Shock absorber structure parameter optimization method based on multi-target genetic algorithm

Abstract

The invention discloses a shock absorber structure parameter optimization method based on a multi-target genetic algorithm. The method comprises the following steps that 1, the piston movement speed of a shock absorber is collected through a sensor; 2, establishing a target function; 3, solving an optimal solution for the target function according to a multi-target genetic algorithm, and obtaining an optimal structure parameter; wherein the structural parameters comprise the length of an effective area of a damping channel, the length of a non-effective area of the damping channel, a damping channel gap, the depth of a coil groove, the radius difference of an iron core coil, the thickness of a cylinder barrel and the radius of a piston rod. The method has the characteristics that the applicability is improved, the damping performance is optimized, and the vehicle driving comfort and the fuel efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of vibration damper technology, and more specifically, to a method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm. Background Technology

[0002] Shock absorbers are an indispensable key component in the automotive structure. They can effectively mitigate the forced vibration of the vehicle body caused by road surface or other factors during driving, thereby ensuring the stability of the vehicle during driving and reducing body sway.

[0003] However, existing shock absorbers are all mass-produced products and cannot be designed according to specific needs. They suffer from problems such as short piston stroke and low efficiency. Existing research on shock absorbers mainly focuses on the study of damping force and directly optimizes the performance of semi-active suspension through damping force. This has the problems of large amount of calculation and poor correlation. Therefore, it is urgent to optimize shock absorbers from other perspectives. Summary of the Invention

[0004] The purpose of this invention is to design and develop a method for optimizing the structural parameters of a shock absorber based on a multi-objective genetic algorithm. By adjusting the structure of the shock absorber based on the damping force according to different piston movement speeds, the overall structure and damping characteristics of the shock absorber are optimized.

[0005] The technical solution provided by this invention is as follows:

[0006] A method for optimizing the structural parameters of a vibration damper based on a multi-objective genetic algorithm includes the following steps:

[0007] Step 1: Collect the piston speed of the shock absorber using sensors;

[0008] Step 2: Establish the objective function:

[0009]

[0010] st700N≤F≤9800N

[0011] 1.5≤α≤5

[0012] 5≤a1≤8

[0013] 10≤a²≤18

[0014] 0.5 ≤ h ≤ 1.5

[0015] 5≤b≤10

[0016] 5≤b1≤10

[0017] 2.5≤D≤4

[0018] 5.5 ≤ r ≤ 7.5

[0019] In the formula, f1 is the first optimization objective, f2 is the second optimization objective, F is the damping force, α is the dynamic adjustable coefficient, and F′ is a multi-objective optimization problem;

[0020] Step 3: Use a multi-objective genetic algorithm to find the optimal solution for the objective function and obtain the optimal structure parameters;

[0021] The structural parameters include: effective region length of the damping channel, ineffective region length of the damping channel, damping channel gap, coil slot depth, core coil radius difference, cylinder thickness, and piston rod radius.

[0022] Preferably, the first optimization objective satisfies:

[0023] f1 = -F;

[0024] In the formula, F is the damping force.

[0025] Preferably, the second optimization objective satisfies:

[0026] f2 = -α;

[0027] In the formula, α is the dynamic adjustable coefficient.

[0028] Preferably, the damping force satisfies:

[0029]

[0030] In the formula, F is the damping force, δ is the dynamic viscosity coefficient of the magnetorheological fluid, a is the length of the electrode, b is the width of the electrode, h is the distance between the electrodes, and A is the distance between the electrodes. i Let be the effective area of ​​the piston, v be the piston speed, τ be the shear strength of the shear zone, and sgn(v) be the velocity symbol.

[0031] Preferably, the velocity symbol satisfies:

[0032]

[0033] Preferably, the power adjustability coefficient satisfies:

[0034]

[0035] In the formula, F δ For the viscous damping force of the liquid, F τ This is the Coulomb damping force.

[0036] Preferably, the first optimization objective satisfies:

[0037]

[0038] In the formula, a1 is the effective length of the damping channel, a2 is the ineffective length of the damping channel, and R1 is the outer diameter of the working cylinder.

[0039] Preferably, the second optimization objective satisfies:

[0040]

[0041] In the formula, b is the depth of the coil slot, b1 is the difference in radius between the iron core coils, r is the radius of the piston rod, h is the damping channel clearance, and D is the cylinder thickness.

[0042] Preferably, the multi-objective genetic algorithm specifically includes the following steps:

[0043] Step 1: Initialize the population;

[0044] Step 2: Determine the fitness function;

[0045] Wherein, the fitness value is the objective function;

[0046] Step 3: Select individuals with high fitness from the initial population using a roulette wheel to serve as parents;

[0047] Step 4: Perform gene crossover and mutation operations on the selected parent generation to generate the offspring population;

[0048] Step 5: Combine the parent and offspring populations, and select the optimal individual using non-dominated sorting, storing it in the Pareto solution set;

[0049] Step 6: Update the Pareto solution set using elite back learning and differential evolution algorithm:

[0050] like Then, the elite reverse learning algorithm is used for updating:

[0051]

[0052] In the formula, As a decision variable, and A random number between 0 and 1. The newly generated solution is t, where t is the current iteration number. Let Q be the original value of the i-th solution, and let Q be a random number. This is the upper bound of the Pareto solution set. It is the lower bound of the Pareto solution set;

[0053] like The value is then updated using the differential evolution algorithm:

[0054]

[0055] In the formula, Let be the mutation value of the i-th solution. β is a random number between 0 and 1. cro Crossover rate, The fitness value when using the mutation value. The fitness value when using the original value;

[0056] The variation value of the i-th solution satisfies:

[0057]

[0058] In the formula, Y is the scaling factor, and r1, r2, and r3 are the scaling factors between 1 and N. p X1(t), X2(t), and X3(t) are three distinct integers randomly selected from i, and are values ​​of three different individuals randomly selected from the current population.

[0059] Step 7: Repeat the above steps until the required number of iterations is reached or all objective functions have converged, thus obtaining the optimal structure parameters.

[0060] The beneficial effects of this invention are as follows:

[0061] This invention presents a method for optimizing the structural parameters of a shock absorber based on a multi-objective genetic algorithm. By adjusting the overall structure of the shock absorber according to different piston movement speeds and combining damping force with the multi-objective genetic algorithm, the method optimizes the damping characteristics of the shock absorber. Supplementary structural parameters can also be adjusted according to actual requirements and usage conditions, providing a more flexible shock absorber design scheme, thereby improving applicability, optimizing damping performance, and enhancing vehicle driving comfort and fuel efficiency. Attached Figure Description

[0062] Figure 1 This is a flowchart illustrating the multi-objective genetic algorithm described in this invention. Detailed Implementation

[0063] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0064] This invention provides a method for optimizing the structural parameters of a vibration damper based on a multi-objective genetic algorithm, which specifically includes the following steps:

[0065] Step 1: Collect the piston speed of the shock absorber using sensors;

[0066] Step 2: Establish the objective function:

[0067]

[0068] st700N≤F≤9800N

[0069] 1.5≤α≤5

[0070] 5≤a1≤8

[0071] 10≤a²≤18

[0072] 0.5 ≤ h ≤ 1.5

[0073] 5≤b≤10

[0074] 5≤b1≤10

[0075] 2.5≤D≤4

[0076] 5.5 ≤ r ≤ 7.5

[0077] In the formula, f1 is the first optimization objective, f2 is the second optimization objective, F is the damping force, α is the dynamic adjustable coefficient, and F′ is a multi-objective optimization problem;

[0078] The first optimization objective and the second optimization objective satisfy:

[0079] f1 = -F;

[0080] f2 = -α;

[0081] Specifically:

[0082] Consider the flow mode of a magnetorheological damper, where the space between two stationary plates is filled with a magnetorheological fluid. The damping force under this mode satisfies:

[0083]

[0084] In the formula, F is the damping force, δ is the dynamic viscosity coefficient of the magnetorheological fluid, a is the length of the electrode, b is the width of the electrode, h is the distance between the electrodes, and A is the distance between the electrodes. i Let be the effective area of ​​the piston, v be the piston velocity, τ be the shear strength of the shear zone, and sgn(v) be the velocity sign, and:

[0085]

[0086] Therefore, the first optimization objective is:

[0087]

[0088] In the formula, a1 is the effective region length of the damping channel, a2 is the ineffective region length of the damping channel, and R1 is the outer diameter of the working cylinder.

[0089] The second optimization objective is as follows:

[0090]

[0091] In the formula, F δ For the viscous damping force of the liquid, Fτ The Coulomb damping force specifically satisfies:

[0092]

[0093] Therefore, the second optimization objective is:

[0094]

[0095] In the formula, b is the depth of the coil slot, b1 is the difference in the radius of the iron core coil, r is the radius of the piston rod, h is the damping channel clearance, and D is the cylinder thickness.

[0096] Step 3: Use a multi-objective genetic algorithm to find the optimal solution for the objective function and obtain the optimal structure parameters;

[0097] The structural parameters include: effective region length of the damping channel, ineffective region length of the damping channel, damping channel gap, coil slot depth, core coil radius difference, cylinder thickness, and piston rod radius.

[0098] like Figure 1 As shown, the multi-objective genetic algorithm specifically includes the following steps:

[0099] Step 1: Initialize the population;

[0100] A population is composed of many individuals, and the number of individuals is also called the population size.

[0101] Step 2: Determine the fitness function;

[0102] Wherein, the fitness value is the objective function;

[0103] Step 3: Select individuals with high fitness from the initial population using a roulette wheel to serve as parents;

[0104]

[0105] In the formula, p i Let p be the fitness value of the i-th individual. j Let be the fitness value of the j-th individual;

[0106] Step 4: Perform gene crossover and mutation operations on the selected parent generation to generate the offspring population;

[0107] Step 5: Combine the parent and offspring populations, and select the optimal individual using non-dominated sorting, storing it in the Pareto solution set;

[0108] Step 6: Update the Pareto solution set using elite back learning and differential evolution algorithm:

[0109] like Then, the elite reverse learning algorithm is used for updating:

[0110]

[0111] In the formula, As a decision variable, and A random number between 0 and 1. The newly generated solution is t, where t is the current iteration number. Let Q be the original value of the i-th solution, and let Q be a random number. This is the upper bound of the Pareto solution set. It is the lower bound of the Pareto solution set;

[0112] like The value is then updated using the differential evolution algorithm:

[0113]

[0114] In the formula, Let be the mutation value of the i-th solution. β is a random number between 0 and 1. cro Crossover rate, The fitness value when using the mutation value. The fitness value when using the original value;

[0115] The variation value of the i-th solution satisfies:

[0116]

[0117] In the formula, Y is the scaling factor, and r1, r2, and r3 are the scaling factors between 1 and N. p X1(t), X2(t), and X3(t) are three distinct integers randomly selected from i, and are values ​​of three different individuals randomly selected from the current population.

[0118] Step 7: Repeat the above steps until the required number of iterations is reached or all objective functions have converged, thus obtaining the optimal structure parameters.

[0119] In this embodiment, the population size is 100, the crossover probability is 0.5–0.9, the mutation probability is 0.01–0.1, the dynamic viscosity coefficient of the magnetorheological fluid is 0.6 Pa·s, and the shear yield strength is 45 kPa. In this embodiment, the damping force and the dynamic adjustable coefficient are inversely correlated. Initially, when the dynamic adjustable coefficient is 2.2, the structural parameters are: effective damping channel length is 6 mm, ineffective damping channel length is 13 mm, damping channel gap is 1.2 mm, coil slot depth is 9 mm, core coil radius difference is 8.55 mm, cylinder thickness is 3.25 mm, and piston rod radius is 6.25 mm. The first optimization objective is... The target value is -6734.03, and the second optimization target is -1.86, both within the objective function constraint range. The optimal solution is: the effective region length of the damping channel is 5mm, the ineffective region length of the damping channel is 15mm, the damping channel gap is 1.5mm, the coil slot depth is 8.5mm, the core coil radius difference is 9mm, the cylinder thickness is 3mm, and the piston rod radius is 6mm. At this time, the first optimization target is -4306.96, and the second optimization target is -2.34, which are also within the objective function constraint range. The maximum damping force after optimization is less than the maximum damping force before optimization, but the maximum damping adjustable coefficient is increased, which can improve the damping characteristics and the adjustment range of the magnetic circuit.

[0120] This invention presents a method for optimizing the structural parameters of a shock absorber based on a multi-objective genetic algorithm. By adjusting the overall structure of the shock absorber according to different piston movement speeds and combining damping force with the multi-objective genetic algorithm, the method optimizes the damping characteristics of the shock absorber. Supplementary structural parameters can also be adjusted according to actual requirements and usage conditions, providing a more flexible shock absorber design scheme, thereby improving applicability, optimizing damping performance, and enhancing vehicle driving comfort and fuel efficiency.

[0121] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A method for optimizing the structural parameters of a vibration damper based on a multi-objective genetic algorithm, characterized in that, Includes the following steps: Step 1: Collect the piston speed of the shock absorber using sensors; Step 2: Establish the objective function: In the formula, f1 is the first optimization objective, f2 is the second optimization objective, F is the damping force, α is the dynamic adjustable coefficient, and F′ is a multi-objective optimization problem; Step 3: Use a multi-objective genetic algorithm to find the optimal solution for the objective function and obtain the optimal structure parameters; The structural parameters include: effective region length of the damping channel, ineffective region length of the damping channel, damping channel gap, coil slot depth, core coil radius difference, cylinder thickness, and piston rod radius.

2. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 1, characterized in that, The first optimization objective satisfies: f1 = -F; In the formula, F is the damping force.

3. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 2, characterized in that, The second optimization objective satisfies: f2＝-α； In the formula, α is the dynamic adjustable coefficient.

4. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 3, characterized in that, The damping force satisfies: In the formula, F is the damping force, δ is the dynamic viscosity coefficient of the magnetorheological fluid, a is the length of the electrode, b is the width of the electrode, h is the distance between the electrodes, and A is the distance between the electrodes. i Let be the effective area of ​​the piston, v be the piston speed, τ be the shear strength of the shear zone, and sgn(v) be the velocity symbol.

5. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 4, characterized in that, The velocity symbol satisfies:

6. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 5, characterized in that, The power adjustability coefficient satisfies: In the formula, F δ For the viscous damping force of the liquid, F τ This is the Coulomb damping force.

7. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 6, characterized in that, The first optimization objective satisfies: In the formula, a1 is the effective length of the damping channel, a2 is the ineffective length of the damping channel, and R1 is the outer diameter of the working cylinder.

8. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 7, characterized in that, The second optimization objective satisfies: In the formula, b is the depth of the coil slot, b1 is the difference in radius between the iron core coils, r is the radius of the piston rod, h is the damping channel clearance, and D is the cylinder thickness.

9. The method for optimizing vibration damper structural parameters based on a multi-objective genetic algorithm as described in claim 8, characterized in that, The multi-objective genetic algorithm specifically includes the following steps: Step 1: Initialize the population; Step 2: Determine the fitness function; Wherein, the fitness value is the objective function; Step 3: Select individuals with high fitness from the initial population using a roulette wheel to serve as parents; Step 4: Perform gene crossover and mutation operations on the selected parent generation to generate the offspring population; Step 5: Combine the parent and offspring populations, and select the optimal individual using non-dominated sorting, storing it in the Pareto solution set; Step 6: Update the Pareto solution set using elite back learning and differential evolution algorithm: like Then, the elite reverse learning algorithm is used for updating: In the formula, As a decision variable, and A random number between 0 and 1. The newly generated solution is t, where t is the current iteration number. Let Q be the original value of the i-th solution, and let Q be a random number. This is the upper bound of the Pareto solution set. It is the lower bound of the Pareto solution set; like The value is then updated using the differential evolution algorithm: Where, Let be the mutation value of the i-th solution. β is a random number between 0 and 1. cro Crossover rate, The fitness value when using the mutation value. The fitness value when using the original value; The variation value of the i-th solution satisfies: In the formula, Y is the scaling factor, and r1, r2, and r3 are the scaling factors between 1 and N. p X1(t), X2(t), and X3(t) are three distinct integers randomly selected from i, and are values ​​of three different individuals randomly selected from the current population. Step 7: Repeat the above steps until the required number of iterations is reached or all objective functions have converged, thus obtaining the optimal structure parameters.

Images