Study On Goat Serum Detection Method Based On Raman Spectroscopy

Serum is an important part of organism,which plays an important role in providing nutrition and maintaining blood osmotic pressure.The health status of organism can be reflected by the changes of serum components.Serum detection is widely used as a method of health status recognition.The existing detection methods have high detection accuracy,but there are some shortcomings,such as the introduction of pollutants,poor timeliness and so on.Therefore,a Raman spectroscopy technology combined with mathematical modeling algorithm is proposed for achieving rapid and accurate detection of serum.In this paper,the background and significance of the research are introduced,the existing Raman spectroscopy technology and the Raman spectroscopy detection of serum research are extensively investigated,the existing problems are raised,and the research objectives and methods are clear.The main research contents are as follows:Firstly,the basic principle of Raman spectrum detection and the mechanism of surface enhanced Raman are described.The preparation method of experimental samples and the collection method of Raman spectrum are introduced.The pretreatment of spectral data is carried out.The liquid nano silver sol and nano gold solid Raman enhanced substrates are made.Secondly,the fast,non-destructive and quantitative detection method of goat serum protein content is studied,and the Raman characteristic peak is assigned;the improved swarm optimization support vector regression algorithm is proposed,and the corresponding mathematical model is established.The classical artificial swarm algorithm is improved to improve the accuracy and timeliness of the model.Compared with the improved algorithm and BP neural network algorithm,the established model is evaluated.Thirdly,it explores the method of rapid quantitative detection of serum albumin by combining Raman spectrum with partial least square method,and the quantitative detection of serum albumin can be realized by selecting appropriate spectral model and spectral pretreatment method,and establishing mathematical model by partial least square method.It introduces the basic theory of algorithm,and compares it with principal component regression algorithm.Moreover,the model is evaluated.Finally,a nano gold solid Raman enhanced substrate is proposed to detect samples with samples with different concentrations of serum,which greatly improves the intensity of the characteristic peaks of the Raman spectrum.By comparing the enhancement effect of liquid and solid Raman enhanced substrate,the solid substrate is selected to obtain the serum Raman enhanced signal with good repeatability and rich information.The partial least square algorithm is used to model and analyze the model Evaluation.