Miniature sawtooth manufacturing process optimization method based on fuzzy logic

Through the micro sawtooth manufacturing process optimization method based on fuzzy logic, the problem of difficult process parameters optimization, real-time regulation and insufficient adaptability in micro sawtooth manufacturing is solved, real-time optimization and dynamic adjustment of process parameters are achieved, and the consistency of product quality is improved. and productivity.

CN119024798AInactive Publication Date: 2024-11-26JIANGSU TIANSHUO ALLOY MATERIAL CO LTD

Patent Information

Application Number
CN202411264529.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing micro sawtooth manufacturing processes have significant challenges in process parameter optimization, including difficult process parameter optimization, lack of real-time regulation mechanism, insufficient adaptability and lagging feedback, resulting in inconsistent product quality and low production efficiency.

Method used

Adopting the micro-saw tooth manufacturing process optimization method based on fuzzy logic, through fuzzy logic systems, hierarchical progressive reasoning strategies and intelligent adaptive feedback control technology, process parameters are collected and optimized in real time, fuzzy rule base is constructed, and membership function and fuzzy are dynamically adjusted. Relationship matrix realizes real-time monitoring and dynamic adjustment of process parameters.

Benefits of technology

It significantly improves the efficiency of process parameter optimization, ensures the consistency of product quality and the stability of the production process, enhances process adaptability, reduces production fluctuations and scrap rates, and improves production flexibility and reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119024798A_ABST
    Figure CN119024798A_ABST
Patent Text Reader

Abstract

The invention discloses a micro saw tooth manufacturing process optimization method based on fuzzy logic, which aims to realize efficient control of the process through multi-level parameter optimization and real-time feedback regulation and control, and comprises the following steps of: collecting key process parameters such as cutting speed and feed rate in the manufacturing process, and a comprehensive data set is formed by combining environment variables. Fuzzy processing is carried out on each parameter, and a preliminary parameter combination scheme is generated by constructing a fuzzy rule base; and optimizing the parameter scheme by adopting a hierarchical progressive reasoning strategy, and finally generating a parameter combination adaptive to complex process conditions. In the production process, a local and global joint optimization method is applied, process parameters are dynamically adjusted according to data monitored in real time, it is ensured that the process parameters can be continuously optimized under different production conditions, and the multi-target production requirement is met. The method has high adaptability and flexibility, and the stability and the product quality in the manufacturing process can be remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of precision manufacturing and micro-machining, and in particular to a method for optimizing the manufacturing process of micro-sawtooth based on fuzzy logic. Background Art

[0002] In the field of precision manufacturing and micromachining, the manufacturing process of micro-sawtooth is crucial to the accuracy and performance of the product. Micro-sawtooth is widely used in high-precision cutting tools, micro-mechanical parts, and biomedical devices, so its processing accuracy and quality are extremely high. However, the existing micro-sawtooth manufacturing process has significant challenges in the optimization of process parameters, which directly affects the performance and consistency of the product.

[0003] The traditional micro-sawtooth manufacturing process mainly relies on the operator's experience and repeated experimental adjustments. This method is not only time-consuming and labor-intensive, but also due to the large number of process parameters involved in the manufacturing process and the complex nonlinear relationship between the parameters, the effect of parameter optimization is often not ideal. The deterministic optimization methods in the prior art can usually only achieve good results under specific process conditions. In the actual production process, when the process conditions change, these methods often find it difficult to ensure the consistency of processing quality. Specifically, the main defects in the prior art include the difficulty of optimizing process parameters, the lack of real-time control mechanism, insufficient adaptability, and feedback lag.

[0004] In the manufacturing process of micro-sawtooth, the optimization of process parameters is an extremely complex task. Key process parameters such as cutting speed, feed rate, tool angle and workpiece material properties have complex nonlinear relationships with each other. Existing deterministic optimization methods are usually unable to effectively handle these complex multivariate relationships. Traditional methods mainly rely on experiments and experience to adjust parameter combinations, which is not only inefficient, but also difficult to find the truly optimal parameter combination, resulting in the inability to fully guarantee the processing accuracy and quality of the product. In addition, due to the lack of sufficient intelligent means in existing methods, it is difficult to flexibly optimize parameter combinations under a wide range of process conditions. Once production conditions change, such as ambient temperature, material batch differences, etc., existing optimization methods are usually unable to respond in time, resulting in product quality fluctuations and even inconsistent quality between batches.

[0005] In terms of real-time control, the existing micro-sawtooth manufacturing process usually relies on pre-set process parameters for production, lacking the ability of real-time monitoring and dynamic adjustment. Once the production conditions change, the traditional process flow cannot adjust the process parameters in time, resulting in the inability to effectively guarantee the stability of product quality when faced with emergencies or fluctuations in the production process. This passive production mode not only increases the uncertainty in the production process, but may also lead to reduced production efficiency and increased scrap rate.

[0006] The existing technology is insufficient in adaptability. Since most traditional optimization methods are based on fixed parameter settings and lack the ability to flexibly adapt to actual process conditions, the process parameters cannot be flexibly adjusted according to actual needs during the processing of different batches or different materials. Parameter imbalance problems caused by material properties or environmental changes often occur in production, and the existing methods cannot effectively cope with these changes, which directly affects the consistency and manufacturing efficiency of the products. This lack of adaptability of the existing technology makes it difficult to maintain stability in the manufacturing process when the production scale is expanded or the product variety increases, affecting the production flexibility and market competitiveness of the enterprise. The problem of feedback lag is also very prominent in the existing technology. Traditional process optimization usually adjusts the process parameters after the production is completed by testing the finished product and finding problems. This post-adjustment method not only lags in response, but also may cause a large number of products to be unqualified or even scrapped, resulting in huge economic losses. In addition, the feedback mechanism lag also limits the frequency and depth of process optimization, and cannot track and optimize the process parameters in the production process in real time, resulting in low efficiency of process improvement and affecting the overall production quality and efficiency.

[0007] Therefore, how to provide a micro-sawtooth manufacturing process optimization method based on fuzzy logic is a problem that technicians in this field urgently need to solve. Summary of the invention

[0008] One purpose of the present invention is to propose a micro-sawtooth manufacturing process optimization method based on fuzzy logic. The present invention makes full use of the fuzzy logic system, hierarchical progressive reasoning strategy and intelligent adaptive feedback control technology, and describes in detail the optimization and real-time regulation process of process parameters. It has the advantages of strong adaptability, high process stability and good product quality consistency.

[0009] A micro-sawtooth manufacturing process optimization method based on fuzzy logic according to an embodiment of the present invention comprises the following steps:

[0010] S1. Collect key process parameters involved in the micro-sawtooth manufacturing process, including cutting speed, feed rate, tool angle and workpiece material properties, and combine environmental variables for data fusion to form a comprehensive process data set;

[0011] S2, fuzzify each parameter in the comprehensive process data set, and map the process parameters to fuzzy sets through a multi-layer membership function adjustment method, which is used to characterize the fuzzy characteristics of the parameters under different process conditions;

[0012] S3. Based on process requirements and historical manufacturing data, a fuzzy rule base is constructed to define the relationship between each process parameter and the corresponding fuzzy rules, and a preliminary parameter combination scheme is generated, which is applicable to the basic requirements under specific process conditions;

[0013] S4. Based on the generated preliminary parameter combination scheme, a hierarchical progressive reasoning strategy is adopted to input the fuzzified process parameters into the fuzzy rule library, and reasoning is performed layer by layer, from basic process parameters to process interaction parameters, and the preliminary parameter combination scheme is further optimized to generate the final parameter combination scheme to cope with complex process conditions;

[0014] S5. Select the final parameter combination scheme, apply the local-global joint optimization method, optimize the single process target, and then integrate the local optimal solution to select the parameter combination that meets multiple process targets;

[0015] S6, applying the selected parameter combination to the micro-sawtooth manufacturing process, and monitoring the process parameters and processing effects in real time during the manufacturing process, dynamically adjusting the parameter combination according to actual feedback to adapt to changes in production;

[0016] S7. Collect feedback data from the manufacturing process, analyze the process performance between batches, update the fuzzy rule base and membership function, and gradually optimize the process parameters to adapt to different process conditions and production requirements.

[0017] Optionally, the S2 specifically includes:

[0018] S21, for each parameter x in the process data set i Perform the initial membership function μ xi (x) setting, describing the fuzzy characteristics of the parameters;

[0019] S22. According to different process conditions, the parameters of the membership function are dynamically adjusted through a multi-layer membership function adjustment method to adapt to real-time process requirements, and the central value c of the membership function is adjusted. i Make adjustments:

[0020]

[0021] Among them, c i ′ is the adjusted center value, α is the adjustment coefficient, w j is the weight of the j-th layer process condition, Δx ij is the parameter x under the process conditions of the jth layer i The deviation of L is the number of layers of process conditions;

[0022] S23, standard deviation σ i Make adjustments to accommodate changes in process conditions:

[0023]

[0024] Among them, σ i ′ is the adjusted standard deviation, β is the adjustment coefficient, v j is the weight of the j-th layer process condition, x ij is the process parameter x i The actual value under the process conditions of the jth layer, x i is the current process parameter x i The actual value of

[0025] S24, shape parameter p of the membership function i Make innovative adjustments to optimize the curve shape of the membership function:

[0026]

[0027] Among them, p i ′ is the adjusted shape parameter, γ is the adjustment coefficient, λ is the sensitivity parameter, c i is the center value before adjustment;

[0028] S25, using the dynamically adjusted membership function, the process parameter x i Mapping to fuzzy sets Generate fuzzy sets that adapt to current process conditions;

[0029] S26, the fuzzy set of all process parameters Comprehensive processing to generate a comprehensive fuzzy process data set containing all process parameters Used to describe the fuzzy characteristics of the entire process system under different conditions;

[0030] S27, the generated comprehensive fuzzy process data set Prepared as input data to achieve optimized control of the micro-sawtooth manufacturing process.

[0031] Optionally, the S3 specifically includes:

[0032] S31. Based on process requirements and combined with historical manufacturing data, the generated comprehensive fuzzy process data set is used Analyze process parameters x i The correlation between them is used to construct the initial fuzzy relationship matrix M ij , used to describe the mutual influence between process parameters;

[0033] S32, the initial fuzzy relationship matrix M ij Make dynamic adjustments using an adaptive fuzzy relationship matrix based on changes in real-time process conditions:

[0034]

[0035] Among them, M′ ij is the updated fuzzy relationship matrix, η is the adaptive adjustment coefficient, Δx i and Δx j is the process parameter x i and x j deviation, σ i and σ j is the standard deviation of the corresponding parameter;

[0036] S33, based on the updated fuzzy relationship matrix M′ ij , build a fuzzy rule base to define the relationship between process parameters, and generate preliminary parameter combination schemes through these rules to meet the basic requirements under specific process conditions;

[0037] S34. In the fuzzy rule base, the process parameters are inferred and verified through a multi-layer fuzzy reasoning algorithm, so that the generated preliminary parameter combination scheme can adapt to various process conditions, and be adjusted when necessary to optimize the parameter combination:

[0038]

[0039] Among them, y k To generate the initial parameter combination scheme, is the parameter x i The normalized membership of i (M′ ij ) is the updated fuzzy relationship matrix M′ ij The output function of

[0040] S35, generate the preliminary parameter combination scheme y k Check and verify it to meet the basic requirements under different process conditions, and store the verified solution in the fuzzy rule base.

[0041] Optionally, the S4 specifically includes:

[0042] S41, based on the generated preliminary parameter combination scheme y k , and combined with the generated comprehensive fuzzy process data set The process parameter x i and its membership Input the fuzzy rule base as the initial input of hierarchical progressive reasoning;

[0043] S42. In the fuzzy rule base, a hierarchical progressive reasoning strategy is adopted. Starting from the basic process parameters, the initial parameter combination scheme of the input is optimized layer by layer. In the reasoning process of each layer, the output of each layer is dynamically adjusted:

[0044]

[0045] Among them, z k is the output result after adjustment of the kth layer, is the output result of the previous layer, is the inference result of the current layer, α k , β k and γ k are adjustment coefficients, which are used to balance the influence of the front and back layers and the interaction between parameters. is the process parameter x i The membership degree, M′ ij is the updated fuzzy relationship matrix, x j For x i Related process parameters, δ k is the weight adjustment factor;

[0046] S43. In the reasoning process, the basic process parameters are gradually extended to the process interaction parameters, and the output results of each layer are optimized layer by layer, so that the reasoning results of each layer are gradually converged in the entire reasoning chain to form a coherent optimization plan;

[0047] S44. In each layer of the reasoning process, steps S42 and S43 are repeated to further optimize the preliminary parameter combination scheme y_k, and finally generate the final parameter combination scheme yk that adapts to complex process conditions. f :

[0048]

[0049] Among them, y f is the final parameter combination solution, z k is the output result after adjustment of the kth layer, θ k is the weight coefficient associated with the kth layer, λ k is the attenuation coefficient, is the normalized membership degree, x i,ref is the process parameter x i The reference value of σ i is the standard deviation of the corresponding parameter.

[0050] Optionally, the S5 specifically includes:

[0051] S51, generate the final parameter combination scheme y f As input, enter the local-global joint optimization stage, optimize each process target separately, and consider the characteristics and target requirements of each process parameter:

[0052]

[0053] in, is the local optimal solution of the i-th process objective, is the weight of the jth parameter in the i-th process target, g j (y f ) is the final parameter combination solution y f The output result under the jth process parameter, is the reference value of the jth process parameter, σ j is the standard deviation of the jth process parameter, ∈ j is the adjustment factor, κ j is the attenuation factor;

[0054] S52. Summarize the local optimal solutions for all process objectives Generate a global optimization solution y through multi-objective integration strategy global , to meet the overall process requirements:

[0055]

[0056] Among them, y global is the integrated global optimization solution, λ i is the weight coefficient of the i-th process target, is the reference value of each local optimal solution;

[0057] S53, the generated global optimization solution y global Conduct multi-dimensional verification, comprehensively consider the weight of each process goal, actual process conditions and the mutual influence between process parameters, so that the global optimization solution can achieve a balance among multiple goals;

[0058] S54, the global optimization solution y global Introduce it into the production process, monitor process parameters and production results in real time, and make dynamic adjustments based on feedback data so that the optimization plan can continue to adapt to production needs under changing process conditions.

[0059] Optionally, the S6 specifically includes:

[0060] S61, the generated global optimization solution y global Import the micro-sawtooth manufacturing process, apply it to the production process, and record the initial process parameters and production data to form a benchmark production status;

[0061] S62. Real-time monitoring of key process parameters during the manufacturing process i and processing effects, and obtain real-time data through sensor networks and data acquisition systems t , calculate the deviation Δg between the actual output and the reference value j :

[0062]

[0063] Among them, g j (x i ) is the current process parameter x i The actual output value under is the reference output value;

[0064] S63, introduce intelligent adaptive nonlinear feedback control mechanism, based on real-time monitoring data D t , dynamically adjust the process parameters and use the nonlinear feedback mechanism to correct the process parameters so that they can adapt to the dynamic changes under complex process conditions:

[0065] x i '=x i -β j Δg j ·(1+α·tanh(λ·Δg j ));

[0066] Among them, x′ i is the adjusted process parameter, β j is the adjustment coefficient, α is the nonlinear gain coefficient, λ is the sensitivity coefficient, tanh(λ·Δg j ) is the hyperbolic tangent function;

[0067] S64, adjust the process parameter x' i Put into production, continuously monitor the processing effect, and constantly adjust the process parameters based on real-time feedback to keep it adaptable and optimized in production;

[0068] S65. During the production process, by continuously collecting feedback data, the global optimization scheme y is optimized based on the intelligent adaptive nonlinear feedback control mechanism. global Refine and improve to cope with dynamic changes in production and maintain the adaptability and stability of process parameter combinations.

[0069] Optionally, the S7 specifically includes:

[0070] S71. During the manufacturing process, based on real-time monitoring data, continuously collect production data for each batch. batch , including process parameters x i , processing effects and production environment conditions to form a batch production data set D batch ;

[0071] S72, collect the batch production data set D batch Analyze and integrate the data of each batch to form a feedback data set D feedback , identify the process parameters x i Variation between batches, by comparing parameter changes between batches to evaluate process consistency and deviation;

[0072] S73. According to the variation of the process parameters, based on the constructed fuzzy rule base, the rules are adjusted and the parameter range of the membership function is redefined to adapt it to the current production conditions and requirements;

[0073] S74, using the updated membership function and fuzzy rule base, generate a new process parameter combination y updated and apply it to production to adapt to actual process conditions;

[0074] S75. In the new production cycle, continue to collect feedback data, gradually adjust and optimize the fuzzy rule base and membership function, so that the process parameter combination remains stable and adaptable under changing production conditions.

[0075] The beneficial effects of the present invention are:

[0076] The present invention effectively solves the complexity problem caused by multivariable and nonlinear relationships in the manufacturing process by introducing fuzzy logic methods. By constructing a fuzzy rule base and adopting a hierarchical progressive reasoning strategy, the present invention can generate the optimal parameter combination under different process conditions, significantly improving the efficiency of process parameter optimization. This breakthrough overcomes the limitation of the difficulty of parameter optimization in traditional methods.

[0077] The present invention realizes real-time monitoring and dynamic adjustment of process parameters through an intelligent adaptive nonlinear feedback control mechanism. This mechanism enables process parameters to respond quickly to changes in production, ensures the stability of the production process and the consistency of product quality, and reduces production fluctuations and scrap rates. Compared with the shortcomings of feedback lag in traditional methods, the present invention greatly enhances the dynamic control capability in the production process.

[0078] The method of the present invention has strong process adaptability. Through the fuzzy logic system and the local-global joint optimization method, the process parameters can be flexibly adjusted under the production conditions of different batches or different materials, ensuring the stability of the production process and the consistency of product quality under variable process conditions. This improvement in adaptability enables the present invention to cope with a wider range of production environments and significantly improves the flexibility and reliability of production.

[0079] The present invention continuously collects and analyzes production data, continuously updates the fuzzy rule base and membership function, and forms a closed-loop optimization feedback mechanism. Compared with traditional methods, this optimization feedback mechanism can adjust the process parameter combination more frequently, so that the process performance in the production process is continuously improved, avoiding the accumulation of quality problems caused by feedback lag. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0081] Figure 1 A flowchart of a micro-sawtooth manufacturing process optimization method based on fuzzy logic proposed by the present invention;

[0082] Figure 2 A schematic diagram of the hierarchical progressive reasoning strategy;

[0083] Figure 3 Schematic diagram of the intelligent adaptive nonlinear feedback control mechanism. DETAILED DESCRIPTION

[0084] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0085] refer to Figure 1-3 , a micro-sawtooth manufacturing process optimization method based on fuzzy logic, comprising the following steps:

[0086] S1. Collect key process parameters involved in the micro-sawtooth manufacturing process, including cutting speed, feed rate, tool angle and workpiece material properties, and combine environmental variables for data fusion to form a comprehensive process data set;

[0087] S2, fuzzify each parameter in the comprehensive process data set, and map the process parameters to fuzzy sets through a multi-layer membership function adjustment method, which is used to characterize the fuzzy characteristics of the parameters under different process conditions;

[0088] S3. Based on process requirements and historical manufacturing data, a fuzzy rule base is constructed to define the relationship between each process parameter and the corresponding fuzzy rules, and a preliminary parameter combination scheme is generated, which is applicable to the basic requirements under specific process conditions;

[0089] S4. Based on the generated preliminary parameter combination scheme, a hierarchical progressive reasoning strategy is adopted to input the fuzzified process parameters into the fuzzy rule library, and reasoning is performed layer by layer, from basic process parameters to process interaction parameters, and the preliminary parameter combination scheme is further optimized to generate the final parameter combination scheme to cope with complex process conditions;

[0090] S5. Select the final parameter combination scheme, apply the local-global joint optimization method, optimize the single process target, and then integrate the local optimal solution to select the parameter combination that meets multiple process targets;

[0091] S6, applying the selected parameter combination to the micro-sawtooth manufacturing process, and monitoring the process parameters and processing effects in real time during the manufacturing process, dynamically adjusting the parameter combination according to actual feedback to adapt to changes in production;

[0092] S7. Collect feedback data from the manufacturing process, analyze the process performance between batches, update the fuzzy rule base and membership function, and gradually optimize the process parameters to adapt to different process conditions and production requirements.

[0093] In this implementation, S2 specifically includes:

[0094] S21, for each parameter x in the process data set i Perform the initial membership function μ xi (x) setting, describing the fuzzy characteristics of the parameters;

[0095] S22. According to different process conditions, the parameters of the membership function are dynamically adjusted through a multi-layer membership function adjustment method to adapt to real-time process requirements, and the central value c of the membership function is adjusted. i Make adjustments:

[0096]

[0097] Among them, c i ′ is the adjusted center value, α is the adjustment coefficient, w j is the weight of the j-th layer process condition, Δx ij is the parameter x under the process conditions of the jth layer i The deviation of L is the number of layers of process conditions;

[0098] S23, standard deviation σ i Make adjustments to accommodate changes in process conditions:

[0099]

[0100] Among them, σ i ′ is the adjusted standard deviation, β is the adjustment coefficient, v j is the weight of the j-th layer process condition, x ij is the process parameter x i The actual value under the process conditions of the jth layer, x i is the current process parameter x i The actual value of

[0101] S24, shape parameter p of the membership function i Make innovative adjustments to optimize the curve shape of the membership function:

[0102]

[0103] Among them, p i ′ is the adjusted shape parameter, γ is the adjustment coefficient, λ is the sensitivity parameter, c i is the center value before adjustment;

[0104] S25, using the dynamically adjusted membership function, the process parameter x i Mapping to fuzzy sets Generate fuzzy sets that adapt to current process conditions;

[0105] S26, the fuzzy set of all process parameters Comprehensive processing to generate a comprehensive fuzzy process data set containing all process parameters Used to describe the fuzzy characteristics of the entire process system under different conditions;

[0106] S27, the generated comprehensive fuzzy process data set Prepared as input data to achieve optimized control of the micro-sawtooth manufacturing process.

[0107] In this implementation, S3 specifically includes:

[0108] S31. Based on process requirements and combined with historical manufacturing data, the generated comprehensive fuzzy process data set is used Analyze process parameters x i The correlation between them is used to construct the initial fuzzy relationship matrix M ij , used to describe the mutual influence between process parameters;

[0109] S32, the initial fuzzy relationship matrix M ij Make dynamic adjustments using an adaptive fuzzy relationship matrix based on changes in real-time process conditions:

[0110]

[0111] Among them, M′ ij is the updated fuzzy relationship matrix, η is the adaptive adjustment coefficient, Δx i and Δx j is the process parameter x i and x j deviation, σ i and σ j is the standard deviation of the corresponding parameter;

[0112] S33, based on the updated fuzzy relationship matrix M′ ij , build a fuzzy rule base to define the relationship between process parameters, and generate preliminary parameter combination schemes through these rules to meet the basic requirements under specific process conditions;

[0113] S34. In the fuzzy rule base, the process parameters are inferred and verified through a multi-layer fuzzy reasoning algorithm, so that the generated preliminary parameter combination scheme can adapt to various process conditions, and be adjusted when necessary to optimize the parameter combination:

[0114]

[0115] Among them, y k To generate the initial parameter combination scheme, is the parameter x i The normalized membership of i (M′ ij ) is the updated fuzzy relationship matrix M′ ij The output function of

[0116] S35, generate the initial parameter combination scheme y k Check and verify it to meet the basic requirements under different process conditions, and store the verified solution in the fuzzy rule base.

[0117] In this implementation manner, the S4 specifically includes:

[0118] S41, based on the generated preliminary parameter combination scheme y k , and combined with the generated comprehensive fuzzy process data set The process parameter x i and its membership Input the fuzzy rule base as the initial input of hierarchical progressive reasoning;

[0119] S42. In the fuzzy rule base, a hierarchical progressive reasoning strategy is adopted. Starting from the basic process parameters, the initial parameter combination scheme of the input is optimized layer by layer. In the reasoning process of each layer, the output of each layer is dynamically adjusted:

[0120]

[0121] Among them, z k is the output result after adjustment of the kth layer, is the output result of the previous layer, y Lk is the inference result of the current layer, α k , β k and γ k are adjustment coefficients, which are used to balance the influence of the front and back layers and the interaction between parameters. is the process parameter x i The membership degree, M′ ij is the updated fuzzy relationship matrix, x j For x i Related process parameters, δ k is the weight adjustment factor;

[0122] S43. In the reasoning process, the basic process parameters are gradually extended to the process interaction parameters, and the output results of each layer are optimized layer by layer, so that the reasoning results of each layer are gradually converged in the entire reasoning chain to form a coherent optimization plan;

[0123] S44. In each layer of the reasoning process, steps S42 and S43 are repeated to further optimize the preliminary parameter combination scheme y_k, and finally generate the final parameter combination scheme yk that adapts to complex process conditions. f :

[0124]

[0125] Among them, y f is the final parameter combination solution, z k is the output result after adjustment of the kth layer, θ k is the weight coefficient associated with the kth layer, λ k is the attenuation coefficient, is the normalized membership degree, x i,ref is the process parameter x i The reference value of σ i is the standard deviation of the corresponding parameter.

[0126] In this implementation manner, S5 specifically includes:

[0127] S51, generate the final parameter combination scheme y f As input, enter the local-global joint optimization stage, optimize each process target separately, and consider the characteristics and target requirements of each process parameter:

[0128]

[0129] in, is the local optimal solution of the i-th process objective, is the weight of the jth parameter in the i-th process target, g j (y f ) is the final parameter combination solution y f The output result under the jth process parameter, is the reference value of the jth process parameter, σ j is the standard deviation of the jth process parameter, ∈ j is the adjustment factor, κ j is the attenuation factor;

[0130] S52. Summarize the local optimal solutions for all process objectives Generate a global optimization solution y through multi-objective integration strategy global, to meet the overall process requirements:

[0131]

[0132] Among them, y global is the integrated global optimization solution, λ i is the weight coefficient of the i-th process target, is the reference value of each local optimal solution;

[0133] S53, the generated global optimization solution y global Conduct multi-dimensional verification, comprehensively consider the weight of each process goal, actual process conditions and the mutual influence between process parameters, so that the global optimization solution can achieve a balance among multiple goals;

[0134] S54, the global optimization solution y global Introduce it into the production process, monitor process parameters and production results in real time, and make dynamic adjustments based on feedback data so that the optimization plan can continue to adapt to production needs under changing process conditions.

[0135] In this implementation manner, S6 specifically includes:

[0136] S61, the generated global optimization solution y global Import the micro-sawtooth manufacturing process, apply it to the production process, and record the initial process parameters and production data to form a benchmark production status;

[0137] S62. Real-time monitoring of key process parameters during the manufacturing process i and processing effects, and obtain real-time data through sensor networks and data acquisition systems t , calculate the deviation Δg between the actual output and the reference value j :

[0138]

[0139] Among them, g j (x i ) is the current process parameter x i The actual output value under is the reference output value;

[0140] S63, introduce intelligent adaptive nonlinear feedback control mechanism, based on real-time monitoring data D t , dynamically adjust the process parameters and use the nonlinear feedback mechanism to correct the process parameters so that they can adapt to the dynamic changes under complex process conditions:

[0141] x i '=x i -β j Δg j·(1+α·tanh(λ·Δg j ));

[0142] Among them, x′ i is the adjusted process parameter, β j is the adjustment coefficient, α is the nonlinear gain coefficient, λ is the sensitivity coefficient, tanh(λ·Δg j ) is the hyperbolic tangent function;

[0143] S64, adjust the process parameter x' i Put into production, continuously monitor the processing effect, and constantly adjust the process parameters based on real-time feedback to keep it adaptable and optimized in production;

[0144] S65. During the production process, by continuously collecting feedback data, the global optimization scheme y is optimized based on the intelligent adaptive nonlinear feedback control mechanism. global Refine and improve to cope with dynamic changes in production and maintain the adaptability and stability of process parameter combinations.

[0145] In this implementation manner, the S7 specifically includes:

[0146] S71. During the manufacturing process, based on real-time monitoring data, continuously collect production data for each batch. batch , including process parameters x i , processing effects and production environment conditions to form a batch production data set D batch ;

[0147] S72, collect the batch production data set D batch Analyze and integrate the data of each batch to form a feedback data set D feedback , identify the process parameters x i Variation between batches, by comparing parameter changes between batches to evaluate process consistency and deviation;

[0148] S73. According to the variation of the process parameters, based on the constructed fuzzy rule base, the rules are adjusted and the parameter range of the membership function is redefined to adapt it to the current production conditions and requirements;

[0149] S74, using the updated membership function and fuzzy rule base, generate a new process parameter combination y updated and apply it to production to adapt to actual process conditions;

[0150] S75. In the new production cycle, continue to collect feedback data, gradually adjust and optimize the fuzzy rule base and membership function, so that the process parameter combination remains stable and adaptable under changing production conditions.

[0151] Embodiment 1:

[0152] In order to verify the feasibility of the present invention in implementation, the present invention is applied in a precision manufacturing enterprise that focuses on the production of micro-sawtooth cutting tools for medical devices. Since medical devices have extremely high requirements for processing accuracy and consistency, the manufacturing quality of micro-sawtooth is directly related to the performance and safety of the product. However, in the past production process, the company faced a series of problems such as difficult process parameter optimization, unstable production process, and large quality fluctuations between product batches. These problems are mainly due to the large number of process parameters involved in the manufacturing process, and the complex nonlinear relationship between the parameters, which is difficult to deal with by traditional deterministic optimization methods. In addition, due to the lack of real-time control and feedback mechanism in the production process, the process parameters cannot be adjusted in time, resulting in low production efficiency and high scrap rate. In order to solve these problems, the company decided to introduce the micro-sawtooth manufacturing process optimization method based on fuzzy logic proposed in the present invention, aiming to improve the overall production efficiency and product quality by optimizing process parameters, enhancing real-time control capabilities and improving the stability of the production process.

[0153] The company applied the method of the present invention on the production line of micro-sawtooth manufacturing. First, through the sensor network installed on the production equipment, the company can collect key process parameters including cutting speed, feed rate, tool angle and workpiece material characteristics in real time, and combine environmental variables such as ambient temperature and humidity to form a comprehensive process data set. These data are mapped to the corresponding fuzzy sets through fuzzy processing to flexibly describe the parameter characteristics under different process conditions.

[0154] Subsequently, the company built a fuzzy rule base covering multivariate relationships based on historical manufacturing data and expert experience. This fuzzy rule base is used to define the mutual influence between process parameters and generate preliminary parameter combination schemes. After generating the preliminary scheme, the company adopted a hierarchical progressive reasoning strategy, which extended simple process parameter reasoning to more complex process interactions through layer-by-layer optimization, and finally generated the optimal parameter combination that adapts to actual production conditions.

[0155] During the production process, the company introduced an intelligent adaptive nonlinear feedback control mechanism. This mechanism monitors the actual output of process parameters in real time and compares it with the expected value. When the actual output is found to deviate from the expected value, the system automatically adjusts the process parameters to ensure the stability of the production process. For example, in one production batch, the sensor detected that the combination of the cutting speed and feed rate of the tool failed to achieve the expected effect due to differences in material batches. The intelligent feedback mechanism responded immediately and adjusted the cutting speed and feed rate in real time, ensuring the continuous operation of the production line while avoiding the accumulation of quality problems.

[0156] By continuously collecting feedback data from each production batch, the company can dynamically adjust the fuzzy rule base and membership function so that the process parameter combination can better adapt to the changes in different batches and materials. Such a closed-loop optimization mechanism not only improves the adaptability of process parameters, but also significantly reduces production fluctuations and scrap rates.

[0157] In order to verify the effectiveness of the method of the present invention, the company conducted detailed data collection and analysis on the production line using the method of the present invention from January to June 2023; before the introduction of the method of the present invention, in the second half of 2022, the average scrap rate of the company's micro-sawtooth cutting tools was 7.8%, the quality consistency of each batch of products was poor, and the processing accuracy fluctuation between batches reached ±10 microns. During the production process, the adjustment of process parameters often relies on manual experience, resulting in a long downtime of the production line, with an average downtime of 12 hours per month due to the adjustment of process parameters.

[0158] After the introduction of the method of the present invention, the scrap rate of the company's micro-sawtooth cutting tools dropped significantly from January to June 2023, with an average scrap rate of 2.3%, of which the scrap rate in March was as low as 1.9%. The fluctuation of machining accuracy between batches was reduced to ±4 microns, and the consistency of product quality was significantly improved. In addition, due to the introduction of the intelligent adaptive feedback control mechanism, the real-time adjustment of process parameters is more efficient, the downtime of the production line is reduced to 4 hours per month, and the production efficiency is significantly improved.

[0159] During this period, the company also conducted several comparative experiments to verify the adaptability of the method of the present invention under different time periods and different environmental conditions. In the case of large temperature changes, the method of the present invention can quickly adjust the process parameters to ensure the stability of the production process and the consistency of product quality. When faced with production tasks with different material properties, the method of the present invention can also automatically optimize the parameter combination based on real-time monitoring data, so that the quality differences between different batches of products are minimized.

[0160] Table 1 Comparison of scrap rate, machining accuracy fluctuation and downtime in the second half of 2022 and January-June 2023

[0161]

[0162] Table 2 Comparison of batch quality consistency and scrap rate under different material characteristics

[0163]

[0164] Table 3 Comparison of process parameter adjustment efficiency and production stability under different environmental conditions

[0165]

[0166]

[0167] Through the comparative analysis of data from the second half of 2022 and January to June 2023, it can be clearly seen that the application of the method of the present invention in the micro-sawtooth manufacturing process has significantly improved various key indicators of the production process. After the introduction of the method of the present invention, the scrap rate was significantly reduced, from an average of 7.8% in 2022 to 2.3% in 2023, with the lowest scrap rate in March, reaching 1.9%. The machining accuracy fluctuation range was also reduced from the previous ±10 microns to ±4 microns, significantly improving the quality consistency of the product. In comparative experiments under different material properties and environmental conditions, the method of the present invention also demonstrated strong adaptability. In 2023, whether in an environment with large temperature fluctuations or in the face of different materials, the adjustment of process parameters was faster and more effective, ensuring production stability and product quality. Specifically, the machining accuracy fluctuations and scrap rates of materials A, B, and C under different temperature fluctuation ranges were significantly better than those in 2022, further proving the advantages of the method of the present invention in dealing with complex production conditions. The method of the present invention significantly reduces the downtime of the production line. Through the intelligent adaptive nonlinear feedback control mechanism, the real-time optimization of process parameters is achieved, reducing the downtime from 12 hours per month in 2022 to about 4 hours in 2023. This not only improves production efficiency, but also reduces production costs, making the overall production process smoother and more efficient.

[0168] In summary, through the implementation of the method of the present invention, the enterprise has successfully solved the problems of difficult process parameter optimization, unstable production process, large fluctuations in product quality, etc., and greatly improved production efficiency and product quality. In practical applications, the method of the present invention has shown strong adaptability, flexibility and reliability, providing strong support for the competitiveness of enterprises in the field of micro-sawtooth manufacturing.

[0169] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the manufacturing process of micro-sawtooth based on fuzzy logic, characterized in that: The steps include: S1. Collect key process parameters involved in the micro-sawtooth manufacturing process, including cutting speed, feed rate, tool angle and workpiece material properties, and combine environmental variables for data fusion to form a comprehensive process data set; S2, fuzzify each parameter in the comprehensive process data set, and map the process parameters to fuzzy sets through a multi-layer membership function adjustment method, which is used to characterize the fuzzy characteristics of the parameters under different process conditions; S3. Based on process requirements and historical manufacturing data, a fuzzy rule base is constructed to define the relationship between each process parameter and the corresponding fuzzy rules, and a preliminary parameter combination scheme is generated, which is applicable to the basic requirements under specific process conditions; S4. Based on the generated preliminary parameter combination scheme, a hierarchical progressive reasoning strategy is adopted to input the fuzzified process parameters into the fuzzy rule library, and reasoning is performed layer by layer, from basic process parameters to process interaction parameters, and the preliminary parameter combination scheme is further optimized to generate the final parameter combination scheme to cope with complex process conditions; S5. Select the final parameter combination scheme, apply the local-global joint optimization method, optimize the single process target, and then integrate the local optimal solution to select the parameter combination that meets multiple process targets; S6, applying the selected parameter combination to the micro-sawtooth manufacturing process, and monitoring the process parameters and processing effects in real time during the manufacturing process, dynamically adjusting the parameter combination according to actual feedback to adapt to changes in production; S7. Collect feedback data from the manufacturing process, analyze the process performance between batches, update the fuzzy rule base and membership function, and gradually optimize the process parameters to adapt to different process conditions and production requirements.

2. The micro-sawtooth manufacturing process optimization method based on fuzzy logic according to claim 1, characterized in that: The S2 specifically includes: S21, for each parameter x in the process data set i Perform the initial membership function μ xi (x) setting, describing the fuzzy characteristics of the parameters; S22. According to different process conditions, the parameters of the membership function are dynamically adjusted through a multi-layer membership function adjustment method to adapt to real-time process requirements, and the central value c of the membership function is adjusted. i Make adjustments: Among them, c i ′ is the adjusted center value, α is the adjustment coefficient, w j is the weight of the j-th layer process condition, Δx ij is the parameter x under the process conditions of the jth layer i The deviation of L is the number of layers of process conditions; S23, standard deviation σ i Make adjustments to accommodate changes in process conditions: Among them, σ i ′ is the adjusted standard deviation, β is the adjustment coefficient, v j is the weight of the j-th layer process condition, x ij is the process parameter x i The actual value under the process conditions of the jth layer, x i is the current process parameter x i The actual value of S24, shape parameter p of the membership function i Make innovative adjustments to optimize the curve shape of the membership function: Among them, p i ′ is the adjusted shape parameter, γ is the adjustment coefficient, λ is the sensitivity parameter, c i is the center value before adjustment; S25, using the dynamically adjusted membership function, the process parameter x i Mapping to fuzzy sets Generate fuzzy sets that adapt to current process conditions; S26, the fuzzy set of all process parameters Comprehensive processing to generate a comprehensive fuzzy process data set containing all process parameters Used to describe the fuzzy characteristics of the entire process system under different conditions; S27, the generated comprehensive fuzzy process data set Prepared as input data to achieve optimized control of the micro-sawtooth manufacturing process.

3. The micro-sawtooth manufacturing process optimization method based on fuzzy logic according to claim 1, characterized in that: The S3 specifically includes: S31. Based on process requirements and combined with historical manufacturing data, the generated comprehensive fuzzy process data set is used Analyze process parameters x i The correlation between them is used to construct the initial fuzzy relationship matrix M ij , used to describe the mutual influence between process parameters; S32, the initial fuzzy relationship matrix M ij Make dynamic adjustments using an adaptive fuzzy relationship matrix based on changes in real-time process conditions: Among them, M′ ij is the updated fuzzy relationship matrix, η is the adaptive adjustment coefficient, Δx i and Δx j is the process parameter x i and x j deviation, σ i and σ j is the standard deviation of the corresponding parameter; S33, based on the updated fuzzy relationship matrix M′ ij , build a fuzzy rule base to define the relationship between process parameters, and generate preliminary parameter combination schemes through these rules to meet the basic requirements under specific process conditions; S34. In the fuzzy rule base, the process parameters are inferred and verified through a multi-layer fuzzy reasoning algorithm, so that the generated preliminary parameter combination scheme can adapt to various process conditions, and be adjusted when necessary to optimize the parameter combination: Among them, y k is the initial parameter combination scheme generated, μ′ xi (x i ) is the parameter x i The normalized membership of i (M′ ij ) is the updated fuzzy relationship matrix M′ ij The output function of S35, generate the preliminary parameter combination scheme y k Check and verify it to meet the basic requirements under different process conditions, and store the verified solution in the fuzzy rule base.

4. The method for optimizing the manufacturing process of micro-sawtooth based on fuzzy logic according to claim 1, characterized in that: The S4 specifically includes: S41, based on the generated preliminary parameter combination scheme y k , and combined with the generated comprehensive fuzzy process data set The process parameter x i and its membership Input the fuzzy rule base as the initial input of hierarchical progressive reasoning; S42. In the fuzzy rule base, a hierarchical progressive reasoning strategy is adopted. Starting from the basic process parameters, the initial parameter combination scheme of the input is optimized layer by layer. In the reasoning process of each layer, the output of each layer is dynamically adjusted: Among them, z k is the output result after adjustment of the kth layer, is the output result of the previous layer, is the inference result of the current layer, α k , β k and γ k are adjustment coefficients, which are used to balance the influence of the front and back layers and the interaction between parameters. is the process parameter x i The membership degree, M′ ij is the updated fuzzy relationship matrix, x j For x i Related process parameters, δ k is the weight adjustment factor; S43. In the reasoning process, the basic process parameters are gradually extended to the process interaction parameters, and the output results of each layer are optimized layer by layer, so that the reasoning results of each layer are gradually converged in the entire reasoning chain to form a coherent optimization plan; S44. In each layer of the reasoning process, steps S42 and S43 are repeated to further optimize the preliminary parameter combination scheme y_k, and finally generate the final parameter combination scheme yk that adapts to complex process conditions. f : Among them, y f is the final parameter combination solution, z k is the output result after adjustment of the kth layer, θ k is the weight coefficient associated with the kth layer, λ k is the attenuation coefficient, is the normalized membership degree, x i,ref is the process parameter x i The reference value of σ i is the standard deviation of the corresponding parameter.

5. The method for optimizing the micro-sawtooth manufacturing process based on fuzzy logic according to claim 1, characterized in that: The S5 specifically includes: S51, generate the final parameter combination scheme y f As input, enter the local-global joint optimization stage, optimize each process target separately, considering the characteristics and target requirements of each process parameter: in, is the local optimal solution of the i-th process objective, is the weight of the jth parameter in the i-th process target, g j (y f ) is the final parameter combination solution y f The output result under the jth process parameter, is the reference value of the jth process parameter, σ j is the standard deviation of the jth process parameter, ∈ j is the adjustment factor, κ j is the attenuation factor; S52. Summarize the local optimal solutions for all process objectives Generate a global optimization solution y through multi-objective integration strategy global , to meet the overall process requirements: Among them, y global is the integrated global optimization solution, λ i is the weight coefficient of the i-th process target, is the reference value of each local optimal solution; S53, the generated global optimization solution y global Conduct multi-dimensional verification, comprehensively consider the weight of each process goal, actual process conditions and the mutual influence between process parameters, so that the global optimization solution can achieve a balance among multiple goals; S54, the global optimization solution y global Introduce it into the production process, monitor process parameters and production results in real time, and make dynamic adjustments based on feedback data so that the optimization plan can continue to adapt to production needs under changing process conditions.

6. The method for optimizing the manufacturing process of micro-sawtooth based on fuzzy logic according to claim 1, characterized in that: The S6 specifically includes: S61, the generated global optimization solution y global Import the micro-sawtooth manufacturing process, apply it to the production process, and record the initial process parameters and production data to form a benchmark production status; S62. Real-time monitoring of key process parameters during the manufacturing process i and processing effects, and obtain real-time data through sensor networks and data acquisition systems t , calculate the deviation Δg between the actual output and the reference value j : Among them, g j (x i ) is the current process parameter x i The actual output value under is the reference output value; S63, introduce intelligent adaptive nonlinear feedback control mechanism, based on real-time monitoring data D t , dynamically adjust the process parameters and use the nonlinear feedback mechanism to correct the process parameters so that they can adapt to the dynamic changes under complex process conditions: Among them, x′ i is the adjusted process parameter, β j is the adjustment coefficient, α is the nonlinear gain coefficient, λ is the sensitivity coefficient, tanh(λ·Δg j ) is the hyperbolic tangent function; S64, adjust the process parameter x' i Put into production, continuously monitor the processing effect, and constantly adjust the process parameters based on real-time feedback to keep it adaptable and optimized in production; S65. During the production process, by continuously collecting feedback data, the global optimization scheme y is optimized based on the intelligent adaptive nonlinear feedback control mechanism. global Refine and improve to cope with dynamic changes in production and maintain the adaptability and stability of process parameter combinations.

7. The method for optimizing the manufacturing process of micro-sawtooth based on fuzzy logic according to claim 1, characterized in that: The S7 specifically includes: S71. During the manufacturing process, based on real-time monitoring data, continuously collect production data for each batch. batch , including process parameters x i , processing effects and production environment conditions to form a batch production data set D batch ; S72, collect the batch production data set D batch Analyze and integrate the data of each batch to form a feedback data set D feedback , identify the process parameters x i Variation between batches, by comparing parameter changes between batches to evaluate process consistency and deviation; S73. According to the variation of the process parameters, based on the constructed fuzzy rule base, the rules are adjusted and the parameter range of the membership function is redefined to adapt it to the current production conditions and requirements; S74, using the updated membership function and fuzzy rule base, generate a new process parameter combination y updated and apply it to production to adapt to actual process conditions; S75. In the new production cycle, continue to collect feedback data, gradually adjust and optimize the fuzzy rule base and membership function, so that the process parameter combination remains stable and adaptable under changing production conditions.

Citation Information

Patent Citations

  • Fuzzy control method and system for cement grinder feed quantity

    CN106950836A

  • Boiler induced draft adjusting system based on self-adaptive fuzzy control and control method

    CN108803327A

  • Turning process parameter adaptive energy efficiency optimization method based on knowledge-driving

    CN109799706A

  • Photovoltaic module production control method and system based on fuzzy control

    CN118550262A

  • Fuzzy tuning device

    JP1993204410A

Cited By

  • Bucket arm robot multi-scene grinding method and system based on adaptive parameter mapping

    CN119772664A

  • Saw chain blade limiting angle automatic optimization method and system

    CN120217872A

  • A method and system for automatically optimizing the limiting angle of a saw chain blade

    CN120217872B

  • Self-adaptive control method, device and equipment for linear cutting machining process

    CN120619499A

  • Standardized optimization method for multi-process-parameter cooperative surface treatment of aeronautical parts

    CN120822282A