Study On The Prediction System Of Silage Corn Of Moisture Based On Visible Spectrum

Silage is a quality animal(mainly ruminant)roughage.With the progress of society,people's demand for meat products and dairy products is increasing.The scale of ruminants such as dairy cows,beef cattle,mutton sheep and domestic cattle continues to expand,and the demand for silage has also increased significantly.In silage,the high or low moisture content of raw materials is not conducive to the normal completion of the silage process.If the moisture content is too high,the soluble nutrients will easily be lost with the exuded juice,which will cause the fermentation of the shuttle and affect the feed quality.If the moisture content is too low,it will not be easily compacted,resulting in the easy accumulation of air in the storage pits.Silage moisture content will also lead to loss of harvest and excessive heat production,so that the amount of fermentation substances or nutrients decreased,resulting in elevated pH,lactic acid and other fermentation products decreased.Therefore,the detection of moisture content of raw materials is of great significance to corn silage.In order to achieve the non-destructive and rapid detection of the moisture content of the silage raw materials,this study designed a prediction system for moisture content of corn silage based on the visible light spectrum,and conducted an experimental study.This paper takes Jialiang No.99 corn as the research object,uses Taiwan's Wuling optical OTO spectrometer and its associated optical components to collect spectral data information in the visible wavelength region of the sample,and uses laboratory drying method to determine the actual moisture content of the corn silage raw material.The upper computer software of the moisture content prediction system of corn silage material was developed in the Microsoft Visual Studio 2010 platform using C++ programming language,and the on-line detection of moisture content of corn silage material was realized.The main tasks of this study are as follows:(1)Combining an OTO spectrometer with its associated optical components to construct a hardware detection system for visible spectrum data.The detection system uses a light emitter to improve the quality of the light source.In the process of receiving the reflected light signal,an optical fiber collimating lens is used together with long and short optical fibers to improve the quality of the optical signal collected by the spectrometer.(2)Analyze system functional requirements and design software-related functional modules.The software system can realize basic functions such as user login,detection environment selection,spectrometer connection,standard whitening,data extraction,moisture content prediction,data analysis,and data storage.(3)Spectral information of silage raw material samples was collected,and the original spectrum was preprocessed by standard normal transformation and other methods.The effect of each pretreatment method on the PLSR prediction model of silage raw material moisture content was analyzed.The results showed that the vector normalization The algorithm is the best spectral pretreatment method.The PLSR prediction model based on the vector normalization algorithm was embedded into the software system to predict the moisture content of corn silage materials.(4)Prepare the communication program of the serial port of the spectrometer and the host computer and debug the data communication function,complete the connection between the software and the spectrometer;test system spectral data extraction,moisture content prediction and data preservation and other functions,the results show that the system can accurately predict the silage The moisture content of corn raw material can basically fulfill the function of this design demand.In this study,a moisture content prediction system for corn silage based on visible light spectrum analysis technology was designed,which can automatically analyze and predict the moisture content of corn silage materials.This study provides a new research idea for quick and non-destructive estimation of the moisture content of silage corn raw materials,and is of great significance for the detection of nutritional quality of silage corn in the future.