| Electronic measurement technology plays an important role in the business of national defense and civilian production. With the development of science and technology in the field of aerospace, the conventional measuring instruments is not able to meet the needs of high-precision test, so the development of precision multimeter is of importance. The technology of digital-to-analog conversion, noise reduction and calibration are the three core technology for high-precision multimeter. This article mainly research the noise reduction and calibration techniques of precision multimeter.The subject derives from the vertical programs "XXX 6?-Digit Multimeter development." This multimeter not only can measure AC and DC voltage, AC and DC current, resistance, and other basic functions, but also has the ability to display the waveform by the digitizer, reaching the technology level of foreign products.In this paper, we analyze the circuit noise and the system error characteristics, and design the filter to restrain the noise and improve the precision of the meter, according to the characteristics that the noise and useful signal is separated in spectrum. Lastly,depending on the characteristics of the system error, the correction methods are designed to improve the accuracy of the multimeter.The main contents of the subject:(1) According to the three measurement circuit, such as measuring the functional circuits, digital to analog conversion circuit and power circuit implementation, I mainly analysis what kind of potential noise and error will be bring by the components, and estimate the noise and the scope of the error of each device, and made a conclusion for the noise and error characteristics.(2) According to the characteristics that measurement system noise is in the high-frequency comparing the measured signal, so I designed the low pass digital filter.Firstly, I designed a filter with four comb, which have 64-fold decimation factor, to reduce the sampling rate and noise. The low-pass filter of 1Hz cutoff frequency is designed by using the Keyser window function method. And the modified particle swarm optimization algorithm is adopted to optimize the low pass filter, to make it has a certain degree of adaptability.(3) As for the non-linear error of amplifier and digital to analog converter, I designuneven piecewise cubic curve fitting to compensate for them. For the voltage ratio changes caused by the AC voltage attenuation network frequency response,nonlinear least square method is used to compute the unknown parameters the voltage ratio expression and the the error is compensated by the voltage ratio expression. The self calibration method is designed to eliminate the overall gain and offset of the measuring circuit, so that the measurement accuracy of the system is only determined by the internal reference source, which ensures the long-term stability.After the treatment method is discussed in each chapter, the comparison is carried out before and after treatment. the results showed that: SNR increased from about 66 dB to 114 dB, nonlinear error increased from 0.0107% to 0.00072%, the voltage ratio changes caused by the AC voltage attenuation network frequency response increased from 0.219% to 0.0202%, overall accuracy from less than four and a half up to six and a half. Finally, the index test is made to verify the effectiveness of the method studied in this paper. |
PDF Full Text Request