Multi-objective Optimization Method for General Operation Power Platform of Machinery in Hilly and Mountainous Areas Based on Regression Model

Abstract

The present invention discloses a multi-objective optimization method for a general operation power platform of hilly and mountainous area machinery based on a regression model, and the present invention relates to the technical field of data optimization. The method includes the following steps: determining each performance parameter and each design parameter of the general operation power platform of hilly and mountainous area machinery, establishing a finite element model of the power platform, and characterizing the performance parameters of the general operation power platform of hilly and mountainous area machinery through the finite element model; establishing a regression model between the performance parameters of the power platform and each design parameter, and at the same time determining the regression coefficients through the Bayesian regression algorithm to generate a linear regression model; constructing a fitness function based on each performance parameter of the linear regression model, and determining the target chromosome through a genetic algorithm; obtaining the environmental parameters of the hilly and mountainous area where the power platform works, and correcting the gene parameters in the target chromosome based on the obtained environmental parameters to obtain the optimal design parameters of the general operation power platform of hilly and mountainous area machinery.

Description

Technical Field

[0001] The present invention relates to the technical field of data optimization, and in particular to a multi-objective optimization method for a general operation power platform of hilly and mountainous area machinery based on a regression model. Background Art

[0002] In the rapidly developing fields of agriculture and engineering machinery today, the operating environment in hilly and mountainous areas poses higher requirements for the performance of mechanical equipment. Hilly areas often face problems such as complex terrain, steep slopes, and variable soil conditions, which directly affect the operating efficiency and stability of mechanical equipment. Traditional mechanical design methods often focus on the optimization of a single performance index and lack comprehensive consideration of multiple performance parameters, resulting in difficulty in meeting diverse operating requirements in specific applications.

[0003] In addition, with the rapid development of intelligent and automated technologies, the design of mechanical equipment also needs to keep up with the times and adopt more advanced design methods to improve performance. The multi-objective optimization technology based on regression models, especially its application in complex environments, is becoming an important trend in the new generation of mechanical design. By quantitatively analyzing the relationship between design parameters and performance parameters, the optimal combination of each performance index can be achieved. Existing design methods often lack real-time feedback and adaptability to environmental factors, resulting in the final design often not conforming to the actual operating conditions, which affects the operating efficiency of machinery in hilly areas.

[0004] At the same time, environmental parameters in hilly areas, such as slope and soil moisture, directly affect the operating performance of machinery, but these factors are often not fully considered in the design stage. This may cause problems such as insufficient power, poor stability, and low operating efficiency during actual operation of the machinery, thereby affecting the overall production efficiency and economic benefits. Therefore, how to effectively integrate environmental parameters and mechanical performance in the design process and improve the operating ability of hilly and mountainous area machinery through scientific optimization methods has become a technical problem to be solved urgently.

[0005] In the prior art, the publication number CN117725764B discloses a multi-objective optimization method, device and medium for a vehicle chassis based on a regression model. By determining each prediction parameter and each design factor, for each prediction parameter in the linear list, a linear regression model is determined based on the corresponding sample set. For each prediction parameter in the non-linear list, the threshold values corresponding to each design factor are determined, and then the corresponding threshold regression model is constructed. The model coefficients therein are determined through the sample set, and a threshold effect and threshold value significance test are performed on each threshold regression model. The threshold regression models that do not pass the test are excluded, and the linear regression model is determined. Finally, each objective function is optimized in sequence according to a preset optimization order. However, this method can only optimize each objective function in sequence according to the preset optimization order, and has weak global optimization ability. At the same time, the influence of environmental conditions on each design factor is not considered, resulting in poor actual performance of the optimization result, unable to meet the expected requirements, and also reducing the accuracy and effectiveness of the system.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The purpose of the present invention is to provide a multi-objective optimization method for a general operation power platform of hilly and mountainous area machinery based on a regression model to solve the problems raised in the above background art.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] A multi-objective optimization method for a general operation power platform of hilly and mountainous area machinery based on a regression model, the specific steps include:

[0010] Determine each performance parameter and each design parameter of the general operation power platform of hilly and mountainous area machinery, establish a finite element model of the general operation power platform of hilly and mountainous area machinery based on each design parameter, and characterize the performance parameters of the general operation power platform of hilly and mountainous area machinery through the finite element models with different design parameters;

[0011] Based on each performance parameter and each design parameter determined by the finite element model, establish a regression model between the performance parameters of the power platform and each design parameter, and at the same time determine the regression coefficients through the Bayesian regression algorithm, and generate a complete linear regression model based on the determined regression coefficients;

[0012] Each design parameter based on the linear regression model constitutes a gene, and the combination of all different types of genes constitutes an individual chromosome. A fitness function is constructed based on each performance parameter of the linear regression model, and the genetic algorithm is used to determine an individual chromosome with the highest fitness, denoted as the target chromosome, where the design parameters include the design parameters of the power platform, and the performance parameters include the performance parameters of the power platform;

[0013] Obtain the environmental parameters of the hilly and mountainous areas where the power platform operates, and correct the gene parameters in the target chromosome based on the obtained environmental parameters to obtain the optimal design parameters of the general-purpose power platform for hilly and mountainous machinery operations, where the environmental parameters of the hilly and mountainous areas include the average slope and the average soil humidity.

[0014] Furthermore, determine the performance parameters and design parameters of the general-purpose power platform for hilly and mountainous machinery operations. The performance parameters of the general-purpose power platform for hilly and mountainous machinery operations include: maximum passing height, suspension stroke, ground contact pressure, turning radius, load capacity, and energy consumption; the design parameters of the general-purpose power platform for hilly and mountainous machinery operations include: engine power, torque, suspension stiffness, platform width, platform height, center of gravity height, and total platform weight; obtain the value range of each design parameter according to the existing public information, and randomly select a set of design parameter values based on the value range of each design parameter of the general-purpose power platform for hilly and mountainous machinery operations, and input them into the established finite element model. Through simulation tests on the finite element model, obtain the performance parameters corresponding to this set of design parameters, and record no less than m groups of test data as the sample set.

[0015] Furthermore, based on the performance parameters and design parameters determined by the finite element model, the specific steps for establishing the regression model between the performance parameters of the power platform and each design parameter include: calibrate the performance parameters of the power platform, and calibrate the maximum passing height as C 1 , calibrate the suspension stroke as C 2 , and calibrate the ground contact pressure, turning radius, load capacity, and energy consumption as C 3 , C 4 , C 5 and C 6 ;

[0016] Calibrate the design parameters of the power platform, and calibrate the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height, and total platform weight as X 1 , X 2 , X 3 , X 4 , X 5 , X 6 and X 7 ;

[0017] Establish a regression model expression between the performance parameters of the power platform and each design parameter. The specific regression model expression is as follows:

[0018] C i = β i，0 + β i，1 * X 1 + β i，2 * X 2 + … + β i，7 * X 7 + ∈ i

[0019] In the formula, C i represents the i-th performance parameter, β i，1 to β i，7 represent the regression coefficients of each design parameter in the regression model expression of the i-th performance parameter, β i，0 represents the intercept term in the regression model expression of the i-th performance parameter, ∈ i represents the error term in the regression model expression of the i-th performance parameter, where i is the index of the performance parameter, i = 1, 2, …, 6;

[0020] Determine the regression coefficients through the Bayesian regression algorithm, and generate a complete linear regression model based on the determined regression coefficients.

[0021] Furthermore, the specific steps for determining the regression coefficients through the Bayesian regression algorithm include:

[0022] First, define a prior distribution for each regression coefficient; based on the collected performance parameter and design parameter data, establish a likelihood function; according to Bayes' theorem, calculate the posterior distribution through the prior distribution and the likelihood function; extract the estimated values of the regression coefficients from the posterior distribution, and substitute the extracted regression coefficient values into the regression model expression to obtain a complete linear regression model.

[0023] Furthermore, each design parameter based on the linear regression model constitutes a gene, which is specifically represented as: calibrate T to represent all optional sets of all design parameters, where T = {P 1 , P 2 , …, P u , …, P n}, where P u = {X u1 , X u2 , …, X u7}, P u represents the design parameter set of the u-th group, n represents the number of parameter combinations, encode each parameter in the optimized parameter combination one by one as a gene, and the genes generated from the same type of data are allelic to each other, that is, X 11 , X21 , …, X n1 are allelic to each other;

[0024] According to the obtained individual chromosome of the gene composition, the individual chromosome is composed of 7 genes. Each individual chromosome S has 7 gene positions, and each gene position corresponds to a gene, that is, X 11 , X 21 , …, X n1 are all optional values of the 1st gene position of the individual chromosome. The 7 gene positions respectively correspond to a certain gene of the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height, and total platform weight parameters.

[0025] Furthermore, a fitness function is constructed based on each performance parameter of the linear regression model, and the expression of the fitness function is:

[0026]

[0027] In the formula, f(S) represents the fitness function, ω 1 , ω 2 , ω 3 , ω 4 , ω 5 and ω 6 , are respectively the weight coefficients of the maximum passing height, suspension stroke, ground contact pressure, turning radius, load capacity, and energy consumption, where ω 1 , ω 2 , ω 3 , ω 4 , ω 5 and ω 6 are all greater than 0:

[0028] Determine an individual chromosome with the highest fitness through the genetic algorithm. The specific steps of the genetic algorithm include: constructing an initial population based on the individual chromosome; performing selection, crossover, and mutation operations on the individual chromosomes in the initial population; performing iterative operations, and setting the mutation probability and iterative termination conditions. When the iterative operation terminates, the individual chromosome with the maximum fitness in the current population is recorded as the target chromosome.

[0029] Furthermore, the logic for correcting the gene parameters in the target chromosome based on the obtained environmental parameters is: correcting the engine power, torque, and suspension stiffness gene parameters in the target chromosome, and the specific formulas for correction are respectively:

[0030] X″ 1 = X′ 1 +(1 + X′ 7 *g*sinθ)

[0031] X″ 2= X' 2 +(1 + X' 7 *g*r*sinθ)

[0032] X″ 3 = X' 3 *(1 + α*sinθ + δ*H)

[0033] wherein, X″ 1 、X″ 2 and X″ 3 respectively represent the corrected engine power, torque and suspension stiffness. X' 1 、X' 2 and X' 3 respectively represent the values of the engine power, torque and suspension stiffness in the target chromosome. X' 7 represents the total platform weight data in the target chromosome, g is the acceleration due to gravity, θ represents the average slope, r is the radius of the power platform wheel, α and δ are respectively the slope influence coefficient and the soil moisture influence coefficient, where α > δ, and both α and δ are greater than 0, and H is the average soil moisture;

[0034] Replace the corrected engine power, torque and suspension stiffness parameters with the engine power, torque and suspension stiffness data in the target chromosome, and finally obtain the optimal design parameters of the general operation power platform for hilly and mountainous machinery, that is, the data included in the replaced target chromosome.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] By establishing a finite element model, the relationship between various performance parameters and design parameters of the power platform can be comprehensively characterized, providing data support for the optimization design. Secondly, the regression coefficients are determined by using the regression model and the Bayesian regression algorithm, which can improve the accuracy of the model and ensure accurate performance prediction under different design parameters. In addition, combining with the genetic algorithm for optimization can quickly find the optimal solution in the multi-dimensional parameter space, significantly improving the design efficiency and achieving comprehensive optimization. After obtaining the environmental parameters, the target chromosome is corrected, so that the finally obtained design parameters not only meet the performance requirements, but also can adapt to the actual operation environment. This optimization design method that comprehensively considers design parameters and environmental factors ensures the stability and efficiency of hilly and mountainous machinery in complex environments, greatly improving the operation ability and economy of the machinery, and thus providing a more reliable technical guarantee for agricultural and engineering operations in hilly areas. Through this innovative design optimization method, the mechanical failure rate can be effectively reduced and the work efficiency can be improved. Brief Description of the Drawings

[0037] Figure 1 It is a schematic diagram of the overall method flow of the present invention. Detailed Embodiment

[0038] In order to make the objectives, technical solutions, and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0039] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second", and similar terms used in the present invention do not indicate any order, quantity, or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0040] Embodiment:

[0041] Please refer to Figure 1 , the present invention provides a technical solution:

[0042] A multi-objective optimization method for a general operation power platform of hilly and mountainous area machinery based on a regression model, the specific steps include:

[0043] Step 1: Determine each performance parameter and each design parameter of the general operation power platform of hilly and mountainous area machinery, establish a finite element model of the general operation power platform of hilly and mountainous area machinery based on each design parameter, and characterize the performance parameters of the general operation power platform of hilly and mountainous area machinery through the finite element models with different design parameters.

[0044] Determine each performance parameter and each design parameter of the general operation power platform of hilly and mountainous area machinery, wherein the performance parameters of the general operation power platform of hilly and mountainous area machinery include: maximum passing height, suspension stroke, ground contact pressure, turning radius, load capacity, and energy consumption; the design parameters of the general operation power platform of hilly and mountainous area machinery include: engine power, torque, suspension stiffness, platform width, platform height, center of gravity height, and total platform weight; obtain the value ranges of each design parameter according to the existing public information, based on the value ranges of each design parameter of the determined general operation power platform of hilly and mountainous area machinery, randomly select a set of design parameter values and input them into the established finite element model, obtain each performance parameter corresponding to this set of design parameters through simulation tests on the finite element model, and record no less than m groups of test data as the sample set.

[0045] The steps of establishing the finite element model include: selecting appropriate finite element analysis (FEA) software (such as ANSYS, Abaqus, SolidWorks, etc.), using modeling software (such as AutoCAD, SolidWorks, etc.) to design a 3D model of the mechanical structure; generating different models based on various design parameters, and characterizing the corresponding performance parameters through simulation tests;

[0046] The specific method for obtaining the value range of each design parameter based on existing public information is: through the existing publicly available design parameter values ​​in related fields, the obtained design parameter values ​​are appropriately enlarged or reduced to form a design parameter value range, wherein the enlargement and reduction ratios are based on what is actually achievable.

[0047] Step 2: Based on the performance parameters and design parameters determined by the finite element model, a regression model between the power platform performance parameters and the design parameters is established. At the same time, the regression coefficient is determined by the Bayesian regression algorithm, and a complete linear regression model is generated based on the determined regression coefficient;

[0048] Based on the performance parameters and design parameters determined by the finite element model, the specific steps of establishing the regression model between the performance parameters of the power platform and the design parameters include: calibrating the performance parameters of the power platform, calibrating the maximum passing height as C 1 , the suspension travel is calibrated as C 2 , ground pressure, turning radius, load capacity and energy consumption are calibrated as C 3 , C 4 , C 5 and C 6 ;

[0049] The design parameters of the power platform are calibrated, and the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height and total platform weight are calibrated as X 1 , X 2 , X 3 , X 4 , X 5 , X 6 and X 7 ;

[0050] A regression model expression between the power platform performance parameters and various design parameters is established, where the specific regression model expression is:

[0051] C i =β i，0 +β i，1 *X 1 +β i，2 *X 2 +…+β i，7 *X7 +∈ i

[0052] Wherein, C i represents the i-th performance parameter, and β i，1 to β i，7 represent the regression coefficients of each design parameter in the regression model expression of the i-th performance parameter, and β i，0 represents the intercept term in the regression model expression of the i-th performance parameter, and ∈ i represents the error term in the regression model expression of the i-th performance parameter, where i is the index of the performance parameter, and i = 1, 2,..., 6;

[0053] Determine the regression coefficients through the Bayesian regression algorithm, and generate a complete linear regression model based on the determined regression coefficients.

[0054] The specific steps of determining the regression coefficients through the Bayesian regression algorithm include:

[0055] First, define a prior distribution for each regression coefficient. The prior distribution can be set based on domain knowledge, expert opinions, or historical data. The commonly used distribution is the normal distribution, which is suitable for most cases of regression coefficients; establish a likelihood function according to the collected performance parameter and design parameter data; according to Bayes' theorem, calculate the posterior distribution through the prior distribution and the likelihood function, and use numerical methods such as Markov Chain Monte Carlo (MCMC) to approximate the posterior distribution. The MCMC method can generate a series of samples, and then estimate the posterior distribution of the regression coefficients; extract the estimated values of the regression coefficients from the posterior distribution, which can be obtained by calculating the posterior mean or median, and usually the posterior mean is selected: substitute the extracted numerical values of the regression coefficients into the regression model expression to obtain a complete linear regression model.

[0056] Step 3: Construct a gene based on each design parameter of the linear regression model. The combination containing all different types of genes constitutes an individual chromosome. Construct a fitness function based on each performance parameter of the linear regression model, and determine an individual chromosome with the highest fitness through the genetic algorithm, denoted as the target chromosome;

[0057] Constructing a gene based on each design parameter of the linear regression model is specifically expressed as: calibrating T to represent all optional sets of all design parameters, where T = {P 1 , P 1 , …, P u , …, P n}, where P u = {X u1 , X u2 , …, X u7}, P uDenote the set of design parameters for the \(u\)-th group, and \(n\) represents the number of parameter combinations. Each parameter in the optimized parameter combination is encoded into a gene one by one. Genes generated from the same type of data are allelic to each other, that is, \(X\) 11 , \(X\) 21 , …, \(X\) n1 are allelic to each other;

[0058] According to the obtained genes, an individual chromosome is formed. The individual chromosome consists of 7 genes. Each individual chromosome \(S\) has 7 gene loci, and each gene locus corresponds to a gene, that is, \(X\) 11 , \(X\) 21 , …, \(X\) n1 are all optional values for the 1st gene locus of the individual chromosome. The 7 gene loci respectively correspond to a certain gene of the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height, and total platform weight parameters.

[0059] Construct a fitness function based on each performance parameter of the linear regression model. The expression of the fitness function is:

[0060]

[0061] In the formula, \(f(S)\) represents the fitness function, \(\omega\) 1 , \(\omega\) 2 , \(\omega\) 3 , \(\omega\) 4 , \(\omega\) 5 and \(\omega\) 6 are the weight coefficients of the maximum passing height, suspension stroke, ground contact pressure, turning radius, load capacity, and energy consumption respectively. Among them, \(\omega\) 1 , \(\omega\) 2 , \(\omega\) 3 , \(\omega\) 4 , \(\omega\) 5 and \(\omega\) 6 are all greater than 0; Since the influence of the ground contact pressure on the performance of the general operation power platform for hilly and mountainous machinery is greater than other parameters, followed by the suspension stroke and turning radius, and the influence of the load capacity, energy consumption, and maximum passing height on the performance of the general operation power platform for hilly and mountainous machinery is weaker compared to other parameters, so \(\omega\) 3 >\(\omega\) 2 ≥\(\omega\) 4 >\(\omega\) 5 ≥\(\omega\) 6 ≥\(\omega\) 1 .

[0062] Determine an individual chromosome with the highest fitness through a genetic algorithm. The specific steps of the genetic algorithm include: constructing an initial population based on the individual chromosomes; performing selection, crossover, and mutation operations on the individual chromosomes within the initial population; performing iterative operations, and setting the mutation probability and the iteration termination condition at the same time. After the iterative operation is terminated, the individual chromosome with the highest fitness in the current population is recorded as the target chromosome.

[0063] Among them, the selection operation uses the roulette method to select individuals. The system randomly generates a random number within the interval [0, 1]. According to the generated probability of being selected, which selection interval of which individual chromosome this random number is distributed in to determine which individual chromosome to select in this round. The process of selecting individuals is based on the fitness of the individuals. The greater the fitness of an individual, the greater the possibility of being selected, and vice versa, the smaller the possibility of being selected. And an individual may be selected multiple times.

[0064] At the same time, set the mutation probability to 0.5. The iteration termination condition can be set by setting the maximum number of iterations. When the number of iterations reaches the set maximum number of iterations, stop the iterative operation and select the individual chromosome with the highest fitness in the current population as the target chromosome; it can also be set by setting a fitness threshold. When an individual chromosome with a fitness greater than the fitness threshold appears during the iteration process, stop the iterative operation and select the individual chromosome with a fitness greater than the fitness threshold as the target chromosome. If there are multiple individual chromosomes with a fitness greater than the fitness threshold in the current population, then select the individual chromosome with the highest fitness among them as the target chromosome.

[0065] As a publicly available and mature optimal solution seeking algorithm, the genetic algorithm will not elaborate on other detailed steps here.

[0066] Step 4: Obtain the environmental parameters of the hilly and mountainous areas where the power platform works, and correct the gene parameters in the target chromosome based on the obtained environmental parameters to obtain the optimal design parameters of the general operation power platform for hilly and mountainous machinery. The environmental parameters of the hilly and mountainous areas include the average slope and the average soil humidity.

[0067] The logic for correcting the gene parameters in the target chromosome based on the obtained environmental parameters is: correct the gene parameters of the engine power, torque, and suspension stiffness in the target chromosome. The specific formulas for the correction are as follows:

[0068] X″ 1 =X′ 1 +(1+X′ 7 *g*sinθ)

[0069] X″ 2 =X′ 2 +(1+X′ 7*g*r*sinθ)

[0070] X″ 3 = X′ 3 *(1 + α*sinθ + δ*H)

[0071] wherein, X″ 1 , X″ 2 and X″ 3 respectively represent the corrected engine power, torque and suspension stiffness. X′ 1 , X′ 2 and X′ 3 respectively represent the values of the engine power, torque and suspension stiffness in the target chromosome. X 7 ′ represents the total platform weight data in the target chromosome. g is the acceleration due to gravity, θ represents the average slope, r is the radius of the power platform wheel, α and δ are the slope influence coefficient and the soil moisture influence coefficient respectively. Since the influence of the slope on the performance requirements of the power platform is greater than that of the soil moisture, α > δ is set, and both α and δ are greater than 0. H is the average soil moisture;

[0072] At a steeper slope, the power and torque requirements of the power system will increase to overcome the influence of gravity. The increase in slope may cause a change in the force state of the suspension, affecting the stiffness of the suspension. Soils with different humidities may have different effects on the response of the suspension. For example, a larger suspension stroke may be required on a slippery road surface to improve stability.

[0073] Replace the corrected engine power, torque and suspension stiffness parameters with the engine power, torque and suspension stiffness data in the target chromosome, and finally obtain the optimal design parameters of the general operation power platform for hilly and mountainous machinery, that is, the data included in the replaced target chromosome.

[0074] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0075] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0076] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, and it may be located in one place or distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] The above is only a specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application.

Claims

1. A multi-objective optimization method for a universal working power platform for hilly and mountainous machinery based on a regression model, characterized in that: The specific steps include: Determine the performance parameters and design parameters of the universal operating power platform for hilly and mountainous machinery, establish a finite element model of the universal operating power platform for hilly and mountainous machinery based on the design parameters, and characterize the performance parameters of the universal operating power platform for hilly and mountainous machinery through finite element models with different design parameters; Based on the performance parameters and design parameters determined by the finite element model, a regression model between the power platform performance parameters and the design parameters is established. At the same time, the regression coefficient is determined by the Bayesian regression algorithm, and a complete linear regression model is generated based on the determined regression coefficient. Based on the performance parameters and design parameters determined by the finite element model, the specific steps of establishing the regression model between the performance parameters of the power platform and the design parameters include: calibrating the performance parameters of the power platform, calibrating the maximum passing height to , the suspension travel is calibrated as , ground pressure, turning radius, load capacity and energy consumption are calibrated as , , and ; The design parameters of the power platform are calibrated, and the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height and total platform weight are calibrated in turn. , , , , , and ; A regression model expression between the power platform performance parameters and various design parameters is established, where the specific regression model expression is: In the formula, Indicates performance parameters, to Indicates The regression coefficients of the design parameters in the performance parameter regression model expression are: Indicates The intercept term in the performance parameter regression model expression, Indicates The error term in the performance parameter regression model expression is is the index of the performance parameter, ; Each design parameter based on the linear regression model constitutes a gene, and the combination of all different types of genes constitutes an individual chromosome. A fitness function is constructed based on each performance parameter of the linear regression model. The individual chromosome with the highest fitness is determined by the genetic algorithm and recorded as the target chromosome. The environmental parameters of the hilly and mountainous areas where the power platform works are obtained, and the gene parameters in the target chromosome are modified based on the obtained environmental parameters to obtain the optimal design parameters of the universal power platform for hilly and mountainous areas machinery, wherein the environmental parameters of the hilly and mountainous areas include the average slope and the average soil moisture.

2. The multi-objective optimization method for a universal working power platform of hilly and mountainous machinery based on a regression model according to claim 1 is characterized by: Determine the performance parameters and design parameters of the universal operating power platform for hilly and mountainous machinery, wherein the performance parameters of the universal operating power platform for hilly and mountainous machinery include: maximum passing height, suspension travel, ground contact pressure, turning radius, load capacity and energy consumption; the design parameters of the universal operating power platform for hilly and mountainous machinery include: engine power, torque, suspension stiffness, platform width, platform height, center of gravity height and total platform weight; obtain the value range of each design parameter based on existing public information, and based on the determined value range of each design parameter of the universal operating power platform for hilly and mountainous machinery, randomly select a group of design parameter values ​​and input them into the established finite element model, obtain the performance parameters corresponding to the group of design parameters by performing simulation tests on the finite element model, and record no less than A group of test data is used as the sample set.

3. The multi-objective optimization method of a universal working power platform for hilly and mountainous machinery based on a regression model according to claim 1 is characterized by: The specific steps of determining the regression coefficient by the Bayesian regression algorithm include: First, a prior distribution is defined for each regression coefficient; a likelihood function is established based on the collected performance parameter and design parameter data; according to Bayes' theorem, the posterior distribution is calculated through the prior distribution and the likelihood function; the estimated value of the regression coefficient is extracted from the posterior distribution, and the extracted regression coefficient value is substituted into the regression model expression to obtain a complete linear regression model.

4. The multi-objective optimization method of a universal working power platform for hilly and mountainous machinery based on a regression model according to claim 3 is characterized by: Each design parameter based on the linear regression model constitutes a gene, which is specifically expressed as: represents all optional sets of all design parameters, where ,in , represents the design parameter set of the uth group, n represents the number of parameter combinations, and the parameters in the optimized parameter combination are encoded one by one as a gene. The genes generated by the same type of data are alleles of each other, that is, are alleles of each other; According to the obtained genes, individual chromosomes are formed, and the individual chromosomes are composed of 7 genes. There are 7 gene positions, each of which corresponds to a gene, namely They are all optional values ​​of the gene position No. 1 of the individual chromosome. The seven gene positions correspond to a gene of the engine power, torque, suspension stiffness, platform width, platform height, center of gravity height and platform total weight parameters respectively.

5. The multi-objective optimization method of a universal working power platform for hilly and mountainous machinery based on a regression model according to claim 4 is characterized by: A fitness function is constructed based on each performance parameter of the linear regression model, where the expression of the fitness function is: In the formula, represents the fitness function, , , , , and , are the weight coefficients of maximum clearance height, suspension travel, ground contact pressure, turning radius, load capacity and energy consumption, respectively. , , , , and All are greater than 0; The genetic algorithm is used to determine an individual chromosome with the highest fitness, wherein the specific steps of the genetic algorithm include: constructing an initial population based on individual chromosomes; performing selection, crossover and mutation operations on individual chromosomes in the initial population; performing iterative operations, while setting mutation probabilities and iteration termination conditions. When the iterative operation is terminated, the individual chromosome with the highest fitness in the current population is recorded as the target chromosome.

6. The multi-objective optimization method of a universal working power platform for hilly and mountainous machinery based on a regression model according to claim 1 is characterized by: The logic for modifying the genetic parameters in the target chromosome based on the acquired environmental parameters is: modifying the genetic parameters of engine power, torque and suspension stiffness in the target chromosome, and the specific formulas for the modification are: In the formula, , and represent the corrected engine power, torque and suspension stiffness respectively, , and Respectively represent the values ​​of engine power, torque and suspension stiffness within the target chromosome, Indicates the total weight data of the platform within the target chromosome, is the acceleration due to gravity, represents the average slope, is the wheel radius of the power platform, and are the slope influence coefficient and soil moisture influence coefficient respectively, where ,and and , is the average soil moisture; The corrected engine power, torque and suspension stiffness parameters are replaced with the engine power, torque and suspension stiffness data in the target chromosome, and finally the optimal design parameters of the universal operating power platform for hilly and mountainous machinery are obtained, that is, the data contained in the replaced target chromosome.

Images