Enhanced Potassium Ion Electrochemical Measurement Using Signals Augmentation and Crucial Features Focusing

Existing methods for potassium ion measurement, like ion chromatography, spectrophotometry, and colorimetry, rely on spectrometric equipment, which is limited by high cost, significant interference, and complex operation. To address these issues, the enhanced electrochemical measurement using the electrical signals augmentation (ESA) and the crucial features focus-utilized neural network (CFNN) is proposed for measuring the potassium ion concentration. The electrical signals dynamically aligning (ESDA) is developed to overcome the electrode drift and enhance the repeatability of signals. The ESA augments both the quantity and trend of signals to address the challenges of learning from a dataset with limited samples. The CFNN is developed to extract crucial features of signals and to establish a robust mapping between signals and concentrations. The proposed method achieves the electrode and exhibits a repeatability error of 1.51% (0.0013 V), and MAPE of 0.88%, and an RMSE of 3.95 in measurements. Experimental results demonstrate the proposed method outperforms in terms of signal repeatability, sample availability, accuracy, and efficiency, compared to conventional electrochemical measurements. The proposed method achieves a balance between accuracy and computational resources, it can also be extended to precise practices in biology, medicine, and agriculture.