首页|加速度计数据采样电路设计及其温度补偿技术研究

加速度计数据采样电路设计及其温度补偿技术研究

扫码查看
加速度计的输出为模拟电流信号,在实际应用中需要一个加速度计数据采样系统将其转化为导航计算机可以处理的数字信号.首先,针对加速度计数据采样电路的小型化要求,设计了基于模数转换原理的加速度计数据采样电路,所设计的电路板尺寸为40 mm×40 mm.其次,针对加速度计数据采样电路的温度补偿问题,对电路进行了高低温实验,得到了电路的零偏和标度因数随温度和输入电流变化的实验数据.采用多项式拟合法、BP神经网络拟合法、PSO-BP神经网络拟合法、ACO-BP神经网络拟合法建立起-20℃ ~40℃之间的静态温度补偿模型,可以实现对电路的实时在线温度补偿.补偿后电路零偏的温度稳定性约为10μg,标度因数的温度稳定性优于5×10-6,标度因数非线性度降低到了4×10-6以下,证明了温度补偿模型的正确性和有效性.
Research on Design of Accelerometer Data Sampling Circuit and Its Temperature Compensation Technology
The output of the accelerometer is analog current signal. In practical application, an accelerometer data sampling system is required to convert analog signal into digital signal which can be processed by the navigation computer. Firstly, in view of the miniaturization requirement of the accelerometer data sampling circuit, an accelerometer data sam-pling circuit based on the analog-to-digital conversion is designed. The designed circuit board size is 40 mm × 40 mm. Sec-ondly, aiming at the problem of the temperature compensation for the accelerometer data sampling, the circuit high and low-temperature experiments are carried out, and the experimental data of the bias and the scale factor with temperature and in-put current variation are obtained. A static temperature compensation model between -20℃ ~40℃ is established by using the polynomial fitting, BP neural network fitting, PSO-BP neural network fitting and ACO-BP neural network fitting, which can achieve real-time on-line temperature compensation for the circuits. After compensation, the temperature stability of ze-ro bias is 10μg, the temperature stability of the scale factor is better than 5 × 10 -6 , and the nonlinearity of the scale factor is reduced to less than 4 × 10 -6 , which demonstrates the correctness and effectiveness of the temperature compensation model.

accelerometerdata samplingA/D conversiontemperature compensationneural network

薛慧兵、陈光

展开 >

北京航天时代激光导航技术有限责任公司,北京100094

加速度计 数据采集 A/D转换 温度补偿 神经网络

2024

导航与控制
北京航天控制仪器研究所

导航与控制

CSTPCD
影响因子:0.133
ISSN:1674-5558
年,卷(期):2024.23(1)