Design of dynamic dip measurement system in rotating state

Design of dynamic dip measurement system in rotating state

[Date:]

Source: China Power Grid Author: Wang Chao, Gao Guowei, Pan Xue, Zhang Jun, Zhao Yu

[Font:Large Medium Small]

　

 

0

The dip measuring module based on MEMS accelerometer has the advantages of small size, light weight, low cost, impact resistance, and high reliability. There is a large error in Inclination Measurement with acceleration interference. This paper studies the sources of this error and the methods for removing this error.

The tilt angle measurement in the rotating state is a kind of dynamic Tilt Angle Measurement with acceleration interference. In the rotating reference system, an object with a weight of m is pulled by a link pointing to the center of rotation, but it is still at rest relative to the reference system, with no acceleration and does not comply with Newton's first law. Therefore, the reference system for constant rotation relative to the inertial system is also a non-inertial system. In this reference system, the form of Newton's second law remains unchanged. When the particle is still in this reference system, centrifugal inertia force should be introduced, which is perpendicular to the rotation axis. In this state, the acceleration that affects the correct output of the Monitored object's skew angle is caused by this force. In order to offset the error caused by acceleration in the rotating state, this paper introduces five different implementation schemes and compares their respective advantages and disadvantages.

1. Construction of dynamic dip measurement system

1.1 single-axis/Dual-axis accelerometer and DC motor Solutions

1. V1 is a dual-axis accelerometer, and V2 is a single-axis accelerometer. V1 is the load of the DC motor, and the output value of AY is always infinitely close to-1.0gn through the control algorithm (in the same direction as the gravity acceleration). At this time, the output of the AX axis perpendicular to AY is just centripetal acceleration-. The output value of V2 is

The biggest advantage of this method is that it is simple in theory, easy to understand, and less demanding on data processing algorithms. However, this method will increase the power consumption because it is driven by a motor. The complexity of system design and debugging and the hardware development cost will be increased due to the need to write control algorithms or use special control units, it is difficult to meet real-time requirements, and the overall feasibility is not strong.

1.2 2 single-axis accelerometer Solution

2. The distance between V1 and the center of rotation is R, the distance between V2 and the center of rotation is R + r, and the sensitivity axis of V1 and V2 is in the same line. Static Measurement
AX = gsin θ

When the object to be tested rotates around the Y axis, the central acceleration is generated on the horizontal plane with the angle θ of the object to be tested. The direction is to the Y axis and the size is a = ω 2T. At this time, V1, the Radial Acceleration of the two accelerometer V2 is

The high g value (dozens of gn) accelerometer is not suitable for detecting the tilt angle, but the low g value (less than 0.5gn) when the rotation angular velocity increases, the output of the accelerometer will quickly reach the nonlinear range or even exceed the range. Therefore, select a low-g acceleration tilt sensor, such as a single axis acceleration tilt sensor produced by VTI.

In the above scheme, when θ = 0 ° and rotate with the highest angular velocity, V2 away from the rotation axis first exceeds the range. Therefore, to ensure that the two sensors work within the range, V1, the absolute position of V2 relative to the rotating axis should be as small as possible, and the relative position r between V1 and V2 should be increased as much as possible within the permitted range, so that the measurement system can be fully sensitive to the difference between the output values of the two sensors due to the different radius of the circular motion.

Because the use of two accelerometer requires a certain distance, the disadvantage of this solution is not conducive to the miniaturization of hardware.

1.3 dual-axis accelerometer Solution

3. During static measurement

This solution uses one dual-axis accelerometer. Compared with the two single-axis accelerometer solutions, it overcomes the disadvantages of making devices less compact; however, the disadvantage of this solution is that it may make the two axes of the accelerometer work in different regions with different output characteristics (for example, the X axis works in the range with high linearity and resolution, however, if the Y axis runs in a range with relatively low linearity and resolution, the system will be in the range (-15 ° <θ <+ 15 °, ω ≤ 100 °/s) the internal precision is reduced, and the overall stability is not high.

1.4 dual-axis accelerometer solution that can reduce zero-Bit Error

Install 1 dual-axis accelerometer 4. The angle between the AX axis and the object to be tested is Beta. Specifically, the specific value of β is related to the output of the two sensitive axes When the tested object is absolute zero, that is

This solution is a generalization of the dual-axis accelerometer solution. While ensuring the miniaturization of hardware, the acceleration module and the installation angle of the tested object are adjusted, the two axes of the accelerometer can work within the same output characteristic range as much as possible, which is helpful for improving the stability of the measurement system, especially in reducing the zero-bit error.

This method can be suitable for applications in a type of environment, but if a higher requirement is put forward to the sensor (for example, the angle measurement range is increased, and the object rotation speed is faster ), it is still possible to make the working range of the two axes of the sensor unsatisfactory, even if the installation angle is adjusted at this time, the output characteristics of the two axes can be similar, however, it may not be guaranteed that the two axes work in the interval with good linearity and high resolution.

1.5 dual-axis accelerometer solution with any angle can be set. This solution is a more general improvement of the dual-axis accelerometer solution. For specific applications, you can select two single-axis accelerometer, install the two in the same position, the angle between the two and the object to be tested, or even the range of the two accelerometer, can be freely set according to this situation, as long as the calculation formula calculated based on the physical ry relationship satisfies the single-value correspondence (each combination of the output values of the two accelerometer sensitive axes in the system corresponds to a unique angle value) you can. Usually select the region with the best linearity and the region with the most sensitive output.
This solution is flexible, but must accurately establish the physical model of the object to be tested.

2 System Structure Design

Based on the above five ideas, a tilt angle measurement system in the rotating state is built. The system structure is shown in Figure 5.

Sensor_A and sensor_ B indicate the two components contained in the measurement system (for the two single-axis accelerometer schemes, 2 single-axis accelerometer schemes; for the double-axis accelerometer scheme, for the Dual Axis accelerometer two vertical sensitive shaft), here, respectively selected VTI SCA61T-FAHH1G (± 0. 5gn) and SCA100T-D02 (± 1. 0gn) two series of sensors. In the figure, to make the two pieces of data collected as values at the same time, the 16-bit high-precision synchronous acquisition analog-to-digital converter AD7654 is used, the A/D converter can collect data from the selected two channels at the same time, theoretically ensuring the concurrency of the two data involved in the calculation; considering that only the collected data must be filtered and sent to the upper computer for fitting and interpolation at present, the eight-bit single-chip microcomputer of the Reduced Instruction Set of Proteus can meet the requirements; the single-chip microcomputer communicates with the data tunnel of the host computer through serial port. For the two solutions, select R = 30 cm, r = 10 cm.

3. Data collection and Error Analysis

In the experiment, the Comprehensive Test Turntable of TZS-74IIA type gyro instrument was adopted, and the speed was 0 °/s ~ Under 100 °/s (interval of 10 °/s), the acquisition angle and the horizontal angle are-15 ° ~ + Angle output within the 15 ° range, such as table 1. (The data in the table is A/D converter output value, counts)

The preceding data take every 3 ° and every 5 ° as the known sample points, and use the interpolation function z = griddata (X0, Y0, Z0, x, y, and 'method' are used for interpolation. (X0, Y0, and Z0 are known sample sets. x and y are new interpolation points. Here, select a point other than a known sample point as a new interpolation point in the preceding table, and use z as the interpolation result. The methods include linear, cubic, nearest, and v4, and draw an error 3D surface. Figure 6 ~ In Figure 9, the X and Y axes indicate the number of digits output after the two components in the dip measurement system are converted to A/D. Unit: counts.

According to the test results, the error gradually decreases as the density of known interpolation sample points increases. At the same time, among the four interpolation methods provided by Matlab, li-near and v4 have the best effect, cubic followed by nearest, and the dual-axis accelerometer solution is far less effective than the two single-axis accelerometer solutions.

4 Conclusion

Five methods were proposed to offset the errors produced by the tilt Measurement System in the rotating state based on the MEMS accelerometer. The formula for calculating the angle of all schemes is of the same complexity and satisfies the unique angle value corresponding to each set of outputs of the sensor. Therefore, these methods are highly feasible. Based on the test results, it is concluded that the dual-axis accelerometer solution has some application value for applications with low angle range and rotation speed. However, for applications with large angle range and high rotation speed, the two single-axis accelerometer solutions have obvious advantages.