A sealing ring grinding process parameter optimization method based on improved genetic algorithm

Abstract

The application relates to a sealing ring grinding process parameter optimization scheme based on an improved genetic algorithm. Firstly, grinding speed, workpiece feeding speed and grinding depth are selected as optimization parameters, grinding specific energy, grinding temperature and surface roughness are selected as optimization targets, and a function model of the optimization parameters and the optimization targets is established. Secondly, according to the function model of the optimization targets and the optimization parameters and the constraint conditions of the optimization targets and the optimization parameters, namely, the optimization target function model and the constraint conditions, an optimization mathematical model of the sealing ring grinding process parameters is established. Finally, multi-objective optimization based on the improved genetic algorithm is adopted, the multi-objective, multi-variable and multi-constraint problem is simplified into a single-objective optimization problem by using the genetic algorithm and a linear weighting method, the optimization model is solved, and after the range solution is obtained, the range solution is used to obtain the accurate optimal solution by using a local search algorithm. Therefore, the application has important significance for the grinding process optimization of the sealing ring.

Description

Technical Field

[0001] This invention relates to a surface grinding process, and more particularly to a method for optimizing the grinding process parameters of a sealing ring. Background Technology

[0002] Construction machinery often operates in harsh environments with complex stresses, which can cause wear and damage to mechanical shaft components. Sealing rings, with their advantages of strong anti-contamination capabilities, wear resistance, impact resistance, reliable sealing performance, automatic end-face wear compensation, simple structure, and long service life, are widely used in the seals of heavy-duty, low-pressure construction machinery such as excavators. Currently, most sealing rings are made from difficult-to-machine materials, significantly reducing processing efficiency and making it difficult to guarantee machining accuracy. Therefore, research on the optimization of sealing ring grinding process parameters is very helpful in guiding the sealing ring grinding process. Only by solving the relationship between grinding process parameters, grinding parameters, and grinding result parameters during sealing ring grinding, and establishing a mathematical model of the sealing ring grinding process, can we better understand the sealing ring grinding process and find the optimal parameters for sealing ring grinding. Summary of the Invention

[0003] To achieve high-quality and efficient grinding of sealing rings, this invention proposes a method for optimizing the grinding process parameters of sealing rings based on an improved genetic algorithm. First, this invention selects grinding speed, workpiece feed speed, and grinding depth as optimization parameters, and grinding specific energy, grinding temperature, and surface roughness as optimization objectives, establishing a functional model between the grinding process parameters and the optimization objectives. Second, based on the functional model between the optimization objectives and the grinding process parameters, as well as the constraints between the optimization objectives and the process parameters (i.e., the optimization objective function model and constraints), this invention establishes an optimization mathematical model for the grinding process parameters of the sealing rings. Finally, this invention employs multi-objective optimization based on an improved genetic algorithm. First, the multi-objective, multi-variable, and multi-constraint problem is simplified into a single-objective optimization problem using a genetic algorithm and a linear weighted method. The multi-objective optimization model for the grinding process parameters of the sealing rings is then solved. After obtaining a range solution using the genetic algorithm, the exact optimal solution is obtained using a local search algorithm.

[0004] Furthermore, the objective function model and constraints specifically include:

[0005] Sub-step 1-1: Calculation formulas for grinding force and grinding specific energy:

[0006] With grinding speed V s Workpiece feed speed V g and grinding depth a pAs an optimized design parameter for the grinding process, the grinding force is established in two ways: empirical and theoretical. This invention establishes it theoretically, using the dynamic number of grinding edges and the grinding cross-sectional area per unit area. The dynamic number of grinding edges is:

[0007]

[0008] In the formula, A n A is a coefficient that is proportional to the number of grinding blades. n ≈1.2, C e d is a coefficient representing the density and shape of abrasive grains on the contact surface between the grinding wheel and the workpiece. e V is the equivalent diameter of the grinding wheel. s V is the grinding speed. g Let a be the workpiece feed rate. p This refers to the grinding depth. α and β are determined by the cutting edge distribution.

[0009] The grinding cross-sectional area within the contact length l between the grinding wheel and the workpiece is:

[0010]

[0011] In the formula, l s For the grinding wheel and the sealing ring to have a long contact arc and l is any contact length on the sealing ring, A n A is a coefficient that is proportional to the number of grinding blades. n ≈1.2, C e d is a coefficient representing the density and shape of abrasive grains on the contact surface between the grinding wheel and the workpiece. e V is the equivalent diameter of the grinding wheel. s V is the grinding speed. g Let a be the workpiece feed rate. p This refers to the grinding depth. α and β are determined by the cutting edge distribution.

[0012] The theoretical formula for calculating its friction force is:

[0013]

[0014] In the formula, F P The unit grinding force is γ for the grinding surface of the sealing ring, and the values ​​of γ and ε are in the range of 0≤γ≤1; 0.5≤ε≤1.

[0015] Sub-steps 1-2 involve the complex grinding process of the sealing ring, characterized by high grinding resistance and the tendency for localized high temperatures to form on the surface. Therefore, minimizing the energy consumed per unit volume of material removed (i.e., the grinding specific energy) is a suitable optimization objective for this grinding process. The formula for calculating the grinding specific energy is:

[0016]

[0017] In the formula, F first t The grinding force is represented by b, which indicates the width of the grinding wheel. V s V is the grinding speed. g For the working feed rate, a p This refers to the grinding depth.

[0018] Formula for calculating grinding temperature

[0019] Grinding temperature is also a crucial factor affecting grinding accuracy during the grinding process, and the appropriate selection of grinding parameters plays a significant role in influencing grinding temperature. Assuming that heat is transferred to the workpiece in a certain proportion during the grinding of the sealing ring, the relationship between the contact area temperature and the grinding force is as follows:

[0020]

[0021] In the formula, F t The grinding force is represented by b, which indicates the width of the grinding wheel. V s For grinding speed, l s For the grinding wheel and the sealing ring to have a long contact arc and The formula for calculating the surface roughness of the workpiece in sub-steps 1-3 is:

[0022]

[0023] In the above formula, The three-dimensional average spacing of the abrasive grains on the grinding wheel is represented by θ, where θ is the semi-apex angle of the abrasive grains, and r is the average spacing of the abrasive grains on the grinding wheel. s V represents the radius of the grinding wheel. s V is the grinding speed. g This represents the workpiece feed rate.

[0024] Step 4: Constraints and Multi-Objective Optimization Model

[0025] In optimizing the grinding process parameters of the sealing ring, constraints need to be added to the optimized process parameters and the objective function model. To achieve the best grinding accuracy, it is necessary to select a small grinding specific energy, a small grinding temperature, and a small surface roughness. Therefore, during the grinding process, the grinding force must not exceed the machine tool's permissible grinding force F. t * The surface roughness must not exceed the permissible surface roughness R required by the workpiece. a * Its expression is as follows:

[0026] F t <F t * ,R a <R a * (7)

[0027] For grinding speed V s Workpiece feed speed V g and grinding depth a p If the grinding process must remain within the permissible range of its parameters, then the constraints are as follows:

[0028]

[0029] To prevent grinding wheel clogging, the chip volume Q of a single abrasive grain is also required. g Less than its chip volume Q c ,Right now:

[0030]

[0031] In the formula, V s V is the grinding speed. g Let a be the workpiece feed rate. p For the grinding depth, d g Where is the diameter of the abrasive grain, V is the volume ratio of the abrasive grain in the grinding wheel, and h is the abrasive grain diameter. i To increase the height of the bonding agent protruding from the abrasive grains of the grinding wheel, N d The number of dynamic grinding edges per unit contact surface.

[0032] In summary, to conduct optimization studies within the range of surface grinding process parameters, the following optimization mathematical model can be established:

[0033]

[0034] Preferably, the optimization problem of the sealing ring grinding parameters can be viewed as a nonlinear optimization problem with multiple objectives, multiple variables, and multiple constraints. This optimization problem has three objective functions: minimum grinding specific energy, minimum grinding temperature, and minimum surface roughness. When solving multi-objective problems, a linear weighting method is often used to transform the multi-objective problem into a single-objective problem, that is, multiplying each objective function by a weight coefficient. Furthermore, to resolve the potentially large differences between each optimization objective due to the weights, this invention introduces the normalization of the optimization objectives, resulting in the weighted objective function to be minimized, as shown in the equation:

[0035] f(x) = w1e s +w2T+w3R a (11)

[0036] In the formula, e s R is the grinding specific energy, T is the grinding temperature, and R is the grinding temperature. a For surface roughness, w1, w2, and w3 are weight coefficients. The operation of the genetic algorithm for optimizing the grinding process parameters of the sealing ring based on the genetic algorithm is as follows:

[0037] (1) Encoding. In optimizing the grinding process parameters of the sealing ring, it is necessary to encode the parameters during the machining process to facilitate calculation. This invention encodes the grinding speed V. s Workpiece feed speed V g and grinding depth a p As fundamental parameters for optimizing the grinding of sealing rings, the grinding process parameters in the sealing ring grinding process are represented using 10-bit binary codes. Therefore, each parameter should contain three optimized parameters for the grinding process, namely V... s V g and a p .

[0038] (2) Generation of the initial population. In the process of optimizing the grinding parameters of the sealing ring, this invention requires selecting a suitable population size M for the genetic algorithm to facilitate subsequent algorithmic operations. After selecting a suitable population size, it is necessary to perform calculations on the randomly generated individuals, globally search for the optimal individual among all generated individuals, and evaluate them to find the best individual. To prevent optimal convergence within a subpopulation, the optimal individuals are migrated in a certain proportion. In the optimization of the grinding parameters of the sealing ring, the initial population is divided into 6 subpopulations, each containing 20 individuals; therefore, the population size M = 120.

[0039] (3) Determine the fitness function. When using a genetic algorithm to optimize the process parameters of the sealing ring grinding process, it is necessary to define the individual fitness function of the objective function. Individuals that do not meet the constraints during computation have a fitness of 0. For individuals that meet the constraints, the fitness F(x) of the objective function f is calculated using the following formula.

[0040]

[0041] (4) Selection. Selection in the genetic algorithm is performed according to a certain proportion. The purpose is to select a certain proportion of superior genes to pass on to the next generation during gene manipulation. When the population size is M, the fitness of individual i is f. i Then the probability p of that individual being selected is... i for:

[0042]

[0043] (5) Crossover. In genetic algorithms, genes from matching chromosomes need to be exchanged to generate new individuals. Because this invention involves multiple optimization parameters, a multi-point crossover method is used to increase the crossover information for V. s V g and a p The three substrings are crossed. To improve the convergence speed, a crossover probability of 0.9 is used for the calculation.

[0044] (6) Mutation. In the operation of the genetic algorithm, it is necessary to replace certain genes on the individual chromosome with their alleles. Select the basic bit mutation operator, that is, according to the mutation probability of 0.08, randomly specify the gene value of one or several loci in the individual's encoding string and invert it.

[0045] (7) Local Search. While genetic algorithms have strong global optimization capabilities, their local optimization capabilities are relatively weak, resulting in a range-based optimal solution with poor accuracy. The improved genetic algorithm selected in this invention first performs global optimization on the sealing ring grinding process, then decodes the range-based solution obtained by the genetic algorithm, and performs a precise search using a local search algorithm. The local search algorithm randomly selects an initial solution within the range-based solution obtained by the genetic algorithm, continuously searches the neighborhood of the initial solution, compares the optimal solutions within the two neighborhoods, sets this optimal solution as the current solution, and then searches the neighborhood of this current solution, iteratively searching to find the optimal solution within the range-based solution obtained by the genetic algorithm.

[0046] The essential feature of this invention lies in establishing an optimized mathematical model of the grinding process through the relationships and constraints between grinding parameters. Combining genetic algorithms and linear weighted methods, the multi-objective process is simplified into a single objective to obtain the optimal parameter range solution. Finally, a local search algorithm is used to determine the precise optimal parameters for grinding the sealing ring. This invention is of great significance for guiding the grinding process of sealing rings, enabling more accurate parameter selection during sealing ring processing, with the aim of obtaining sealing rings with lower surface roughness and higher precision. Attached Figure Description

[0047] Figure 1 This is a diagram illustrating the grinding force.

[0048] Figure 2 This is a flowchart of the genetic algorithm;

[0049] 1. Grinding wheel; 2. Workpiece. Detailed Implementation

[0050] The invention will now be further described with reference to the accompanying drawings.

[0051] This invention discloses a method for optimizing the grinding process parameters of sealing rings based on a genetic algorithm. First, three processing parameters are selected: grinding speed V... s Workpiece feed speed V g and grinding depth a p As the starting variable for optimizing the sealing ring grinding process, this optimization comprehensively considers other processing requirements and related constraints, such as grinding specific energy, grinding temperature, and surface roughness. Specifically, it includes: the objective function model and constraints, and multi-objective optimization based on an improved genetic algorithm.

[0052] Grinding force is caused by elastic deformation, plastic deformation, chip formation, and friction between the abrasive grains and bonding agent and the workpiece surface after the workpiece 2 comes into contact with the grinding wheel 1. When grinding difficult-to-machine materials, grinding resistance and energy consumption are often high. Therefore, minimizing the energy consumed per unit volume of material removed (grinding specific energy) is a suitable optimization objective. During grinding, almost all the work done is converted into heat energy, with only a small amount carried away by the chips. This unremoved heat often forms localized high temperatures on the workpiece surface, causing thermal damage and affecting the dimensional and shape accuracy of the workpiece. Surface roughness is an important indicator of workpiece surface quality. Therefore, when studying the optimal grinding parameters, grinding specific energy, grinding temperature, and surface roughness can be used as objective functions.

[0053] according to Figure 1 A schematic diagram of grinding force can divide the grinding force into three mutually perpendicular components, namely the tangential grinding force F along the tangential direction of the grinding wheel. t The normal grinding force F along the radial direction of the grinding wheel n and the grinding force F along the grinding wheel axis a In surface grinding, the axial force is small and can be ignored.

[0054] The specific method for optimizing the mathematical model of the sealing ring grinding process parameters is as follows:

[0055] Step 1: Formulas for calculating grinding force and grinding specific energy:

[0056] With grinding speed V s Workpiece feed speed V g and grinding depth a p As an optimized design parameter for the grinding process, the grinding force is established in two ways: empirical and theoretical. This invention establishes it theoretically, using the dynamic number of grinding edges and the grinding cross-sectional area per unit area. The dynamic number of grinding edges is:

[0057]

[0058] In the formula, A n A is a coefficient that is proportional to the number of grinding blades. n ≈1.2, C e d is a coefficient representing the density and shape of abrasive grains on the contact surface between the grinding wheel and the workpiece. e V is the equivalent diameter of the grinding wheel. s V is the grinding speed. g Let a be the workpiece feed rate. p This refers to the grinding depth. α and β are determined by the cutting edge distribution.

[0059] The grinding cross-sectional area within the contact length l between the grinding wheel and the workpiece is:

[0060]

[0061] In the formula, l s For the grinding wheel and the sealing ring to have a long contact arc and l is any contact length on the sealing ring, A n A is a coefficient that is proportional to the number of grinding blades. n ≈1.2, C e d is a coefficient representing the density and shape of abrasive grains on the contact surface between the grinding wheel and the workpiece. e V is the equivalent diameter of the grinding wheel. s V is the grinding speed. g Let a be the workpiece feed rate. p This refers to the grinding depth. α and β are determined by the cutting edge distribution.

[0062] The theoretical formula for calculating its friction force is:

[0063]

[0064] In the formula, F P The unit grinding force is γ for the grinding surface of the sealing ring, and the values ​​of γ and ε are in the range of 0≤γ≤1; 0.5≤ε≤1.

[0065] The grinding process of sealing rings is complex, with high grinding resistance and a tendency for localized high temperatures to form on the surface. Therefore, minimizing the energy consumed per unit volume of material removed (i.e., the grinding specific energy) is a suitable optimization objective. The formula for calculating the grinding specific energy is:

[0066]

[0067] In the formula, F first t The grinding force is represented by b, which indicates the width of the grinding wheel. V s V is the grinding speed. g For the working feed rate, a p This refers to the grinding depth.

[0068] Step 2: Calculation formula for grinding temperature

[0069] Grinding temperature is also a crucial factor affecting grinding accuracy during the grinding process, and the appropriate selection of grinding parameters plays a significant role in influencing grinding temperature. Assuming that heat is transferred to the workpiece in a certain proportion during the grinding of the sealing ring, the relationship between the contact area temperature and the grinding force is as follows:

[0070]

[0071] In the formula, F t The grinding force is represented by b, which indicates the width of the grinding wheel. Vs For grinding speed, l s For the grinding wheel and the sealing ring to have a long contact arc and Step 3: The formula for calculating the surface roughness of the workpiece is:

[0072]

[0073] In the above formula, The three-dimensional average spacing of the abrasive grains on the grinding wheel is represented by θ, where θ is the semi-apex angle of the abrasive grains, and r is the average spacing of the abrasive grains on the grinding wheel. s V represents the radius of the grinding wheel. s V is the grinding speed. g This represents the workpiece feed rate.

[0074] Step 4: Constraints and Multi-Objective Optimization Model

[0075] In optimizing the grinding process parameters of the sealing ring, constraints need to be added to the optimized process parameters and the objective function model. To achieve the best grinding accuracy, it is necessary to select a small grinding specific energy, a small grinding temperature, and a small surface roughness. Therefore, during the grinding process, the grinding force must not exceed the machine tool's permissible grinding force F. t * The surface roughness must not exceed the permissible surface roughness R required by the workpiece. a * Its expression is as follows:

[0076] F t <F t * ,R a <R a * (7)

[0077] For grinding speed V s Workpiece feed speed V g and grinding depth a p If the grinding process must remain within the permissible range of its parameters, then the constraints are as follows:

[0078]

[0079] To prevent grinding wheel clogging, the chip volume Q of a single abrasive grain is also required. g Less than its chip volume Q c ,Right now:

[0080]

[0081] In the formula, V s V is the grinding speed. g Let a be the workpiece feed rate. p For the grinding depth, d gWhere is the diameter of the abrasive grain, V is the volume ratio of the abrasive grain in the grinding wheel, and h is the abrasive grain diameter. i To increase the height of the bonding agent protruding from the abrasive grains of the grinding wheel, N d The number of dynamic grinding edges per unit contact surface.

[0082] In summary, to conduct optimization studies within the range of surface grinding process parameters, the following optimization mathematical model can be established:

[0083]

[0084] The optimization problem of sealing ring grinding parameters can be viewed as a nonlinear optimization problem with multiple objectives, multiple variables, and multiple constraints. This optimization problem has three objective functions: minimum grinding specific energy, minimum grinding temperature, and minimum surface roughness. When solving multi-objective problems, a linear weighting method is often used to transform the multi-objective problem into a single-objective problem, that is, multiplying each objective function by a weight coefficient. Furthermore, to resolve the potentially large differences between each optimization objective due to the weights, this invention introduces normalization of the optimization objectives, resulting in the weighted objective function to be minimized, as shown in the equation:

[0085] f(x) = w1e s +w2T+w3R a (11)

[0086] In the formula, e s R is the grinding specific energy, T is the grinding temperature, and R is the grinding temperature. a For surface roughness, w1, w2, w3 are weighting coefficients, such as... Figure 2 The operation of the genetic algorithm for optimizing the grinding process parameters of the sealing ring based on the genetic algorithm is as follows: (1) Encoding. In order to facilitate its calculation, the parameters in the grinding process of the sealing ring need to be encoded. In this invention, the grinding speed V s Workpiece feed speed V g and grinding depth a p As fundamental parameters for optimizing the grinding of sealing rings, the grinding process parameters in the sealing ring grinding process are represented using 10-bit binary codes. Therefore, each parameter should contain three optimized parameters for the grinding process, namely V... s V g and a p .

[0087] (2) Generation of the initial population. In the process of optimizing the grinding parameters of the sealing ring, this invention requires selecting a suitable population size M for the genetic algorithm to facilitate subsequent algorithmic operations. After selecting a suitable population size, it is necessary to perform calculations on the randomly generated individuals, globally search for the optimal individual among all generated individuals, and evaluate them to find the best individual. To prevent optimal convergence within a subpopulation, the optimal individuals are migrated in a certain proportion. In the optimization of the grinding parameters of the sealing ring, the initial population is divided into 6 subpopulations, each containing 20 individuals; therefore, the population size M = 120.

[0088] (3) Determine the fitness function. When using a genetic algorithm to optimize the process parameters of the sealing ring grinding process, it is necessary to define the individual fitness function of the objective function. Individuals that do not meet the constraints during computation have a fitness of 0. For individuals that meet the constraints, the fitness F(x) of the objective function f is calculated using the following formula.

[0089]

[0090] (4) Selection. Selection in the genetic algorithm is performed according to a certain proportion. The purpose is to select a certain proportion of superior genes to pass on to the next generation during gene manipulation. When the population size is M, the fitness of individual i is f. i Then the probability p of that individual being selected is... i for:

[0091]

[0092] (5) Crossover. In genetic algorithms, genes from matching chromosomes need to be exchanged to generate new individuals. Because this invention involves multiple optimization parameters, a multi-point crossover method is used to increase the crossover information for V. s V g and a p The three substrings are crossed. To improve the convergence speed, a crossover probability of 0.9 is used for the calculation.

[0093] (6) Mutation. In the operation of the genetic algorithm, it is necessary to replace certain genes on the individual chromosome with their alleles. Select the basic bit mutation operator, that is, according to the mutation probability of 0.08, randomly specify the gene value of one or several loci in the individual's encoding string and invert it.

[0094] (7) Local Search. While genetic algorithms have strong global optimization capabilities, their local optimization capabilities are relatively weak. The optimal solution they calculate is a range solution, resulting in poor accuracy. The improved genetic algorithm selected in this invention first uses a genetic algorithm to perform global optimization on the sealing ring grinding process. Then, it decodes the range solution obtained by the genetic algorithm and performs a precise search using a local search algorithm. The local search algorithm randomly selects an initial solution within the range solution obtained by the genetic algorithm, continuously searches the neighborhood of the initial solution, compares the optimal solutions within the two neighborhoods, sets this optimal solution as the current solution, and then searches the neighborhood of this current solution, iteratively searching to find the optimal solution within the range solution obtained by the genetic algorithm.

Claims

1. A method for optimizing the grinding process parameters of sealing rings based on an improved genetic algorithm, characterized in that, Includes the following steps: Step 1: Establish an optimization objective function model with grinding speed, workpiece feed speed, and grinding depth as the grinding process parameters for the sealing ring, and grinding specific energy, grinding temperature, and surface roughness as the optimization objectives. Step 2: Based on the functional model between the optimization objective and the grinding process parameters of the sealing ring, as well as the constraints between the optimization objective and the process parameters, i.e. the optimization objective function model and constraints, a multi-objective optimization model for the grinding process parameters of the sealing ring is established. Step 3: Using a multi-objective optimization based on an improved genetic algorithm, the multi-objective, multi-variable, and multi-constraint problem is first simplified into a single-objective optimization problem using a genetic algorithm and a linear weighted method. The multi-objective optimization model for the sealing ring grinding process parameters is then solved. After obtaining the range solution using the genetic algorithm, the local search algorithm is then used to find the exact optimal solution of the range solution. Step two includes the following sub-steps: Sub-step 1-1, formulas for calculating grinding force and grinding specific energy: With grinding speed V s Workpiece feed speed V g and grinding depth a p The grinding process parameters for the sealing ring are established using the dynamic number of grinding edges and the grinding cross-sectional area per unit area; the dynamic number of grinding edges is: （1） In the formula, A coefficient that is proportional to the number of grinding blades. C e d is a coefficient representing the density and shape of abrasive grains on the contact surface between the grinding wheel and the workpiece. e V is the equivalent diameter of the grinding wheel; s V is the grinding speed. g Let a be the workpiece feed rate. p Grinding depth; refers to and Determined by the distribution of the cutting edge; Contact length between grinding wheel and workpiece The grinding cross-sectional area inside is: （2） In the formula, For the grinding wheel and the sealing ring to have a long contact arc and , For any contact length on the sealing ring, The theoretical formula for calculating its friction force is: （3） In the formula, The unit grinding force on the grinding surface of the sealing ring. and The range of values ​​is ; ; Sub-steps 1-2 aim to minimize the energy consumed in removing a unit volume of material, i.e., the grinding specific energy. The formula for calculating grinding specific energy is: （4） In the formula, b represents the width of the grinding wheel; Formula for calculating grinding temperature Assuming that heat is transferred to the workpiece at a fixed ratio during the grinding process of the sealing ring, the relationship between the contact area temperature and the grinding force is as follows: （5） Sub-steps 1-3, the formula for calculating the surface roughness of the workpiece is: （6） In the above formula, This represents the three-dimensional average spacing of the abrasive grains on the grinding wheel. For the abrasive grain's half-apex angle, Where is the radius of the grinding wheel; Sub-steps 1-4: Constraints and Multi-objective Optimization Model Selecting a small grinding energy, a small grinding temperature, and a small surface roughness requires that the grinding force during the grinding process does not exceed the machine tool's permissible grinding force. The surface roughness must not exceed the permissible surface roughness required by the workpiece. Its expression is as follows: （7） For grinding speed V s Workpiece feed speed V g and grinding depth a p If the grinding process must remain within the permissible range of its parameters, then the constraints are as follows: （8） To prevent grinding wheel clogging, the chip volume of a single abrasive grain is also required. Smaller than its chip volume ,Right now: (9) In the formula, Where V is the diameter of the abrasive grain, and V is the volume ratio of the abrasive grain in the grinding wheel. The height of the bonding agent protruding from the abrasive grains of the grinding wheel; Within the permissible range of the grinding process parameters for the sealing ring, the following multi-objective optimization model is established: （10）。 2. The method for optimizing the grinding process parameters of sealing rings based on an improved genetic algorithm according to claim 1, characterized in that, Step three, the linear weighted method, includes: normalizing the optimization objective to obtain the weighted objective function to be minimized, as shown in the equation: （11） In the formula, For grinding specific energy, For grinding temperature, For surface roughness, Weighting coefficients Genetic algorithms include the following sub-steps: Sub-step 3-1: Encode the parameters during the machining process; grinding speed V s Workpiece feed speed V g and grinding depth a p As the grinding process parameters for the sealing ring; the grinding process parameters in the sealing ring grinding process are each represented by a 10-bit binary code, then each individual contains three sealing ring grinding process parameters, V s V g and a p ; Sub-step 3-2: Generation of the initial population; Selecting the population size M for the genetic algorithm and finding the optimal individual; To prevent optimal convergence within a subpopulation, the optimal individuals are migrated in a certain proportion; The initial population is divided into 6 subpopulations, each containing 20 individuals, therefore, the population size M = 120; Sub-step 3-3: Determine the fitness function; calculate the fitness F(x) of the objective function f using the following formula. （12） Sub-steps 3-4: Selection; when the population size is M, the fitness of individual i is f. i Then the probability p of that individual being selected is... i for: （13） Sub-steps 3-5 involve cross-referencing; to increase the amount of cross-referencing information, a multi-point cross-referencing method is used for V. s V g and a p The three substrings are crossed; to improve the convergence speed, a crossover probability of 0.9 is used for the calculation. Sub-steps 3-6, mutation; In the operation of the genetic algorithm, it is necessary to replace some genes of the individual chromosome with their alleles; Select the basic position mutation operator, that is, according to the mutation probability of 0.08, randomly specify the gene value of one or several loci in the individual encoding string and invert it; Sub-steps 3-7, Local Search: The improved genetic algorithm first uses a genetic algorithm to globally optimize the grinding process parameters of the sealing ring. The range solution obtained by the genetic algorithm is then decoded, and a local search algorithm is used again for precise searching. The local search algorithm randomly selects an initial solution within the range solution obtained by the genetic algorithm, and then continuously searches the neighborhood of the initial solution. The optimal solution in the two neighborhoods is compared and set as the current solution. Then the neighborhood of this current solution is searched again, and the optimal solution of the range solution obtained by the genetic algorithm is searched in a loop.

Images