Classical and quantum regression analysis for the optoelectronic performance of NTCDA/p-Si UV photodiode

Due to the pivotal role of UV photodiodes in many technological applications in tandem with the high efficiency achieved by machine learning techniques in regression and classification problems, different artificial intelligence (AI) techniques are adopted to simulate and model the performance of organic/inorganic heterojunction UV photodiode. Herein, the performance of a fabricated Au/NTCDA/p-Si/Al photodiode was explained in details and showed an excellent responsivity, and detectivity for UV light of intensities ranges from 20 to 80 ${mW/cm^2}$. The fabricated photodiodes exhibited a linear current-irradiance relationship under illumination up to 65 ${mW/cm^2}$. It also exhibits good response times of t_rise = 408 ms and t_fall = 490 ms. Furthermore, we have not only fitted the characteristic I-V curve at the highlighted intensities but also evaluated three classical algorithms; k-nearest neighbour (KNN), artificial neural network (ANN), and genetic programming (GP) besides using a quantum neural network (QNN) to predict the behaviour of the fabricated device inside and outside the studied illumination intensity range. The ANN method achieved the highest accuracy under dark and illumination conditions with a mean squared error ${~1.6812\times10^{-11}}$ on the testing data. On the other hand, the Continuous-Variable (CV) QNN - using only one Qumode - has been used for the first time to model the performance of the fabricated photodiode and achieved an acceptable result with the lowest parameter set. Besides, we have utilised the full space of only one Qumode for encoding the two features using the Displaced Squeezed state preparation routine.