Optimization Method for the Production Process of the Security Interface of Electronic Devices Based on Fuzzy Logic

Abstract

The present invention discloses an optimization method for the production process of electronic device security interfaces based on fuzzy logic, which includes the following steps: S1, designing and constructing a multi-level fuzzy rule base, where the multi-level fuzzy rule base includes a first level, a second level, and a third level; S2, using a genetic algorithm in the first level to generate and optimize the initial fuzzy rule base; S3, using a particle swarm optimization algorithm in the second level to optimize the fuzzy rule base; S4, using an ant colony algorithm in the third level to optimize the specific process parameters in the fuzzy rule base; S5, through data acquisition and monitoring technology, obtaining the data in the production process in real time, constructing a data set, and transmitting the data set to the central processing unit; S6, constructing a personalized fuzzy rule base; S7, displaying the key data and optimization results in the production process through visualization technology. The present invention utilizes a multi-level fuzzy rule base and intelligent optimization algorithms to achieve efficient optimization of the production process of electronic device security interfaces.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic device manufacturing, and particularly to an optimization method for the production process of a safety interface of an electronic device based on fuzzy logic. Background Art

[0002] With the continuous development of electronic device manufacturing technology, the functional complexity and integration degree of electronic devices are also increasing continuously. Especially in the production process involving safety interfaces, the requirements for process control are becoming more and more strict. These safety interfaces not only need to ensure the stable operation of electronic devices under various environmental conditions, but also need to meet the requirements of high reliability, high performance and low energy consumption. To achieve these goals, the optimization of the production process is particularly important. However, in the prior art, there are still many deficiencies in the optimization methods for the production process of electronic device safety interfaces, and they cannot fully meet the needs of modern electronic manufacturing for efficient and high-precision production control.

[0003] In traditional production processes, fixed rules or simple experience-based adjustment methods are usually used to control production parameters. Although these methods can play a role in some cases, since the process parameters in the production process involve many variables, such as temperature, pressure, material flow rate, equipment load, etc., and the relationships between these variables are complex and uncertain, it is difficult for traditional methods to maintain the best process control state in a dynamically changing production environment. In addition, when dealing with changes in production requirements, fluctuations in equipment status or external environmental disturbances, traditional methods often lack flexibility and adaptability, which easily lead to problems such as low production efficiency, unstable product quality and excessive energy consumption.

[0004] To make up for these deficiencies, in recent years, some intelligent process control methods have been gradually applied to the field of electronic device manufacturing, such as intelligent optimization algorithms like fuzzy logic control, genetic algorithm optimization, particle swarm optimization and ant colony optimization. However, these methods usually only optimize at a single level or within a local range, lacking a global consideration of the entire production process. For example, although fuzzy logic can handle uncertainty and ambiguity, the design and optimization of its rule base often rely on expert experience and it is difficult to adapt to a complex and changeable production environment. Although intelligent algorithms such as genetic algorithms, particle swarm optimization and ant colony optimization have strong global search capabilities, in the production process, due to the lack of comprehensive consideration of multiple levels and multiple objectives, it is difficult to achieve comprehensive optimization from the global to the details. In addition, the data processing methods in the prior art are mostly static processing and cannot dynamically adapt to the volatility and real-time nature of data in the production process, resulting in limited response speed and control accuracy of the system.

[0005] In the prior art, there are also some limitations in data acquisition and monitoring technologies. Although the development of industrial Internet of Things and sensor technologies has made real-time data acquisition possible, this data is usually not optimized before being directly applied to the production control system, which may lead to information overload or data noise affecting system decision-making. In addition, most data standardization processes adopt fixed standardization formulas and cannot adapt to the dynamic characteristics of data changing over time. When the production environment changes rapidly, this is likely to result in inadaptability of control parameters, thus affecting production efficiency and product quality.

[0006] Based on the above analysis, the prior art mainly has the following defects in the production process optimization of electronic device safety interfaces. First, there is a lack of a multi-level optimization strategy that combines the global and local aspects, making it difficult to achieve global optimization in a complex production environment. Second, the application of existing intelligent optimization algorithms is mostly limited to single-objective optimization, lacking the ability of multi-objective comprehensive optimization and unable to balance production speed, product quality, and energy consumption. Third, data processing technologies are insufficient in dealing with the volatility and dynamic characteristics of real-time data, and traditional static data processing methods are difficult to provide real-time and accurate production control support. Finally, existing production control systems lack personalized optimization strategies when dealing with equipment differences and production task diversity and cannot be flexibly adjusted according to the status of specific equipment or the requirements of production tasks.

[0007] Therefore, how to provide a production process optimization method for electronic device safety interfaces based on fuzzy logic is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0008] An object of the present invention is to propose a production process optimization method for electronic device safety interfaces based on fuzzy logic. By introducing intelligent optimization technologies such as a multi-level fuzzy rule base, genetic algorithm, particle swarm optimization algorithm, and ant colony algorithm, the present invention realizes the global and local optimization of the production process of electronic device safety interfaces. The system can dynamically adapt to changes in the production environment and accurately control key parameters in the production process. At the same time, the present invention adopts an adaptive dynamic data processing technology and the construction of a personalized rule base to ensure the efficient operation of the production process under multi-objective optimization conditions, and has the advantages of strong self-adaptability, high precision, low energy consumption, and high production efficiency.

[0009] According to an embodiment of the present invention, a production process optimization method for electronic device safety interfaces based on fuzzy logic includes the following steps:

[0010] S1. Design and construct a multi-level fuzzy rule base, where the multi-level fuzzy rule base includes a first level, a second level, and a third level;

[0011] S2. The first layer uses a genetic algorithm to generate and optimize an initial fuzzy rule base. The input variables include the total production demand, the overall health status of the equipment, the energy consumption rate, and the production duration. The output variables include the global production speed setting, the equipment utilization rate allocation, and the energy consumption optimization strategy, which are used to achieve global control of the overall production speed, energy consumption, and output balance.

[0012] S3. The second layer uses a particle swarm optimization algorithm to optimize the fuzzy rule base. The input variables include the temperature, pressure, material flow rate, and production line load of each production line. The output variables include the production line speed adjustment, temperature setting, and pressure regulation, which are used to optimize the operating parameters of a single production line and ensure that each production line operates in the best state.

[0013] S4. The third layer uses an ant colony algorithm to optimize the specific process parameters in the fuzzy rule base. The input variables include the batching ratio, the fine-tuning parameters of the equipment, and the process error. The output variables include the ratio optimization, the equipment fine-tuning instruction, and the process parameter correction, which are used to precisely control the production process parameters and ensure further improvement of product quality and production efficiency.

[0014] S5. Through data acquisition and monitoring technology, data in the production process is obtained in real time, a multi-dimensional data set is constructed, and the multi-dimensional data set is transmitted to the central processing unit.

[0015] S6. Use the multi-dimensional data set to construct a personalized fuzzy rule base, automatically generate and optimize an exclusive rule base according to the individual differences of the equipment and task classification, and perform adaptive adjustment according to real-time data during the production process.

[0016] S7. Display the key data and optimization results in the production process through visualization technology, generate a real-time graphical interface, and support quick decision-making and problem troubleshooting.

[0017] Optionally, S2 specifically includes:

[0018] S21. Initialize the population to generate a number of candidate rule sets. Each rule in the candidate rule set is composed of a combination of input variables and output variables. The input variables include the total production demand D total , the overall health status H equipment , the energy consumption rate E rate , and the production duration T production , and the output variables include the global production speed setting S global , the equipment utilization rate allocation U utilization , and the energy consumption optimization strategy C consumption ;

[0019] S22. Evaluate the fitness of each candidate rule set, and calculate the fitness value of each rule set through the fitness function f(x):

[0020]

[0021] Among them, w 1 、w 2 、w 3 and w 4 represent pre-set weight coefficients, and f(x) represents the fitness value of the rule set in terms of balancing production speed, energy consumption, and output. The larger the value, the better the optimization effect of the rule set;

[0022] S23. Perform a selection operation based on the fitness value, retain the rule set with a higher fitness value, and perform a crossover operation. The crossover operation generates a new rule set by exchanging the values of some input variables and output variables in two rule sets;

[0023] S24. Perform a mutation operation on the rule set after the selection and crossover operations. The mutation operation dynamically adjusts the mutation amplitude based on the population fitness:

[0024]

[0025] Among them, X i,j represents the j-th variable of the i-th rule set after mutation, X i,j represents the j-th variable of the i-th rule set before mutation, δ(t) represents the mutation amplitude coefficient, which is dynamically adjusted with the generation number t, rand(-1, 1) represents a random number between -1 and 1, f avg represents the average fitness of the current population, and f max represents the maximum fitness in the current population;

[0026] is used to control the mutation intensity. When the overall fitness of the population tends to be stable, the mutation intensity will gradually weaken, thereby improving the convergence and stability of the algorithm;

[0027] S25. Repeat steps S22 - S24 until the fitness value of the rule set in the population converges to a preset threshold or reaches a predetermined number of iterations, and finally obtain an optimized fuzzy rule base;

[0028] S26. Apply the optimized fuzzy rule base to the production process to adjust the global production speed setting S global 、equipment utilization rate allocation U utilization and energy consumption optimization strategy C consumption .

[0029] Optionally, the specific content of S3 includes:

[0030] S31. Initialize the particle swarm, set the size of the particle swarm to N, and randomly generate the initial position X of each particle iand velocity V i , where i represents the particle number, and the initial position X i represents the temperature T of each production line i , pressure P i , material flow rate F i and production line load L i of the initial values, and the initial velocity V i represents the temperature T of each production line i , pressure P i , material flow rate F i and production line load L i of the initial change rate;

[0031] S32. Define the fitness function f 1 (X i ), which is used to evaluate the operating state of the production line corresponding to the current position X i of each particle:

[0032]

[0033] where T opt represents the optimal target value of temperature, P opt represents the optimal target value of pressure, F opt represents the optimal target value of material flow rate, L opt represents the optimal target value of production line load, v 1 , v 2 , v 3 and v 4 represent the weight coefficients;

[0034] S33. According to the fitness function value, update the individual optimal position A i (t) and the global optimal position G(t) of each particle, record the fitness value of each particle in the current iteration, and update the position of the particle with a better fitness value to the individual optimal position A i (t), and at the same time update the optimal position among all particles to the global optimal position G(t);

[0035] S34. Update the position and velocity of the particle. The update formulas for the velocity V i (t + 1) and the position X i (t + 1) are:

[0036] V i (t + 1) = ω·V i (t) + c 1 ·r 1 ·(A i (t) - X i (t)) + c 2 ·r2 ·(G(t) - X i (t));

[0037] X i (t + 1) = X i (t) + V i (t + 1);

[0038] Where ω represents the inertia weight, which is used to control the influence of the current velocity of the particle, and c 1 and c 2 represent the learning factors, which represent the influence of the particle's own experience and the group experience respectively, and r 1 and r 2 represent random numbers between 0 and 1;

[0039] S35. In each iteration, the inertia weight ω is dynamically adjusted, and as the number of iterations increases, ω is gradually decreased. At the same time, c 1 and c 2 are adaptively adjusted to enhance the collaborative optimization ability of the particle swarm;

[0040] S36. Iteratively update the particle swarm. Through multiple iterations, the particle swarm gradually converges until the fitness function f 1 (X i ) converges to the set threshold or reaches the maximum number of iterations, and finally finds the particle position X 1 (X i ) that minimizes the fitness function f i , which is used as the optimized production line operation parameters, including the optimal production line speed adjustment, temperature setting, and pressure regulation.

[0041] Optionally, the S4 specifically includes:

[0042] S41. Initialize the ant colony, set the scale M of the ant colony and the maximum number of iterations C max , randomly distribute the initial pheromone concentration τ ij (0). Each ant represents a process parameter combination X j , including the ingredient ratio R j , the equipment fine-tuning parameter A j and the process error E j , where i represents the ant number and j represents the parameter combination number;

[0043] S42. Define the multi-objective fitness function f 2 (X j ), which is used to evaluate the influence of each parameter combination X j on the production process:

[0044]

[0045] Among them, R opt represents the optimal target value of the ingredient ratio, A opt represents the optimal target value of the equipment fine-tuning parameter, E opt represents the optimal target value of the process error, α 1 , β 1 and γ 1 represent the weight coefficients;

[0046] S43. Calculate the selection probability P ij (t) of each ant on each path based on the entropy-based adaptive selection mechanism. The selection probability is based on the pheromone concentration τ ij (t), the heuristic factor η ij and the uncertainty H ij (t) of the current path as follows:

[0047]

[0048] Among them, α(t) represents the weight coefficient of the pheromone concentration, β(t) represents the weight coefficient of the heuristic factor, γ(t) represents the weight factor of the information entropy, k represents the index of other paths available for the ant to choose, allowed represents the set of paths not yet selected by the current ant, t represents the iteration number, τ ij (t) represents the current pheromone concentration on path ij, η ij represents the heuristic factor, H ij (t) represents the information entropy on path ij, which is used to describe the uncertainty of path selection:

[0049]

[0050] Among them, l represents the index variable, P il (t) represents the probability that the ant selects path il;

[0051] S44. Each ant selects a path according to the selection probability P ij (t), determines a process parameter combination X j , and evaluates the effect through the multi-objective fitness function f 2 (X j ), and records the best process parameter combination found by each ant;

[0052] S45. Update the pheromone concentration, and adopt an adaptive pheromone evaporation mechanism and dynamic incremental update to update the pheromone concentration τ ij (t + 1) on each path:

[0053] τ ij (t + 1) = (1 - ρ(t))·τ ij (t) + Δτ ij (t);

[0054] Among them, ρ(t) represents the pheromone evaporation coefficient, and Δτ ij (t) represents the pheromone increment on path ij, and τ ij (t) represents the pheromone on path ij;

[0055]

[0056] Among them, ∈ represents the stability coefficient, and σ(X j ) represents the volatility of the current parameter combination;

[0057] S46. Repeat steps S43 - S45 until the maximum number of iterations C max or the fitness function converges, and finally determine the optimal process parameter combination including the optimal ingredient ratio equipment fine - tuning parameters and process error control and apply the optimization results to the actual production process to ensure further improvement of product quality and production efficiency.

[0058] Optionally, the specific steps of S5 include:

[0059] S51. Arrange a sensor network, install various sensors at key positions on the production line, including temperature sensors, pressure sensors, speed sensors, flow sensors, and vibration sensors, to collect multi - dimensional data of temperature, pressure, speed, flow, and equipment vibration in real - time during the production process;

[0060] S52. Define the symbolic representation of the multi - dimensional data collected by the sensors. T c (t) represents the temperature data stream, P c (t) represents the pressure data stream, V c (t) represents the speed data stream, F c (t) represents the flow data stream, S c (t) represents the equipment vibration data stream, where t represents the time index;

[0061] S53. Pre - process the multi - dimensional data collected in real - time, including data denoising, missing value filling, and outlier detection. Use a low - pass filter to perform denoising on the original data to obtain the filtered temperature data stream T f (t), the filtered pressure data stream P f (t), the filtered speed data stream V f (t), the filtered flow data stream F f (t), and the filtered equipment vibration data stream S f (t);

[0062] S54. Normalize the preprocessed multi-dimensional data by using the adaptive dynamic normalization technique:

[0063]

[0064] Among them, X(t) represents any preprocessed data stream, μ X (t - τ, t) represents the dynamic mean of the data stream within the time window [t - τ, t], τ represents the size of the time window, σ X (t - τ, t) represents the dynamic standard deviation of the data flow within the time window [t - τ, t], σ ΔX (t) represents the volatility of the data stream X(t) at the time point t, λ 3 represents the weight coefficient for adjusting the influence of volatility;

[0065] Obtain the normalized temperature data stream T dyn_norm (t), the normalized pressure data stream P dyn_norm (t), the normalized velocity data stream V dyn_norm (t), the normalized flow data stream F dyn_norm (t) and the normalized equipment vibration data stream S dyn_norm (t);

[0066] S55. Construct a multi-dimensional data set D(t) from the data stream after dynamic normalization according to the time index t:

[0067] D(t) = {T dyn_norm (t), P dyn_norm (t), V dyn_norm (t), F dyn_norm (t), S dyn_norm (t)};

[0068] S56. Perform high-speed data transmission through industrial Ethernet or wireless network, and transmit the constructed multi-dimensional data set D(t) to the central processing unit in real time.

[0069] Optionally, the specific content of S6 includes:

[0070] S61. Extract key features from the multi-dimensional data set D(t) transmitted to the central processing unit in real time, and determine that the input variable set of the personalized fuzzy rule base is the temperature data stream T dyn_norm (t), the pressure data stream P dyn_norm (t), the velocity data stream V dyn_norm (t), the flow data stream F dyn_norm (t) and the equipment vibration data stream S dyn_norm (t);

[0071] S62. Define the output variable set of the personalized fuzzy rule base. The output variables include the production speed setting S output (t), the equipment utilization rate adjustment U output (t), the energy consumption optimization strategy C output (t), the process parameter fine-tuning A output (t) and the production quality target Q output (t);

[0072] S63. Construct a dynamic adaptive fuzzy membership function to map the input variables into the fuzzy sets. Each input variable X dyn_norm (t) is mapped into the fuzzy set A Ai through the dynamic membership function μ i as follows:

[0073]

[0074] where c i (t) represents the dynamic center value of the fuzzy set A i and b i (t) represents the dynamic width parameter of the fuzzy set, σ X (t) represents the volatility of the input variable, λ 4 represents the weight coefficient, and m(t) represents the shape control parameter of the dynamic membership function;

[0075] S64. Generate the initial fuzzy rule base according to the membership degrees of the input variables and the initial rule set of the output variables;

[0076] S65. Optimize the initial fuzzy rule base using the multi-objective genetic algorithm. For multiple optimization objectives, evaluate and evolve the fuzzy rule base through the complex fitness function f multi_obj (R):

[0077]

[0078] where R represents the fuzzy rule, T represents the evaluation time period, ω 1 (t) represents the dynamic weight coefficient of the production speed at time t, ω 2 (t) represents the dynamic weight coefficient of the equipment utilization rate at time t, ω 3 (t) represents the dynamic weight coefficient of the energy optimization at time t, ω 4 (t) represents the dynamic weight coefficient of the production quality target at time t, S target (t) represents the target value of the production speed setting at time t, U target (t) represents the target value of the equipment utilization rate adjustment at time t, C target (t) represents the target value of the energy consumption optimization strategy at time t, Qtarget (t) represents the production quality target value at time t;

[0079] S66. According to the fuzzy rule base optimized by the multi-objective genetic algorithm, the process parameters in the production process are adjusted in real time to adapt to the individual differences of the equipment and the task requirements, ensuring that each piece of equipment and each task operate in the optimal state, thereby improving production efficiency, reducing energy consumption, and enhancing product quality.

[0080] The beneficial effects of the present invention are as follows:

[0081] First of all, by designing and constructing a multi-level fuzzy rule base, we have achieved comprehensive optimization control from the global to the local, from the macroscopic to the microscopic. The multi-level structure enables the system to separately handle the global production speed setting, the optimization of single production line parameters, and the refined control of specific process parameters, thus ensuring optimal control at different levels. This hierarchical optimization method can not only maintain the overall coordination of the system in a complex production environment but also accurately regulate the key nodes in production, effectively coping with the challenges of dynamic changes and uncertainties.

[0082] Secondly, the present invention introduces three intelligent optimization algorithms, namely the genetic algorithm, the particle swarm optimization algorithm, and the ant colony algorithm. By combining the advantages of these algorithms, the efficient generation and dynamic optimization of the rule base are achieved. The genetic algorithm is responsible for the global optimization of the initial rule base to ensure that the system has strong adaptability and evolutionary ability in the initial stage; the particle swarm optimization algorithm is used to adjust the production line parameters in real time to ensure that each production line operates in the best state; the ant colony algorithm further optimizes the specific process parameters to ensure that the microscopic control in the production process reaches the best level. Through the organic combination of these intelligent optimization algorithms, the present invention not only improves the global search ability and convergence speed of the system but also enhances the adaptability and flexibility of the rule base, enabling the system to operate efficiently under multi-objective optimization.

[0083] In addition, the present invention overcomes the limitations of traditional static data processing methods by introducing adaptive dynamic data processing technology. The adaptive dynamic standardization technology can dynamically adjust according to the volatility of real-time data and the characteristics of time changes, ensuring the accuracy and timeliness of the data processing results. This data processing method not only improves the response speed of the system to real-time changes but also enhances the accuracy of control decisions, thereby effectively improving the stability and consistency of the production process.

[0084] More importantly, the present invention proposes a method for constructing a personalized fuzzy rule base, which can automatically generate and optimize an exclusive rule base according to the individual differences of each device and the requirements of specific production tasks. Through this innovative design, the system can customize the optimal control strategy for different devices and tasks, ensuring the best production effect under various conditions. This personalized optimization method greatly enhances the adaptability and flexibility of the system, enabling the production process to better cope with the complex and changeable production environment and meet diverse product requirements at the same time.

[0085] Finally, through visualization technology, the present invention presents the key data and optimization results in the production process in a graphical interface, supporting rapid decision-making and problem troubleshooting. This real-time visualization function not only improves the operability and user experience of the system, but also provides production managers with a more intuitive and convenient means of production monitoring and adjustment, further ensuring the efficiency and controllability of the production process. Description of the Drawings

[0086] The 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 to the present invention. In the drawings:

[0087] Figure 1 is a flowchart of the optimization method for the production process of the electronic device safety interface based on fuzzy logic proposed by the present invention;

[0088] Figure 2 is a schematic diagram of the multi-level fuzzy rule base structure of the optimization method for the production process of the electronic device safety interface based on fuzzy logic proposed by the present invention;

[0089] Figure 3 is a flowchart of the construction and optimization of the personalized fuzzy rule base of the optimization method for the production process of the electronic device safety interface based on fuzzy logic proposed by the present invention. Detailed Embodiments

[0090] Now, the present invention will be further described in detail with reference to the drawings. These drawings are all simplified schematic diagrams, only illustrating the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0091] Refer to Figures 1-3 , the optimization method for the production process of the electronic device safety interface based on fuzzy logic includes the following steps:

[0092] S1. Design and construct a multi-level fuzzy rule base, where the multi-level fuzzy rule base includes a first level, a second level, and a third level;

[0093] S2. The first layer uses a genetic algorithm to generate and optimize an initial fuzzy rule base. The input variables include the total production demand, the overall health status of the equipment, the energy consumption rate, and the production duration. The output variables include the global production speed setting, the equipment utilization rate allocation, and the energy consumption optimization strategy, which are used to achieve global control of the overall production speed, energy consumption, and output balance.

[0094] S3. The second layer uses a particle swarm optimization algorithm to optimize the fuzzy rule base. The input variables include the temperature, pressure, material flow rate, and production line load of each production line. The output variables include the production line speed adjustment, temperature setting, and pressure regulation, which are used to optimize the operating parameters of a single production line and ensure that each production line operates in the best state.

[0095] S4. The third layer uses an ant colony algorithm to optimize the specific process parameters in the fuzzy rule base. The input variables include the batching ratio, the fine-tuning parameters of the equipment, and the process error. The output variables include the ratio optimization, the equipment fine-tuning instruction, and the process parameter correction, which are used to precisely control the production process parameters and ensure further improvement of product quality and production efficiency.

[0096] S5. Through data acquisition and monitoring technology, data in the production process is obtained in real time, a multi-dimensional data set is constructed, and the multi-dimensional data set is transmitted to the central processing unit.

[0097] S6. Use the multi-dimensional data set to construct a personalized fuzzy rule base, automatically generate and optimize an exclusive rule base according to the individual differences of the equipment and task classification, and perform adaptive adjustment according to real-time data during the production process.

[0098] S7. Display the key data and optimization results in the production process through visualization technology, generate a real-time graphical interface, and support quick decision-making and problem troubleshooting.

[0099] In this embodiment, S2 specifically includes:

[0100] S21. Initialize the population to generate a number of candidate rule sets. Each rule in the candidate rule set is composed of a combination of input variables and output variables. The input variables include the total production demand D total , the overall health status H of the equipment equipment , the energy consumption rate E rate , and the production duration T production , and the output variables include the global production speed setting S global , the equipment utilization rate allocation U utilization , and the energy consumption optimization strategy C consumption ;

[0101] S22. Evaluate the fitness of each candidate rule set, and calculate the fitness value of each rule set through the fitness function f(x):

[0102]

[0103] Among them, w 1 、w 2 、w 3 and w 4 represent pre-set weight coefficients, f(x) represents the fitness value of the rule set in terms of balancing production speed, energy consumption and output. The larger the value, the better the optimization effect of the rule set;

[0104] S23. Perform a selection operation based on the fitness value, retain the rule set with a higher fitness value, and perform a crossover operation. The crossover operation generates a new rule set by exchanging the values of some input variables and output variables in two rule sets;

[0105] S24. Perform a mutation operation on the rule set after the selection and crossover operations. The mutation operation dynamically adjusts the mutation amplitude based on the population fitness:

[0106]

[0107] Among them, X i,j represents the j-th variable of the i-th rule set after mutation, X i,j represents the j-th variable of the i-th rule set before mutation, δ(t) represents the mutation amplitude coefficient, which is dynamically adjusted with the generation number t, rand(-1, 1) represents a random number between -1 and 1, f avg represents the average fitness of the current population, f max represents the maximum fitness in the current population;

[0108] is used to control the mutation intensity. When the overall fitness of the population tends to be stable, the mutation intensity will gradually weaken, thereby improving the convergence and stability of the algorithm;

[0109] S25. Repeat steps S22 - S24 until the fitness value of the rule set in the population converges to a preset threshold or reaches a predetermined number of iterations, and finally obtain an optimized fuzzy rule base;

[0110] S26. Apply the optimized fuzzy rule base to the production process to adjust the global production speed setting S global 、equipment utilization rate allocation U utilization and energy consumption optimization strategy C consumption .

[0111] In this embodiment, the specific steps of S3 include:

[0112] S31. Initialize the particle swarm, set the size of the particle swarm to N, and randomly generate the initial position X of each particlei and velocity V i , where i represents the particle number, and the initial position X i represents the temperature T of each production line i , pressure P i , material flow rate F i and production line load L i of the initial value, and the initial velocity V i represents the temperature T of each production line i , pressure P i , material flow rate F i and production line load L i of the initial change rate;

[0113] S32. Define the fitness function f 1 (X i ) for evaluating the operating state of the production line corresponding to the current position X i of each particle:

[0114]

[0115] where T opt represents the optimal target value of temperature, P opt represents the optimal target value of pressure, F opt represents the optimal target value of material flow rate, L opt represents the optimal target value of production line load, v 1 , v 2 , v 3 and v 4 represent the weight coefficients;

[0116] S33. Update the individual optimal position A i (t) and the global optimal position G(t) of each particle according to the fitness function value, record the fitness value of each particle in the current iteration, and update the position of the particle with a better fitness value to the individual optimal position A i (t), and at the same time update the optimal position among all particles to the global optimal position G(t);

[0117] S34. Update the position and velocity of the particle. The update formulas for the velocity V i (t + 1) and the position X i (t + 1) of the particle are:

[0118] V i (t + 1) = ω·V i (t) + c 1 ·r 1 ·(A i (t) - X i (t)) + c2 ·r 2 ·(G(t) - X i (t));

[0119] X i (t + 1) = X i (t) + V i (t + 1);

[0120] Among them, ω represents the inertia weight, which is used to control the influence of the current velocity of the particle, and c 1 and c 2 represent the learning factors, which represent the influence of the particle's own experience and the group experience respectively. r 1 and r 2 represent random numbers between 0 and 1;

[0121] S35. In each iteration, dynamically adjust the inertia weight ω, and gradually decrease ω as the number of iterations increases. At the same time, perform adaptive adjustment on c 1 and c 2 to enhance the collaborative optimization ability of the particle swarm;

[0122] S36. Iteratively update the particle swarm, and make the particle swarm gradually converge through multiple iterations until the fitness function f 1 (X i ) converges to the set threshold or reaches the maximum number of iterations, and finally find the particle position X 1 (X i ) that minimizes the fitness function f i , as the optimized production line operation parameters, including the optimal production line speed adjustment, temperature setting, and pressure regulation.

[0123] In this embodiment, the S4 specifically includes:

[0124] S41. Initialize the ant colony, set the scale M of the ant colony and the maximum number of iterations C max , randomly distribute the initial pheromone concentration τ ij (0). Each ant represents a process parameter combination X j , including the ingredient ratio R j , the equipment fine-tuning parameter A j and the process error E j , where i represents the ant number and j represents the parameter combination number;

[0125] S42. Define the multi-objective fitness function f 2 (X j ), which is used to evaluate the influence of each parameter combination X j on the production process:

[0126]

[0127] Among them, R opt represents the optimal target value of the ingredient ratio, A opt represents the optimal target value of the equipment fine-tuning parameter, E opt represents the optimal target value of the process error, α 1 , β 1 and γ 1 represent the weight coefficients;

[0128] S43. Calculate the selection probability P ij (t) of each ant on each path based on the entropy-based adaptive selection mechanism. The selection probability is based on the pheromone concentration τ ij (t), the heuristic factor η ij and the uncertainty H ij (t) of the current path:

[0129]

[0130] Among them, α(t) represents the weight coefficient of the pheromone concentration, β(t) represents the weight coefficient of the heuristic factor, γ(t) represents the weight factor of the information entropy, k represents the index of other paths available for the ant to choose, allowed represents the set of paths not yet selected by the current ant, t represents the number of iterations, τ ij (t) represents the current pheromone concentration on path ij, η ij represents the heuristic factor, H ij (t) represents the information entropy on path ij, which is used to describe the uncertainty of path selection:

[0131]

[0132] Among them, l represents the index variable, P il (t) represents the probability that the ant selects path il;

[0133] S44. Each ant selects a path according to the selection probability P ij (t), determines a process parameter combination X j , and evaluates the effect through the multi-objective fitness function f 2 (X j ), and records the best process parameter combination found by each ant;

[0134] S45. Update the pheromone concentration, and use the adaptive pheromone evaporation mechanism and dynamic increment to update the pheromone concentration τ ij (t + 1) on each path:

[0135] τ ij (t + 1) = (1 - ρ(t))·τ ij (t) + Δτij (t);

[0136] Among them, ρ(t) represents the pheromone evaporation coefficient, and Δτ ij (t) represents the pheromone increment on path ij, and τ ij (t) represents the pheromone on path ij;

[0137]

[0138] Among them, ∈ represents the stability coefficient, and σ(X j ) represents the volatility of the current parameter combination;

[0139] S46. Repeat steps S43 - S45 until the maximum number of iterations C max or the fitness function converges, and finally determine the optimal process parameter combination including the optimal ingredient ratio equipment fine - tuning parameters and process error control and apply the optimization results to the actual production process to ensure further improvement of product quality and production efficiency.

[0140] In this embodiment, the specific steps of S5 include:

[0141] S51. Arrange a sensor network, install various sensors at key positions on the production line, including temperature sensors, pressure sensors, speed sensors, flow sensors, and vibration sensors, to collect multi - dimensional data of temperature, pressure, speed, flow, and equipment vibration in real - time during the production process;

[0142] S52. Define the symbolic representation of the multi - dimensional data collected by the sensors. T c (t) represents the temperature data stream, P c (t) represents the pressure data stream, V c (t) represents the speed data stream, F c (t) represents the flow data stream, S c (t) represents the equipment vibration data stream, where t represents the time index;

[0143] S53. Pre - process the multi - dimensional data collected in real - time, including data denoising, missing value filling, and outlier detection. Use a low - pass filter to perform denoising on the original data to obtain the filtered temperature data stream T f (t), the filtered pressure data stream P f (t), the filtered speed data stream V f (t), the filtered flow data stream F f (t), and the filtered equipment vibration data stream S f (t);

[0144] S54. Normalize the preprocessed multi-dimensional data using the adaptive dynamic normalization technique:

[0145]

[0146] Among them, X(t) represents any preprocessed data stream, μ X (t - τ, t) represents the dynamic mean of the data stream within the time window [t - τ, t], τ represents the size of the time window, σ X (t - τ, t) represents the dynamic standard deviation of the data flow within the time window [t - τ, t], σ ΔX (t) represents the volatility of the data stream X(t) at time point t, λ 3 represents the weight coefficient for adjusting the influence of volatility;

[0147] Obtain the normalized temperature data stream T dyn_norm (t), the normalized pressure data stream P dyn_norm (t), the normalized velocity data stream V dyn_norm (t), the normalized flow data stream F dyn_norm (t) and the normalized equipment vibration data stream S dyn_norm (t);

[0148] S55. Construct a multi-dimensional data set D(t) based on the time index t for the data stream after dynamic normalization:

[0149] D(t) = {T dyn_norm (t), P dyn_norm (t), V dyn_norm (t), F dyn_norm (t), S dyn_norm (t)};

[0150] S56. Perform high-speed data transmission through industrial Ethernet or wireless network, and transmit the constructed multi-dimensional data set D(t) to the central processing unit in real time.

[0151] In this embodiment, the specific content of S6 includes:

[0152] S61. Extract key features from the multi-dimensional data set D(t) transmitted to the central processing unit in real time. According to the historical operation data, current status and task requirements of each device, determine that the input variable set of the personalized fuzzy rule base is the temperature data stream T dyn_norm (t), the pressure data stream P dyn_norm (t), the velocity data stream V dyn_norm (t), the flow data stream F dyn_norm (t) and the equipment vibration data stream S dyn_norm(t);

[0153] S62. Define the output variable set of the personalized fuzzy rule base. The output variables include production speed setting S output (t), equipment utilization rate adjustment U output (t), energy consumption optimization strategy C output (t), process parameter fine-tuning A output (t) and production quality target Q output (t);

[0154] S63. Construct a dynamic adaptive fuzzy membership function to map the input variables into fuzzy sets. Each input variable X dyn_norm (t) is mapped into the fuzzy set A Ai through the dynamic membership function μ i as follows:

[0155]

[0156] where c i (t) represents the dynamic center value of the fuzzy set A i and b i (t) represents the dynamic width parameter of the fuzzy set, σ X (t) represents the volatility of the input variable, λ 4 represents the weight coefficient, and m(t) represents the shape control parameter of the dynamic membership function;

[0157] S64. Generate an initial fuzzy rule base according to the membership degrees of the input variables and the initial rule set of the output variables;

[0158] S65. Use the multi-objective genetic algorithm to optimize the initial fuzzy rule base. For multiple optimization objectives, evaluate and evolve the fuzzy rule base through the complex fitness function f multi_obj (R):

[0159] f multi_obj (R)

[0160]

[0161] where R represents the fuzzy rule, T represents the evaluation time period, ω 1 (t) represents the dynamic weight coefficient of the production speed at time t, ω 2 (t) represents the dynamic weight coefficient of the equipment utilization rate at time t, ω 3 (t) represents the dynamic weight coefficient of energy optimization at time t, ω 4 (t) represents the dynamic weight coefficient of the production quality target at time t, S target (t) represents the production speed setting target value at time t, Utarget (t) represents the target value of equipment utilization rate adjustment at time t, C target (t) represents the target value of energy consumption optimization strategy at time t, Q target (t) represents the production quality target value at time t;

[0162] S66. According to the fuzzy rule base optimized by the multi-objective genetic algorithm, the process parameters in the production process are adjusted in real time to adapt to the individual differences of equipment and task requirements, ensuring that each piece of equipment and each task operate in the optimal state, thereby improving production efficiency, reducing energy consumption, and enhancing product quality.

[0163] Example 1:

[0164] To verify the feasibility of the present invention in implementation, the present invention is applied to an electronic equipment manufacturing factory, which mainly produces high-performance safety interfaces for smartphones and industrial control equipment. Due to the volatility of market demand and the diversity of product models, the production line needs to have a high degree of flexibility and adaptability to cope with the production of products in different batches and specifications. The problem faced by this manufacturer is how to improve production efficiency, reduce energy consumption, and address the challenges brought by equipment aging and production environment changes while ensuring product quality.

[0165] In practical applications, the manufacturer first designed and constructed a multi-level fuzzy rule base, and optimized the production process from global to local through this rule base. The generation and optimization of the first-level rule base mainly rely on genetic algorithms. Based on input variables such as total production demand, overall equipment health status, energy consumption rate, and production duration, global production speed setting, equipment utilization rate allocation, and energy consumption optimization strategies are output. After optimization, the global production speed setting has increased by an average of 12%, the equipment utilization rate has increased by 15%, and the energy consumption has decreased by about 8%.

[0166] In the second-level optimization, the manufacturer used the particle swarm optimization algorithm to dynamically optimize key parameters such as temperature, pressure, material flow rate, and production line load of each production line. Through this optimization, the operating parameters of each production line can be dynamically adjusted according to real-time data to ensure that the production line operates in the best state. Specifically, within a one-month production cycle, the product qualification rate has increased from the original 95% to 97.5%, and at the same time, the failure rate of the production line has decreased by about 20%, further reducing uncontrollable factors in the production process.

[0167] The optimization at the third level applies the ant colony algorithm for fine control of specific process parameters such as ingredient ratios, equipment fine-tuning parameters, and process errors. Through these optimizations, the micro-quality indicators of the product, such as the connection strength, wear resistance, and oxidation resistance of the interface, have been significantly improved. After comparing the production data for two months, the average connection strength of the optimized interface has increased by 10%, the wear resistance has increased by 8%, and the oxidation resistance has been improved by approximately 7%. The improvement of these indicators has greatly enhanced the competitiveness of the product in the market.

[0168] In terms of data collection and monitoring, the manufacturer obtains multi-dimensional data during the production process in real time through temperature, pressure, speed, flow, and vibration sensors arranged throughout the production line. Through the adaptive dynamic standardization technology proposed in the present invention, the accuracy and timeliness of data collection are ensured. Specifically, the manufacturer's data analysis team compared the data processing efficiency before and after implementing the present invention and found that the data transmission delay has been reduced by approximately 15%, and the impact of data noise has been reduced by more than 20%, thus significantly improving the response speed and decision-making accuracy of the production control system.

[0169] In addition, the manufacturer attaches great importance to the construction of a personalized fuzzy rule base. Due to the different service lives and states of production equipment, the personalized rule base proposed in the present invention can automatically generate fuzzy rules adapted to its specific needs based on the historical data and real-time status of each piece of equipment. Taking two pieces of equipment with different states in the factory as an example, Equipment A is a new equipment with good operating conditions; Equipment B has been used for five years and has certain aging phenomena. In actual production, the personalized rule base can set a more aggressive production speed and lower energy consumption strategy for Equipment A, while setting a more conservative production speed and more stable operating parameters for Equipment B. After three months of production practice, the production efficiency of Equipment A has increased by 18%, and the energy consumption has been reduced by approximately 10%; while for Equipment B, the product quality remains stable when the production speed is reduced by 5%, and the failure rate of the equipment has been reduced by approximately 25%.

[0170] To verify the beneficial effects of the present invention, the manufacturer compared the production data before and after applying the present invention within one year.

[0171] Table 1 Comparison Table of Implementation Effect Data of the Intelligent Production Optimization System

[0172]

[0173]

[0174] Table 1 clearly shows the significant advantages and improvement effects of the present invention in the production process of the electronic device safety interface. By comparing the key indicators before and after implementation, we can comprehensively understand the actual benefits brought by the present invention.

[0175] First, from the perspectives of global production speed setting and equipment utilization rate, after implementing the present invention, these two indicators have increased by 12% and 15% respectively. This shows that by introducing a multi-level fuzzy rule base and an intelligent optimization algorithm, the overall operating efficiency of the production line has been greatly improved, and the utilization rate of equipment has been better optimized, avoiding production bottlenecks caused by uneven production line load or unreasonable allocation of equipment resources.

[0176] In terms of energy consumption, the energy consumption after implementation has been reduced by 8%, demonstrating the strong ability of the present invention in optimizing energy utilization and reducing production costs. This improvement not only helps to reduce production costs but also reflects the potential of the present invention in supporting green manufacturing and sustainable development.

[0177] In terms of product quality, the product qualification rate has increased from 95% to 97.5% after implementation, and at the same time, the failure rate of the production line has been reduced by 20%. This shows that the present invention effectively improves the consistency and reliability of products through refined process parameter optimization, reducing unplanned downtime and fault repair time during the production process.

[0178] From the specific performance indicators of the product, the interface connection strength, wear resistance, and oxidation resistance have increased by 10%, 8%, and 7% respectively. These improvements directly enhance the quality and service life of the safety interface of electronic devices, enabling them to better cope with harsh usage environments and long-term working pressures.

[0179] In terms of data processing and control response, the data transmission delay has been reduced by 15%, and the impact of data noise has been reduced by 20%. The response speed of the production control system has been significantly improved. This shows that the present invention has strong capabilities in real-time data processing and control optimization, and can maintain efficient and precise control in a dynamically changing production environment.

[0180] In the application of specific equipment, the production efficiency of Equipment A has increased by 18%, and the energy consumption has been reduced by 10%; the failure rate of Equipment B has been reduced by 25%. This shows that the personalized fuzzy rule base of the present invention can be customized and optimized according to the status of different equipment, enabling new and old equipment to operate at their respective best states and giving full play to their potential.

[0181] Finally, from the perspective of overall production efficiency, after implementing the present invention, the production efficiency has increased by 20%, the product qualification rate has further increased by 2.5%, the energy consumption has been reduced by 15%, and the production adjustment time has been reduced by 30%. These data show that the present invention can not only significantly improve the overall operating efficiency of the production line but also effectively reduce energy consumption and the time cost of production adjustment, thus greatly enhancing the competitiveness and economic benefits of the production line at the comprehensive level.

[0182] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, shall be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the production process of electronic equipment security interfaces based on fuzzy logic, characterized in that: The steps include: S1. Design and construct a multi-level fuzzy rule base, wherein the multi-level fuzzy rule base includes a first level, a second level and a third level; S2. The first level uses genetic algorithm to generate and optimize the initial fuzzy rule base. The input variables include total production demand, overall equipment health status, energy consumption rate and production period. The output variables include global production speed setting, equipment utilization allocation and energy consumption optimization strategy. S3. The second level uses particle swarm optimization algorithm to optimize the fuzzy rule base. The input variables include the temperature, pressure, material flow rate and production line load of each production line. The output variables include production line speed adjustment, temperature setting and pressure regulation. S4, the third level uses ant colony algorithm to optimize specific process parameters in the fuzzy rule base, the input variables include ingredient ratio, equipment fine-tuning parameters and process errors, and the output variables include ratio optimization, equipment fine-tuning instructions and process parameter correction; S5. Through data acquisition and monitoring technology, real-time data from the production process is obtained, a multidimensional data set is constructed, and the multidimensional data set is transmitted to the central processing unit; S6. Use multidimensional data sets to build a personalized fuzzy rule base, automatically generate and optimize the exclusive rule base according to the individual differences of equipment and task classification, and make adaptive adjustments based on real-time data during the production process; Extract key features from the multidimensional data set D(t) transmitted to the central processing unit in real time, and determine the input variable set of the personalized fuzzy rule base as the temperature data stream T according to the historical operation data, current status and task requirements of each device. dyn_norm (t), pressure data stream P dyn_norm (t), speed data flow V dyn_norm (t), traffic data flow F dyn_norm (t) and equipment vibration data stream S dyn_norm (t); Define the output variable set of the personalized fuzzy rule base, the output variable includes the production speed setting S output (t), Equipment utilization adjustment U output (t), Energy consumption optimization strategy C output (t), process parameter fine-tuning A output (t) and production quality target Q output (t); Construct a dynamically adaptive fuzzy membership function, map the input variables to the fuzzy set, and generate an initial fuzzy rule base based on the membership of the input variables and the initial rule set of the output variables; S7. Display key data and optimization results in the production process through visualization technology, generate a real-time graphical interface, and support quick decision-making and problem troubleshooting.

2. The method for optimizing the production process of electronic device security interface based on fuzzy logic according to claim 1, characterized in that: The S2 specifically includes: S21, initialize the population and generate several candidate rule sets, each rule in the candidate rule set is composed of a combination of input variables and output variables, where the input variables include the total output demand D total 、Overall health status of equipment equipment , energy consumption rate E rate and production period T production , the output variables include the global production speed setting S global , equipment utilization allocation U utilization and energy consumption optimization strategy C consumption ; S22. Perform fitness evaluation on each candidate rule set, and calculate the fitness value of each rule set through the fitness function f(x): Wherein, w1, w2, w3 and w4 represent pre-set weight coefficients; S23, performing a selection operation according to the fitness value, retaining the rule set with a higher fitness value, and performing a crossover operation, wherein the crossover operation generates a new rule set by exchanging the values ​​of some input variables and output variables in two rule sets; S24, performing a mutation operation on the rule set after the selection and crossover operations, wherein the mutation operation dynamically adjusts the mutation range based on the population fitness: Among them, X′ i,j represents the jth variable of the ith rule set after mutation, X i,j represents the jth variable of the i-th rule set before mutation, δ(t) represents the coefficient of variation, which is dynamically adjusted with the number of generations t, rand(-1,1) represents a random number between -1 and 1, and f avg represents the average fitness of the current population, f max Represents the maximum fitness in the current population; S25, repeating steps S22-S24 until the fitness value of the rule set in the population converges to a preset threshold or reaches a predetermined number of iterations, and finally obtaining an optimized fuzzy rule base; S26. Apply the optimized fuzzy rule base to the production process and adjust the global production speed setting S in real time global , equipment utilization allocation U utilization and energy consumption optimization strategy C consumption .

3. The method for optimizing the production process of electronic device security interface based on fuzzy logic according to claim 1, characterized in that: The S3 specifically includes: S31, initialize the particle swarm, set the size of the particle swarm to N, and randomly generate the initial position X of each particle i and speed V i , where i represents the number of the particle and the initial position X i Indicates the temperature T of each production line i 、Pressure P i , material flow rate F i and production line load L i The initial value, initial velocity V i Indicates the temperature T of each production line i 、Pressure P i , material flow rate F i and production line load L i The initial rate of change of S32, define the fitness function f1(X i ), used to evaluate the current position X of each particle i The corresponding production line operation status: Among them, T opt represents the optimal target value of temperature, P opt Indicates the optimal target value of pressure, F opt Indicates the optimal target value of material flow rate, L opt represents the optimal target value of the production line load, and v1, v2, v3 and v4 represent weight coefficients; S33, according to the fitness function value, the individual optimal position A of each particle i (t) and the global optimal position G(t), record the fitness value of each particle in the current iteration, and update the position of the particle with a better fitness value to the individual optimal position A i (t), and update the optimal position of all particles to the global optimal position G(t); S34. Update the position and speed of the particle, the speed of the particle V i (t+1) and position X i The update formula for (t+1) is: V i (t+1)=ω·V i (t)+c1·r1·(A i (t)-X i (t))+c2·r2·(G(t)-X i (t)); X i (t+1)=X i (t)+V i (t+1); Among them, ω represents the inertia weight, c1 and c2 represent learning factors, representing the influence of the particle's own experience and the group's experience respectively, and r1 and r2 represent random numbers between 0 and 1; S35, in each iteration, dynamically adjust the inertia weight ω, and gradually reduce ω as the number of iterations increases, while adaptively adjusting c1 and c2; S36, iteratively update the particle swarm, and make the particle swarm gradually converge through multiple iterations until the fitness function f1(X i ) converges to the set threshold or reaches the maximum number of iterations, and finally finds the fitness function f1(X i )The smallest particle position X i , as the optimized production line operating parameters, including optimal production line speed adjustment, temperature setting and pressure regulation.

4. The method for optimizing the production process of electronic device security interface based on fuzzy logic according to claim 1, characterized in that: The S4 specifically includes: S41, initialize the ant colony, set the size M of the ant colony and the maximum number of iterations C max , randomly distributed initial pheromone concentration τ ij (0), each ant represents a process parameter combination X j , including the ingredient ratio R j 、Equipment fine-tuning parameters A j and process error E j , where i represents the ant number and j represents the parameter combination number; S42, define the multi-objective fitness function f2(X j ), used to evaluate each parameter combination X j Impact on production process: Among them, R opt Indicates the optimal target value of the ingredient ratio, A opt represents the optimal target value of the device fine-tuning parameters, E opt represents the optimal target value of the process error, α1, β1 and γ1 represent weight coefficients; S43, entropy-based adaptive selection mechanism calculates the selection probability P of each ant on each path ij (t), the selection probability is based on the pheromone concentration τ ij (t), heuristic factor η ij and the uncertainty of the current path H ij (t) Calculation: Among them, α(t) represents the weight coefficient of pheromone concentration, β(t) represents the weight coefficient of heuristic factor, γ(t) represents the weight factor of information entropy, k represents the index of other paths available for ants to choose, allowed represents the set of paths that the current ant has not chosen, t represents the number of iterations, τ ij (t) represents the current pheromone concentration on path ij, η ij represents the heuristic factor, H ij (t) represents the information entropy on path ij: Among them, l represents the index variable, P il (t) represents the probability that the ant chooses path il; S44, each ant chooses a ij (t) Select a path and determine a process parameter combination X j , through the multi-objective fitness function f2(X j ) Evaluate the effect and record the best process parameter combination found by each ant; S45, update the pheromone concentration, use the adaptive pheromone volatilization mechanism and dynamic increment to update the pheromone concentration τ on each path ij (t+1): t ij (t+1)=(1-ρ(t))·τ ij (t)+Δτ ij (t); Among them, ρ(t) represents the pheromone volatility coefficient, Δτ ij (t) represents the pheromone increment on path ij, τ ij (t) represents the pheromone on path ij; Among them, ∈ represents the stability coefficient, σ(X j ) represents the volatility of the current parameter combination; S46, repeat steps S43-S45 until the maximum number of iterations C is reached max Or the fitness function converges, and finally determines the optimal process parameter combination Including the optimal ratio of ingredients Equipment fine-tuning parameters and process error control 5. The method for optimizing the production process of electronic device security interface based on fuzzy logic according to claim 1, characterized in that: The S5 specifically includes: S51. Arrange a sensor network and install a variety of sensors at key locations of the production line, including temperature sensors, pressure sensors, speed sensors, flow sensors and vibration sensors, to collect multi-dimensional data of temperature, pressure, speed, flow and equipment vibration in real time during the production process; S52. Define the symbolic representation of multidimensional data collected by sensors, T c (t) represents the temperature data stream, P c (t) represents the pressure data flow, V c (t) represents the velocity data flow, F c (t) represents the traffic data flow, S c (t) represents the equipment vibration data stream, where t represents the time index; S53, preprocessing the multidimensional data collected in real time, including data denoising, missing value filling and outlier detection, using a low-pass filter to denoise the original data to obtain a filtered temperature data stream T f (t), filtered pressure data stream P f (t), filtered velocity data stream V f (t), filtered traffic data stream F f (t) and the filtered equipment vibration data stream S f (t); S54, using adaptive dynamic normalization technology to normalize the pre-processed multidimensional data: Where X(t) represents any preprocessed data stream, μ X (t-τ, t) represents the dynamic mean of the data stream in the time window [t-τ, t], τ represents the size of the time window, σ X (t-τ,t) represents the dynamic standard deviation of the data flow in the time window [t-τ,t], σ ΔX (t) represents the volatility of the data stream X(t) at time point t, and λ3 represents the weight coefficient for adjusting the impact of volatility; Get the normalized temperature data stream T dyn_norm (t), normalized pressure data stream P dyn_norm (t), normalized velocity data stream V dyn_norm (t), normalized traffic data flow F dyn_norm (t) and the normalized equipment vibration data stream S dyn_norm (t); S55. Construct a multidimensional data set D(t) according to the time index t of the data stream after dynamic standardization processing: D(t)={T dyn_norm (t),P dyn_norm (t),V dyn_norm (t),F dyn_norm (t),S dyn_norm (t)}; S56. Perform high-speed data transmission via industrial Ethernet or wireless network to transmit the constructed multidimensional data set D(t) to the central processing unit in real time.

6. The method for optimizing the production process of electronic device security interface based on fuzzy logic according to claim 1, characterized in that: The S6 specifically includes: S61, each input variable X dyn_norm (t) through the dynamic membership function μ Ai Mapping to fuzzy set A i middle: Among them, c i (t) represents the fuzzy set A i The dynamic center value, b i (t) represents the dynamic width parameter of the fuzzy set, σ X (t) represents the volatility of the input variable, λ4 represents the weight coefficient, and m(t) represents the shape control parameter of the dynamic membership function; S62, using a multi-objective genetic algorithm to optimize the initial fuzzy rule base, targeting multiple optimization objectives, through a complex fitness function f multi_obj (R) Evaluate and evolve the fuzzy rule base: Where R represents the fuzzy rule, T represents the evaluation time period, ω1(t) represents the dynamic weight coefficient of the production speed at time t, ω2(t) represents the dynamic weight coefficient of the equipment utilization at time t, ω3(t) represents the dynamic weight coefficient of the energy optimization at time t, ω4(t) represents the dynamic weight coefficient of the production quality target at time t, S target (t) represents the production speed setting target value at time t, U target (t) represents the target value of equipment utilization adjustment at time t, C target (t) represents the target value of the energy consumption optimization strategy at time t, Q target (t) represents the production quality target value at time t; S63. According to the fuzzy rule base optimized by the multi-objective genetic algorithm, the process parameters in the production process are adjusted in real time.

Images