Integrated Optimization Design System for Tracked Vehicles Based on FMI

Abstract

An FMI-based tracked vehicle integrated optimization design system includes: a tracked vehicle simulation model unit, an FMU model conversion unit, a data interface interaction unit, and an optimization design unit. Based on the integrated integration of vehicle joint simulation models based on the FMI standard model interface, the present invention eliminates the nested iterations of the complex system optimization design process through a parallel optimization solution for multi-domain subsystems, thereby improving the efficiency of the tracked vehicle system optimization design process and product quality.

Description

Technical Field

[0001] The present invention relates to a technology in the field of tracked vehicle optimization design, specifically to an integrated optimization design system for tracked vehicles based on the FMI (Functional Mock-up Interface) standard model interface. Background Art

[0002] Tracked vehicles are large and complex equipment that include multi-domain subsystems. During the design optimization process, they face difficulties such as long iteration cycles, low optimization efficiency, and unsatisfactory optimization results. The above problems are mainly caused by the following two aspects: (1) Tracked vehicles involve multiple fields such as mechanics, hydraulics, control, and electronics. The simulation models of systems in different fields, such as mechanical structure models, multi-body dynamics models, and control system models, need to be established in different simulation tools. In the process of tracked vehicle research and development and optimization design, it is necessary to combine the models of software platforms in various fields to form a complete tracked vehicle system simulation analysis model. However, in the traditional cross-platform joint simulation method based on secondary development interfaces, the models of various fields are still solved on their own simulation tools, with a weak degree of coupling, resulting in tool fragmentation, and unable to solve the problems of compatibility and integration between software, which seriously hinders the efficiency of integrated optimization design of tracked vehicles. (2) The optimization design of complex systems such as traditional tracked vehicles often adopts a serial nested iteration method in each field, which causes the number of iterations in the optimization process to increase exponentially, the time cost is high, and the optimization results are prone to fall into local optimality, resulting in limited product quality. However, the existing FMI-based equipment collaborative simulation technology only uses the main control program to solve the system model, and lacks consideration of multi-domain optimization methods for complex systems. Summary of the Invention

[0003] In response to the problems of low integration of multi-domain models and low efficiency of the optimization process in the existing tracked vehicle design optimization process, the present invention proposes an FMI-based tracked vehicle integrated optimization design system. Based on the integrated integration of vehicle joint simulation models based on the FMI standard model interface, the parallel optimization scheme of multi-domain subsystems is used to eliminate the nested iterations of the complex system optimization design process, so as to improve the efficiency of the tracked vehicle system optimization design process and product quality.

[0004] The present invention is achieved through the following technical solutions:

[0005] The present invention relates to an FMI-based tracked vehicle integrated optimization design system, comprising: a tracked vehicle simulation model unit, an FMU model conversion unit, a data interface interaction unit, and an optimization design unit, wherein: the tracked vehicle simulation model unit provides a basis for subsequent conversion and packaging into an FMU model based on a unified FMI interface; the FMU model conversion unit converts and packages the tracked vehicle simulation model into an FMU model with a unified FMI interface, providing a basis for the subsequent integration of a tracked vehicle integrated joint simulation system model; the data interface interaction unit determines the data interaction interface between each FMU model, and realizes a one-to-one correspondence connection of FMU models in various fields through a unified FMI model interface. Realize the integrated integration of the tracked vehicle joint simulation model; secondly, the data interface interaction unit also includes the interface interaction between the tracked vehicle joint simulation model and the optimization design unit. The parallel optimization design unit divides the optimization design problem of the tracked vehicle into a two-layer optimization strategy of a system level and multiple domain subsystem levels, and simultaneously determines the coupling design variables of each subsystem. At the system-level optimization layer, the conflicts of the coupling variables of each subsystem are coordinated by constructing consistency constraints of the coupling variables, so that the various domain subsystems of the subsystem optimization layer can execute the optimization process completely independently. The optimization design unit passes the optimization variables to the integrated joint simulation model through interface interaction, and dynamically changes the optimization variables to find the optimal value.

[0006] The tracked vehicle simulation model unit includes: a tracked vehicle multi-body dynamics model, a control system model, a three-dimensional structure model, a finite element analysis model and other domain subsystem models established on different simulation software.

[0007] The FMU model conversion unit includes: a multi-body dynamics FMU model conversion module, a control system FMU model conversion module, a three-dimensional structure FMU model conversion module and a finite element analysis FMU model conversion module, wherein: the multi-body dynamics FMU model conversion module converts the established multi-body dynamics model information based on a unified FMI interface and converts it into a multi-body dynamics FMU model; the control system FMU model conversion module converts it into a control system FMU model based on the established control system model information; the three-dimensional structure FMU model conversion module converts it into a three-dimensional structure FMU model based on the established three-dimensional structure model information; and the finite element analysis FMU model conversion module converts it into a finite element analysis FMU model based on the established finite element analysis model information.

[0008] The data interface interaction unit includes an FMU model interface module and an optimization interface module. The FMU model interface module transfers information between parameters of the FMU models to establish interface interaction relationships between the FMU models, thereby achieving integrated integration of the tracked vehicle co-simulation model. The optimization interface module establishes an optimized interface interaction relationship between the optimization design unit and the tracked vehicle co-simulation model based on information such as the tracked vehicle's optimization objectives and optimization variables.

[0009] The parallel optimization design unit comprises a system-level optimization module and a subsystem-level optimization module. The system-level module establishes consistency constraints on coupling variables to coordinate conflicts between coupling variables across subsystems, enabling autonomous optimization of subsystems in each domain. The subsystem optimization layer independently constructs optimization problems for each subsystem in each domain, enabling parallel optimization of these subsystems.

[0010] The present invention relates to a tracked vehicle integrated optimization method based on FMI of the above system, comprising the following steps:

[0011] Step S1, using different simulation tools to establish simulation models of subsystems in different fields such as tracked vehicle multi-body dynamics, control system, hydraulic transmission, three-dimensional structure, and finite element analysis;

[0012] Step S2: Based on the unified FMI model interface, each domain model is converted and encapsulated into a corresponding FMU model;

[0013] Step S3: Extract the interface data attribute information between FMU modules to form a model interface module, completing the integrated integration of the tracked vehicle model. At the same time, according to the specific requirements of the tracked vehicle optimization design, complete the tracked vehicle optimization design process, set the optimization objectives and optimization variables in each field, and form the optimization interface module to form the final data interface interaction unit.

[0014] Step S4: Decompose the tracked vehicle design optimization process into two levels: system-level optimization and multi-domain subsystem-level optimization. Determine the optimization objectives, optimization variables, and coupling variables of the system-level and multi-domain subsystem optimization processes, respectively. Simultaneously, establish consistency constraints for the coupling variables in each domain, achieve parallel optimization of the multi-domain subsystems of the tracked vehicle, and establish an optimization design unit for the tracked vehicle.

[0015] Step S5: Run the connected tracked vehicle integrated optimization design system, determine the simulation step size, synchronize the clocks between multiple models, and complete the initialization of the integrated optimization system;

[0016] Step S6: Utilize the tracked vehicle integrated optimization design system to perform cyclic iterative optimization on the tracked vehicle to determine whether the optimization result meets the optimization target. If not, continue iterating until the optimization result meets the optimization target.

[0017] Technical Effects

[0018] Compared with existing conventional technical means, the present invention first uses the FMU model conversion unit to convert the subsystem models of different domain software in the multi-domain simulation model unit into FMU models based on the FMI unified interface, thereby realizing the integrated integration of the multi-domain joint simulation model of the tracked vehicle, and solving the problem of insufficient simulation efficiency and accuracy caused by the incompatibility between multi-domain simulation software; secondly, the parallel optimization strategy of the multi-domain subsystem of the optimization design unit is used to eliminate the nested loops between the systems in each domain, further improve the efficiency of the tracked vehicle design optimization, adapt to the multi-domain parameter coupling characteristics of the tracked vehicle system optimization design, and improve the optimization iterative design efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the system structure of the present invention;

[0020] Figure 2 Flowchart of specific operation steps of the system of the present invention;

[0021] Figure 3 This is a schematic diagram of the overall process of optimizing the smoothness of a tracked vehicle according to an embodiment;

[0022] Figure 4 is the E-level pavement excitation map obtained according to the national standard GB / T 7031 2005;

[0023] Figure 5 A comparison chart of integrated optimization results of tracked vehicles using the present invention;

[0024] Figure 6 This is a schematic diagram of the vertical acceleration optimization effect;

[0025] Figure 7 Schematic diagram of the vertical pitch angle optimization effect. DETAILED DESCRIPTION

[0026] like Figure 1 As shown, this embodiment relates to an FMI-based tracked vehicle integrated optimization design system, including: a tracked vehicle multi-domain simulation model unit, an FMU model conversion unit, a data interface interaction unit and an optimization design unit.

[0027] like Figure 2 FIG. 1 is an example of a tracked vehicle ride comfort optimization design based on the above-mentioned system according to this embodiment. The skyhook drag control coefficient in the vehicle suspension control system, which affects the ride comfort of the vehicle, is optimized to achieve the optimization goal. Specifically, the following steps are performed:

[0028] Step S1, based on the specific optimization design requirements of the tracked vehicle, establish corresponding domain system models in Pro / E, AMESim, RecurDyn, and Simulink domain software respectively. Specifically, the three-dimensional structure model of the tracked vehicle is established in Pro / E and imported into the RecurDyn software to establish a tracked vehicle multi-body dynamics system model, the tracked vehicle hydraulic system model is established in AMESim, and the tracked vehicle suspension control system model is established in Simulink.

[0029] Step S2: Through the FMI standard model interface, the FMU conversion modules of various field software, including AMESim-to-FMU conversion module, RecurDyn-to-FMU conversion module, and Simulink-to-FMU conversion module, are used to convert and encapsulate the simulation models in various field software into FMU models with a unified interface standard.

[0030] Step S3: construct a data interface interaction module in the integrated optimization design system, specifically including the data interaction interface between FMU models in various fields, and the interface interaction between the integrated tracked vehicle joint simulation model and the performance design optimization unit.

[0031] The data interaction interfaces between the FMU models in various fields include: the interface interaction between the hydraulic subsystem FMU model and the tracked vehicle multi-body dynamics FMU model, and the interface interaction between the suspension control subsystem FMU model and the tracked vehicle multi-body dynamics FMU model. Specifically, the hydraulic subsystem FMU model transmits the driving torque of the tracked vehicle drive wheel to the vehicle multi-body dynamics FMU model, and the control subsystem FMU model transmits the adjustable damping force calculated by the control skyhook damping system to the vehicle multi-body dynamics FMU model. The vibration displacement x in the vertical direction of the vehicle body center of mass calculated by the multi-body dynamics FMU model b The data is then transferred to the suspension control system FMU model. By connecting the above FMU model interfaces one by one, the integrated integration of the tracked vehicle joint simulation model is completed.

[0032] The interface interaction between the tracked vehicle system joint simulation model and the performance design optimization unit is as follows: the optimization variables are the skyhook damping coefficient L1 and the skyhook damping coefficient L2 of the key wheels; the optimization target is the smoothness index of the tracked vehicle, specifically: σ acc = and Where: acc is the vertical acceleration of the vehicle's center of mass The root mean square, σ α is the root mean square of the pitch angle α, and the specific optimization process is as follows Figure 3 shown.

[0033] Step S4, establish the optimization design unit of the tracked vehicle, determine the separate optimization objectives, optimization variables and coupling variables in each field of the subsystem optimization layer, determine the system-level optimization objectives and optimization variables and consistency constraints, coordinate the conflicts of the coupling variables of each subsystem, and realize the parallel optimization of subsystems in each field of the subsystem optimization layer.

[0034] The system-level optimization problem of the optimization method is:

[0035] min F sys (X, Y, Z)

[0036] stJ(Z)=||ZY||=0,

[0037] Where: Y is the given subsystem optimization state variable, {X, Z} is the system-level optimization variable, Z is the coupling variable of the corresponding subsystem and Z L ≤Z≤Z U , J is the system-level consistency constraint to eliminate the conflicts caused by coupling variables in the separate optimization process of each sub-domain system.

[0038] The subsystem-level optimization problem is:

[0039] min F i (Y i , X, Z)i=1→S3

[0040] stJ(Y i , X, Z)

[0041] Where: {X, Z} are variables for given system-level optimization and At the subsystem optimization level, multiple domain subsystems are optimized separately. The optimization method is: while keeping the global variables and coupling variables {X, Z} constant, find the optimal value of the subsystem under given constraints and the subsystem optimization variable Y i , and passed back to the system-level optimization layer.

[0042] Step S5: Determine the boundary conditions of the optimization design problem, including the initial values of the optimization design variables and system state variables, external environment parameters, etc. At the same time, determine the simulation step size of the tracked vehicle joint simulation model, synchronize the clocks of the FMU models in each field, and complete the initialization of the optimization design process;

[0043] Step S6, based on the smoothness optimization target of the tracked vehicle, starts to optimize the skyhook damping control system of the key wheels of the tracked vehicle suspension control system until the specified optimization target is met and the optimization result is obtained.

[0044] For the boundary conditions of the road surface excitation of the tracked vehicle, the road surface roughness classification level of the national standard GB / T 7031 2005 is adopted, and the E-level road surface roughness is selected as the boundary condition of the road surface environment of the embodiment, such as Figure 4 shown.

[0045] Through specific actual simulation experiments, according to the national standard GB / T 7031-2005, the road roughness classification is divided into levels. Under the setting of E-level road roughness as the specific road environment, the skyhook damping coefficient L1 and skyhook damping coefficient L2 of the key wheel are used as optimization parameters to run the above method. The results of the smoothness optimization of the tracked vehicle system are obtained. The verification data obtained are:

[0046] like Figure 5 As shown in the figure, by comparing the number of iterations of this system with that of the traditional system, it can be seen that compared with the traditional overall optimization method, this system reaches the convergence condition with fewer iteration steps, thereby improving the efficiency of the integrated optimization design of tracked vehicles.

[0047] like Figure 6 and Figure 7 As shown in the figure, the smoothness index effect after the tracked vehicle optimization system is optimized. This system can effectively improve the smoothness of the vehicle. After the optimization design, the vertical acceleration is more stable, the peak value is smaller, and the acceleration root mean square value σ acc Compared with the pre-optimization, it decreased by 20.63%. From the pitch angle results, the pitch angle changes of the vehicle after optimization are also more stable, and the root mean square value of the pitch angle σ α Compared with before optimization, it decreased by 22.14%.

[0048] Compared with the existing technology, this system first uses a unified FMI model interface to integrate multi-domain models to improve the operating efficiency of the tracked vehicle joint simulation model. At the same time, it uses the optimization design unit to eliminate the nested loops of the sub-domain system to achieve parallel optimization of multi-domain subsystems, reduce the computational complexity of the optimization process, and improve the optimization design efficiency and product design quality.

[0049] The above-mentioned specific implementation can be partially adjusted in different ways by those skilled in the art without departing from the principles and purpose of the present invention. The scope of protection of the present invention shall be based on the claims and shall not be limited by the above-mentioned specific implementation. All implementation schemes within its scope shall be subject to the constraints of the present invention.

Claims

1. An integrated optimization design system for tracked vehicles based on FMI, characterized by: include: Tracked vehicle simulation model unit, FMU model conversion unit, data interface interaction unit and optimization design unit, among which: the tracked vehicle simulation model unit provides the basis for subsequent conversion and packaging into an FMU model based on a unified FMI interface, the FMU model conversion unit converts and packages the tracked vehicle simulation model into an FMU model with a unified FMI interface, providing the basis for the subsequent integration of the tracked vehicle integrated joint simulation system model, the data interface interaction unit determines the data interaction interface between each FMU model, and realizes the one-to-one correspondence connection of FMU models in various fields through a unified FMI model interface, realizing the unified tracked vehicle joint simulation model. Integrated integration; secondly, the data interface interaction unit also includes the interface interaction between the tracked vehicle joint simulation model and the optimization design unit. The parallel optimization design unit divides the optimization design problem of the tracked vehicle into a two-layer optimization strategy of a system level and multiple domain subsystem levels, and determines the coupling design variables of each subsystem at the same time. At the system-level optimization layer, the conflicts of the coupling variables of each subsystem are coordinated by building consistency constraints of the coupling variables, so that the domain subsystems of the subsystem optimization layer can execute the optimization process completely independently. The optimization design unit then passes the optimization variables to the integrated joint simulation model through interface interaction and dynamically changes the optimization variables to find the optimal value. The FMI-based integrated optimization of tracked vehicles refers to: Step S1, using different simulation tools to establish simulation models of tracked vehicle multi-body dynamics, control system, hydraulic transmission, three-dimensional structure, and finite element analysis; Step S2: Based on the unified FMI model interface, each domain model is converted and encapsulated into a corresponding FMU model; Step S3: extracting the interface data attribute information between FMU modules to form a model interface module, completing the integrated integration of the tracked vehicle model, and at the same time, completing the tracked vehicle optimization design process according to the specific requirements of the tracked vehicle optimization design, setting the optimization objectives and optimization variables in each field, forming the optimization interface module, and forming the final data interface interaction unit; Step S4: Decompose the tracked vehicle design optimization process into two levels: system-level optimization and multi-domain subsystem-level optimization. Determine the optimization objectives, optimization variables, and coupling variables of the system-level and multi-domain subsystem optimization processes, respectively. Simultaneously, establish consistency constraints for the coupling variables in each domain, achieve parallel optimization of the multi-domain subsystems of the tracked vehicle, and establish an optimization design unit for the tracked vehicle. Step S5: Run the connected tracked vehicle integrated optimization design system, determine the simulation step size, synchronize the clocks between multiple models, and complete the initialization of the integrated optimization system; Step S6, using the tracked vehicle integrated optimization design system to perform cyclic iterative optimization on the tracked vehicle, and judging whether the optimization result meets the optimization target. If not, continue iterating until the optimization result meets the optimization target. The system-level optimization problem of the optimization method is: , , where: Y is the given subsystem optimization state variable, {X, Z} is the system-level optimization variable, Z is the coupling variable of the corresponding subsystem and , J() is the consistency constraint at the system level to eliminate the conflict caused by the coupling variables in the separate optimization process of each sub-domain system; The subsystem-level optimization problem is: , , where: {X, Z} are variables for given system-level optimization and , in the subsystem optimization level, multiple domain subsystems are optimized separately. The optimization method is: while keeping the global variables and coupling variables {X, Z} constant, find the optimal value of the subsystem under given constraints. Subsystem optimization variables , and pass it back to the system-level optimization level; The tracked vehicle simulation model unit includes: a tracked vehicle multi-body dynamics model, a control system model, a three-dimensional structure model, and a domain subsystem model established on the finite element analysis model; The FMU model conversion unit includes: a multi-body dynamics FMU model conversion module, a control system FMU model conversion module, a three-dimensional structure FMU model conversion module and a finite element analysis FMU model conversion module, wherein: the multi-body dynamics FMU model conversion module converts the established multi-body dynamics model information based on a unified FMI interface and converts it into a multi-body dynamics FMU model; the control system FMU model conversion module converts it into a control system FMU model based on the established control system model information; the three-dimensional structure FMU model conversion module converts it into a three-dimensional structure FMU model based on the established three-dimensional structure model information; and the finite element analysis FMU model conversion module converts it into a finite element analysis FMU model based on the established finite element analysis model information; The data interface interaction unit includes: an FMU model interface module and an optimization interface module, wherein: the FMU model interface module obtains the interface interaction relationship between the FMU models according to the parameter transmission information between the various FMU models, thereby realizing the integrated integration of the tracked vehicle joint simulation model; the optimization interface module obtains the optimization interface interaction relationship between the optimization design unit and the tracked vehicle joint simulation model according to the optimization target and optimization variables of the tracked vehicle; The parallel optimization design unit includes: a system-level optimization module and a subsystem-level optimization module, wherein: the system-level module constructs consistency constraints on coupling variables to coordinate the coupling variable conflicts of each subsystem, so that the subsystems in each field can be optimized independently; the subsystem optimization layer independently constructs the optimization problems of the subsystems in each field to realize parallel optimization of the subsystems in each field.

Images