Intelligent matching design of characteristic curves for semiconductor characteristic testing in oceanographic instruments

Abstract Wide band gap devices have a wide application prospect in Oceanographic instruments, and the demand for their performance evaluation and characteristic testing is becoming increasingly prominent. The traditional device tester is not ideal for testing some unique electrical characteristics of wide band gap devices. Therefore, the design of wide band gap device characteristic tester has practical technical significance. This paper focuses on designing an intelligent matching algorithm and scheme for processing large amounts of data measured by wide band gap device characteristic testers and fitting standard wide band gap device characteristic curves. Firstly, the standard wideband gap device characteristic curve is generated based on the device model equation in MATLAB environment, and the error is added and the data set is extended to simulate the real environment. Then, in a Python environment, the data is processed using clustering and linear regression methods. Finally, polynomial fitting is carried out to the data, and the characteristic curve of the wide-band gap device is drawn. This paper provides an efficient and accurate data processing method and intelligent matching scheme for characteristic testing of wide band gap semiconductors used in Oceanographic instruments, which is expected to play an important role in device performance evaluation and production quality control.