Diesel engine connecting rod structure parameter optimization method

Abstract

The invention discloses a diesel engine connecting rod structure parameter optimization method which specifically comprises the following steps: (1) obtaining load values, weights and stress values which can be borne by an oil extraction machine connecting rod under different connecting rod body structure parameters through finite element analysis, and taking the load values, the weights and the stress values as samples; (2) completing the training of a Kriging agent model based on the constructed sample; and (3) searching an optimal connecting rod body structure parameter in a feasible space based on a particle swarm algorithm by taking the connecting rod body structure parameter as a position code, taking a maximum load value which can be borne by the connecting rod and a lightweight design requirement as constraint conditions and taking a minimum stress value as a target function. The trained Kriging proxy model is used for stress values, bearable load values and weight corresponding to connecting rod body structure parameters of all position codes in the particle swarm algorithm, and then the maximum load value is found quickly, and the connecting rod body structure parameters corresponding to the minimum stress value are found under the lightweight design requirement.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of parameter optimization, and more particularly, the present application relates to a diesel engine connecting rod structure parameter optimization method. BACKGROUND

[0002] As a key moving part of the internal combustion engine power transmission system, the structural strength and lightweight level of the diesel engine connecting rod directly affect the reliability, energy efficiency conversion and service life of the diesel engine. Under dynamic conditions, the connecting rod needs to withstand the combined action of periodic gas explosion pressure and inertial load, and its core structural parameters directly determine the stress distribution pattern and fatigue damage evolution law.

[0003] The explosion pressure of modern high-power density diesel engines has exceeded 25MPa, and the traditional design method based on empirical coefficients faces the challenge of lightweight and high reliability optimization. The current parameter optimization field of key components of diesel engines still has the following problems:

[0004] Traditional finite element parameter optimization relies on full-factor iterative calculation, and a single simulation takes several hours. When the optimized finite element parameters are used for prediction in the nonlinear region by the response surface method, there is a problem of large prediction error. SUMMARY

[0005] The present application provides a diesel engine connecting rod structure parameter optimization method, which aims to improve at least one of the above problems.

[0006] The present application is implemented as follows: a diesel engine connecting rod structure parameter optimization method, the method is as follows:

[0007] (1) Obtain the load value, weight and stress value that the diesel engine connecting rod can withstand under different connecting rod body structure parameters through finite element analysis, and use them as samples;

[0008] (2) Train the Kriging surrogate model based on the constructed samples;

[0009] (3) Take the connecting rod body structure parameters as position encoding, take the maximum load value that the connecting rod can withstand and the lightweight design requirement as constraint conditions, take the minimum stress value as the objective function, and search for the optimal connecting rod body structure parameters in the feasible space based on the particle swarm algorithm.

[0010] Further, the construction process of the sample is as follows:

[0011] (11) Construct a three-dimensional model of the diesel engine connecting rod, import the constructed three-dimensional model of the diesel engine connecting rod into the Ansys software, set the material parameters of the connecting rod body, divide the three-dimensional model of the diesel engine connecting rod into grids, and apply constraints and working load;

[0012] (12) input the connecting rod body structure parameters of each connecting rod into the three-dimensional model of the connecting rod of the diesel engine, and obtain the load value, weight and stress value that the oil pumping machine connecting rod can bear under the corresponding connecting rod body structure parameters.

[0013] Further, the connecting rod body structure parameters of the position of the particle are input into the trained Kriging surrogate model, and the Kriging surrogate model outputs the stress value, the load value that can be borne and the weight corresponding to the connecting rod body structure parameters of the position of the particle, and finds the connecting rod body structure parameters corresponding to the minimum stress value under the maximum load value and the lightweight design requirement.

[0014] Further, all samples are divided into training samples and test samples according to a set proportion, the Kriging surrogate model is trained based on the training samples, the Kriging surrogate model is tested based on the test samples, and the Kriging surrogate model training is completed after the test accuracy of the Kriging surrogate model reaches a set requirement.

[0015] Further, the velocity update formula of the particle is as follows:

[0016] v i (t+1)=ωv i (t)+c1r1(pbest i -x i (t))+c2r2(gbest i -x i (t));

[0017] Wherein, ω is an inertia weight, c1 and c2 are respectively an individual learning factor and a social learning factor, r1 and r2 are random numbers, x i (t), x i (t+1) respectively represent the position of particle i at time t and time t+1, v i (t), v i (t+1) respectively represent the speed of particle i at time t, pbest i represents the current optimal position of particle i, and gbest i represents the current optimal position of all particles.

[0018] Further, the position update formula of the particle is as follows:

[0019] x i (t+1)=x i (t)+v i (t+1)

[0020] Wherein, x i (t), x i(t+1) respectively represent the velocity of particle i at time t+1.

[0021] Further, the optimal Latin hypercube method is used for sampling multiple groups of connecting rod body structure parameters.

[0022] Further, a three-dimensional model of the connecting rod of the diesel engine is constructed by using Solidworks software.

[0023] Further, the length L, thickness M and fillet radius R of the connecting rod body are taken as the connecting rod body structure parameters

[0024] The trained Kriging surrogate model is used for the stress value, the load value that can be borne and the weight of the connecting rod body structure parameters coded by each position in the particle swarm algorithm, so that the maximum load value and the minimum stress value corresponding to the connecting rod body structure parameters under the lightweight design requirement are quickly found. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flowchart of the diesel engine connecting rod structure parameter optimization method provided by the embodiment of the present application is provided.

[0026] Figure 2 A schematic diagram of the key structure parameters of the connecting rod body is provided by the embodiment of the present application, wherein (a) is a top view of the connecting rod, and (b) is a side view of the connecting rod.

[0027] Figure 3 A schematic diagram of the stress corresponding to the connecting rod body structure parameters obtained by finite element analysis is provided by the embodiment of the present application. DETAILED DESCRIPTION

[0028] The specific embodiments of the present application are further described in detail below by comparing the drawings and the embodiments, to help the skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present application.

[0029] Figure 1 A flowchart of the diesel engine connecting rod structure parameter optimization method provided by the embodiment of the present application is provided, and the method is specifically as follows:

[0030] (1) The load value, weight and stress value of the connecting rod of the oil pumping machine under different connecting rod body structure parameters are obtained by finite element analysis, and are taken as samples;

[0031] In the embodiment of the present application, the construction process of the sample is specifically as follows:

[0032] (11) using Solidworks software to build a three-dimensional model of the connecting rod of the diesel engine, importing the three-dimensional model of the connecting rod of the diesel engine into the Ansys software, setting the material parameters of the connecting rod body, meshing the three-dimensional model of the connecting rod of the diesel engine, and applying constraints and working load;

[0033] (22) using the optimal Latin hypercube method to sample multiple groups of connecting rod body structure parameters;

[0034] (23) inputting each group of connecting rod body structure parameters into the three-dimensional model of the connecting rod of the diesel engine to obtain the load value, weight and stress value that the connecting rod of the diesel engine can withstand under the corresponding connecting rod body structure parameters.

[0035] The present application selects the length L, thickness M and over-round radius R of the connecting rod body as the connecting rod body structure parameters, as shown in Figure 2 (a) connecting rod top view and Figure 2 (b) connecting rod side view, Figure 3 The connecting rod body structure parameters provided by the embodiment of the present application correspond to the schematic diagram of stress analysis.

[0036] The simulation model of the connecting rod of the diesel engine is constructed using Solidworks software, the mass of the connecting rod is 20.8 kg, the material of the connecting rod is set to be alloy steel, the number of meshing is 33948 units, the small end of the connecting rod is fixed, and the working load of 15 MPa is applied to the large end of the connecting rod to obtain the stress value of the connecting rod.

[0037] The optimal Latin hypercube method is used to design 60 groups of data points for the structure parameters of the connecting rod body, and the core target is to improve the uniformity and orthogonality of the samples in the parameter space through mathematical optimization, and to ensure that the structure parameters of the 60 groups of data sets are modeled and simulated to obtain the stress values of the 60 groups of connecting rod models.

[0038] (2) based on the sample, the Kriging surrogate model is trained;

[0039] In the embodiment of the present application, all samples are divided into training samples and test samples according to the ratio of 2:1, the Kriging surrogate model is trained based on the training samples, the Kriging surrogate model is tested based on the test samples, and the Kriging surrogate model training is completed after the test accuracy of the Kriging surrogate model reaches the set requirement.

[0040] (3) taking the connecting rod body structure parameters as the position coding, taking the maximum load value that the connecting rod can withstand and the lightweight design requirement as the constraint condition, taking the minimum stress value as the objective function, and searching for the optimal solution of the connecting rod structure parameters in the feasible space based on the particle swarm algorithm.

[0041] In the embodiment of the present application, when the minimum stress value corresponding to the connecting rod body structure parameter is searched based on the particle swarm algorithm under the maximum load value and the lightweight design requirement, the stress value corresponding to the connecting rod body structure parameter at the position of the particle needs to be determined, the connecting rod body structure parameter at the position of the particle is input into the trained Kriging surrogate model, the Kriging surrogate model outputs the stress value, the load value that can be borne and the weight corresponding to the connecting rod body structure parameter at the position of the particle, and then the minimum stress value corresponding to the connecting rod body structure parameter (global optimal position) is found.

[0042] In the embodiment of the present application, the particle swarm algorithm (PSO) is an optimization algorithm based on swarm intelligence, which gradually approaches the global optimal solution by simulating the search behavior of individuals (particles) in the solution space, and is widely used in function optimization, machine learning, engineering design and other fields. Each particle represents a candidate solution in the solution space, and has two attributes of current solution and search direction. The particle adjusts its motion by tracking the following two optimal values: individual historical optimal position pbest and global optimal position gbest. The particle dynamically adjusts the motion direction according to the current speed, the difference between the individual optimal and the global optimal:

[0043] v i (t+1)=ωv i (t)+c1r1(pbest i -x i (t))+c2r2(gbest i -x i (t));

[0044] x i (t+1)=x i (t)+v i (t+1);

[0045] Wherein, ω is the inertia weight, c1 and c2 are the individual learning factor and social learning factor respectively, r1 and r2 are random numbers, x i (t), x i (t+1) represent the position of particle i at time t and time t+1 respectively, v i (t), v i (t+1) represent the speed of particle i at time t and time t+1 respectively, pbest i represents the current optimal position of particle i, and gbest i represents the current optimal position of all particles.

[0046] The application adopts the connecting rod rod body structure parameters by optimal Latin hypercube, obtains the stress values corresponding to the connecting rod rod body structure parameters based on the Ansys software simulation, forms samples of the load values and weights that can be borne, completes the training of the riging proxy model based on the samples, uses the trained Kriging proxy model for the stress values corresponding to the position coding connecting rod rod body structure parameters in the particle swarm algorithm, the load values and weights that can be borne, and then quickly finds the maximum load value and the minimum stress value corresponding to the connecting rod rod body structure parameters under the lightweight design requirement.

[0047] The application is described exemplarily, and it is obvious that the specific implementation of the application is not limited by the above mode, as long as various non-essential improvements are made by adopting the method concept and technical scheme of the application, or the concept and technical scheme of the application is directly applied to other occasions without improvement, which are all within the protection scope of the application.

Claims

1. A method for optimizing the structural parameters of a diesel engine connecting rod, characterized in that: The method is specifically as follows: (1) The load, weight, and stress values ​​that the connecting rod of the oil production machine can withstand under different connecting rod body structural parameters are obtained through finite element analysis and used as samples; (2) Complete the training of the Kriging proxy model based on the constructed samples; (3) The connecting rod body structural parameters are used as position codes, the maximum load value that the connecting rod can withstand and the lightweight design requirements are used as constraints, and the minimum stress value is used as the objective function. The optimal connecting rod body structural parameters are searched in the feasible space based on the particle swarm algorithm.

2. The diesel engine connecting rod structural parameter optimization method according to claim 1, characterized in that: The sample construction process is as follows: (11) Construct a three-dimensional model of the diesel engine connecting rod, import the constructed three-dimensional model of the diesel engine connecting rod into the Ansys software, set the material parameters of the connecting rod body, mesh the three-dimensional model of the diesel engine connecting rod, and apply constraints and working loads; (12) The structural parameters of each connecting rod body are input into the three-dimensional model of the diesel engine connecting rod to obtain the load value, weight and stress value that the oil production machine connecting rod can withstand under the corresponding connecting rod body structural parameters.

3. The diesel engine connecting rod structural parameter optimization method according to claim 1, characterized in that: The connecting rod shaft structural parameters encoded at the particle location are input into the trained Kriging proxy model. The Kriging proxy model outputs the stress value, load value, and weight corresponding to the connecting rod shaft structural parameters encoded at the particle location, and finds the connecting rod shaft structural parameters corresponding to the minimum stress value under the maximum load value and lightweight design requirements.

4. The diesel engine connecting rod structural parameter optimization method according to claim 1, characterized in that: All samples are divided into training samples and test samples according to the set ratio. The Kriging proxy model is trained based on the training samples and tested based on the test samples. When the test accuracy of the Kriging proxy model reaches the set requirements, the Kriging proxy model training is completed.

5. The diesel engine connecting rod structural parameter optimization method according to claim 1, characterized in that: The particle velocity update formula is as follows: v i (t+1)=ωv i (t)+c1r1(pbest i -x i (t))+c2r2(gbest i -x i (t)); Among them, ω is the inertia weight, c1 and c2 are individual learning factors and social learning factors respectively, r1 and r2 are random numbers, and x i (t), x i (t+1) represents the position of particle i at time t and time t+1, v i (t), represents the velocity of particle i at time t, pbest i Indicates the current optimal position of particle i, gbest i Indicates the current optimal position of all particles.

6. The diesel engine connecting rod structural parameter optimization method according to claim 5, characterized in that: The particle position update formula is as follows: x i (t+1)=x i (t)+v i (t+1) Among them, x i (t), x i (t+1) represents the position of particle i at time t and time t+1, v i (t+1) represents the velocity of particle i at time t+1.

7. The diesel engine connecting rod structural parameter optimization method according to claim 1, characterized in that: The optimal Latin hypercube method is used to sample multiple groups of connecting rod shaft structural parameters.

8. The diesel engine connecting rod structural parameter optimization method according to claim 2, characterized in that: Use Solidworks software to build a 3D model of the diesel engine connecting rod.

9. The diesel engine connecting rod structural parameter optimization method according to claim 2, characterized in that: The length L, thickness M and transition circle radius R of the connecting rod shaft are used as the connecting rod shaft structural parameters.

Images