| Electronic tongue is an intelligent analytical instrument which consists of a cross-sensitive sensor array, a signal conditioning circuit and a pattern recognition algorithm. It has wide application in food quality and safety, biology, and environment, because it demonstrates the features of fast detection, simple operation and safe disposal. However, with the extensive application of electronic tongues, such problems as deficient flexibility and weak reconfiguration appear gradually which may lead to performance degradation or even performance loss when complicated samples are analyzed. It is obvious practical significance to develop a reconfigurable and customized electronic tongue to satisfy different requirements.A voltammetric electronic tongue experiment platform-Smart Tongue based on virtual instrument was constructed to deal with the above situation in this paper. Application development software model based on LabVIEW programming tool was also designed. Smart Tongue platform comprised a particular electrochemical sensor array module, a self-designed signal conditioning unit, a data acquisition card and a personal computer with LabVIEW programming environment. The system implemented such functions as generating excitation signal, acquiring and real-time displaying multi-channel signal, and cleaning electrodes. To testify the feasibility and reliability of operating platform, classification and discrimination experiments of tea beverages were conducted by taking combinatorial frequency pulse as excitation signal, collecting the responding signal from sensor array, extracting the eigenvalue, and processing the data with pattern recognition algorithm.The main research work and results are as follows.(1) A six-channel signal conditioning circuit was developed as the electronic core module for Smart Tongue. Problems of signal noise, excitation signal distortion, and dynamic responding signal output range were successfully solved by debugging and improving, which greatly enhanced the detection precision and stabilization.(2) A Smart Tongue experiment platform based on virtual instrument was constructed. This platform features a module hardware design which makes it convenient for the later debugging and maintaining, and a LabVIEW programming environment-based measure module which makes it possible to configure the model rapidly according to different demands. (3) A suitable algorithm for extracting eigenvalue was designed by analyzing the responding signal from sensor array under the excitation of combinatorial frequency pulse. At the same time, a sample database based on LabSQL was developed which lead to a seamless communication with pattern recognition module.(4) Principle components analysis was found applicable for the fast classification of Masterkong tea beverages. SIMCA method produced a recognition rate of 100% when used for identifying different brands of black tea and their diluted counterparts. Results show that the construction of Smart Tongue platform based on virtual instrument is feasible and reliable.In conclusion, taking virtual instrument as development environment, Smart Tongue experiment platform was constructed successfully with due expectancy and greatly enhanced design process. Furthermore, it also provides good foundation for further application of electronic tongue. |
PDF Full Text Request