A modelica-based real-time calibration and semi-physical simulation test method for loader model

Abstract

The application discloses a loader model real-time calibration and semi-physical simulation test method based on Modelica, which comprises the following steps: firstly, a loader system is decomposed to obtain a system model, a subsystem model, a component model and an element model, and a loader system model is established; then, external function C code is encapsulated and called by using Modelica language, and a data communication interface model is established; subsequently, connection and real-time communication between a loader physical system and a simulation model are established based on calibration parameters and simulation data; finally, real-time calibration and semi-physical simulation test are carried out on a loader simulation model under the condition of actual loading operation of the loader under multi-working condition variable load. The loader simulation model constructed based on Modelica language has the characteristics of hierarchy, reusability and extensibility, and greatly improves the modeling efficiency. The model after real-time calibration has higher reliability, the test result of the semi-physical simulation is closer to the actual situation, and the performance analysis and test of the loader can be more effectively realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of loader system modeling simulation, and particularly relates to a Modelica-based loader model real-time calibration and semi-physical simulation test method. BACKGROUND

[0002] Loaders are mainly applied to earthwork construction of railways, buildings, mines and other projects, that is, they can shovel and load soil, sand, lime, coal and other bulk materials, and can also moderately shovel and excavate stone mines, hard soil and other materials, and thus play an important role in engineering machinery. It is particularly important to perform system modeling and simulation on loaders to realize product performance analysis and testing. Loaders are usually complex systems composed of subsystems in different fields such as machinery, electronics, hydraulics and control. Existing modeling and simulation methods usually use different software platforms to model and simulate specific fields. For example, three-dimensional simulation software Pro / E or Solidworks is used to establish a loader mechanism entity model, which is imported into ADAMS to perform dynamics and kinematics simulation of the mechanism, and to study the load characteristics of the loader arm, bucket and other components. AMESim simulation software is used to build hydraulic pump, hydraulic valve and hydraulic actuator models and system models to study the system response under different model parameters. MATLAB / Simulink is used to physically connect the loader power system module, hydraulic pump module and load module to establish a model, and to use an optimal control algorithm to realize energy-saving control system simulation. However, when using complex system modeling and simulation technology to design and analyze the whole loader, the subsystems in different fields exhibit tight coupling characteristics, and higher requirements are put forward for multi-field unified modeling and consistent simulation. In this case, traditional single-field simulation analysis tools cannot perform the task of overall performance analysis of such complex systems, and simple integration of these single-field simulation tools cannot fundamentally meet the requirements of multi-field coupling and continuous-discrete hybrid system design and analysis. Modelica, as a multi-field unified modeling language for object-oriented non-causal relationships, describes systems in different fields in a unified way, realizes seamless integration and data exchange between models in different fields, is suitable for modeling large-scale complex heterogeneous physical systems, and greatly facilitates loader system modeling and simulation. SUMMARY

[0003] The Modelica-based loader model real-time calibration and semi-physical simulation test method aims at overcoming the deficiencies of the prior art and providing a Modelica-based loader model real-time calibration and semi-physical simulation test method. By adopting the modeling mode of the multi-field unified modeling Modelica language, the loader simulation model has the characteristics of hierarchy, reusability and scalability, and the modeling efficiency is greatly improved. By establishing a data communication interface model, the loader physical system is connected with the loader simulation model, the model is calibrated in real time, and a semi-physical simulation system is further constructed. The model calibrated in real time has higher reliability, and since the physical object is connected in the simulation loop, the test results of the semi-physical simulation are closer to the actual situation than the offline digital simulation, and the performance analysis and test of the loader can be more effectively realized.

[0004] The Modelica-based loader model real-time calibration and semi-physical simulation test method adopts the following technical solutions:

[0005] A Modelica-based loader model real-time calibration and semi-physical simulation test method comprises the following steps:

[0006] S1: The loader is decomposed into a system to obtain a model library architecture, wherein the model library architecture comprises a system model, a subsystem model, a component model and an element model from top to bottom;

[0007] S2: The element is mathematically modeled and encapsulated to obtain an element model library by using the Modelica language according to the model library architecture of the loader system, the element model library is integrated and extended to encapsulate a component model library with specific functions, the element model library and the component model library form a subsystem model, the subsystem model transmits data through a coupling interface, and the loader system model is established by using the drag-and-drop and wiring mode;

[0008] S3: The Modelica language is used for encapsulation and calling of external function C code to establish a data communication interface model; the calibration parameters and simulation data of the loader system model are analyzed, the corresponding data values in the calibration parameter and simulation data transmission are analyzed, and the mapping relationship between the data values and the loader system model is established, so as to establish a loader simulation model; and the constructed loader simulation model and the running environment are arranged in an industrial PC;

[0009] S4: The calibration parameters and simulation data are obtained by externally connecting sensors on the loader physical system under multiple working conditions and variable loads, and by obtaining the calibration parameters and simulation data through LIN bus communication and CAN bus communication; the calibration parameters and simulation data obtained on the loader physical system are connected with the loader simulation model through a LIN communication adapter and a CAN communication adapter and are communicated in real time;

[0010] S5: the loader simulation model is calibrated in real time and tested in semi-physical simulation; for the model calibration stage, the calibration parameters obtained from the loader physical system are used for real-time calibration of the loader simulation model, and the loader simulation model calibrated in real time is used for semi-physical simulation test; for the simulation test stage, the simulation data obtained from the loader physical system are used as control signals to drive the loader simulation model to realize simulation and calculation of the model, and output corresponding simulation results; the simulation and calculation results are sent to the loader physical system as feedback signals to realize real-time communication and semi-physical joint simulation between the loader physical system and the loader simulation model.

[0011] Specifically, the subsystem model in step S1 includes but is not limited to: power system model, running system model, steering system model, main valve system model, actuator system model, hydraulic auxiliary system model, mechanical system model and control system model; the component model includes but is not limited to: engine model, torque converter model, clutch model, gearbox model, drive axle model, reducer model, tire model, steering model, various hydraulic valve models, hydraulic cylinder model, hydraulic pipe model, filter model, energy storage model, boom model, bucket model, vehicle body base model, control handle model and PID control model.

[0012] Specifically, the encapsulation and calling mode of the external function C code in step S3 using the Modelica language includes but is not limited to:

[0013] Use C files, and use Include annotation in annotation to include C files implementing the called functions

[0014] Use link library files, and use Library annotation in annotation to specify link libraries, so as to call functions in the specified libraries;

[0015] No external files, and directly embed C code in annotation using Include annotation.

[0016] Specifically, the external function in step S3 refers to a related function for realizing reliable operation of the communication interface model and data communication, including but not limited to: opening device function, reading device information function, sending and receiving data frame function, initializing and starting channel function.

[0017] Specifically, in step S3, a mapping relationship between data values and the loader system model is established, and the mapping relationship is specifically:

[0018] According to the communication protocol, each ID value in the frame data is mapped to the corresponding parameter interface of the loader simulation model, so as to realize data communication and interaction between the loader physical system and the loader simulation model.

[0019] Specifically, the running environment in step S3 includes but is not limited to MWorks, Dymola and OpenModelica.

[0020] Specifically, the calibration parameters in step S4 include but are not limited to engine external characteristic parameters, torque converter characteristic parameters, hydraulic pump characteristic parameters, valve element characteristic parameters and actuator characteristic parameters.

[0021] Specifically, the simulation data in step S4 include but are not limited to:

[0022] Control signals: engine throttle opening degree signal and electromagnetic valve control signal;

[0023] Feedback signals: working device handle control signal, boom cylinder signal, bucket cylinder signal, working pump signal, steering pump signal, engine signal and gearbox signal.

[0024] Specifically, the multi-working condition variable load in step S4 refers to excavating work materials with different compactness and particle size.

[0025] From the above description of the present application, compared with the prior art, the present application has the following beneficial effects:

[0026] The loader model real-time calibration and semi-physical simulation test method based on Modelica provided by the present application has the following beneficial effects: BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 The figure is a decomposition block diagram of the loader system of the present application.

[0028] Figure 2 The figure is an external function called and encapsulated based on Modelica of the present application.

[0029] Figure 3 The figure is a sensor arrangement schematic diagram of the loader physical system of the present application.

[0030] Figure 4The schematic diagram of real-time calibration and semi-physical simulation test of the loader model based on Modelica of the application.

[0031] The application will be further described below in combination with the drawings and specific embodiments. DETAILED DESCRIPTION

[0032] The application will be further described below in combination with the drawings and specific embodiments.

[0033] The application provides a kind of real-time calibration and semi-physical simulation test method of loader model based on Modelica.The modeling mode of the multi-field unified modeling Modelica language is used, and the simulation model of the loader constructed has hierarchical, reusable and extensibility, which greatly improves the modeling efficiency.The data communication interface model is established, the physical system of the loader is connected with the simulation model of the loader, the model constructed is calibrated in real time, and a semi-physical simulation system is further constructed.The model calibrated in real time has higher reliability, and since the real object is connected in the simulation loop, the test result of the semi-physical simulation is closer to the actual situation than offline digital simulation, and the performance analysis and test of the loader can be more effectively realized.

[0034] In the process of modeling and simulation of complex system, the accurate calibration of model parameters directly affects the reliability of the constructed model, and the model parameters used in the existing modeling method are usually obtained by bench test and manually calibrated.However, the system of the loader in actual operation process is a dynamic change process with multiple working conditions, variable load and strong coupling, and the offline digital simulation model constructed by bench test and manual calibration is still static, which cannot fully reflect the actual operating characteristics of the system, such as the fuel consumption characteristics of the engine, the speed and torque characteristics of the transmission system, the pressure and flow characteristics of the hydraulic system, and the load characteristics of the working device, etc.The performance of the system is related to the performance of the components and the control strategy of the whole vehicle, and is closely related to the dynamic and static characteristics of the controlled object under different working conditions and loads.Semi-physical simulation is to replace part of the actual controlled object or system component with a high-speed real-time simulation model, and the control system uses the actual physical system, so the real-time calibration and semi-physical simulation of the loader model realized by Modelica provides a new technical method to solve the above problems.

[0035] The application provides a kind of real-time calibration and semi-physical simulation test method of loader model based on Modelica, and the test method comprises the following steps:

[0036] 1) the system of the loader is decomposed to obtain a model library architecture, as shown in the figure, Figure 1 The model library architecture comprises system model, subsystem model, component model and element model from top to bottom.

[0037] Specifically, the subsystem model includes a power system model, a driving system model, a steering system model, a master valve system model, an actuator system model, a hydraulic auxiliary system model, a mechanical system model, and a control system model.

[0038] Specifically, the component model includes an engine model, a torque converter model, a clutch model, a gearbox model, a drive axle model, a reducer model, a tire model, a steering model, various hydraulic valve models, a hydraulic cylinder model, a hydraulic pipe model, a filter model, an accumulator model, a boom model, a bucket model, a vehicle body base model, a control handle model, and a PID control model.

[0039] 2) The elements involved in the loader system are mathematically modeled using the Modelica language to obtain an element model library. The element model library is integrated and expanded to form a component model library with specific functions. The element model library and the component model library form a subsystem model library. The subsystem model transmits data through a coupling interface. The obtained subsystem model is established by dragging and wiring, thereby establishing a loader system model.

[0040] Specifically, the element model refers to the mathematical modeling of key elements involved in the modeling of the loader system. The construction method includes: constructing the element model based on model principles and basic theorems through mathematical equations, constructing the element model based on physical data, equipment data, or constructing the element model based on empirical formulas.

[0041] Specifically, the element model is encapsulated as a component model by programming with the Modelica language according to the physical principles of the component. The component model can transmit data and can complete specific functions according to the real components of the loader.

[0042] Specifically, the coupling interface includes a mechanical interface, a hydraulic interface, and a control signal interface. Since the Modelica language has a non-causal modeling feature, the data flow in the interface is bidirectional, allowing mutual influence and matching between upstream and downstream models.

[0043] 3) As shown in Figure 2 the Modelica language is used to encapsulate and call external function C code to establish a data communication interface model. According to the calibration parameters and simulation data requirements of the loader system model, the corresponding data values in data transmission are analyzed, and the mapping relationship between the data values and the loader system model is established, thereby establishing a loader simulation model. The constructed loader simulation model and the operating environment are arranged in an industrial PC.

[0044] Specifically, the encapsulation and calling mode of the external function C code mainly contains three kinds: 1) using C files, such as.c,.h files, including the C file of the called function implementation in the annotation with Include annotation; 2) using link library files, such as.lib,.dll files, specifying the link library in the annotation with Library annotation, so as to call the function in the specified library; 3) without external files, directly embedding C code in the annotation with Include annotation.

[0045] Specifically, the external function refers to the related function for realizing reliable operation of the communication interface model and data communication, such as opening device function, reading device information function, sending and receiving data frame function, initializing and starting channel function, etc. And calling in the algorithm of the loader simulation model.

[0046] Specifically, the mapping relationship refers to parsing each ID value in the frame data according to the actual communication protocol, and mapping the corresponding physical information to the corresponding parameter interface of the loader simulation model, so as to realize the data communication and interaction between the loader physical system and the loader simulation model. The data value refers to the calibration parameter and simulation data of the loader simulation model in the actual operation and working process.

[0047] Specifically, the communication protocol includes LIN communication protocol and CAN communication protocol.

[0048] Specifically, the running environment refers to the software platform that ensures the stable operation of the Modelica model, such as MWorks, Dymola, OpenModelica, etc.

[0049] 4) as shown in Figure 3 The calibration parameter and simulation data source contains two parts: external sensors on the loader physical system and obtained through LIN bus communication method, and CAN bus data of the loader physical system obtained through CAN bus communication method. The calibration parameter and simulation data are connected and communicated in real time with the loader simulation model in the industrial PC through the LIN communication adapter and the CAN communication adapter.

[0050] Specifically, corresponding sensors are arranged on the loader physical system, and data transmission is carried out through the LIN communication method. The CAN bus data is directly obtained from the CAN bus of the loader physical system.

[0051] Specifically, the calibration parameter and simulation data are connected and communicated in real time with the loader simulation model in the industrial PC through the LIN communication adapter and the CAN communication adapter.

[0052] Specifically, the calibration parameters mainly include: engine external characteristic parameters, hydraulic torque converter characteristic parameters, hydraulic pump characteristic parameters, valve component characteristic parameters, and actuator characteristic parameters. Their corresponding data values ​​are shown in Table 1.

[0053] 4.2) The simulation data mainly includes: ① Control signals: engine throttle opening signal and solenoid valve control signal, etc. ② Feedback signals: working device operating handle control signal, boom cylinder signal, bucket cylinder signal, working pump signal, steering pump signal, engine signal, and gearbox signal, etc. Their corresponding data values ​​are shown in Table 1.

[0054] Table 1. Communication data between real-time calibration and hardware-in-the-loop simulation of the loader model based on Modelica.

[0055]

[0056]

[0057] 5) Conduct actual loading operations of the loader under multiple working conditions and varying loads, and perform real-time calibration and hardware-in-the-loop simulation tests on the loader simulation model, such as... Figure 4 As shown. During the model calibration phase, the received and parsed model parameter values ​​are used for real-time calibration of the loader simulation model, and the real-time calibrated model is then used for hardware-in-the-loop simulation testing. During the simulation testing phase, the received and parsed simulation data values ​​are used as control signals to drive the loader simulation model to perform simulation and calculations, and output the corresponding simulation results. The simulation and calculation results are then sent as feedback signals to the loader physical system, achieving real-time communication and hardware-in-the-loop co-simulation between the loader physical system and the loader simulation model.

[0058] Specifically, the multi-condition variable load refers to the large variation in the load of the loader when shoveling materials of different densities and particle sizes. In actual operation, four typical working conditions, namely shoveling earth, gravel, sand, and cinder, are selected for shoveling operations and semi-physical simulation tests are conducted.

[0059] The above are merely specific embodiments of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial modifications made to the present invention using this concept shall be considered as infringing upon the protection scope of the present invention.

Claims

1. A method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica, characterized in that, Includes the following steps: S1: Decompose the loader into a system architecture to obtain a model library architecture, which includes system model, subsystem model, component model and element model from top to bottom. S2: Using the Modelica language, the components of the loader system are mathematically modeled and encapsulated according to the model library architecture to obtain a component model library. The component model library is then integrated and extended to encapsulate a part model library with specific functions. The component model library and the part model library form a subsystem model. The subsystem model transmits data through a coupling interface and establishes the loader system model through drag-and-drop and connection methods. S3: The Modelica language is used to encapsulate and call external function C code, and a data communication interface model is established; based on the calibration parameters and simulation data of the loader system model, the corresponding data values ​​in the calibration parameters and simulation data transmission are analyzed, and the mapping relationship between the data values ​​and the loader system model is established, thereby establishing the loader simulation model; and the constructed loader simulation model and running environment are deployed on an industrial PC. S4: Obtain calibration parameters and simulation data by connecting external sensors to the loader physical system under multiple working conditions and varying loads, and obtaining calibration parameters and simulation data through LIN bus communication and CAN bus communication; connect and communicate in real time with the loader simulation model through LIN communication adapter and CAN communication adapter using the calibration parameters and simulation data obtained from the loader physical system. S5: The loader simulation model undergoes real-time calibration and hardware-in-the-loop (HIL) simulation testing. During the model calibration phase, calibration parameters obtained from the loader physical system are used for real-time calibration of the loader simulation model, and the calibrated loader simulation model is then used for HIL simulation testing. During the simulation testing phase, simulation data obtained from the loader physical system is used as control signals to drive the loader simulation model to perform simulation and calculation, and outputs corresponding simulation results. The simulation and calculation results are then sent as feedback signals to the loader physical system, enabling real-time communication and HIL co-simulation between the loader physical system and the loader simulation model. The calibration parameters mentioned in step S4 include: engine external characteristic parameters, hydraulic torque converter characteristic parameters, hydraulic pump characteristic parameters, valve component characteristic parameters, and actuator characteristic parameters; The simulation data mentioned in step S4 includes: Control signals: engine throttle opening signal and solenoid valve control signal; Feedback signals: control signals from the working device operating handle, boom cylinder, bucket cylinder, working pump, steering pump, engine, and gearbox.

2. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, The subsystem models mentioned in step S1 include: power system model, driving system model, steering system model, main valve system model, actuator system model, hydraulic auxiliary system model, mechanical system model, and control system model; the component models include: engine model, hydraulic torque converter model, clutch model, gearbox model, drive axle model, reducer model, tire model, steering gear model, various hydraulic valve models, hydraulic cylinder models, hydraulic pipe models, filter models, energy accumulator models, boom model, bucket model, vehicle body base model, control handle model, and PID control model.

3. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, Step S3 involves using Modelica to encapsulate and call external function C code, including: Use a C file, and include the C file implementing the called function in the annotation using the Include annotation; Using a linked library file, specify the linked library in the annotation with the Library annotation, thereby calling the functions in the specified library; No external files are needed; C code can be directly embedded in the annotation using the Include annotation.

4. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, The external functions mentioned in step S3 refer to the functions related to ensuring reliable operation of the communication interface model and data communication, including: device start function, device information read function, data frame send and receive function, and channel initialization and start function.

5. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, In step S3, a mapping relationship is established between the data values ​​and the loader system model. Specifically, the mapping relationship is as follows: Each ID value in the frame data is parsed according to the communication protocol, and the corresponding physical information is mapped to the corresponding parameter interface of the loader simulation model, thereby realizing data communication and interaction between the loader physical system and the loader simulation model.

6. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, The operating environment mentioned in step S3 includes MWorks, Dymola, and OpenModelica.

7. The method for real-time calibration and hardware-in-the-loop simulation testing of a loader model based on Modelica according to claim 1, characterized in that, The multi-condition variable load mentioned in step S4 refers to shoveling work materials with different densities and particle sizes.

Images