Design And Research Of Crop Ion Detection System Based On All-solid Ion Selective Electrodes

Inorganic ions play an important role in crop growth and are the main source of crop nutrients.Accurate detection of inorganic ions in agricultural production processes is of great significance for promoting the modernization of agriculture.At present,there are many methods for detecting inorganic ions,but they still cannot meet the ion detection needs of crops in different directions and application scenarios.Especially,the development of sensing technologies related to nutrient detection in hydroponic environments is of great significance for accurately controlling nutrients.This thesis investigated the preparation method of all solid-state ion selective electrodes,which overcame the vulnerability of glass electrodes.At the same time,a neural network model for nutrient solution ion concentration was established based on machine learning related knowledge,effectively suppressing environmental interference in nutrient solution.Finally,a ion signal detection system platform was developed,achieving real-time monitoring of crop ion signals.The main research work includes the following:（1）The potassium ion selective electrode with solid contact layer was prepared by combining carboxylated multiwalled carbon nanotubes（CMWCNT）and carboxylated graphite oxide oxide（CGO）,And compared with the common electrodeposited Mn O₂modified all solid-state potassium ion selective electrode,the CMWCNT/CGO solid contact layer effectively improves the detection accuracy of the potassium ion selective electrode.The optimal mixing ratio of CMWCNT and CGO is 1:3,the electrode response slope is 54.4±1 m V·decade^-1,and the detection limit is 10^-5.5mol·L^-1;At the same time,a back propagation（BP）neural network model optimized by particle swarm optimization（PSO）and Gravitational Search Algorithm（GSA）was constructed based on nutrient solution samples.The recovery error of the optimized prediction model was less than 1%,and the recovery error of the improved nonlinear inertia factor PSO-ANN was less than 0.12%;（2）Six ion concentrations of Ca²⁺,K⁺,Mg²⁺,NH₄⁺,NO₃^-,and HPO₄^2-were collected from different nutrient solutions using a commercial multi ion analyzer,expanding the research from single sensors to multi sensor arrays.At the same time,the artificial neural network（ANN）method was used to accurately calculate the true element concentration,and possible error sources during the collection process of commercial multi ion analyzers were analyzed;The study compared BP and Radial Basis Function（RBF）neural networks optimized based on PSO,Genetic Algorithm（GA）,and GSA.The optimization effect was different for the two neural networks,but the optimization operation rate was GA>PSO>GSA.At the same time,the study further proved that the structure of BP neural networks is worse than RBF neural networks in terms of generalization ability in this situation;（3）A software and hardware acquisition system based on all solid-state ion selective electrodes were developed.The electrical signal of the all solid-state ion sensor was input to the NI-9205 acquisition card through a designed amplification and conditioning circuit,including a low-pass filter,in-phase proportional operation circuit,and voltage follower.The system was driven by NI-DAQmx and loaded into the computer.The software system was then built through LabVIEW,which includes a data acquisition module,a data analysis module,a data storage module,human-computer interaction to achieve ion concentration collection and storage;Afterwards,a testing platform was built to test the accuracy and stability of the collection system;（4）The crop ion signal detection system was prospected in terms of the performance of all solid-state ion selective electrodes,crop in vivo detection,machine learning model embedding software,and their integration as a part of ion signal detection.We look forward to the attention and research of integrated crop ion signal detection systems,and the development of all solid-state ion selective electrodes in the agricultural field.