CN-121979020-A - Multi-level collaborative verification method, platform and storage medium for aircraft power generation system controller
Abstract
The invention relates to the field of aviation electrical engineering and real-time simulation test, and provides a multi-level collaborative verification method, a platform and a storage medium for a control strategy of an aircraft power generation system, wherein the method comprises an offline simulation stage for preliminarily designing and verifying the control strategy; the method comprises a real-time simulation stage, a rapid prototyping stage, a real-time simulation stage, a control strategy generation stage and a real-time simulation stage, wherein the real-time simulation stage performs closed-loop real-time verification and parameter optimization on the control strategy, the rapid prototyping stage disposes an optimized control strategy code to a rapid prototyping controller and is connected with real or high-precision simulated power generation system hardware to perform hardware-in-loop-real object hybrid verification, a unified performance reference and error feedback chain which penetrates through three stages of off-line simulation, real-time simulation and rapid prototyping is established, a verification result of each stage is used as an expected reference of the next stage, and error monitoring and backtracking correction are performed. The invention greatly improves the confidence coefficient and the primary success rate of the final physical verification through the three-layer progressive verification of the off-line-real-time-prototype and the model precision progressive calibration mechanism.
Inventors
- LV JIAYI
- WU LIN
- WEI ZEPING
- JI CHENLU
- Zou Fanshu
Assignees
- 中国商用飞机有限责任公司北京民用飞机技术研究中心
- 中国商用飞机有限责任公司
Dates
- Publication Date
- 20260505
- Application Date
- 20251224
Claims (10)
- 1. A multi-level collaborative verification method for an aircraft power generation system controller, comprising: Off-line simulation, namely establishing a high-precision off-line model based on physical parameters of an aircraft power generation system, and primarily designing and verifying a control strategy; Real-time simulation, namely carrying out real-time order reduction treatment on the offline model to obtain a real-time order reduction model, deploying the real-time order reduction model to a real-time simulator to form a hardware-in-loop simulation environment, and carrying out closed-loop real-time verification and parameter optimization on a control strategy; quick prototype verification, namely deploying the optimized control strategy code to a quick prototype controller, connecting the control strategy code with real or high-precision simulated power generation system hardware, and carrying out hardware-in-loop-physical hybrid verification; And simultaneously, establishing a unified performance benchmark and error feedback chain penetrating through the three stages of off-line simulation, real-time simulation and rapid prototype verification, taking the verification result of each stage as a reference benchmark of the next stage, and carrying out real-time error monitoring and backtracking correction.
- 2. The method of multi-level collaborative verification of an aircraft power generation system controller of claim 1, wherein the offline model is an aircraft three-level synchronous generator system model comprising: auxiliary exciter based on permanent magnet synchronous generator model; an excitation power supply with an asymmetric H-bridge topology is adopted; a main exciter and a main generator based on a rotary rectifier; The control strategy comprises a double closed-loop control structure for voltage regulation, wherein the outer ring is a generator output voltage ring, and the inner ring is an excitation current ring.
- 3. The method for multi-level collaborative verification of an aircraft power generation system controller according to claim 1, wherein during the real-time simulation, by performing a model accuracy progressive calibration on an offline model, it is ensured that a real-time reduced order model remains consistent with the offline model in terms of key dynamics, thereby ensuring reliability of the transfer of the unified performance benchmark between phases, the model accuracy progressive calibration specifically comprising: (1) Parameterized calibration, namely taking high-precision dynamic output data generated under standard test excitation in the off-line simulation as a reference standard, adopting a parameter identification or optimization algorithm to adjust one or more key parameters in the real-time reduced model, and minimizing the error between the output response of the real-time reduced model under the same excitation and the reference standard; (2) And in the parameter calibration or subsequent real-time simulation operation process, if the difference between the output response of the real-time reduced model and the reference standard exceeds a preset threshold value, triggering an error feedback chain, wherein the error feedback chain guides the correction of the parameters of the offline model or the current real-time reduced model or the parameters of the control strategy.
- 4. The method of multi-level collaborative verification of an aircraft power generation system controller according to claim 1, wherein the rapid prototyping further comprises: the self-adaptive interface simulation comprises the steps of automatically identifying or configuring a physical interface protocol between a rapid prototyping controller and real power generation system hardware, wherein the interface protocol comprises at least one of CAN, ARINC429 and discrete quantity I/O; The self-adaptive interface simulation supports on-line modification of interface parameters, and does not affect the running of deployed control strategy codes.
- 5. The method of claim 1, wherein the method comprises the steps of load disturbance and evaluation during the real-time simulation or rapid prototyping, specifically: Selecting or programming a disturbance sequence from a predefined load disturbance scene library, wherein the scene library comprises load step mutation, nonlinear load access, short circuit fault and parallel switching; And loading the disturbance sequence into the offline model or the real-time reduced order model, automatically executing a disturbance test, evaluating the robustness of a control strategy according to a preset performance index, and generating an evaluation report.
- 6. An aircraft power generation system control strategy verification platform for implementing the method of any one of claims 1-5, comprising: the upper computer subsystem is used for off-line modeling, simulation monitoring and code generation; the real-time simulation subsystem comprises a first target machine and a second target machine which are respectively used for running a controlled object model and a control strategy model; the rapid prototyping subsystem comprises a rapid prototyping controller and real power generation system hardware; And the collaborative verification management unit is used for managing the unified performance benchmark and the error feedback chain and coordinating the data transmission and verification flow among all subsystems.
- 7. The aircraft power generation system control strategy verification platform of claim 6, wherein the collaborative verification management unit comprises: The benchmark management module is used for defining and maintaining a key performance index reference benchmark of each verification stage; the error analysis module is used for calculating the deviation between the output of each stage and the corresponding reference standard in real time and judging whether backtracking correction is triggered or not; And the flow scheduling module is used for automatically scheduling re-execution or parameter adjustment of the verification stage according to the error analysis result.
- 8. The aircraft power generation system control strategy verification platform of claim 6, wherein the rapid prototyping subsystem comprises an adaptive interface simulation module for automatically identifying or configuring a physical interface protocol between a rapid prototyping controller and real power generation system hardware, wherein the adaptive interface simulation module is a pluggable hardware board card or a configurable FPGA logic unit.
- 9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any of claims 1-5.
- 10. An electronic device comprising a processor and a memory, the memory storing a computer program, wherein the processor implements the method of any of claims 1-5 when executing the program.
Description
Multi-level collaborative verification method, platform and storage medium for aircraft power generation system controller Technical Field The invention relates to the technical field of aviation electrical engineering and real-time simulation test, in particular to a multi-level collaborative verification method, a platform and a storage medium for an aircraft power generation system control strategy, which are suitable for rapid prototype development and high-confidence verification of an aircraft three-level brushless alternating current power generation system controller. Background Modern aircraft power generation systems are the core of the avionics system, the reliability of their control strategies being directly relevant to flight safety. Traditional control strategy development and verification relies mainly on pure digital simulation and final physical bench testing. The pure digital simulation cannot fully consider real factors such as instantaneity, hardware interfaces, electromagnetic compatibility and the like, has low confidence, and directly carries out physical bench test, which has high risk, huge cost and long period, and is difficult to cover all critical and fault working conditions. In the prior art, hardware-in-loop (HIL) simulation and Rapid Control Prototyping (RCP) techniques have been introduced into the development flow of control systems. For example, some patents disclose aircraft system testing methods based on semi-physical simulation. However, most of these schemes focus on a single simulation level (e.g., pure HIL or pure RCP), or simple two-stage integration, lacking a full-flow seamless-join verification architecture from offline design to physical docking, with closed-loop quality control. The following problems are presented in detail: 1. The model consistency is poor, namely, the offline design model, the real-time simulation model and the real hardware behavior are cracked, the model precision is inconsistent in different stages, and a great deal of time is consumed for backtracking and positioning to the problems found in the later test. 2. The verification process is disjoint, that is, each verification stage is relatively independent, data and conclusions cannot be effectively transferred and mutually verified, and continuous and progressive evaluation on the performance of the control strategy is difficult to form. 3. The interface configuration is complex, when the RCP stage is in butt joint with the real hardware, complicated manual configuration and debugging are required to be carried out aiming at a specific interface protocol, and the adaptation efficiency is low. 4. And the test scene is on one side, and the robustness of the control strategy under complex and extreme load disturbance is not enough due to the lack of a systematic automatic test scene library. Therefore, an integrated verification method and platform capable of integrating offline design, real-time simulation and rapid prototyping and realizing model, data and flow collaboration are urgently needed, so as to improve the development efficiency, quality and safety of an aircraft power generation system controller. Disclosure of Invention Aiming at the defects of the prior art, the invention aims to provide a multi-level collaborative verification method, a platform and a storage medium of an aircraft power generation system control strategy, which are characterized in that a collaborative verification system which comprises three levels of off-line simulation, real-time simulation and rapid prototyping and has model precision progressive calibration, cross-level consistency reference and error feedback chains, self-adaptive interface simulation and intelligent load disturbance injection capability is constructed. The invention adopts the following technical scheme: in one aspect, the invention provides a multi-level collaborative verification method for an aircraft power generation system controller, comprising the steps of: Off-line simulation, namely establishing a high-precision off-line model based on physical parameters of an aircraft power generation system, and primarily designing and verifying a control strategy; Real-time simulation, namely carrying out real-time order reduction treatment on the offline model to obtain a real-time order reduction model, deploying the real-time order reduction model to a real-time simulator to form a hardware-in-loop simulation environment, and carrying out closed-loop real-time verification and parameter optimization on a control strategy; quick prototype verification, namely deploying the optimized control strategy code to a quick prototype controller, connecting the control strategy code with real or high-precision simulated power generation system hardware, and carrying out hardware-in-loop-physical hybrid verification; and simultaneously, establishing a unified performance benchmark and error feedback chain penetrating through the offline simulat