Heavy-load vehicle configuration multi-objective optimization design method and system

Abstract

The invention provides a multi-objective optimization design method and system for a heavy-load vehicle configuration, and belongs to the technical field of vehicle manufacturing. The optimization design method provided by the invention comprises the following steps: defining design parameters of a heavy-load vehicle, and constructing a heavy-load vehicle configuration multi-objective optimization model by taking system efficiency and performance requirements of the heavy-load vehicle as optimization objectives based on a parametric modeling method; a staged fuzzy evolution mechanism is introduced on the basis of a traditional NSGA-II algorithm, an iteration process is divided into a plurality of stages, different fuzzy constraint rules and sliding window smoothing processing strategies are designed for each stage, weights of different optimization targets are dynamically adjusted in combination with the convergence speed of a target function, and the optimal optimization target is obtained. And solving the heavy-load vehicle configuration multi-objective optimization model, and outputting the optimal value of the heavy-load vehicle design parameter. According to the invention, an NSGA-II optimization algorithm of staged fuzzy evolution is provided, the robustness and global stability of an optimization solution set are effectively improved, and reliable operation of a vehicle in a changeable mine environment is ensured.

Description

Technical Field

[0001] This invention belongs to the field of vehicle manufacturing technology, specifically relating to a multi-objective optimization design method and system for heavy-duty vehicle configurations. Background Technology

[0002] Currently, most intelligent vehicles are retrofitted based on traditional drive and steering configurations. The vehicle's maneuverability and handling are limited by these traditional configurations, making it difficult to adapt to complex, unstructured scenarios characterized by narrow congestion and numerous obstacles. Distributed drive and steering configurations, however, due to their high flexibility and maneuverability, and their ability to facilitate integrated, lightweight, and modular design, will become the ideal platform for future intelligent vehicles.

[0003] With continuous technological advancements, the intelligentization and automation of heavy-duty mining vehicles have become a future development trend. In mining scenarios, heavy-duty vehicles need to meet the basic requirements of high load capacity, high stability, and durability, while also considering safety, economy, and environmental friendliness in complex environments. However, vehicles typically face complex terrain, heavy transport loads, and harsh weather conditions. The transportation environment in mines is highly uncertain, requiring vehicle designs to not only possess sufficient load-bearing capacity but also maintain efficient and stable performance on various surfaces such as rugged roads, steep slopes, mud, or sand, to meet the diverse needs of mining operations and the pursuit of efficient transportation. Therefore, it is necessary to establish a parametric modeling and optimization system for actual working conditions, coordinate the balance between multiple performance objectives, and select different heavy-duty vehicle parameter design schemes according to operational requirements. This will ensure vehicle performance and safety while minimizing operating costs and maximizing economic benefits.

[0004] In conclusion, how to design heavy-duty vehicle configurations that can meet the requirements of high load, high stability, durability and efficient transportation under complex mining conditions, while also taking into account safety, economy and environmental friendliness, is an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a multi-objective optimization design method and system for heavy-duty vehicle configurations, and proposes a staged fuzzy evolution NSGA-II optimization algorithm, which effectively improves the robustness and global stability of the optimization solution set, ensuring the reliable operation of the vehicle in a variable mining environment, thereby solving at least one of the technical problems involved in the background art.

[0006] To solve the above-mentioned technical problems, the present invention is implemented as follows: A multi-objective optimization design method for heavy-duty vehicle configurations includes the following steps: Step S1: Define the design parameters of the heavy-duty vehicle. Based on the parametric modeling method, construct a multi-objective optimization model for the configuration of the heavy-duty vehicle with the system efficiency and performance requirements of the heavy-duty vehicle as the optimization objectives. Step S2 introduces a phased fuzzy evolution mechanism based on the traditional NSGA-II algorithm, dividing the iteration process into several stages. Different fuzzy constraint rules and sliding window smoothing strategies are designed for each stage. The weights of different optimization objectives are dynamically adjusted in combination with the convergence speed of the objective function to solve the multi-objective optimization model of heavy-duty vehicle configuration and output the optimal values ​​of heavy-duty vehicle design parameters.

[0007] As a preferred improvement, the design parameters of heavy-duty vehicles include the number of drive shafts, tire radius, drive motor power, wheelbase, and suspension stiffness; performance requirements include cost, climbing ability, reliability, and lateral stability.

[0008] As a preferred improvement, in the construction of the multi-objective optimization model for heavy-duty vehicle configuration, system efficiency is taken as the main optimization objective, performance requirements are converted into constraints, and performance thresholds are set for control.

[0009] As a preferred improvement, the system efficiency of heavy-duty vehicles Represented as: In the formula, This indicates the power requirement of heavy-duty vehicles; This represents the input power of a heavy-duty vehicle; where: In the formula, Indicates the vehicle's resistance to movement; Indicates the vehicle's speed; In the formula, Indicates the total weight of the vehicle; Indicates the rolling resistance coefficient; Indicates the air drag coefficient; Indicates the windward area; In the formula, Indicates vehicle number The torque of the root drive shaft; The rotational speed of the drive shaft; This represents the number of drive shafts.

[0010] As a preferred improvement, the climbing ability constraint is expressed as: In the formula, This represents the vehicle's driving force coefficient. The coefficient of friction; The minimum climbing angle; where: In the formula, This represents the power of a single drive shaft; the sum of the power of all drive shafts is the input power. ; Indicates the tire radius; and These are parameters related to the transmission system, expressed as: In the formula, Indicates the gear ratio of the transmission; Indicates the transmission ratio of the main reducer; This indicates the mechanical efficiency of the transmission system.

[0011] As a preferred improvement, during the cost constraint construction process, the heavy-duty vehicle is simplified into a system consisting of a driveshaft, tires, and a drive motor. Each driveshaft is equipped with two tires, and each tire is individually equipped with a drive motor, forming an independent wheel-side drive structure. The cost constraint then becomes... Represented as: In the formula, This indicates the unit price of the tire; This indicates the unit price of the drive motor; Indicates the first The unit price of a drive shaft; This indicates the maximum permissible cost.

[0012] As a preferred improvement, reliability constraints include failure constraints and mean service life constraints, wherein: Fault constraints are represented as: In the formula, Indicates fault reliability; This represents the required failure reliability threshold; where: In the formula, Represents probability; Indicates the time when the component failed; Indicates the normal operating time required for a component to complete its task; This indicates that the component is not faulty; Indicates the probability that a component is fault-free; Service life reliability constraints are expressed as follows: In the formula, Indicates service life reliability; This represents the required service life reliability threshold, where: In the formula, , , These represent the service life and reliability of the drive shaft, tires, and drive motor, respectively. These represent the average service life of the drive shaft, tires, and drive motor, respectively. This indicates the threshold required for the service life of heavy-duty vehicles. Indicates the service life of heavy-duty vehicles. As a preferred improvement, the lateral stability constraint is expressed as: In the formula, Represents the static stability coefficient; and These are the minimum and maximum values ​​allowed for the static stability coefficient, respectively; where: In the formula, It represents the lateral acceleration of a vehicle, and the magnitude of the acceleration when the vehicle moves laterally. Represents gravitational acceleration; Indicates wheel track; Indicates the height of the vehicle's center of gravity; Indicates the maximum travel of the suspension. Indicates suspension stiffness; This is the empty vehicle weight. For load capacity.

[0013] As a preferred improvement, step S2 specifically includes the following steps: Step S21: Based on the traditional NSGA-II algorithm, a staged fuzzy evolution mechanism is introduced to reduce the overall number of iterations. Divided into Stages: Step S22: In each stage, different fuzzy constraint rules and sliding window smoothing strategies are designed according to the search feature requirements to generate stage-specific fuzzy constraint penalties. The fuzzy rules are defined as follows: In the Generation, for constraints degree of violation Represented as: In the formula, This represents the design variable vector for the current individual, containing all design parameters. Indicates the first Each constraint on the individual The constraint function value; Indicates the first The upper limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual This constraint is completely violated, with a violation level of 1. Indicates the first The lower limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual If this constraint is satisfied, the degree of violation is 0; The sliding window smoothing strategy is expressed as follows: In the formula, Indicates the first Individual For the first The degree of smooth violation of each constraint; This indicates the window length, which is adaptively adjusted based on the current optimization stage. Indicates the first Individual For the first The degree of violation of a constraint, with a numerical range between 0 and 1; The index variable represents the sliding window, and the algebra offset represents the number of algebras to backtrack from the current algebra. Then, the fuzzy membership function is expressed as: In the formula, Represents an individual For the first The membership degree of a constraint's minor violation reflects the degree to which the constraint is slightly violated; the closer the value is to 1, the less severe the violation. Indicates parameters For the first The degree of severe violation of a constraint reflects the extent to which the constraint is severely violated; the closer the value is to 1, the more severe the violation. Phased flexible penalty function Represented as: In the formula, Indicates the set number of rules; Indicates the index of the rule being set; Indicates the weights of fuzzy rule inference; Indicates the first The penalty function value corresponding to each fuzzy rule is used to describe the individual. The severity of punishment under this rule; Step S23: Considering the differences in convergence speed among different optimization objectives, a dynamic weight adjustment strategy based on the convergence speed is defined to update the weights of each optimization objective. This process solves the multi-objective optimization model for the heavy-duty vehicle configuration, outputting the optimal values ​​for the heavy-duty vehicle design parameters, where: No. The first goal in convergence speed of the generation Defined as: In the formula, Indicates the first The generation The value of each objective function, i.e., the objective evaluation result of the current generation; Indicates the first The generation The value of the objective function, i.e. the objective evaluation result of the previous generation; This represents a very small positive number, used to avoid the denominator being zero; Based on stage parameters during the evolution process Update the weights of each optimization objective. The update process is represented as follows: In the formula, Indicates the first The first goal in The weight of generations; express The first goal in The weight of generations; The target quantity.

[0014] A system for the above-described multi-objective optimization design method for heavy-duty vehicle configurations includes: The optimization model building module is used to define the design parameters of heavy-duty vehicles. Based on the parametric modeling method, it constructs a multi-objective optimization model for the configuration of heavy-duty vehicles with the system efficiency and performance requirements of heavy-duty vehicles as optimization objectives. The optimization solution module introduces a phased fuzzy evolution mechanism on the basis of the traditional NSGA-II algorithm. It divides the iterative process into several stages, designs different fuzzy constraint rules and sliding window smoothing strategies for each stage, and dynamically adjusts the weights of different optimization objectives in combination with the convergence speed of the objective function. It solves the multi-objective optimization model of heavy-duty vehicle configuration and outputs the optimal values ​​of heavy-duty vehicle design parameters.

[0015] The beneficial effects of this invention are as follows: This invention can more accurately address the diverse needs under complex mining conditions. By constructing a multi-dimensional evaluation index system, it incorporates functionality, economy, reliability, and other dimensions into consideration, thereby achieving a more comprehensive optimized design. Simultaneously, it proposes a staged fuzzy evolution NSGA-II optimization algorithm, effectively improving the robustness and global stability of the optimized solution set, ensuring reliable vehicle operation in the variable mining environment. Compared to single numerical optimization methods, this invention not only improves computational speed but also significantly increases the optimization success rate, providing a more efficient and reliable solution for practical applications. Ultimately, this invention achieves efficient, stable, and economical multi-objective optimization design of heavy-duty vehicle drive systems under complex mining conditions, providing strong technical support for improving the performance of mining vehicles. Detailed Implementation

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] This embodiment provides a multi-objective optimization design method for heavy-duty vehicle configurations, including the following steps: Step S1: Define the design parameters of the heavy-duty vehicle. Based on the parametric modeling method, construct a multi-objective optimization model for the configuration of the heavy-duty vehicle with the system efficiency and performance requirements of the heavy-duty vehicle as the optimization objectives.

[0018] Design parameters for heavy-duty vehicles include the number of drive shafts, tire radius, drive motor power, wheelbase, and suspension stiffness; performance requirements include cost, climbing ability, reliability, and lateral stability.

[0019] In the modeling process, this application takes maximizing system efficiency as the primary optimization objective. Simultaneously, considering the diverse performance requirements of heavy-duty vehicles in engineering applications (cost, climbing ability, reliability, and lateral stability) also have clear optimization objective characteristics, this application uses system efficiency as the primary optimization objective to improve optimization efficiency. Performance requirements are transformed into constraints and controlled by setting performance thresholds. This method essentially still reflects the collaborative optimization characteristics of multiple performance indicators and belongs to a multi-objective optimization modeling strategy of "integrated objective and constraint."

[0020] The optimization objective for heavy-duty vehicles is to maximize system efficiency while satisfying all constraints. While ensuring vehicle performance, energy loss is reduced and energy utilization is improved.

[0021] System efficiency Defined as demand power With input power The ratio is expressed as: Among them, the required power It is the power required for the vehicle to run, expressed as: In the formula, For vehicle driving resistance; Vehicle speed; driving resistance It consists of rolling resistance and air resistance, specifically expressed as: In the formula, The total weight of the vehicle; This is the rolling resistance coefficient; This refers to the air drag coefficient; This refers to the windward area.

[0022] Input power It is the power provided by the vehicle's powertrain (such as an engine or electric motor), calculated using the following formula: In the formula, For the vehicle powertrain system The torque of the root drive shaft; The rotational speed of the drive shaft; This represents the number of drive shafts.

[0023] During operation, the vehicle must meet the following constraints: (1) Climbing ability constraint Climbing ability is an important indicator of a vehicle's ability to drive on inclines. To ensure a vehicle can overcome a certain gradient, the following constraints apply to climbing ability: In the formula, This represents the vehicle's driving force coefficient. The coefficient of friction; The minimum climbing angle; where: In the formula, This represents the power of a single drive shaft; the sum of the power of all drive shafts is the input power. ; Indicates the tire radius; and These are parameters related to the transmission system, specifically expressed as: In the formula, This refers to the gear ratio of the transmission. The main reducer transmission ratio; This refers to the mechanical efficiency of the transmission system.

[0024] (2) Cost constraints The cost of heavy-duty vehicles mainly lies in the procurement and manufacturing of components such as drive shafts, tires, and drive motors. Therefore, this application simplifies heavy-duty vehicles into a system consisting of drive shafts, tires, and drive motors, where each drive shaft is equipped with two tires, and each tire is individually equipped with a drive motor, forming an independent wheel-side drive structure. This simplifies the cost constraints. Represented as: In the formula, This refers to the unit price of the tire; This refers to the unit price of the drive motor; For the first The unit price of a drive shaft. This represents the maximum permissible cost.

[0025] (3) Reliability constraints Reliability quantification indicators include failure reliability and average service life. The configuration design involves components such as drive shafts, tires, and drive motors. These components are integrated on the vehicle. Without considering mechanical reliability issues caused by installation and manufacturing processes, the system reliability is determined by these components. If one or more of these components fail, the entire vehicle cannot function properly. Since the components are connected in series, it is a reliability series system.

[0026] Failure reliability of each component Represented as: In the formula, Represents probability; Indicates the time when the component failed; Indicates the normal operating time required for a component to complete its task; This indicates that the component is not faulty; This indicates the probability that a component is fault-free.

[0027] The failure reliability constraints for each component are then expressed as follows: In the formula, This indicates the required fault reliability threshold.

[0028] The logic for determining fault reliability is as follows: before maintenance intervention, the component is in a normal operating state; over time, the component fails and ceases normal operation. Therefore, this application uses the time of component failure and the normal operating time required for the component to complete its task as the criteria for determination. If the normal operating time required for a component to complete its task is 5000 hours and the time before a failure occurs is 5100 hours, then the component can successfully complete its task before a failure occurs, and the component's reliability meets the task requirements.

[0029] In addition, this application introduces the concept of probability distribution to represent the probability of a component working properly, which can more intuitively represent the reliability of the component.

[0030] Mean service life can be defined as the mathematical expectation of the interval between two failures. These represent the average service life of the drive shaft, tires, and drive motor, respectively.

[0031] Service life of heavy-duty vehicles It equals the minimum lifespan of all components, expressed as: Therefore, service life reliability Represented as: In the formula, , , These represent the service life and reliability of the drive shaft, tires, and drive motor, respectively. This indicates the threshold required for the service life of heavy-duty vehicles.

[0032] The service life reliability constraint is then expressed as: In the formula, This indicates the required service life reliability threshold for heavy-duty vehicles.

[0033] This application quantifies the service life reliability of heavy-duty vehicles as the product of the service life reliability of core sub-components (drive shaft, tires, drive motor), and introduces the concept of probability distribution to represent the probability distribution of the service life of each core sub-component in probabilistic form. This can more accurately reflect the probability of heavy-duty vehicles reaching the minimum service life standard and reflect the collaborative reliability of multiple components.

[0034] (4) Lateral stability constraints Lateral stability is a vehicle's ability to resist rollover in the lateral plane, characterized by the static stability coefficient. Represented as: In the formula, It represents the lateral acceleration of a vehicle, and the magnitude of the acceleration when the vehicle moves laterally. Represents gravitational acceleration; Indicates wheel track; Indicates the height of the vehicle's center of gravity; Indicates the maximum travel of the suspension. Indicates suspension stiffness; This is the empty vehicle weight. For load capacity.

[0035] To ensure the lateral stability of the vehicle, the static stability coefficient The following constraints must be met: in, and These are the minimum and maximum values ​​of the static stability coefficient, respectively.

[0036] Step S2 introduces a phased fuzzy evolution mechanism based on the traditional NSGA-II algorithm to iteratively solve the configuration optimization model of heavy-duty vehicles. The iterative process is divided into several stages, and different fuzzy constraint rules and sliding window smoothing strategies are designed for each stage. The weights of different optimization objectives are dynamically adjusted in combination with the convergence speed of the objective function.

[0037] In solving multi-objective optimization problems, a common approach is to weight and combine the objective functions according to preset weight coefficients, thereby transforming the multi-objective problem into a single-objective optimization problem and finding its unique optimal solution. This method is simple to implement and computationally efficient, but it has significant limitations: the optimization result is highly dependent on the setting of the weight coefficients and cannot fully reflect the trade-offs between multiple objectives. Another more commonly used method employs heuristic search techniques such as evolutionary algorithms to directly search for an approximate Pareto optimal front in the objective space, thereby obtaining a set of non-dominated solutions. This provides decision-makers with a basis for choosing among various optimization schemes, such as the NSGA-II multi-objective optimization algorithm. However, the traditional NSGA-II algorithm faces early overexploration or premature convergence in complex multi-constraint optimization scenarios. To address this issue, this invention proposes a staged fuzzy evolutionary NSGA-II optimization algorithm. This method introduces stage division and adaptive adjustment of fuzzy logic constraint processing into the evolutionary process of the traditional NSGA-II algorithm, achieving a more flexible search strategy and better solution set diversity.

[0038] Step S2 specifically includes the following process: Step S21: Based on the traditional NSGA-II algorithm, a staged fuzzy evolution mechanism is introduced to reduce the overall number of iterations. Divided into Stages: Step S22: In each stage, different fuzzy constraint rules and sliding window smoothing strategies are designed according to the search feature requirements to generate stage-specific fuzzy constraint penalties.

[0039] Exploration phase: Focus on maintaining diversity, with lenient constraints and penalties, and a long smoothing period for the sliding window; Equilibrium phase: Gradually tighten the constraint strength to guide the solution distribution to equilibrium; Convergence Phase: Enhanced feasibility convergence, short sliding window period, and strict penalty; Specifically: The definition of fuzzy rules is as follows: In the Generation, for constraints degree of violation Represented as: In the formula, This represents the design variable vector for the current individual, containing all design parameters. Indicates the first Each constraint on the individual The constraint function value; Indicates the first The upper limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual This constraint is completely violated, with a violation level of 1. Indicates the first The lower limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual If this constraint is satisfied, the degree of violation is 0; The sliding window smoothing strategy is expressed as follows: In the formula, Indicates the first Individual For the first The degree of smooth violation of each constraint; This indicates the window length, which is adaptively adjusted based on the current optimization stage. Indicates the first Individual For the first The degree of violation of a constraint, with a numerical range between 0 and 1; The index variable represents the sliding window, and the algebra offset represents the number of algebras to backtrack from the current algebra. Then, the fuzzy membership function is expressed as: In the formula, Represents an individual For the first The membership degree of a constraint's minor violation reflects the degree to which the constraint is slightly violated; the closer the value is to 1, the less severe the violation. Indicates parameters For the first The degree of severe violation of a constraint reflects the extent to which the constraint is severely violated; the closer the value is to 1, the more severe the violation. Phased flexible penalty function Represented as: In the formula, Indicates the set number of rules; Indicates the index of the rule being set; Indicates the weights of fuzzy rule inference; Indicates the first The penalty function value corresponding to each fuzzy rule is used to describe the individual. The severity of punishment under this rule; Step S23: Considering the differences in convergence speed among different optimization objectives, a dynamic weight adjustment strategy based on the convergence speed is defined to update the weights of each optimization objective. This process solves the multi-objective optimization model for the heavy-duty vehicle configuration, outputting the optimal values ​​for the heavy-duty vehicle design parameters, where: No. The first goal in convergence speed of the generation Defined as: In the formula, Indicates the first The generation The value of the objective function, i.e., the objective evaluation result of the current generation. Indicates the first The generation The value of the objective function, i.e., the objective evaluation result of the previous generation. This represents a very small positive number, used to avoid the denominator being zero and to ensure the numerical stability of the calculation; its value is 0.0001. Based on stage parameters during the evolution process Update the weights of each optimization objective. The update process is represented as follows: In the formula, Indicates the first The first goal in The weight of generations; express The first goal in The weight of generations; The target quantity.

[0040] In summary, the overall process of the NSGA-II optimization algorithm for staged fuzzy evolution is as follows: Step 1: Initialize population coding and set phase division parameters; Step 2: In the current stage, perform genetic operations (selection, crossover, mutation), non-dominated sorting, and crowding calculation; Step 3: Based on the stage-specific sliding window length and fuzzy rules, calculate the flexible constraint penalty function and adjust the individual fitness; Step 4: Dynamically adjust the weights of multiple objectives based on the convergence speed to enhance the search intensity of weak convergence objectives; Step 5: Determine whether to proceed to the next stage, and automatically switch based on the threshold or population diversity index; Step 6: Iterate until the maximum algebra or other termination condition is met.

[0041] By introducing staged fuzzy logic processing, sliding window smoothing, and dynamic weight adjustment into NSGA-II, high search diversity can be maintained in the early stages, while the focus gradually shifts to feasible solutions in the later stages, achieving balanced and efficient search of the Pareto front under complex constraints. Based on the convergence speed of the objective function, a dynamic weight adjustment method enhances global search capabilities in the early stages of optimization, balances solution set diversity and feasibility in the middle stages, and focuses on adaptive optimization to converge and approach the Pareto front in the later stages. This strategy effectively overcomes the shortcomings of traditional fixed-weight methods in handling feasibility and objective conflicts in multi-constraint, multi-objective optimization, improving the diversity, distribution balance, and convergence efficiency of the Pareto solution set.

[0042] In addition, other multi-objective optimization algorithms (such as the MOEA / D algorithm and the SPEA2 algorithm) can be used to replace the NSGA-II algorithm in other embodiments, or similar optimization effects can be achieved by adjusting the weights of the objective function and changing the form of the constraints. However, the core idea of ​​these alternatives is still based on multi-objective optimization and flexible constraint handling mechanisms. Regardless of how the algorithm or model structure is adjusted, its essence is still to improve the optimization effect of the heavy-duty vehicle drive system by coordinating the balance between multiple performance objectives and combining flexible constraint mechanisms. Therefore, these alternatives do not deviate from the multi-objective optimization design framework proposed in this paper.

[0043] A system for the above-described multi-objective optimization design method for heavy-duty vehicle configurations includes: The optimization model building module is used to define the design parameters of heavy-duty vehicles. Based on the parametric modeling method, it constructs a multi-objective optimization model for the configuration of heavy-duty vehicles with the system efficiency and performance requirements of heavy-duty vehicles as optimization objectives. The optimization solution module introduces a phased fuzzy evolution mechanism on the basis of the traditional NSGA-II algorithm. It divides the iterative process into several stages, designs different fuzzy constraint rules and sliding window smoothing strategies for each stage, and dynamically adjusts the weights of different optimization objectives in combination with the convergence speed of the objective function. It solves the multi-objective optimization model of heavy-duty vehicle configuration and outputs the optimal values ​​of heavy-duty vehicle design parameters.

[0044] The embodiments of the present invention have been described above. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A multi-objective optimization design method for heavy-duty vehicle configurations, characterized in that, Includes the following steps: Step S1: Define the design parameters of the heavy-duty vehicle. Based on the parametric modeling method, construct a multi-objective optimization model for the configuration of the heavy-duty vehicle with the system efficiency and performance requirements of the heavy-duty vehicle as the optimization objectives. Step S2 introduces a phased fuzzy evolution mechanism based on the traditional NSGA-II algorithm, dividing the iteration process into several stages. Different fuzzy constraint rules and sliding window smoothing strategies are designed for each stage. The weights of different optimization objectives are dynamically adjusted in combination with the convergence speed of the objective function to solve the multi-objective optimization model of heavy-duty vehicle configuration and output the optimal values ​​of heavy-duty vehicle design parameters.

2. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 1, characterized in that, The design parameters for heavy-duty vehicles include the number of drive shafts, tire radius, drive motor power, wheelbase, and suspension stiffness. Performance requirements include cost, climbing ability, reliability, and lateral stability.

3. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 1, characterized in that, In constructing a multi-objective optimization model for heavy-duty vehicle configurations, system efficiency is taken as the main optimization objective, performance requirements are converted into constraints, and performance thresholds are set for control.

4. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 2, characterized in that, System efficiency of heavy-duty vehicles Represented as: In the formula, This indicates the power requirement of heavy-duty vehicles; This represents the input power of a heavy-duty vehicle; where: In the formula, Indicates the vehicle's resistance to movement; Indicates the vehicle's speed; In the formula, Indicates the total weight of the vehicle; Indicates the rolling resistance coefficient; Indicates the air drag coefficient; Indicates the windward area; In the formula, Indicates vehicle number The torque of the root drive shaft; The rotational speed of the drive shaft; This represents the number of drive shafts.

5. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 4, characterized in that, The climbing ability constraint is expressed as follows: In the formula, This represents the vehicle's driving force coefficient. The coefficient of friction; The minimum climbing angle; where: In the formula, This represents the power of a single drive shaft; the sum of the power of all drive shafts is the input power. ; Indicates the tire radius; and These are parameters related to the transmission system, expressed as: In the formula, Indicates the gear ratio of the transmission; Indicates the transmission ratio of the main reducer; This indicates the mechanical efficiency of the transmission system.

6. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 5, characterized in that, In constructing the cost constraint, the heavy-duty vehicle is simplified into a system consisting of a driveshaft, tires, and a drive motor. Each driveshaft has two tires, and each tire is equipped with its own drive motor, forming an independent wheel-side drive structure. The cost constraint then becomes... Represented as: In the formula, This indicates the unit price of the tire; This indicates the unit price of the drive motor; Indicates the first The unit price of a drive shaft; This indicates the maximum permissible cost.

7. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 6, characterized in that, Reliability constraints include failure constraints and mean service life constraints, among which: Fault constraints are represented as: In the formula, Indicates fault reliability; This represents the required failure reliability threshold; where: In the formula, Represents probability; Indicates the time when the component failed; Indicates the normal operating time required for a component to complete its task; This indicates that the component is not faulty; Indicates the probability that a component is fault-free; Service life reliability constraints are expressed as follows: In the formula, Indicates service life reliability; This represents the required service life reliability threshold, where: In the formula, , , These represent the service life and reliability of the drive shaft, tires, and drive motor, respectively. These represent the average service life of the drive shaft, tires, and drive motor, respectively. This indicates the threshold required for the service life of heavy-duty vehicles. Indicates the service life of heavy-duty vehicles.

8. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 2, characterized in that, Lateral stability constraints are expressed as: In the formula, Represents the static stability coefficient; and These are the minimum and maximum values ​​allowed for the static stability coefficient, respectively; where: In the formula, It represents the lateral acceleration of a vehicle, and the magnitude of the acceleration when the vehicle moves laterally. Represents gravitational acceleration; Indicates wheel track; Indicates the height of the vehicle's center of gravity; Indicates the maximum travel of the suspension. Indicates suspension stiffness; This is the empty vehicle weight. For load capacity.

9. The multi-objective optimization design method for heavy-duty vehicle configuration according to claim 1, characterized in that, Step S2 specifically includes the following steps: Step S21: Based on the traditional NSGA-II algorithm, a staged fuzzy evolution mechanism is introduced to reduce the overall number of iterations. Divided into Stages: Step S22: In each stage, different fuzzy constraint rules and sliding window smoothing strategies are designed according to the search feature requirements to generate stage-specific fuzzy constraint penalties. The fuzzy rules are defined as follows: In the Generation, for constraints degree of violation Represented as: In the formula, This represents the design variable vector for the current individual, containing all design parameters. Indicates the first Each constraint on the individual The constraint function value; Indicates the first The upper limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual This constraint is completely violated, with a violation level of 1. Indicates the first The lower limit of the fuzzy tolerance interval of a constraint, when At that time, it was believed that the individual If this constraint is satisfied, the degree of violation is 0; The sliding window smoothing strategy is expressed as follows: In the formula, Indicates the first Individual For the first The degree of smooth violation of each constraint; This indicates the window length, which is adaptively adjusted based on the current optimization stage. Indicates the first Individual For the first The degree of violation of a constraint, with a numerical range between 0 and 1; The index variable represents the sliding window, and the algebra offset represents the number of algebras to backtrack from the current algebra. Then, the fuzzy membership function is expressed as: In the formula, Represents an individual For the first The membership degree of a constraint's minor violation reflects the degree to which the constraint is slightly violated; the closer the value is to 1, the less severe the violation. Indicates parameters For the first The degree of severe violation of a constraint reflects the extent to which the constraint is severely violated; the closer the value is to 1, the more severe the violation. Phased flexible penalty function Represented as: In the formula, Indicates the set number of rules; Indicates the index of the rule being set; Indicates the weights of fuzzy rule inference; Indicates the first The penalty function value corresponding to each fuzzy rule is used to describe the individual. The severity of punishment under this rule; Step S23: Considering the differences in convergence speed among different optimization objectives, a dynamic weight adjustment strategy based on the convergence speed is defined to update the weights of each optimization objective. This process solves the multi-objective optimization model for the heavy-duty vehicle configuration, outputting the optimal values ​​for the heavy-duty vehicle design parameters. No. The first goal in convergence speed of the generation Defined as: In the formula, Indicates the first The generation The value of each objective function, i.e., the objective evaluation result of the current generation; Indicates the first The generation The value of the objective function, i.e. the objective evaluation result of the previous generation; This represents a very small positive number, used to avoid the denominator being zero; Based on stage parameters during the evolution process Update the weights of each optimization objective. The update process is represented as follows: In the formula, Indicates the first The first goal in The weight of generations; express The first goal in The weight of generations; The target quantity.

10. A system for executing the multi-objective optimization design method for heavy-duty vehicle configurations according to any one of claims 1-9, characterized in that, include: The optimization model building module is used to define the design parameters of heavy-duty vehicles. Based on the parametric modeling method, it constructs a multi-objective optimization model for the configuration of heavy-duty vehicles with the system efficiency and performance requirements of heavy-duty vehicles as optimization objectives. The optimization solution module introduces a phased fuzzy evolution mechanism on the basis of the traditional NSGA-II algorithm. It divides the iterative process into several stages, designs different fuzzy constraint rules and sliding window smoothing strategies for each stage, and dynamically adjusts the weights of different optimization objectives in combination with the convergence speed of the objective function. It solves the multi-objective optimization model of heavy-duty vehicle configuration and outputs the optimal values ​​of heavy-duty vehicle design parameters.

Images