CN-121997738-A - Gasoline engine silencer design method based on multi-target particle swarm optimization

CN121997738ACN 121997738 ACN121997738 ACN 121997738ACN-121997738-A

Abstract

The invention relates to the technical field of design of gasoline engine silencers, and discloses a design method of a gasoline engine silencer based on multi-target particle swarm optimization, which comprises the steps of collecting core parameters of a gasoline engine and setting silencer design constraint; the method comprises the steps of determining a double objective function with maximized optimized variables and minimized transmission loss and minimized total back pressure of a silencer design, initializing multi-objective particle swarm optimization algorithm parameters, generating initial particle swarm, performing non-dominant sorting and layering on the particle swarm, storing a 1 st layer non-dominant solution into an external archive, updating particle speed and position, repeating the non-dominant solution sorting and updating iteration of the particles, outputting the non-dominant solution in the external archive as an optimal solution set, and combining dynamic operation parameters of a gasoline engine through a fuzzy logic dynamic matching mechanism to obtain a dynamic working condition optimal solution so as to solve the technical problems that the prior art cannot break through single objective design limitation and cannot cope with contradiction between sound transmission loss and back pressure coupling.

Inventors

Mao Zhanxin
QU HANSHI
Yi Shouchun
DUAN JIAQUAN
FU NA
LIU MINGLI
HU SHIYI

Assignees

中国汽车工程研究院股份有限公司
中国第一汽车股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (7)

1. A design method of a gasoline engine silencer based on multi-target particle swarm optimization is characterized by comprising the following steps: core parameters of the gasoline engine are collected, muffler design constraints are set, and the design constraints comprise noise reduction performance, back pressure, muffler space and muffler structure manufacturing cost; Determining optimization variables and multiple optimization targets of the silencer design, constructing objective functions of quantitative mapping of the optimization variables and the optimization targets, wherein the multiple optimization targets comprise transmission loss TL maximization and total back pressure Minimizing; Initializing multi-target particle swarm optimization algorithm parameters, including particle numbers, iteration times and external archiving capacity, uniformly and randomly generating initial particle swarm in a constraint space, and executing feasibility test, wherein only particles meeting constraint conditions are reserved; The method comprises the steps of performing non-dominant sorting on particle groups, dividing the particle groups into multiple layers according to dominant relations, storing a layer 1 non-dominant solution into an external archive, taking the particle with the greatest crowding degree in the external archive as a global leader g_best, updating the particle speed and the position, performing constraint verification again on the updated particle, and updating an individual optimal solution p_best if the constraint verification is satisfied, otherwise, giving up; Repeating the particle non-dominant solution sorting and updating iteration until the iteration round reaches the maximum iteration round of initial setting, removing overflow inferior solutions through non-dominant sorting according to the external archiving capacity, and outputting non-dominant solutions in external archiving as Pareto optimal solution sets; and the optimal solution of the dynamic working condition is obtained by combining the dynamic operation parameters of the gasoline engine through a fuzzy logic dynamic matching mechanism.
2. The method for designing the gasoline engine silencer based on the multi-target particle swarm optimization according to claim 1, wherein the constraint conditions comprise transmission loss TL not less than 15dB and total back pressure of the silencer Muffler volume less than or equal to 3kPa Less than or equal to 2.5L, the number of muffler chambers ≤4。
3. The method for designing a muffler for a gasoline engine based on multi-objective particle swarm optimization according to claim 2, wherein the objective functions comprise a first objective function and a second objective function, First objective function: ; Wherein 2 pi f/c is the basic composition of wave number k, and is expressed completely by multiplying the geometric relation of exhaust pipe length L, wherein f is the sound wave frequency; Is the inlet sectional area of the exhaust pipe; The number of muffler chambers; Is the muffler volume; Second objective function: ; Wherein K is a constant, and epsilon is the threading rate.
4. The method for designing the gasoline engine silencer based on the multi-target particle swarm optimization is characterized by initializing parameters of a multi-target particle swarm optimization algorithm, further comprising an inertia weight w and a learning factor, specifically, the particle number is 80, the iteration number is 80, the inertia weight w linearly decreases by 3 from 0.7 to 0.3 according to the iteration number, the setting of the individual optimal weight and the global optimal weight is equal to 2, the external archiving capacity is 50, and the speed boundary is +/-20% variable range.
5. The method for designing a muffler of a gasoline engine based on multi-objective particle swarm optimization according to claim 4, wherein the non-dominant order of the particles is divided into 3 layers, and the 1 st layer non-dominant solution is set to 22, the 2 nd layer 35 and the 3 rd layer 23.
6. The method for designing the gasoline engine silencer based on multi-target particle swarm optimization according to claim 5, wherein in the core parameter acquisition process, the displacement is calculated The error of (2) is less than or equal to +/-5%, the error of the rated rotation speed n is less than or equal to +/-100 rpm.
7. The method for designing a muffler for a gasoline engine based on multi-objective particle swarm optimization according to claim 6, wherein said core parameters include displacement The rated rotation speed N, the stroke number T, the cylinder number N, the sound velocity c, the exhaust gas density rho and the inlet cross section area of the exhaust pipe 。

Description

Gasoline engine silencer design method based on multi-target particle swarm optimization Technical Field The invention relates to the technical field of muffler design, in particular to a method for designing a gasoline engine muffler based on multi-target particle swarm optimization. Background Under the dual driving of automobile noise control and environmental protection regulations, the design of a gasoline engine silencer has become a core technical field for improving the NVH performance of a vehicle and reducing the emission pollution. However, the existing design method of the gasoline engine silencer has multi-dimensional technical limitation, and is difficult to meet the comprehensive performance requirement under complex working conditions: The single-target/step design contradiction is that the traditional scheme mostly adopts a step strategy of silencing priority-back pressure check, and after parameters such as the volume, the expansion ratio and the like of the silencer are determined through an empirical formula or single-target optimization, the pipeline size is adjusted through fluid simulation so as to reduce the back pressure. This mode results in a strong coupling contradiction between noise cancellation and back pressure that cannot be reconciled-for example, increasing the expansion ratio or the number of chambers increases the sound Transmission Loss (TL), but increases the local drag coefficient and the perforated pipe resistance simultaneously, resulting in an out-of-specification back pressure, whereas decreasing the back pressure may sacrifice noise cancellation performance, creating a zero and game of "noise cancellation-back pressure". The objective function is incomplete, namely the existing optimization model focuses on double targets of volume-sound transmission loss, or only considers the balance of noise reduction performance and volume, the back pressure threshold value is not used as a constraint condition, and an explicit mathematical model of back pressure and size parameters is not established. This results in the possibility that there may be "volume and transmission loss up to standard but back pressure out of standard" solutions in Pareto solutions, or "optimal transmission loss but cost too high" engineering infeasible solutions, e.g. an increase in cavity diameter results in a square increase in material cost, an increase in the number of perforations resulting in an increase in processing man-hours. The parameter coupling characteristic ignores that part of the method adopts test logic of fixing other parameters and only optimizing single variable, and therefore, adopts an electric digital data processing technology to construct a silencer design optimization platform based on physical field coupling. For example, independent tuning of the main pipe inner diameter, throat length, ignores the strong interactions between the design parameters of the muffler entirely. The local optimization can only acquire the parameter optimal value under the fixed working condition, and cannot capture the global optimal solution under the multi-parameter coupling, so that the performance fluctuation in the actual working condition is obvious. The cost-performance balance is lost, namely, the objective function and the constraint condition are not included into cost factors, so that the partial Pareto optimal solution meets the performance index, but the engineering cannot be landed due to the too high cost. Disclosure of Invention The invention aims to provide a design method of a gasoline engine silencer based on multi-target particle swarm optimization, which aims to solve the technical problems that the prior art cannot break through the single-target design limitation and cannot cope with the contradiction between sound transmission loss and back pressure coupling. In order to achieve the aim, the invention adopts the following technical scheme that the design method of the gasoline engine silencer based on multi-target particle swarm optimization comprises the following steps: core parameters of the gasoline engine are collected, muffler design constraints are set, and the design constraints comprise noise reduction performance, back pressure, muffler space and muffler structure manufacturing cost; Determining optimization variables and multiple optimization targets of the silencer design, constructing objective functions of quantitative mapping of the optimization variables and the optimization targets, wherein the multiple optimization targets comprise transmission loss TL maximization and total back pressure Minimizing; Initializing multi-target particle swarm optimization algorithm parameters, including particle numbers, iteration times and external archiving capacity, uniformly and randomly generating initial particle swarm in a constraint space, and executing feasibility test, wherein only particles meeting constraint conditions are reserved; The method comprises the