Linear calibration method

Abstract

The invention discloses a linear calibration method. The linear calibration method comprises the following steps: when data of a system to be calibrated have curve characters in a two-dimensional coordinate system, firstly selecting a group of sampling data, determining a basic line with two points at the two ends of the group of data, then amending the intercept and gradient in sequence on the basic line, distributing the points of the group of sampling data to one side of the amended straight line as many as possible, and finally amending the point with greater deviation integrally. At the condition of the curve characters during calibration, the rated output value is approached by adopting a universal straight line manner directly, so that the curve character that numerical values are usually convex or concave in practice is solved, the precision of actual output value is improved, the control operation difficulty is reduced, and meanwhile the method is convenient to understand and realize; the method is simple in principle and higher in universality, and is suitable for all numerical values with linear transformation theoretically.

Description

technical field [0001] The invention relates to a linear calibration method, which belongs to a processing method when a measurement system, a measurement instrument and the like need linear calibration. Background technique [0002] At present, when people use DC power modules, they usually have very strict requirements on the accuracy of output voltage, output current, etc., but due to the differences between the numerous electronic components inside the power supply, the rated output voltage and current are equal to their actual output values. There is a big difference between them, the hardware consistency is poor, and the accuracy index of the final output and detection requirements in use cannot be achieved. Therefore, in production, it is often necessary to calibrate the rated output values ​​of the power module, such as voltage, current, temperature, power, etc., and the actual output value with high precision. [0003] Calibration methods vary according to complexi...

AI Technical Summary

Problems solved by technology

As far as the DC power supply module is concerned, a simple and easy method is usually adopted. For example, it is generally considered that the sampling circuit and operational amplifier circuit used in the power supply work are linear, and the actual output value of each module satisfies the basic functional relationship of the straight line by default. y=kx+b, and sampling two sets of values ​​into the function, the slope k and intercept b of the function can be determined, so as to determine the only straight line; but after using this method to calibrate, due to the relationship between components in practical applications, It is not ideal linearity, and there are often many curve characteristics, which cause deviations in various quantities in some cases

[0004] In view of the above-mentioned curve characteristics, people have proposed to use a complex algorithm to collect more sample points to fit the specific curve equation of the module. Although the accuracy of this method is improved, the algorithm is difficult, and in the actual production process. Sampling more points affects production efficiency

Images