Design method of high-precision multi-channel aircraft structural load measurement system

Abstract: Aircraft is a device that flies in the atmosphere or outside the atmosphere, including aircraft, spacecraft, rockets and missiles. During the maneuvering process of the aircraft, the section load of the aircraft structure is an important basis for the structural design, and the accuracy of the measurement results of the section load is directly related to the reliability of the structural design. Aiming at the problem that the internal stress of the missile structure is difficult to measure during flight, a method of using the load tester to measure the strain load and LS-DYNA software combined with simulation analysis is proposed. Verify and confirm the layout of measuring points for each strip analysis component, and design a load tester for high-precision load measurement. The tester is composed of an 8-channel strain measurement module and a 6-channel temperature measurement module. The system is jointly controlled by a single-chip microcomputer and FPCA. Through the strain signal The conditioning circuit and the temperature signal conditioning circuit improve the linearity of the signal, use Flash to store a large amount of data, and upload the data to the host computer through RS422 to USB. Through the above work, the high-precision acquisition scheme of the load tester is verified. The results show that the strain data and temperature data collected by the load tester are highly linear, which has a certain reference value for the missile body structure design.
Keywords: finite element simulation: high-precision acquisition; multi-channel measurement: projectile structure load design

0. Introduction
        At present, the identification of flight load is a very complicated research, which not only involves the modeling of missile structure, the measurement of load, but also includes the more difficult problem of load inversion. Including the selection of inversion algorithm, the selection and quantity matching of sensors, the layout of points, the coupling relationship between winter loads, the calibration algorithm of load data, and the influence of sensor layout position errors, etc. The missile structure is used as the head-carrying and protecting part, and it is subjected to severe and complex load conditions during the air movement. In order to optimize the dynamic and static characteristics of the projectile structure and optimize the projectile structure, it is necessary to obtain the load distribution of the projectile structure when it moves in the air. At present, the main way to obtain the flight load of the missile is to use simulation calculation to calculate the load of each section as the basis of structure design and strength check. The load is reasonable in theory. However, the real load data size of the flight state needs further verification, and the direct measurement of the load during the flight is the most direct and effective method. This article discusses the design scheme of a high-precision load tester. Through the finite element simulation verification of the stress performance of a single strip component, the layout of the measuring points of each strip component is confirmed, and the Wheatstone bridge full bridge composed of strain gauges is used. The strain data is collected in the form of a 4-wire Pt100 platinum resistance temperature sensor to measure the small strain on the aircraft structure. The platinum resistance is a precise component with a large temperature measurement range [2]. Most of the traditional multi-channel acquisition systems only collect one physical quantity, and cannot collect and analyze multiple sensor data at the same time, and cannot store them for a long time. In order to solve this problem, the signal conditioning circuit and the data storage circuit. For high-precision acquisition requirements, the traditional bridge sub-balance method of manually adjusting the potentiometer is abandoned [3], and the self-adjustment of the bridge balance is realized through the control of the single-chip microcomputer [4], which is more in line with the application scene on the bullet: platinum resistance temperature sensor The 4-wire system is used for measurement, and the power supply through the current source can effectively eliminate the influence of lead resistance: adding a filter circuit to further improve the accuracy of data acquisition. Through the above work, the high-precision acquisition scheme of the load tester is verified, which has certain reference value for the missile body structure design.

1. The overall design scheme of the tester
        The high-precision load tester is mainly composed of a power management module, a main control module, a signal acquisition module, a signal conditioning module, a data storage module, a power supply excitation module and a communication module. Among them, the signal acquisition module is responsible for collecting the analog signal input by the sensor, and the signal conditioning module is responsible for a series of signal processing, including DC bias circuit, signal attenuation circuit, differential operation amplifier circuit, bridge Pingde self-regulation circuit, signal gain circuit and controllable filter circuits, etc.; the power supply excitation module is used to generate corresponding excitation signals for the tester, and the power supply excitation module includes three-channel output signals to provide excitation requirements for the tester respectively. The combination of single-chip microcomputer and FPGA forms the main control module of the system, which is connected with the PC terminal according to the requirements of the tester. Since the strain signal is a transient signal and the temperature signal is a slow-changing signal[5], a reliable strain signal conditioning module and a temperature signal conditioning module are designed respectively for the different signal characteristics of the strain signal and the temperature signal. After signal conditioning, it enters the high-precision A/D conversion module, and the quantized code is stored in the data storage module through FPGA control, and the RS422 communication module is controlled by the single-chip microcomputer to upload to the host computer [5]. Communication to realize erasing and reading and writing of the data storage module. The overall block diagram of the system is shown in Figure 1.

2. Finite element simulation verification
The force condition of a single label component is analyzed by finite element method to determine that in the actual test, the layout of the strain-sensitive points of a single analysis bar is conducive to better acquisition of load data.

2.1 Analysis of model structure
Using CATIA modeling software, based on a certain type of missile structure prototype, a single poplar component model was established. The model structure diagram is shown in Figure 2. In order to facilitate the simulation calculation, the single bar model is designed as a scaled simplified model compared with the actual missile body structure, that is, the length and width are adjusted, while the fluid aerodynamic shape of the missile body remains unchanged.

 

 

 According to the analysis situation, the middle/side elevation of the poplar is selected for the position of the measuring point of a single poplar, as shown in Figure 7, and four patch locations are selected near the same measuring point to form the measuring points of the Wheatstone full bridge. Through simulation. This test point can truly reflect the section load situation.

3. Key technology research
3.1 Design of strain signal conditioning circuit

The strain signal conditioning circuit is composed of ADG507 analog switch

 

         AD8221 instrumentation amplifier, MAX7400 low-pass bridge circular filter AD7192 high-precision analog-to-digital converter, OPA140 precision operational amplifier and LT2602 digital-to-analog converter. The front end of the circuit is a Wheatstone bridge composed of 350 2 strain gauges, which is measured in the form of a full bridge. The bridge circuit uses the 5 V voltage output by the LDO power supply as the excitation for the bridge. The winter channel strain gauge signal is gated by the analog switch and enters the instrument amplifier for amplification. After the signal is filtered by the low-pass elliptic filter, it enters the high-precision A/D conversion module.
        The strain signals of 8 channels are fed into AD8221, an instrument amplifier with adjustable gain, through a piece of ADG507 analog switch, and the adjustment of the magnification can be realized by adjusting the gain resistor. Adopt MAX7400 to realize the function of 8th-order low-pass elliptic filter [6], calculate by formula (1) [7], adjust the switched capacitor to filter out the noise above the frequency of 100 Hz. Figure 8 shows the schematic diagram of the strain signal conditioning circuit.

        The strain gauge will generate an initial bias voltage when it is not deformed, which will cause the bridge to shift. In order to improve the measurement accuracy, a certain method must be used to drive the bridge to balance [8]. The traditional potentiometer method is difficult to adjust, and the external structure is easily damaged. In this paper, a single-chip microcomputer is used to control the DAC output, adjust the VREF bow pin of the instrumentation amplifier, and then eliminate the output DC bias of the instrumentation amplifier. In order to realize the function of adjusting the positive and negative offset of the bridge, a subtraction circuit is composed of a precision amplifier and 4 precision resistors. The bridge balance self-adjustment scheme is shown in Fig.9.
Set the DAC output voltage U, 0 ~ 4.096 V, the reference voltage source output voltage U. is 2.5 V[10], the above subtraction circuit converts the U output into a balanced voltage U. , assuming that the positive input voltage of the precision amplifier is U., and the negative input voltage is U_, according to the principle of "virtual short" and "virtual break" [11]:

 

 The schematic diagram of the bridge balance self-regulating circuit is shown in Figure 10. The voltage reference of LTC2602 is provided by REF3040, the reference voltage is 4.096 V, and the subtraction circuit is composed of OPA140 and 4 precision resistors to realize the bipolar output voltage of DAC.

 3.2 Design of storage module
        Due to the particularity of the measurement environment of this design, it is suitable to choose the non-volatile memory Fash as the storage unit [14]. This article selects 2 pieces of NAND type K9K8GO8 Flash to store strain and temperature data. Since the bad blocks must be judged first and then the data is stored, in order to improve the storage rate, the ping-pong storage structure is adopted, that is, the bad block detection operation is performed on Flash 2 when Flash 1 is storing data, so that the two Flash blocks are stored in bad blocks. The detection and data storage operations are alternated to realize continuous storage of signal data. Figure 11 shows the schematic diagram of the memory module circuit.

3.3 System software design scheme
        The software design of the high-precision load measurement system is based on the practicability of function realization, and at the same time, it is developed by taking into account the reliability design method. System software design includes single-chip programming and FPGA programming. The software design scheme is shown in Figure 12. 

 

        The MCU program includes: DAC driver module, RS422 driver module, SPI data communication module and built-in DAC current output, etc. Among them, the DAC driver module is responsible for driving the DAC and controlling the corresponding voltage value of the DAC output; the RS422 driver module is responsible for driving the RS422 and uploading the stored data To the host computer; data communication between the MCU and the FPGA through the SPI bus; the built-in DAC current output is responsible for providing power supply to the temperature sensor.
        The FPGA program includes: AD conversion module, storage control module and SPI data communication module. Among them, the AD conversion module controls the high-precision AD conversion chip to convert the voltage signal corresponding to the strain and temperature into a digital signal. And store the data in Fash by controlling the data storage module; communicate with the microcontroller through the SPI bus.
4. Design verification
For the design of the bridge sub-balance self-regulating circuit above, use Tina-TI to carry out modeling and simulation, as shown in Figure 13:

 

 

        It can be seen from Table 2 that the self-balancing circuit can eliminate positive and negative voltage values, has better circuit performance, and meets the design requirements. Use a standard signal source to apply a standard voltage signal as the input end of the strain conditioning circuit, use a standard resistance box as the input end of the temperature signal conditioning circuit, measure the linearity of the output signal of each channel after the signal conditioning circuit, and use it as a signal Validation basis for conditioning circuits. Figure 15 shows the linearity analysis of the strain signal conditioning circuit, and Figure 16 shows the linearity analysis of the temperature signal conditioning circuit.

         Table 3 is the comparison of standard strain and system measured strain. It can be seen from Table 3 that the highest strain measurement accuracy of the load measurement system is 0.33%. Table 4 is the comparison between the standard temperature and the system measured temperature. It can be seen from Table 4 that the temperature measurement accuracy of the load measurement system is +0.5°C. From the measured results, it can be concluded that both the strain signal and the temperature signal have high linearity. 

5. Conclusion
        A high-precision aircraft load measurement system is designed. Through the design of the strain and temperature signal conditioning circuit, the bridge sub-balance self-regulation circuit signal acquisition circuit, and the storage circuit, the high-precision acquisition and storage of strain and temperature signals is realized. Communicate with the host computer via RS422. The tester gives priority to sending real-time data, and the design of the storage circuit greatly ensures the continuous operation of the tester in harsh environments and avoids data loss due to harsh environments. The experimental results show that the accuracy of the strain data collected by the high-precision aircraft load measurement system can reach 0.3%, and the accuracy of the temperature data can reach +0.5C, which meets the expected indicators and has certain reference value for optimizing the structural design of the aircraft.