Mechanical element parameter optimization method and device based on artificial intelligence algorithm

Abstract

According to the mechanical element parameter optimization method and device based on the artificial intelligence algorithm provided by the invention, when the optimal mechanical element parameter is solved, the eagle algorithm is introduced based on the probability to update the position of the chimpanzee population, and the update depth is controlled based on the diving depth factor, so that the updating of the solution is more flexible, and the jumping out of local optimum is facilitated; therefore, the global search capability and the convergence speed of the algorithm are improved. A golden sine algorithm is introduced, in each iteration, by adjusting a random angle and utilizing periodic changes of a sine function, updating of an individual position has randomness and directionality, the global search capability is enhanced, the algorithm is helped to jump out of a local optimal solution, and along with iteration proceeding, by dynamically adjusting a theta value, a local optimal solution is obtained. The algorithm can flexibly adjust the proportion of exploration and utilization according to the current search state, effectively balance the relationship between the two, and improve the convergence speed. Complete random position reset is introduced to avoid excessive aggregation around a local optimal solution, Gaussian noise disturbance is introduced to more carefully explore in a local area, premature convergence is avoided, and the continuity and effectiveness of search are prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical component design, and particularly to a method and device for optimizing mechanical component parameters based on artificial intelligence algorithms. Background Art

[0002] The design process of mechanical components is a highly complex engineering problem with multi-parameter interactions, which is particularly prominent in fields requiring high precision such as precision instruments and aerospace components. Designers must comprehensively consider various factors such as material properties, mechanical properties, manufacturing processes, and environmental conditions. For example, turbine blades in aircraft engines need to maintain structural integrity under high temperature and high pressure. This not only requires the material to have high strength and heat resistance but also the influence of thermal expansion must be considered. Otherwise, temperature fluctuations may cause changes in precise fit clearances, affecting the blade performance. In addition, in precision instruments, tiny geometric errors or surface defects may lead to unacceptable deviations during equipment measurement. Therefore, in the design process, not only advanced simulation technologies are needed to predict physical phenomena, but also strict manufacturing process control and high-precision detection means are required to ensure that each component can meet the demanding design requirements. This complexity and high-precision requirement make the design of high-end mechanical components a major challenge in the engineering field, and at the same time promote the development of related disciplines such as materials science, manufacturing technology, and computational mechanics.

[0003] Current technologies establish optimization objectives and constraint conditions for mechanical parameters and use methods such as genetic algorithms, particle swarm optimization algorithms, ant colony algorithms, and sand cat swarm algorithms to find the optimal solutions for these parameters to improve the performance of mechanical components. However, these optimization algorithms face challenges in the diversity and uniformity of initial solutions when applied to practical engineering problems. This may lead to the initial population not fully covering the solution space, affecting the efficiency and effectiveness of optimization. For example, during the design of high-speed bearings, if the initial solution distribution is uneven, it may cause vibrations in the bearings during high-speed rotation, which will affect the stability and lifespan of the machine. In addition, existing algorithms also have limitations in convergence speed and accuracy and are prone to falling into local optimal solutions. For example, in the design of aircraft wings, local optimal solutions may lead to poor aerodynamic performance, thus affecting the fuel efficiency and flight stability of the aircraft. To overcome these problems, there is an urgent need to develop new optimization schemes to improve the diversity and uniformity of initial solutions, enhance the global search ability of the algorithms, and increase the convergence speed and accuracy. This may include combining multiple optimization strategies, using more advanced initialization techniques, or introducing new intelligent algorithms to ensure that the optimization process can find solutions close to the global optimal solution, thereby greatly improving the overall performance of mechanical components. Therefore, there is an urgent need for a new optimization scheme for the parameter design of mechanical components. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and device for optimizing mechanical element parameters based on artificial intelligence algorithms to eliminate or improve one or more defects existing in the prior art and solve the problems that traditional artificial gorilla algorithms may face slow convergence speed, being easily trapped in local optima, and insufficient global search ability when dealing with complex multi-modal optimization problems.

[0005] One aspect of the present invention provides a method for optimizing mechanical element parameters based on artificial intelligence algorithms, the method comprising the following steps:

[0006] Based on the artificial gorilla population optimization algorithm, for multiple parameters to be optimized of the target mechanical element, combined with the constraint conditions between the parameters to be optimized, perform random initialization to construct the positions of multiple gorilla individuals;

[0007] In the exploration stage, by calculating the fitness of each gorilla individual, select the optimal individual as the silverback gorilla, search and update the position globally, where the fitness introduces a penalty function using the optimization objective and the constraint conditions; in the exploitation stage, search and update the positions of gorilla members locally based on the mechanism of following the silverback gorilla or the mechanism of gorilla competing for adult females; wherein, in the exploration stage and the exploitation stage, each gorilla individual updates its position by introducing the osprey optimization strategy according to a first preset probability; in the exploration stage and the exploitation stage, the golden sine algorithm is also introduced to adjust the update path of each gorilla position using the golden ratio and sine volatility; in the exploration stage and the exploitation stage, two mechanisms of completely random position reset and Gaussian noise perturbation are also introduced to update the positions of each gorilla;

[0008] When the set number of iterations is reached or the fitness of the optimal individual reaches the set value, output the parameters corresponding to the optimal individual as the target values of the parameters to be optimized of the target mechanical element.

[0009] In some embodiments, the position update model of the silverback gorilla in the exploration stage is:

[0010]

[0011] Wherein, F = cos(2r4)+1; L = Cl; H = ZX(t); Z ∈ [-C, C];

[0012] X(t) represents the position of the silverback gorilla in the t-th iteration, GX(t + 1) represents the updated position of the silverback gorilla; UB represents the upper bound of the search space, LB represents the lower bound of the search space; r1, r2, r3, r4 and rand are random numbers between (0, 1); X r and GX rThey are all the positions of gorillas in the randomly selected t-th iteration; p ∈ (0, 1) is a given parameter used to simulate the influencing factors of the dominant gorilla individual's exploration of unknown positions; l is a random number in the interval (0, 1); Z is a random number in the interval [-C, C]; MaxIt represents the maximum number of iterations.

[0013] In some embodiments, the position update model of each gorilla member in the exploitation stage is:

[0014] When C ≥ W, the position of each gorilla member is updated by selecting the mechanism of following the dominant gorilla, and the expression is:

[0015]

[0016] When C < W, the position of each gorilla member is updated by selecting the mechanism of competing adult females, and the expression is:

[0017]

[0018] Among them, X silveriack is the position of the dominant gorilla, L represents the scaling factor, M represents the fitness adjustment factor, g represents the fitness distribution exponent, N represents the population size, N1 represents a random number in the normal distribution and the problem dimension, N2 represents a random number in the normal distribution, β and W are given parameters, and r5 is a random number in the interval (0, 1).

[0019] In some embodiments, in the exploration stage and the exploitation stage, each gorilla individual updates its position by introducing the osprey optimization strategy according to the first preset probability, and the position update expression is:

[0020] GX(i, :) = X(observed i dx, :) - R × D × (X(observed i dx, :) - X(i, :));

[0021] Among them, R represents a random number in the interval (0, 1), D is the diving depth factor that controls the update depth, X(i, :) represents the position of the current individual in the solution space, and X(observed i dx, :) represents the position of the observed individual in the solution space.

[0022] In some embodiments, in the exploration stage and the exploitation stage, the golden sine algorithm is also introduced to adjust the update path of each gorilla position by using the golden ratio and sine volatility, and the position update strategy is:

[0023] When the generated random number is greater than or equal to 0.5, the position of each gorilla individual approaches the optimal solution:

[0024] GX(i,:) = X(i,:) + φ·sin(θ)×(Silverback - X(i,:));

[0025] When the generated random number is less than 0.5, move away from the optimal solution to increase the exploration range:

[0026] GX(i,:) = X(i,:) + φ·sin(θ)×(Silverback - X(i,:));

[0027] G is defined as the product of the golden ratio φ and the sine function, expressed as G = φ·sin(θ), where φ is the golden ratio with a value of 1.618, and θ is a randomly generated angle with a value range in [0, π];

[0028] Silverback represents the position of the current optimal solution; X(i,:) represents the position of the i-th gorilla individual; GX(i,:) represents the new position of the i-th gorilla individual after update.

[0029] In some embodiments, during the exploration phase and the exploitation phase, when the generated random number is less than the set perturbation rate k, the method introduces the complete random position reset to update the positions of the gorillas, and the expression is:

[0030] GX(i,:) = LB + (UB - LB)·rand(1, variables_no);

[0031] When the generated random number is in the interval (k, 2k), introduce the Gaussian noise perturbation to update the positions of the gorillas, and the expression is:

[0032] GX(i,:) = X(i,:) + normrnd(0, 1, [1, variables_no]);

[0033] Where, UB represents the upper bound of the search space, and LB represents the lower bound of the search space;

[0034] rand(1, variables_no) represents generating a vector with a dimension of variables_no, and each element is a uniform random number in the interval [0, 1]; normrnd(0, 1, [1, variables_no]) represents generating a vector with a dimension of variables_no, and each element is a normally distributed random number with a mean of 0 and a standard deviation of 1;

[0035] X(i,:) represents the position of the i-th gorilla individual; GX(i,:) represents the new position of the i-th gorilla individual after update.

[0036] In some embodiments, the target mechanical element is a multi-disc clutch brake, and the parameters to be optimized include: inner disc radius, outer disc radius, disc thickness, driving force, and number of friction surfaces;

[0037] The optimization objective is established as:

[0038]

[0039] where x1 represents the inner disc radius in millimeters; x2 represents the outer disc radius in millimeters; x3 represents the disc thickness in millimeters; x4 represents the driving force in Newtons; x5 represents the number of friction surfaces; the density ρ of the clutch disc has a value of 0.0000078 kg / mm 3 ;

[0040] The constraint conditions are constructed as follows:

[0041]

[0042] where the average contact pressure on the friction surface total friction surface area the linear velocity of the clutch contact surface the average radius of the clutch contact surface

[0043] where the difference ΔR between the outer diameter of the outer disc and the inner diameter of the inner disc is 20 mm;

[0044] where the maximum axial length L of the clutch max = 30 mm, the total operating clearance δ between the clutch friction plates is 0.5 mm,

[0045] where the theoretical maximum torque transmitted by the clutch the friction coefficient μ between the friction plates is 0.6; the safety factor s for the minimum torque requirement is 1.5; the minimum torque M required for the multi-disc clutch brake s = 40 Nm;

[0046] where the actual torque transmitted by the clutch angular velocity the moment of inertia I of the multi-disc clutch brake z = 55 Kg·m 2 ; fixed friction torque M f = 3 Nm;

[0047] where the maximum linear velocity V sr,max= 10 m / s, linear velocity

[0048] Wherein, the maximum operating time T allowed by the multi-disc clutch brake max = 15 s;

[0049] Rotational speed n = 250 rpm;

[0050] The maximum working pressure p that the multi-disc clutch brake can withstand max = 1 mpa;

[0051] And, the inner disc radius, the outer disc radius, the disc thickness, the driving force, and the number of friction surfaces conform to a preset boundary range.

[0052] In some embodiments, the target mechanical element is a rolling bearing, and the parameters to be optimized include: ball diameter, pitch diameter, inner raceway curvature coefficient, outer raceway curvature coefficient, and number of balls;

[0053] The optimization objective is established as:

[0054]

[0055] Wherein,

[0056]

[0057] D b represents the ball diameter, D m represents the pitch diameter, f i represents the inner raceway curvature coefficient, f o represents the outer raceway curvature coefficient, Z represents the number of balls; α represents the angle of the contact point between the inner raceway and the outer raceway relative to the bearing center line; f c is the load coefficient; γ is a parameter characterizing the internal geometry of the bearing;

[0058] The constraint conditions are constructed to include:

[0059]

[0060] K Dmin is a design parameter to ensure that the bearing design meets the minimum size requirements and safety standards;

[0061] K Dmax is a design parameter to ensure that the bearing design meets the maximum size requirements and safety standards;

[0062]

[0063] Among them, r i represents the effective radius of the inner ring; r0 represents the effective radius of the outer ring;

[0064] Angle parameter

[0065] The spatial constraint parameter T = D - d - 2D b , the diameter D of the bearing outer ring is 160 mm, and the diameter d of the bearing inner ring is 90 mm;

[0066]

[0067] Setting boundary conditions includes:

[0068] 0.5(D + d) ≤ D m ≤ 0.6(D + d);

[0069] 0.515(D - d) ≤ D b ≤ 0.45(D - d);

[0070] 4 ≤ 2 ≤ 50; 0.515 ≤ f i ≤ 0.6; 0.515 ≤ f0 ≤ 0.6;

[0071] 0.4 ≤ K D min ≤ 0.5; 0.6 ≤ K Dmax ≤ 0.7;

[0072] The contact elastic deformation parameter of the bearing is 0.3 ≤ ∈ ≤ 0.4;

[0073] The dimensionless coefficient related to clearance or preload is 0.02 ≤ e ≤ 0.1;

[0074] The damping ratio parameter is 0.6 ≤ ξ ≤ 0.85.

[0075] On the other hand, the present invention also provides a mechanical element parameter optimization device based on the artificial gorilla algorithm, including a processor, a memory, and a computer program / instructions stored on the memory. The processor is used to execute the computer program / instructions, and when the computer program / instructions are executed, the device implements the steps of the above method.

[0076] On the other hand, the present invention also provides a computer-readable storage medium, on which computer program / instructions are stored. It is characterized in that when the computer program / instructions are executed by a processor, the steps of the above method are implemented.

[0077] The beneficial effects of the present invention are at least:

[0078] The mechanical component parameter optimization method and device based on artificial intelligence algorithms according to the present invention, when solving for the optimal mechanical component parameters, updates the positions of the gorilla population based on the osprey algorithm introduced with probability, and controls the update depth based on the diving depth factor, making the update of the solution more flexible, helping to jump out of local optima, thereby improving the global search ability and convergence speed of the algorithm. The golden sine algorithm is introduced. In each iteration, by adjusting the random angle and using the periodic change of the sine function, the update of the individual position is both random and directional, enhancing the global search ability and helping the algorithm to jump out of local optimal solutions. As the iteration progresses, by dynamically adjusting the θ value, the algorithm can flexibly adjust the proportion of exploration and exploitation according to the current search state, effectively balancing the relationship between the two and improving the convergence speed. Introducing complete random position reset to avoid excessive aggregation around local optimal solutions, and introducing Gaussian noise perturbation to explore more carefully in the local area to avoid premature convergence, extending the persistence and effectiveness of the search.

[0079] Additional advantages, objects, and features of the present invention will be partly described below and will partly become apparent to those of ordinary skill in the art after studying the following, or can be learned from the practice of the present invention. The objects and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the specification and the drawings.

[0080] Those skilled in the art will understand that the objects and advantages that can be achieved by the present invention are not limited to the above specifically described, and the above and other objects that the present invention can achieve will be more clearly understood according to the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] The drawings described herein are used to provide a further understanding of the present invention, form a part of this application, and do not limit the present invention. In the drawings:

[0082] Figure 1 is a schematic flowchart of the mechanical component parameter optimization method based on artificial intelligence algorithms according to an embodiment of the present invention.

[0083] Figure 2 is a schematic structural diagram of a multi-disc clutch brake according to an embodiment of the present invention.

[0084] Figure 3 is the convergence graph of the best fitness with the number of iterations when the GSA, WOA, AO, GTO, and IGTO algorithms optimize the parameters of the Figure 2 multi-disc clutch brake described therein.

[0085] Figure 4 is a schematic structural diagram of a rolling bearing according to an embodiment of the present invention.

[0086] Figure 5The convergence graphs of the best fitness with the number of iterations when the GSA, WOA, AO, GTO, and IGTO algorithms optimize the Figure 4 rolling bearings described in

[0087] Figure 6 Stereogram of the planetary gear train according to an embodiment of the present invention.

[0088] Figure 7 For Figure 6 Cross-sectional view of the planetary gear train described above.

[0089] Figure 8 The convergence graphs of the best fitness with the number of iterations when the GSA, WOA, AO, GTO, and IGTO algorithms optimize the Figure 6 planetary gear train described above. Detailed implementation manners

[0090] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in combination with the implementation manners and the drawings. Herein, the illustrative implementation manners of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention.

[0091] Herein, it also needs to be noted that in order to avoid obscuring the present invention due to unnecessary details, only the structures and / or processing steps closely related to the solution of the present invention are shown in the drawings, while other details less related to the present invention are omitted.

[0092] It should be emphasized that the term "comprising / including" when used herein refers to the presence of features, elements, steps, or components, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0093] Herein, it also needs to be noted that if not otherwise specified, the term "connection" in this document can not only refer to direct connection, but also represent indirect connection with an intermediate.

[0094] During the design process of mechanical components, an optimization algorithm is used to search for the optimal solution. The Gorilla Troops Optimization (GTO) algorithm is a swarm intelligence optimization algorithm that simulates the foraging and social cooperation behaviors of gorilla troops to solve complex optimization problems. By initializing the population, selecting leaders, updating followers, cooperative foraging, and iterative updating, GTO effectively explores the search space to find the global optimal solution and is applicable to various application fields such as machine learning, engineering design, and optimization problems. However, the Gorilla Troops Optimization algorithm also has defects. For example, the performance of GTO is very sensitive to the setting of initial parameters, including population size, number of iterations, learning rate, etc. Inappropriate parameter settings may lead to slow convergence of the algorithm or failure to find the global optimal solution. During the search process, over-reliance on the direction of the leader may result in insufficient diversity of the search space, making it difficult to jump out of the local optimum. Excessive cooperative behavior and following strategies may limit the exploration of new solution space regions. In the later iteration process, the population diversity may be insufficient, resulting in individuals tending to be similar.

[0095] The present invention provides a method for optimizing mechanical component parameters based on an artificial intelligence algorithm, and the method includes the following steps S101 to S103:

[0096] Step S101: Based on the Gorilla Troops Optimization algorithm, for multiple parameters to be optimized of the target mechanical component, combined with the constraint conditions between the parameters to be optimized, perform random initialization to construct the positions of multiple gorilla individuals.

[0097] Step S102: In the exploration stage, by calculating the fitness of each gorilla individual, select the optimal individual as the silverback gorilla, search and update the position globally, where the fitness introduces a penalty function using the optimization objective and constraint conditions; in the exploitation stage, search and update the positions of gorilla members locally based on the mechanism of following the silverback gorilla or the mechanism of gorilla competition for adult females; wherein, in the exploration stage and the exploitation stage, each gorilla individual updates its position by introducing the osprey optimization strategy according to the first preset probability; in the exploration stage and the exploitation stage, the golden sine algorithm is also introduced to adjust the update path of each gorilla's position using the golden ratio and sine volatility; in the exploration stage and the exploitation stage, two mechanisms of completely random position reset and Gaussian noise perturbation are also introduced to update the positions of each gorilla.

[0098] Step S103: When the set number of iterations is reached or the fitness of the optimal individual reaches the set value, output the parameters corresponding to the optimal individual as the target values of the parameters to be optimized of the target mechanical component.

[0099] In step S101, in the parameter optimization problem of mechanical components, it is first necessary to construct it as a mathematical problem to clarify the optimization objectives and constraints. The optimization objectives can be set according to specific requirements. For example, by improving performance indicators such as strength, stiffness, wear resistance, or sealing performance, ensure that the performance of mechanical components meets the standards under actual working conditions. In addition, the optimization objectives can also include cost factors, such as reducing manufacturing costs, material costs, and maintenance costs, to minimize expenses while ensuring performance. At the same time, improving material utilization and reducing material waste should be considered to ensure that the components have the necessary strength and stability. Under the modern design trend, lightweight design is also an important objective, aiming to minimize the component weight while maintaining or improving its performance indicators, thereby saving energy and resources. Improving production efficiency is also a direction of optimization, which can be achieved by simplifying the manufacturing process and shortening the production cycle, thus improving efficiency and reducing costs. When setting constraints, it is necessary to be based on the functional requirements, material characteristics, manufacturing process, environmental conditions, and relevant standards and specifications of mechanical components. This includes ensuring the role of the component in the system, such as bearing loads and motion requirements, and conforming to the physical and mechanical properties of the material to ensure its safety within the expected service life. The limitations in the manufacturing process, such as machining accuracy, cost, and equipment availability, and its adaptability to work under specific environmental conditions (such as temperature, humidity, corrosive environment) also need to be considered. Finally, follow industry standards and specifications to ensure the rationality and safety of the design. These standards usually include regulations on some specific constraint conditions for parameters.

[0100] In order to introduce the artificial gorilla troop optimization algorithm for parameter optimization and solution, in this application, multiple solutions of each parameter to be optimized are constructed by random initialization as the initial positions of gorilla individuals to initiate optimization. A penalty function is constructed according to the constraints and optimization directions and incorporated into the fitness calculation to guide the optimization direction.

[0101] In step S102, the gorilla troop optimization algorithm (GTO) solves complex optimization problems by simulating the foraging behavior and group cooperation of gorillas in nature. The operation of this algorithm includes two main stages: the exploration stage and the exploitation stage.

[0102] In the exploration stage, the GTO is mainly responsible for searching the problem space to discover potential excellent solutions. This stage emphasizes the diversity of the population and the global search ability to avoid falling into local optima. When the population is initialized, each individual (solution) is randomly distributed in the search space, representing different possible solutions. Each individual is evaluated for its quality through a fitness function, which is usually based on the optimization objective, such as a performance metric or cost. Individuals in the gorilla tribe explore new solutions by simulating natural foraging behavior. The information exchange and mutual influence among individuals are particularly crucial at this stage, mimicking gorillas sharing foraging locations through visual and auditory signals. During this process, some individuals are selected as leaders (usually those with higher fitness), responsible for guiding the population towards potential optimal regions. Other individuals act as followers, relying on the guidance of the leaders for updates. This role assignment of leaders and followers improves the search efficiency while maintaining the diversity of the population. In the exploration stage, the population continuously adjusts the positions of individuals to expand the search scope and enhance the exploration ability for the global optimal solution. The exploration ability at this stage depends on how to balance the guidance of the leaders and the maintenance of population diversity, usually achieved through randomness and mutation strategies.

[0103] In some embodiments, the position update model of the silverback gorilla in the exploration stage is as follows:

[0104]

[0105] Where, F = cos(2r4) + 1; L = Cl; H = ZX(t); Z ∈ [-C, C];

[0106] X(t) represents the position of the silverback gorilla in the t-th iteration, and GX(t + 1) represents the updated position of the silverback gorilla; UB represents the upper bound of the search space, and LB represents the lower bound of the search space; r1, r2, r3, r4, and rand are random numbers between (0, 1); X r and GX r are both randomly selected gorilla positions in the t-th iteration; p ∈ (0, 1) is a given parameter used to simulate the influencing factor of the silverback gorilla individual's exploration of unknown positions; l is a random number in the interval (0, 1); Z is a random number in the interval [-C, C]; MaxIt represents the maximum number of iterations.

[0107] After entering the development stage, the focus of GTO shifts from extensive search to local development and fine-tuning to better approximate the optimal solution. In the development stage, individuals (solutions) with higher fitness in the population are regarded as potential local optimal solutions, and the algorithm will conduct more detailed exploration around these excellent solutions. Individuals will utilize their own experience and the guidance of the leader to further optimize their positions. The main task in the development stage is to refine the good solutions discovered in the previous exploration stage, reduce the gap between solutions, and improve the accuracy of solutions. In this stage, the leader continues to play a key role, but more local search strategies are adopted to explore the subtle changes in the solution space. The algorithm will gradually reduce the diversity of the population, making individuals tend to concentrate in a narrower region of the solution space to ensure gradual convergence to the global optimal solution. The search process in the development stage is usually accompanied by cooperative behaviors among individuals, simulating the cooperation of gorillas during foraging to improve the foraging success rate. The information exchange among individuals is more frequent to share each other's local optimal solution information, thereby more effectively conducting fine-tuning and optimization. The development stage emphasizes the depth rather than the breadth of optimization, and gradually approaches the global optimal solution by conducting fine-tuning and adjustment on the local area of the solution space.

[0108] In some embodiments, the position update model of each gorilla member in the development stage is as follows:

[0109] When C ≥ W, the silverback gorilla following mechanism is selected to update the positions of each gorilla member, and the expression is:

[0110]

[0111] When C < W, the competing adult female mechanism is selected to update the positions of each gorilla member, and the expression is:

[0112]

[0113] Among them, X silveriack is the position of the silverback gorilla, L represents the scaling factor, M represents the fitness adjustment factor, g represents the fitness distribution exponent, N represents the population size, N1 represents a random number in the normal distribution and the problem dimension, N2 represents a random number in the normal distribution, β and W are given parameters, and r5 is a random number between (0, 1).

[0114] In this process, the original Gorilla Troop Optimization (GTO) algorithm may face the problems of slow convergence speed and being easily trapped in local optima when dealing with complex multimodal optimization problems. To overcome these drawbacks, the improved algorithm in this application introduces the Osprey Optimization Algorithm (OOA), which is a strategy that mimics the foraging behavior of ospreys and is particularly suitable for enhancing global search ability and avoiding local optima. The specific improvement methods are as follows: In each iteration, each individual has a certain probability p to update its position according to the OOA strategy, which is used to enhance the global search ability and avoid local optima. Specifically, a random number can be generated before each update. If the random number is less than the first preset probability value, the osprey optimization algorithm is executed to update the position.

[0115] In some embodiments, in the exploration stage and the exploitation stage, each gorilla individual updates its position by introducing the osprey optimization strategy according to the first preset probability, and the position update expression is:

[0116] GX(i, :) = X(observed i dx, :) - R × D × (X(observed i dx, :) - X(i, :));

[0117] where R represents a random number between (0, 1), D is the diving depth factor that controls the update depth, X(i, :) represents the position of the current individual in the solution space, and X(observed i dx, :) represents the position of the observed individual in the solution space.

[0118] The benefits of introducing this strategy include increasing the diversity of the algorithm, enabling individuals in the population to explore a wider search space, and being able to quickly approach a better solution by simulating the positions of other individuals. In addition, the introduction of the diving depth factor makes the update of the solution more flexible, helps to jump out of local optima, and thus improves the global search ability and convergence speed of the algorithm. By this method, the improved algorithm in this application (labeled as IGTO) can effectively improve the optimization efficiency and the quality of the solution while maintaining the diversity of the solution.

[0119] Furthermore, the original Gorilla Troop Optimization (GTO) algorithm may be insufficient in terms of global search ability. Especially when exploring complex or extensive search spaces, it is difficult to effectively balance the relationship between exploration and exploitation, which may lead the algorithm to fall into local optimal solutions or have a slow convergence rate. To address these issues, the Golden Sine Algorithm (GSA) is introduced in the improved algorithm. GSA is a search strategy based on the golden ratio and sine function, aiming to optimize the search process of the algorithm. The core idea of GSA is to utilize the golden ratio and sine volatility to adjust the update path of the solution, thereby achieving a dynamic balance between exploration and exploitation.

[0120] In some embodiments, during the exploration phase and the exploitation phase, the Golden Sine Algorithm is also introduced to utilize the golden ratio and sine volatility to adjust the update path of the positions of each gorilla. The position update strategy is as follows:

[0121] When the generated random number is greater than or equal to 0.5, the positions of each gorilla individual approach the optimal solution:

[0122] GX(i, :) = Silverback + φ · sin(θ) × (Silverback - X(i, :));

[0123] When the generated random number is less than 0.5, move away from the optimal solution to increase the exploration range:

[0124] GX(i, :) = X(i, :) + φ · sin(θ) × (Silverback - X(i, :));

[0125] G is defined as the product of the golden ratio φ and the sine function, expressed as G = φ · sin(θ). φ is the golden ratio, with a value of 1.618, and θ is a randomly generated angle, with a value range in [0, π];

[0126] Silverback represents the position of the current optimal solution; X(i, :) represents the position of the i-th gorilla individual; GX(i, :) represents the new position of the i-th gorilla individual after update.

[0127] By introducing the GSA, in each iteration, the algorithm adjusts θ (random angle) and utilizes the periodic change of the sine function, enabling the update of individual positions to be both random and directional. Such changes bring the following benefits: 1) Enhance the global search ability: The volatility of the sine function helps the algorithm jump out of local optimal solutions and improve the diversity of solutions. 2) Dynamically adjust the search strategy: As the iteration progresses, by dynamically adjusting the value of θ, the algorithm can flexibly adjust the proportion of exploration and exploitation according to the current search state, effectively balancing the relationship between the two. 3) Improve the convergence speed: The golden ratio provides a mathematically ideal ratio, helping the algorithm to more precisely adjust the step size when approaching the global optimal solution.

[0128] Furthermore, the present application also introduces a dual perturbation strategy. The original gorilla troop optimization algorithm (GTO) may exhibit the problem of excessive convergence when dealing with high-dimensional and complex optimization problems, resulting in insufficient diversity of solutions and making the algorithm prone to falling into local optimal solutions. To enhance the ability of the algorithm to maintain diversity in a wide search space, a dual random perturbation strategy is introduced in the improved algorithm. The dual random perturbation strategy introduces two mechanisms, namely, complete random position reset and Gaussian noise perturbation, with the aim of introducing additional randomness during the search process to enhance the exploration ability of the population and avoid premature convergence.

[0129] Specifically, in the exploration stage and the development stage, when the generated random number is less than the set perturbation rate k, the method introduces a complete random position reset to update the positions of each gorilla. The expression is:

[0130] GX(i,:) = LB + (UB - LB)·rand[1, variables_no]);

[0131] When the generated random number is in the interval (k, 2k), Gaussian noise perturbation is introduced to update the positions of each gorilla. The expression is:

[0132] GX(i,:) = X(i,:) + normrnd(0, 1, [1, variables_no]);

[0133] Among them, UB represents the upper bound of the search space, and LB represents the lower bound of the search space;

[0134] rand(1, variables_no) represents generating a vector with the dimension of variables_no, and each element is a uniform random number in the interval [0, 1]; normrnd(0, 1, [1, variables_no]) represents generating a vector with the dimension of variables_no, and each element is a random number following a normal distribution with a mean of 0 and a standard deviation of 1;

[0135] X(i, :) represents the position of the i-th gorilla individual; GX(i, :) represents the new position of the i-th gorilla individual after update.

[0136] After introducing these perturbation mechanisms, the algorithm has obtained the following advantages: 1) Enhanced solution diversity: By completely randomly resetting the positions, the algorithm can explore new regions, avoid over-accumulating around local optimal solutions, and thus improve the global search ability. 2) Improved local search effect: The addition of Gaussian noise enables the algorithm to explore more meticulously in the local area, contributing to finding better solutions. 3) Avoid premature convergence: The introduction of the perturbation strategy reduces the risk of the algorithm prematurely converging to non-global optimal solutions, prolonging the persistence and effectiveness of the search.

[0137] Specifically, the present invention is illustrated with the following engineering case:

[0138] Embodiment 1

[0139] Multi-disc clutch brake design problem: The multi-disc clutch is an important mechanical transmission component, widely used in occasions that require transmitting large torques, such as automobiles, construction machinery, etc. The main problem in this scenario is to reduce the mass of the multi-disc clutch brake, aiming to improve fuel economy and performance through lightweight design.

[0140] In this scenario, the target mechanical element is the multi-disc clutch brake, and the parameters to be optimized include: inner disc radius, outer disc radius, disc thickness, driving force, and number of friction surfaces;

[0141] The established optimization objective is:

[0142]

[0143] Among them, as Figure 2 shown, x1 represents the inner disc radius, with the unit of millimeter; x2 represents the outer disc radius, with the unit of millimeter; x3 represents the disc thickness, with the unit of millimeter; x4 represents the driving force, with the unit of Newton; x5 represents the number of friction surfaces; the value of the density ρ of the clutch disc is 0.0000078 kg / mm 3 .

[0144] The constructed constraint conditions include:

[0145]

[0146] Among them, the average contact pressure on the friction surface The total friction surface area The linear velocity of the clutch contact surface The average radius of the clutch contact surface

[0147] Among them, the difference ΔR between the outer diameter of the outer disk and the inner diameter of the inner disk is 20 mm;

[0148] The total operating clearance δ between the clutch friction plates is 0.5 mm,

[0149] Among them, the theoretical maximum torque transmitted by the clutch The friction coefficient δ between the friction plates is 0.6; the safety factor s for the minimum torque requirement is 1.5; the minimum torque M required for the multi-disc clutch brake s = 40 Nm;

[0150] Among them, the actual torque transmitted by the clutch Angular velocity The moment of inertia I of the multi-disc clutch brake z = 55 Kg·m 2 ; The fixed friction torque M f = 3 Nm;

[0151] Among them, the maximum linear velocity V sr，max = 10 m / s, the linear velocity

[0152] Among them, the maximum operating time T allowed for the multi-disc clutch brake max = 15 s;

[0153] The rotational speed n = 250 rpm;

[0154] The maximum working pressure p that the multi-disc clutch brake can withstand max = 1 mpa;

[0155] In addition, the inner disk radius, outer disk radius, disk thickness, driving force, and number of friction surfaces conform to the preset boundary ranges.

[0156] Specifically, 60 ≤ x1 ≤ 80, 90 ≤ x2 ≤ 110, 1 ≤ x3 ≤ 3, 0 ≤ x4 ≤ 1000, and 2 ≤ x5 ≤ 9.

[0157] Based on the above parameters to be optimized, optimization objectives, and constraint conditions, optimization solutions are obtained respectively based on the Gravitational Search Algorithm (GSA), Whale Optimization Algorithm (WOA), Archimedes Optimization Algorithm (AO), traditional Gorilla Troop Optimization Algorithm (GTO), and the improved Gorilla Troop Optimization Algorithm (IGTO) of the above steps S101 - S103. The results are shown in Table 1:

[0158] Table 1 Optimal solutions obtained by each algorithm for the multi-disc clutch brake design problem

[0159]

[0160]

[0161] Meanwhile, the convergence curves of the five types of algorithms are as Figure 3 shown. It can be seen that compared with other traditional algorithms, the improved artificial gorilla population optimization algorithm achieves better results while ensuring the convergence speed.

[0162] Example 2

[0163] Rolling bearings are basic components in mechanical equipment. By optimizing design variables and parameters to improve the load-carrying capacity, the service life of the bearings can be extended, maintenance costs can be reduced, and the reliability of the equipment can be improved.

[0164] In this scenario, as Figure 4 shown, the target mechanical component is a rolling bearing, and the parameters to be optimized include: ball diameter, pitch diameter, inner raceway curvature coefficient, outer raceway curvature coefficient, and number of balls.

[0165] The optimization objective is established as:

[0166]

[0167] Among them,

[0168]

[0169] D b represents the ball diameter, D m represents the pitch diameter, f i represents the inner raceway curvature coefficient, f o represents the outer raceway curvature coefficient, Z represents the number of balls; α represents the angle of the contact point between the inner and outer raceways relative to the bearing center line; f c is the load coefficient; γ is a parameter characterizing the internal geometry of the bearing.

[0170] The constraint conditions are constructed as follows:

[0171]

[0172] K Dmin is a design parameter to ensure that the bearing design meets the minimum size requirements and safety standards;

[0173] K Dmax is a design parameter to ensure that the bearing design meets the maximum size requirements and safety standards;

[0174]

[0175]

[0176] Among them, r i represents the effective radius of the inner ring; r0 represents the effective radius of the outer ring;

[0177] Angle parameter

[0178] The space constraint parameter T = D - d - 2D b , the diameter D of the outer ring of the bearing is 160 mm, and the diameter d of the inner ring of the bearing is 90 mm;

[0179] Setting boundary conditions includes:

[0180] 0.5(D + d) ≤ D m ≤ 0.6(D + d);

[0181] 0.15(D - d) ≤ D b ≤ 0.45(D - d);

[0182] 4 ≤ Z ≤ 50; 0.515 ≤ f i ≤ 0.6; 0.515 ≤ f0 ≤ 0.6;

[0183] 0.4 ≤ K Dmin ≤ 0.5; 0.6 ≤ K Dmax ≤ 0.7;

[0184] The contact elastic deformation parameter of the bearing is 0.3 ≤ ∈ ≤ 0.4;

[0185] The dimensionless coefficient related to clearance or preload is 0.02 ≤ e ≤ 0.1;

[0186] The damping ratio parameter is 0.6 ≤ ζ ≤ 0.85.

[0187] Based on the above parameters to be optimized, optimization objectives and constraint conditions, optimization solutions are obtained respectively based on the Gravitational Search Algorithm (GSA), Whale Optimization Algorithm (WOA), Archimedes Optimization Algorithm (AO), traditional Gorilla Troop Optimization Algorithm (GTO), and the improved Gorilla Troop Optimization Algorithm (IGTO) of the above steps S101 - S103. The results are shown in Table 2:

[0188] Table 2 Optimal solutions obtained by each algorithm for the rolling bearing design problem

[0189]

[0190] At the same time, the convergence curves of the five algorithms are as Figure 5As shown, it can be seen that the improved artificial gorilla population optimization algorithm achieves better final results while ensuring the convergence speed compared to other traditional algorithms.

[0191] Embodiment 3

[0192] The planetary gear train is an efficient transmission method, commonly found in automotive gearboxes. Optimizing its design to minimize the maximum error of the gear ratio can improve transmission efficiency and performance, and reduce energy consumption and noise.

[0193] In this scenario, as Figure 6 and 7 shown, for the number of teeth N1, N2, N3, N4, N5, and N6 of gears 1 to 6 in the planetary gear train (where 2 and 3 are stepped transmission gears), in addition, there are three discrete design variables: the number of planetary gears (P) and two gear modules (m1 and m2).

[0194] The optimization objective is established as:

[0195] f(X) = max|i k - i 0k |, k = {1, 2, R}, where R represents the reverse gear ratio;

[0196] Among them, the gear ratio between the sun gear and the planetary gear The target gear ratio i 01 = 3.11;

[0197] The ratio from the planet carrier to the sun gear The target gear ratio i 0R = -3.11; The gear ratio of the planetary gear in the reverse mode The target gear ratio i 02 = 1.84;

[0198] The vector of design variables

[0199] The constructed constraint conditions include:

[0200] Among them, m3 represents the tooth face width; D max is the maximum gear diameter that can be accommodated within the interval;

[0201]

[0202] The meshing angle adjustment parameter δ 22 = δ 33 = δ 55 = δ 35 = δ 56 = 0.5;

[0203] The helix angle

[0204] Setting boundary conditions includes:

[0205] The module p of the gear = (3, 4, 5);

[0206] m1 = (1.75, 2.0, 2.25, 2.5, 2.75, 3.0); m3 = (1.75, 2.0, 2.25, 2.5, 2.75, 3.0);

[0207] 17 ≤ N1 ≤ 96; 14 ≤ N2 ≤ 54; 14 ≤ N3 ≤ 51; 17 ≤ N4 ≤ 46; 14 ≤ N5 ≤ 51; 48 ≤ N6 ≤ 124; N i = integer.

[0208] Based on the above parameters to be optimized, optimization objectives and constraint conditions, optimization solutions are obtained respectively based on the Gravitational Search Algorithm (GSA), Whale Optimization Algorithm (WOA), Archimedes Optimization Algorithm (AO), traditional Gorilla Troop Optimization Algorithm (GTO), and the improved Gorilla Troop Optimization Algorithm (IGTO) of the above steps S101 - S103. The results are shown in Table 3:

[0209] Table 3 Optimal solutions obtained by each algorithm for the planetary gear train design problem

[0210] Algorithm type N1 N2 N3 N4 N5 GSA 79.41399 38.26205 28.71733 42.10227 27.74959 WOA 25.8987 16.6013 16.5936 20.4397 17.0315 AO 56.3978 14.1309 14.581 35.1322 22.6731 GTO 36.8506 29.1215 29.0496 27.4814 21.4227 IGTO 34.2775 14.0719 16.1793 27.5211 19.5042 Algorithm type N6 P m1 m2 Optimal value of objective function GSA 101.0173 1.498923 4.593437 0.7814371 1.3896675e+20 WOA 61.5999 0.656143 0.980032 1.47611 0.77172 AO 82.1184 1.19565 2.4739 1.11086 3.81e+17 GTO 93.218 0.898603 1.05641 1.03293 0.59735 IGTO 99.7357 1.73991 6.49 0.51 0.54971

[0211] During the application process, the parameters of N1 - N6 and P in Table 3 are rounded to integers.

[0212] Meanwhile, the convergence curves of the five algorithms are as Figure 8 shown. It can be seen that compared with other traditional algorithms, the improved Gorilla Troop Optimization Algorithm has a better final effect while ensuring the convergence speed.

[0213] Correspondingly to the above method, the present invention also provides a device / system. The device / system includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the device / system implements the steps of the method described above.

[0214] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the foregoing edge computing server deployment method are implemented. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, register, floppy disk, hard disk, removable storage disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0215] On the other hand, the present invention also provides a computer program product, including a computer program / instructions, characterized in that when the computer program / instructions are executed by a processor, the steps of the above method are implemented.

[0216] In summary, for the method and device for optimizing mechanical element parameters based on an artificial intelligence algorithm according to the present invention, when solving for the optimal mechanical element parameters, the osprey algorithm is introduced based on probability to update the positions of the gorilla population, and the update depth is controlled based on the diving depth factor, making the update of the solution more flexible, helping to jump out of local optima, thereby improving the global search ability and convergence speed of the algorithm. The golden sine algorithm is introduced. In each iteration, by adjusting the random angle and using the periodic change of the sine function, the update of the individual position is both random and directional, enhancing the global search ability and helping the algorithm to jump out of the local optimal solution. As the iteration progresses, by dynamically adjusting the θ value, the algorithm can flexibly adjust the proportion of exploration and exploitation according to the current search state, effectively balancing the relationship between the two and improving the convergence speed. Introducing complete random position reset avoids excessive aggregation around the local optimal solution, and introducing Gaussian noise perturbation can explore more carefully in the local area, avoiding premature convergence and prolonging the persistence and effectiveness of the search.

[0217] Those of ordinary skill in the art should understand that the various exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Specifically, whether to implement in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, etc. When implemented in software, the elements of the present invention are programs or code segments used to execute the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave on a transmission medium or a communication link.

[0218] It should be clear that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present invention.

[0219] In the present invention, features described and / or illustrated for one embodiment can be used in the same or a similar manner in one or more other embodiments, and / or combined with the features of other embodiments or replace the features of other embodiments.

[0220] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and modifications can be made to the embodiments of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for optimizing mechanical component parameters based on artificial intelligence algorithm, characterized in that: The method comprises the following steps: Based on the artificial gorilla population optimization algorithm, multiple parameters to be optimized of the target mechanical component are randomly initialized in combination with the constraints between the parameters to be optimized to construct multiple gorilla individual positions; In the exploration phase, the fitness of each individual gorilla is calculated, and the best individual is selected as the silverback gorilla, and the position is searched and updated in the global scope, wherein the fitness introduces a penalty function using the optimization target and the constraint condition; in the development phase, the positions of gorilla members are searched and updated in the local scope based on the follow-the-silverback-gorilla mechanism or the gorilla competition for adult females mechanism; wherein, in the exploration phase and the development phase, each individual gorilla introduces the osprey optimization strategy according to the first preset probability to update the position; in the exploration phase and the development phase, the golden sine algorithm is also introduced to adjust the update path of the position of each gorilla using the golden ratio and sine volatility; in the exploration phase and the development phase, two mechanisms, namely, completely random position reset and Gaussian noise perturbation, are also introduced to update the position of each gorilla; When the set number of iterations is reached or the fitness of the optimal individual reaches a set value, the parameters corresponding to the optimal individual are output as target values ​​of the parameters to be optimized of the target mechanical element.

2. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 1, characterized in that: The position update model of the silverback gorilla in the exploration phase is: in, F=cos(2r4)+1; L=Cl; H=ZX(t); Z∈[-C, C]; X(t) represents the position of the silverback gorilla in the tth iteration, GX(t+1) represents the position of the silverback gorilla after the update; UB represents the upper bound of the search space, LB represents the lower bound of the search space; r1, r2, r3, r4 and rand are random numbers between (0,1); X r and GX r are all randomly selected gorilla positions in the tth iteration; p∈(0,1) is a given parameter used to simulate the influencing factors of the silverback gorilla individual's exploration of unknown positions; l is a random number in the interval (0,1); Z is a random number in the interval [-C,C]; MaxIt represents the maximum number of iterations.

3. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 2, characterized in that: The position update model of each gorilla member in the development stage is: When C≥W, the following silverback gorilla mechanism is selected to just update the positions of the gorilla members, and the expression is: When C<W, the competitive adult female mechanism is selected to just update the positions of the gorilla members, and the expression is: Among them, X silveriack is the position of the silverback gorilla, L represents the scaling factor, M represents the fitness adjustment factor, g represents the fitness distribution index, N represents the population size, N1 represents the random number in the normal distribution and the problem dimension, N2 represents the random number in the normal distribution, β and W are given parameters, and r5 is a random number between (0,1).

4. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 3, characterized in that: In the exploration stage and the development stage, each gorilla individual introduces the osprey optimization strategy to update its position according to the first preset probability. The position update expression is: GX(i,:)=X(observed i dx,:)-R×D×(X(observed i dx,:)-X(i,:)); Among them, R represents a random number between (0,1), D is the diving depth factor that controls the update depth, X(i,:) represents the position of the current individual in the solution space, and X(observed i dx,:) represents the position of the observed individual in the solution space.

5. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 4, characterized in that: In the exploration stage and the development stage, the golden sine algorithm is also introduced to use the golden ratio and sine volatility to adjust the update path of the positions of the various gorillas. The position update strategy is: When the generated random number is greater than or equal to 0.5, the positions of the individual gorillas are close to the optimal solution: GX(i,:)=Silverback+φ·sin(θ)×(Silverback-X(i,:)); When the generated random number is less than 0.5, move away from the optimal solution to increase the exploration range: GX(i,:)=X(i,:)+φ·sin(θ)×(Silverback-X(i,:)); G is defined as the product of the golden ratio φ and the sine function, expressed as G = φ·sin(θ), where φ is the golden ratio, with a value of 1.618, and θ is a randomly generated angle, with a value range of [0,π]; Silverback represents the position of the current optimal solution; X(i,:) represents the position of the i-th gorilla individual; GX(i,:) represents the new position of the i-th gorilla individual after updating.

6. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 5, characterized in that: In the exploration phase and the development phase, when the generated random number is less than the set disturbance rate k, the method introduces the completely random position reset to update the positions of the gorillas, and the expression is: GX(i,:)=LB+(UB-LB)·rand(1, variables_no); When the generated random number is in the interval (k, 2k), the Gaussian noise disturbance is introduced to update the positions of the gorillas. The expression is: GX(i,:)=X(i,:)+normmd(0,1,[1,variables_no]); Wherein, UB represents the upper bound of the search space, and LB represents the lower bound of the search space; rand(1,variables_no) means generating a vector with dimension variables_no, each element of which is a uniform random number in the interval [0,1]; normrnd(0,1,[1,variables_no]) means generating a vector with dimension variables_no, each element of which is a normally distributed random number with mean 0 and standard deviation 1; X(i,:) represents the position of the i-th gorilla individual; GX(i,:) represents the new position of the i-th gorilla individual after updating.

7. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 6, characterized in that: The target mechanical element is a multi-disc clutch brake, and the parameters to be optimized include: inner disk radius, outer disk radius, disk thickness, driving force, and number of friction surfaces; The optimization goal is established as: Wherein, x1 represents the inner disk radius, in millimeters; x2 represents the outer disk radius, in millimeters; x3 represents the disk thickness, in millimeters; x4 represents the driving force, in Newtons; x5 represents the number of friction surfaces; the clutch disk density ρ is 0.0000078 kg / mm 3 ; Constructing the constraints includes: The average contact pressure on the friction surface is Total friction surface area Linear speed of the clutch contact surface The average radius of the clutch contact surface Among them, the difference between the outer diameter of the outer disk and the inner diameter of the inner disk is ΔR = 20 mm; Among them, the maximum axial length of the clutch is L max =30mm, total running clearance between clutch friction plates δ = 0.5mm, The theoretical maximum torque transmitted by the clutch is The friction coefficient between the friction plates μ = 0.6; the safety factor on the minimum torque requirement s = 1.5; the minimum torque required for the multi-disc clutch brake M s =40Nm; The actual torque transmitted by the clutch is Angular velocity The moment of inertia of the multi-disc clutch brake is z =55Kg·m 2 ; Fixed friction torque M f =3Nm; Among them, the maximum linear velocity V sr，max =10m / s, linear speed The maximum operating time allowed by the multi-disc clutch brake is T max =15s; Rotation speed n = 250 rpm; The maximum working pressure p that the multi-disc clutch brake can withstand max =1mpa; Furthermore, the inner disk radius, the outer disk radius, the disk thickness, the driving force and the number of friction surfaces conform to a preset boundary range.

8. The method for optimizing mechanical component parameters based on artificial intelligence algorithm according to claim 6, characterized in that: The target mechanical element is a rolling bearing, and the parameters to be optimized include: ball diameter, pitch diameter, inner raceway curvature coefficient, outer raceway curvature coefficient and number of balls; The optimization goal is established as: in, D b Denotes the ball diameter, D m represents the pitch diameter, f i represents the inner raceway curvature coefficient, f o represents the curvature coefficient of the outer raceway, Z represents the number of balls; α represents the angle between the contact point of the inner raceway and the outer raceway relative to the center line of the bearing; f c is the load factor; γ is a parameter that characterizes the internal geometry of the bearing; Constructing the constraints includes: K Dmin Design parameters to ensure that the bearing design meets minimum size requirements and safety standards; K Dmax Design parameters to ensure that the bearing design meets the maximum size requirements and safety standards; in, r i Indicates the effective radius of the inner circle; r0 represents the effective radius of the outer ring; Angle parameters Spatial constraint parameter T = Dd-2D b , the diameter of the bearing outer ring is D = 160mm, the diameter of the bearing inner ring is d = 90mm; Setting boundary conditions includes: 0.5(D+d)≤D m ≤0.6(D+d); 0.15(D-d)≤D b ≤0.45(D-d); 4≤Z≤50;0.515≤f i ≤0.6;0.515≤f0≤0.6; 0.4≤K Dmin ≤0.5;0.6≤K Dmax ≤0.7; The contact elastic deformation parameter of the bearing is 0.3≤∈≤0.4; Dimensionless coefficient related to clearance or preload 0.02≤e≤0.1; Damping ratio parameter 0.6≤ζ≤0.

85.

9. A device for optimizing mechanical component parameters based on an artificial gorilla algorithm, comprising a processor, a memory, and a computer program / instruction stored in the memory, characterized in that: The processor is used to execute the computer program / instructions. When the computer program / instructions are executed, the device implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.

Images