+1-th stage inspection) of this parameter combination be triggered. Furthermore, only after the feedback mechanism for the next stage inspection of a parameter combination is triggered can the processor 11 of the optimization system 1 pass through the triggered feedback mechanism (e.g., Figure 5 (The reported feedback field) further obtains the test results of the product corresponding to this parameter combination in the next stage of testing. In other words, only after a product has passed all test items 123 in the i-th stage of testing does machine 2 need to perform the i+1-th stage of testing on the product, and only then does the optimization system 1 need to receive the test results of the product after the i+1-th stage of testing.
[0071] Therefore, after machine 2 completes the i-th stage inspection of K products, the processor 11 of the optimization system 1 first determines whether the current stage number of the i-th stage inspection has reached the number of inspection stages (i.e., whether the inspection is completed). If the current stage number of the i-th stage inspection has not reached the number of inspection stages, for one or more products that meet the stage criteria of all inspection items 123 of the i-th stage inspection, the processor 11 of the optimization system 1 can obtain the parameter combinations corresponding to these products respectively, and perform the next stage inspection (i.e., the i+1-th stage inspection) on the products produced according to these parameter combinations, and obtain the inspection results generated after these products have undergone the next stage inspection of one or more inspection items 123.
[0072] In one embodiment, the optimization system 1 uses the processor 11 to generate corresponding production instructions based on the K sets of parameter combinations, and transmits the production instructions to the machine tool 2 via the communication unit 13. Thus, the machine tool 2 can automatically produce the corresponding K products. In this embodiment, the optimization system 1 also automatically receives the test results generated by the machine tool 2 after performing tests on these products (e.g., stage i test, stage i+1 test, etc.) via the communication unit 13. In another embodiment, after the processor 11 generates the K sets of parameter combinations, the user manually controls the machine tool 2 to produce one or more corresponding products based on the K sets of parameter combinations. After the tests are completed, the user also manually inputs the test results of these products into the optimization system 1.
[0073] In this application, the optimization system 1 and the testing machine 2 will continuously repeat the above actions. That is, for one or more products that have passed all test items 123 in the previous stage of testing, the testing machine 2 will perform one or more test items 123 in the next stage of testing, and the optimization system 1 will obtain the test results of these products in the next stage of testing, until the current stage of testing reaches the number of testing stages set by the user. It is worth mentioning that if all products fail the i-th stage of testing, the testing machine 2 will not perform the i+1-th stage of testing, but will directly end the current testing program. Similarly, if all products fail the i+1-th stage of testing, the testing machine 2 will not perform the i+2-th stage of testing.
[0074] During the testing process and obtaining test results, the processor 11 continuously determines whether the current stage of the phased test has reached the required number of testing stages, that is, whether all phased tests have been completed. If the current stage of the phased test has reached the required number of testing stages, it means that all K sets of parameter combinations generated this time have completed phased testing. At this time, the processor 11 determines whether the preset stop conditions input by the user are met, such as whether the number of execution rounds has been reached, or whether any product meets the quality standards 124 of all test items 123 in all testing stages, without imposing any restrictions.
[0075] In one embodiment, the preset stopping condition is the number of execution rounds for the action of recommending K sets of parameter combinations. In this embodiment, when the processor 11 determines that the current stage detection stage has reached the detection stage number, it will further determine whether the current recommendation count of the K sets of parameter combinations has reached the execution round number. If the recommendation count of the K sets of parameter combinations (e.g., 2 times) has not reached the execution round number (e.g., 3 times), the processor 11 will continue to the next round of the program, and in the next round, recommend a new K sets of parameter combinations (i.e., the 3rd time) and perform detection.
[0076] In one embodiment, the preset stopping condition is that the stage test result of at least one product corresponding to any set of parameter combinations meets the quality standard 124 of all test items in all stages of testing. In this embodiment, when the processor 11 determines that the number of stages of the current stage test has reached the required number of test stages, it will further determine whether any product's stage test result meets the quality standard 124 of all test items in all stages of testing (e.g., the test value specifications, optimization goals, and specification requirements of each stage test). If any product meets the above condition, the processor 11 can directly take the parameter combination corresponding to this product as the optimal parameter combination and stop recommending new K sets of parameter combinations.
[0077] When it is determined that the current stage of the phased inspection has reached the required number of inspection stages but the preset stop condition has not yet been met, the processor 11 will refer to all the inspection results obtained so far and selectively refer to the historical inspection data of the machine 2 to automatically generate and recommend a new set of K parameter combinations. At this time, the optimization system 1 and the machine 2 will repeat the above actions to perform phased inspections on the K products corresponding to the new set of K parameter combinations and obtain inspection results in stages.
[0078] When the current stage of the inspection reaches the required number of stages and the preset stop condition is met, the current inspection process can be terminated. At this point, the optimization system 1 can automatically generate and recommend a set of optimal parameter combinations based on all inspection results obtained in all stages and selectively referencing historical inspection data from machine 2.
[0079] In one embodiment, the processor 11 generates the parameter combination and the optimized parameter combination based on one or more detection results using statistical methods, machine learning methods, or Bayesian optimization methods. However, the above is merely one specific embodiment of this application and is not intended to limit the scope of the invention.
[0080] Please refer to the following for further details. Figures 2 to 9 ,in Figure 2 This is an embodiment of the flowchart of the optimization method of this application. Figure 3 This is a schematic diagram of the input interface for this application. Figure 4 This is a first embodiment of a schematic diagram of the recommended parameters page of this application. Figures 5 to 8 The images shown are a first embodiment, a second embodiment, a third embodiment, and a fourth embodiment, respectively, illustrating the detection result reporting page of this application. Figure 9 This is a second embodiment of a schematic diagram of the recommended parameters page of this application.
[0081] Figure 2 The optimization method of this application is disclosed, and this optimization method is mainly applied to Figure 1 The optimized system 1 shown is not limited to this.
[0082] like Figure 2 As shown, when using the optimization system 1 of this application, the user first triggers the optimization system 1 (e.g., opens the parameter recommendation interface of the optimization system 1), and the processor 11 obtains multiple parameter categories 121 from the database 12 (step S20). In one embodiment, the processor 11 may also obtain the adjustment factors 122 of each of the multiple parameter categories 121 from the database 12, as well as one or more detection items for each detection stage. In one embodiment, the processor 11 may also obtain the quality standards 124 of each detection item 123 from the database 12.
[0083] In one embodiment, a user can open a new project on the optimization system 1, enter a project name, and select the template corresponding to the product to be tested. In this application, each template corresponds to a different product category, and records the parameter category 121, adjustment factor 122, and test item 123 corresponding to this product category. Therefore, when a user selects a specific template, the processor 11 automatically retrieves the required parameter category 121, adjustment factor 122, and test item 123 from the database 12, without requiring the user to input them manually.
[0084] It is worth noting that if a new experiment or product needs to be tested, the optimization system 1 may not have the corresponding template, parameter category 121, adjustment factor 122, and test item 123. In this case, the optimization system 1 can activate the human-machine interface 14, allowing the user to directly input the product to be tested, the parameter category 121 that the product should possess, the adjustment factor 122 for each parameter category 121, and the test items 123 that the product should undergo. Therefore, the optimization system 1 must adhere to the above information when automatically generating recommended parameter combinations.
[0085] Next, the processor 11 obtains the recommended batch number, the number of detection stages, and the preset stop condition (step S21). In one embodiment, the processor 11 receives external settings through the human-machine interface 14 to obtain the recommended batch number, the number of detection stages, and the preset stop condition. In another embodiment, the recommended batch number, the number of detection stages, and the preset stop condition are stored in the database 12 via user input, and in step S21, the processor 11 directly obtains the recommended batch number, the number of detection stages, and the preset stop condition from the database 12.
[0086] In one embodiment, the processor 11 can also obtain the number of repeated detections for each parameter combination through the human-machine interface 14. The number of repeated detections represents the number of times each parameter combination recommended by the processor 11 needs to be detected. By repeatedly detecting the same set of parameter combinations, the stability of these parameter combinations can be effectively detected.
[0087] like Figure 3 As shown, when setting experimental objectives, the optimization system 1 can accept user input via the human-machine interface 14 for the recommended rounds R (i.e., the number of rounds to be executed, which can be used as one of the preset stopping conditions), the number of repetitions (i.e., the number of repeated detections M for each parameter combination), and the number of recommended batches per round (i.e., how many parameter combinations need to be recommended simultaneously in each round, which is the number of batch recommended groups K). For example, if the user inputs 6 recommended rounds, 2 repetitions, and 3 recommended batches per round (i.e., R=6, M=2, K=3), it means that the optimization system 1 needs to execute a recommendation program for six rounds, and each round of the recommendation program needs to recommend three parameter combinations in batches, and each parameter combination needs to be detected twice.
[0088] After step S21, the processor 11 then automatically generates and recommends K sets of parameter combinations based on each parameter category 121 and the number of batch recommendation groups required by the user (step S22). More specifically, the processor 11 generates K sets of parameter combinations based on one or more adjustment factors 122 for each parameter category 121 and the number of batch recommendation groups required by the user. In this application, K is a positive integer and K≥1. In other words, based on the number of batch recommendation groups set by the user, the processor 11 will automatically generate the corresponding number of parameter combinations based on the adjustment factor 122, wherein each set of parameter combinations includes the recommended values of multiple parameter categories 121 included in this template (i.e., Figure 4 The parameter values shown are X1 to Xn. Furthermore, in one embodiment, the recommended values of each parameter category 121 in each parameter combination recommended by the processor 11 are not repeated.
[0089] like Figure 4As shown, assuming the template selected by the user includes n parameter categories 121, and the user requests a batch recommendation group number K of 3, then when the user triggers the "Start Recommendation Parameters" button on the human-machine interface 14, the processor 11 will automatically generate three sets of parameter combinations according to the adjustment factors 122 of each of these parameter categories 121. Figure 4 Taking Group 1, Group 2, and Group 3 as examples, each group's parameter combination contains n parameter categories 121 ( Figure 4 (Taking the first to nth parameters as examples), and the recommended values for each parameter category 121 ( Figure 4 (Taking X1 to Xn as an example).
[0090] For machine 2, the difference in experimental time per round between recommending multiple parameter combinations in batches and recommending a single parameter combination in a single run is not significant. However, by using the batch recommendation method, machine 2 can try more parameter combinations per unit of time, thus shortening the overall experimental time (e.g., finding the optimal parameter combination in fewer rounds).
[0091] After step S22, the optimization system 1 of this application, in conjunction with the machine tool 2, performs detection for each parameter combination recommended by the processor 11 (step S23). In one embodiment, the optimization system 1 can automatically transmit multiple parameter combinations recommended by the processor 11 and the number of repeated detections set by the user to the machine tool 2, so that the machine tool 2 produces the corresponding number of products. For example, if the processor 11 recommends three parameter combinations and the user sets the number of repeated detections to two, then the machine tool 2 will produce multiple products (e.g., six products, but not limited to this). In another embodiment, the user can also manually input the multiple parameter combinations and the number of repeated detections into the machine tool 2, so that the machine tool 2 produces the corresponding products.
[0092] It is worth mentioning that after the product is produced by machine 2, machine 2 itself can perform phased testing on these products and generate corresponding test results, or other testing equipment (including the optimization system 1) can perform phased testing on these products and generate corresponding test results. Finally, the optimization system 1 receives these test results.
[0093] In the following explanation, P0 refers to all K sets of parameter combinations, P1 represents one or more parameter combinations that pass the first stage of detection, P2 represents one or more parameter combinations that pass the second stage of detection, and so on. In other words, K sets of parameter combinations P will be used in the first stage of detection. i-1 Where i represents the stage number, meaning that all K sets of parameter combinations (P) will be used in the first stage of detection. i-1 =P0) is detected, and the first stage of detection is performed on one or more sets of parameter combinations (P)i =P1). Similarly, the second-stage detection will use one or more parameter combinations P1(P) that passed the first-stage detection. i-1 =P1) is used for detection, and the combination of one or more parameters (P) passes the second stage of detection. i =P2). For example Figure 2 As shown, in the i-th stage of detection, the K sets of parameter combinations P are tested by machine 2 or other detection equipment. i-1 The corresponding product undergoes one or more testing items 123 (hereinafter referred to as testing items S) in the i-th stage of testing. i The optimization system 1 obtains P based on the K sets of parameter combinations. i-1 At least one product produced separately underwent one or more testing items S in the i-th stage of testing. i The detection results are then generated (step S24). In one embodiment, i is a positive integer and 1≤i≤L, where L is the number of detection stages set by the user.
[0094] It is worth mentioning that if the number of repeated tests M is greater than or equal to 1, then in step S24, the processor 11 needs to obtain all the test items S corresponding to at least one product that has undergone the i-th stage test for each set of parameters. i The system then generates M test results. In one embodiment, repeated testing refers to testing multiple products corresponding to the same parameter combination separately. In another embodiment, repeated testing refers to testing the same product multiple times.
[0095] Please see Figure 5 and Figure 6 .At Figure 5 and Figure 6 In this embodiment, the number of detection stages set by the user is 2 ( Figure 5 (Taking phases #1 and #2 as examples), phase #1 includes three detection items S. i ( Figure 6 (Taking items 1, 2, and 3 as examples). Furthermore, the user sets the number of repeated tests M to 2. In this embodiment, for the first set of parameter combinations, the user or machine 2 needs to report the test results of product 1 and product 2 to the optimization system 1 respectively.
[0096] In one embodiment, the optimization system 1 can enter the process after generating multiple sets of recommended parameter combinations for the current round. Figure 5 The test results page is shown, and the test results (such as the test results of item 1, item 2, and item 3) are filled in manually by the user or automatically by machine 2 (or the testing equipment).
[0097] After obtaining all the test results of the product in the i-th stage of testing, the processor 11 of the optimization system 1 optimizes the system according to each test item S. i The quality standard 124 determines whether the test results of each product meet the standard, thereby determining which products meet all the test items S of the i-th stage of testing. i Quality standard 124, namely, determines one or more products and their corresponding one or more parameter combinations P that will proceed to the next stage of testing. i (Step S25). For example, the i-th stage of detection includes three detection items S. i These three testing items S i The test values are all specified as 0-5 points, the optimization target is to minimize the score, and the specification requirement is 0 points for all. In this embodiment, a product scores 0 points in these three test items S. i All test results must be 0 points to be considered as meeting all test items S in the i-th stage of testing. i The quality standard is 124.
[0098] At Figure 7 In this embodiment, the test results of both products corresponding to the first set of parameter combinations fail to meet the standards. In this case, the optimization system 1 will determine that the first set of parameter combinations does not meet all the test items of the i-th stage of testing. At this time, the optimization system 1 will exclude the first set of parameter combinations from the next stage of testing.
[0099] At Figure 8 In this embodiment, the test results of the two products corresponding to the second set of parameters are both deemed compliant by the optimization system 1. In this case, the optimization system 1 will determine that the second set of parameters meets all test items S of the i-th stage of testing. i The quality standard is 124. At this point, the optimization system 1 can request the machine 2 or other testing equipment to perform the next stage of testing on the two products corresponding to the second set of parameters, and wait to receive the test results of the next stage of testing.
[0100] It is worth mentioning that in this application, the processor 11 only requires that all products corresponding to any given parameter combination have passed all test items S of the previous stage of testing. i Only then will the feedback mechanism for the next stage of detection be triggered (such as...). Figure 8 As shown, this is the reward field for stage #2. Furthermore, the feedback mechanism for the next stage detection is only triggered (e.g., Figure 8After the return button in [ ] is displayed, the user or the machine 2 can input the test results of multiple products corresponding to this parameter combination in the next stage of testing into the optimization system 1. That is to say, only after the feedback mechanism in the next stage of testing is triggered, can the processor 11 receive the test results of the next stage of testing through the triggered feedback mechanism (i.e., the return field). Specifically, in this application, the feedback mechanism includes indicating the parameter combination to be used in the next stage of testing, and waiting to receive the stage test results of at least one product corresponding to the parameter combination after the next stage of testing is completed.
[0101] More specifically, the staged testing in this application distinguishes multiple test items 124 according to types, and groups the test items 124 of the same or similar types into the same test stage. For example, the first test stage only includes appearance quality test items (such as charring, burrs, or insufficient filling, etc.), and the second test stage only includes dimensional quality test items (such as unbalance angle, unbalance amount, inner circle perpendicularity, or outer circle perpendicularity, etc.). If a parameter combination fails the first stage of testing, it means that the products produced according to this parameter combination do not meet the quality requirements, so there is no need to perform the second stage of testing on the products corresponding to this parameter combination.
[0102] The staged testing applied in this application can effectively avoid wasting testing time, save the time cost of repeating unnecessary test items, and improve the efficiency of the parameter adjustment process. For example, the first stage of testing can include appearance test items that take less time, while the second stage of testing can include detailed test items that require precision instruments and take more time. When a product fails even the appearance test items, it is not necessary for the optimization system 1 to perform detailed test items that take longer. Therefore, through the technical means of staged testing, this application can effectively shorten the testing time required for traditional complete testing.
[0103] In Figure 2 in step S25, before determining one or more parameter combinations P that can enter the next stage of testing for testing i the processor 11 first determines whether the number of stages of the current stage of testing (i.e., i) reaches the number of testing stages L set by the user. When i < L (i.e., the recommended parameter combinations in this round have not completed all the testing stages), the processor 11 sets i = i + 1 (step S26), and executes steps S23 to S25 again. Thus, in the next stage of testing (i.e., the (i + 1)-th stage of testing), all test items S that meet the previous stage of testing (i.e., the i-th stage of testing) are tested by the machine 2 or other testing devices by the processor 11 iOne or more products of quality standard 124 are tested, and the optimization system 1 obtains one or more test items S of these products that have undergone stage i+1 testing. i+1 The test results are then generated separately. In other words, when the processor 11 determines that the number of stages of the current stage test has not reached the number of test stages set by the user, it will obtain one or more sets of parameter combinations corresponding to the stage test results that meet the stage standards of the current test, and then perform the next stage test on at least one product produced based on these one or more sets of parameter combinations.
[0104] Similarly, after the (i+1)th stage of detection ends and the user or machine 2 reports the detection results of the product to the optimization system 1, the processor 11 determines, based on these detection results, one or more products and their corresponding one or more parameter combinations P to proceed to the next stage of detection (i.e., the (i+2)th stage of detection). i+1 .
[0105] When i = L, it indicates that the recommended parameter combination for this round has completed all stage detections, meaning that the processor 11 has completed the detection of all K sets of parameter combinations recommended in this round. At this time, the processor 11 further determines whether the preset stop condition set by the user is met (step S27). In one embodiment, the preset stop condition is the number of execution rounds of the processor 11 recommending the K sets of parameter combinations. In another embodiment, the preset stop condition is that the detection result of the product corresponding to any parameter combination meets the quality standard 124 of all detection items 123 in all detection stages.
[0106] In another embodiment, the preset stopping condition may be the total number of experiments. The total number of experiments refers to the total number of parameter combinations that need to be tested. In this embodiment, when the processor 11 determines that the current stage detection stage number has reached the detection stage number, but the total number of currently recommended parameter combinations has not yet reached the total number of experiments, it will execute steps S22 to S26 again to recommend one or more new parameter combinations (not necessarily K groups) in the next round and perform testing until the total number of already recommended parameter combinations is equal to the total number of experiments, at which point the entire program will stop.
[0107] For example, if the preset stop condition sets the total number of experiments to 10, the batch recommended number of groups K to 3, and the number of repeated tests M to 1, then because the optimization system 1 will recommend three parameter combinations in each round (i.e., it will experiment three times), in the fourth round, the optimization system 1 will only recommend one parameter combination, and the machine 2 will only perform the last experiment. However, the above is only one specific implementation example of this application, and is not limited thereto.
[0108] If i = L (i.e., the current stage number has reached the detection stage number) but the preset stopping condition has not yet been met, then the optimization system 1 returns to step S22. At this time, the processor 11 will automatically generate and recommend new K sets of parameter combinations by referring to the detection results of one or more rounds obtained previously.
[0109] like Figure 9 As shown, after all three parameter combinations recommended in the first round have been tested, since the preset stopping condition has not yet been met (e.g., the user sets the number of rounds to 6), the processor 11 will refer to all previously obtained test results (and may selectively refer to the historical data of optimization system 1 and / or machine 2), and recommend new K parameter combinations again. Figure 9 (Taking the recommended parameter combinations 1, 2, and 3 in the second round as an example). In one embodiment, after receiving the detection results of each parameter combination from the previous round, the processor 11 scores each parameter combination. Then, based on the scores and combined with the historical data (i.e., the experimental data accumulated in each execution round or the past experimental data of the machine 2), the processor 11 generates a new set of K parameter combinations to be recommended in the next round. Thus, the machine 2 can produce one or more new products based on the new set of K parameter combinations and perform phased testing on one or more products, while the optimization system 1 can also obtain the detection results of these products in stages.
[0110] If i = L (i.e., the current stage number has reached the detection stage number) and the preset stopping condition is met, then the processor 11 can generate and recommend a set of optimized parameter combinations based on all detection results obtained in all previous stages (step S28). In one embodiment, the processor 11 uses a statistical method, a machine learning method, or a Bayesian optimization method to calculate and generate the optimized parameter combinations based on all previous detection results. When the machine 2 uses the optimized parameter combinations to produce products, it can achieve the best yield in the production process or the best quality in the produced products.
[0111] Compared to previous experimental planning methods for machine parameter tuning, this application uses an optimization system to recommend multiple parameter combinations in batches. Through an interactive approach, after the recommended parameter combinations for one round are tested, multiple parameter combinations for the next round are recommended in batches based on the test results. Furthermore, a staged testing method is adopted; only parameter combinations that pass the previous stage's test need to proceed to the next stage. This effectively allows for obtaining the optimal parameter combinations recommended by the system in the shortest possible time.
Claims
1. An interactive machine parameter optimization system for batch and staged testing, the interactive machine parameter optimization system being connected to an external machine, comprising: The database records multiple parameter categories; as well as The processor, coupled to the database, executes the following program: (a) Obtain the recommended number of batches, the number of testing stages, and the preset stop conditions; (b) Based on the multiple parameter categories and the number of recommended batch groups, generate at least one parameter combination; (c) Based on at least one product produced according to each set of the parameter combinations, instruct the machine to perform at least one stage of inspection on the at least one product, and obtain the stage inspection result corresponding to each stage of inspection from the machine; (d) Determine whether the number of stages currently being detected has reached the number of detection stages; (e) When the number of stages reaches the number of detection stages, determine whether the preset stop condition has been met; (f) When it is determined that the preset stop condition has been met, an optimal parameter combination is generated based on all the previously obtained stage detection results; (g) If it is determined that the preset stop condition has not been met, repeat the procedure (b) to (f); The procedure (c) includes: when the number of stages has not reached the number of detection stages, obtaining the parameter combination corresponding to the stage detection result that meets the stage standard, and performing the next stage detection on at least one of the products produced according to the corresponding parameter combination.
2. The batch and phased interactive machine parameter optimization system according to claim 1, wherein the database further records one or more adjustment factors for each of the plurality of parameter categories, and in the procedure (b), the processor is configured to generate the at least one set of parameter combinations based on the one or more adjustment factors for each of the plurality of parameter categories and the batch recommended group number.
3. The batch and phased interactive machine parameter optimization system according to claim 1, wherein each phase of the detection phase includes at least one detection item.
4. The interactive machine parameter optimization system for batch and phased testing according to claim 3, wherein in the program (c), the processor is configured to determine at least one product that meets the phase criteria for all the test items of the currently performed phase test, and to use the at least one set of parameter combinations corresponding to the at least one product as the at least one set of parameter combinations in the next phase test.
5. The interactive machine parameter optimization system for batch and phased testing according to claim 1, wherein each group of parameter combinations includes recommended values for the plurality of parameter categories, and the recommended values for each parameter category in each group of parameter combinations are not repeated.
6. The interactive machine parameter optimization system for batch and phased detection according to claim 1, further comprising a human-machine interface coupled to the processor, the human-machine interface receiving the recommended number of batch groups, the number of detection phases, and the preset stop condition, wherein the preset stop condition is the number of execution rounds for recommending the at least one set of parameter combinations.
7. The interactive machine parameter optimization system for batch and phased testing according to claim 1, wherein the processor further executes the following program: Obtain the number of repeated detections, where the number of repeated detections represents the number of times each set of parameters needs to be detected; In the procedure (c), the processor is configured to repeatedly perform the stage detection on at least one product corresponding to each set of parameter combinations based on the number of repeated detections, so as to generate multiple stage detection results.
8. The batch and phased testing interactive machine parameter optimization system according to claim 3, wherein in the program (c), the processor is configured to trigger a feedback mechanism for the next phase test when the at least one product corresponding to any one of the parameter combinations passes all the test items of the currently performed phase test, wherein the feedback mechanism includes indicating the parameter combination to be used in the next phase test, and waiting to receive the phase test result of the at least one product corresponding to the parameter combination in the next phase test.
9. The interactive machine parameter optimization system for batch and phased testing according to claim 1, wherein the database further records the quality standards of each testing item, and the preset stop condition is that the phased testing result of at least one product corresponding to any one of the parameter combinations meets the quality standards of all phased testing, wherein the quality standards include the test value specifications, optimization targets and specification requirements of each phased testing.
10. A method for optimizing interactive machine parameters for batch and staged testing, comprising: a) Retrieve multiple parameter categories from the database; b) The processor sets the recommended number of batches, the number of detection stages, and the preset stop conditions; c) The processor generates at least one set of parameter combinations based on the plurality of parameter categories and the batch recommendation group number; d) The processor performs at least one stage detection on at least one product produced according to each set of parameter combinations to generate a stage detection result corresponding to each stage detection; e) The processor determines whether the number of stages currently being detected has reached the required number of detection stages; f) When the number of stages reaches the number of detection stages, determine whether the preset stop condition has been met; g) When it is determined that the preset stopping condition has been met, the processor generates a set of optimized parameter combinations based on all the previously obtained stage detection results; and h) If it is determined that the preset stop condition has not been met, repeat steps c) to g); Step d) includes: when the number of stages has not reached the number of detection stages, obtaining the parameter combination corresponding to the stage detection result that meets the stage standard, and performing the next stage detection on at least one of the products produced based on the corresponding parameter combination.
11. The batch and phased interactive machine parameter optimization method according to claim 10, wherein step a) includes obtaining one or more adjustment factors for each of the plurality of parameter categories from the database, and step c) includes generating the at least one set of parameter combinations based on the one or more adjustment factors for each of the plurality of parameter categories and the batch recommended group number.
12. The interactive machine parameter optimization method for batch and phased testing according to claim 10, wherein each phase of the testing phases includes at least one testing item, and step d) includes: Identify at least one product that meets the stage criteria for all the test items in the current stage of testing, and use the at least one set of parameters corresponding to the at least one product as the at least one set of parameters in the next stage of testing.
13. The interactive machine parameter optimization method for batch and phased testing according to claim 10, wherein each group of parameter combinations includes recommended values for the plurality of parameter categories, and the recommended values for each parameter category in each group of parameter combinations are not repeated.
14. The interactive machine parameter optimization method for batch and phased detection according to claim 10, wherein the preset stopping condition is the number of execution rounds for recommending the at least one set of parameter combinations, and step h) includes repeating steps c) to g) when it is determined that the number of phases has reached the number of detection phases, but the number of executions in step d) has not yet reached the number of execution rounds, so as to recommend a new set of at least one set of parameter combinations and perform detection in the next round.
15. The interactive machine parameter optimization method for batch and phased testing according to claim 10, wherein step a) further includes obtaining the quality standards of each of the testing items from the database, wherein the preset stopping condition is that the phased testing result of at least one product corresponding to any one of the parameter combinations meets the quality standards of all phased testing, wherein the quality standards include the test value specifications, optimization targets and specification requirements of each phased testing.
16. The interactive machine parameter optimization method for batch and phased testing according to claim 10, wherein the preset stopping condition is the total number of experiments, the total number of experiments is the total number of parameter combinations to be tested, and step h) includes repeating steps c) to g) when it is determined that the phase number has reached the testing phase number, but the total number of the recommended at least one set of parameter combinations has not yet reached the total number of experiments, to recommend new at least one set of parameter combinations and perform testing in the next round, until the total number of the recommended at least one set of parameter combinations is equal to the total number of experiments.
17. The interactive machine parameter optimization method for batch and phased testing according to claim 10, further comprising: b1) The processor obtains the number of repeated detections, wherein the number of repeated detections represents the number of times each set of the parameter combinations needs to be detected; In step d), the processor performs the stage detection repeatedly on at least one product corresponding to each set of parameters based on the number of repeated detections, so as to generate multiple stage detection results.
18. The interactive machine parameter optimization method for batch and phased testing according to claim 17, wherein in step d), when the processor triggers a feedback mechanism for the next phase of testing when the at least one product corresponding to any one of the parameter combinations passes all the testing items of the currently performed testing phase, wherein the feedback mechanism includes indicating the parameter combination to be used in the next phase of testing, and waiting to receive the phase testing result of the at least one product corresponding to the parameter combination in the next phase of testing.
19. The interactive machine parameter optimization method for batch and phased testing according to claim 11, wherein step b) is achieved by the processor receiving external settings for the recommended number of batches, the number of testing phases, and the preset stop condition, or by obtaining the recommended number of batches, the number of testing phases, and the preset stop condition from the database.