| With the rapid economic development and population growth,the consumption of food and energy is increasing,and the environmental pollution is becoming increasingly serious.All countries in the world are seeking solutions.Microalgae are aquatic pho-totrophic unicellular microorganisms with the characteristics of easy cultivation,short growth cycle,high yield per unit area and unoccupied agricultural land,which have been widely used in many fields,such as additives,biosolid carbon,and renewable bioen-ergy.However,the higher cost of microalgal culture has always been the difficulty of restricting the development of its industry,but also the lack of rapid and nondestructive detection methods for monitoring the growth of microalgae and their metabolite changes in real time,obtaining timely changes in the characteristics of these important indica-tors in the production process of microalgal culture,and then effectively adjusting the culture conditions to guarantee the production of microalgae with high efficiency and quality.In this paper,the effects of magnetic field and light quality on the growth of mi-croalgae and their metabolites(photosynthetic pigments,carbohydrates,proteins,oils,and fatty acid contents)were investigated in C.pyrenoidosa and T.obliquus;Optimiza-tion of magnetic field and light quality conditions to promote the growth of microalgae,and microalgae were cultured under optimized magnetic field and light quality condi-tions.The real-time dynamic changes of growth information(biomass,photosynthetic pigments,carbohydrate and protein contents)of the microalgae were visualized by vis-ible/near infrared hyperspectral techniques combined with chemometric methods,in the hope of providing a certain theoretical reference and technical support for efficient and superior production of microalgae.The main work and results of this paper are as follows:1.Firstly,the effects of magnetic fields on the growth of microalgae and their metabolites were investigated.The results showed that cultured under a 30 m T mag-netic field significantly enhanced biomass growth of C.pyrenoidosa and T.obliquus(p<0.05),increases the protein content of C.pyrenoidosa while inhibiting its carbohy-drate and oil accumulation(p<0.05),significantly reduced carbohydrate content of T.obliquus,meanwhile,the magnetic field was also unfavorable for the synthesis of two microalgal fatty acids.Secondly,compared the bioeffects of different surround light quality(red,green,yellow,blue,violet).The results showed green and blue light had the best effect on biomass growth whereas violet light had the negative effect;Red light is most beneficial for carbohydrate synthesis in both microalgae while had side nega-tive effect on oil accumulation(p<0.05);Violet and blue light significantly promote protein content of C.pyrenoidosa and T.obliquus(p<0.05),while blue and yellow light are the optimal light sources for oil synthesis.Finally,the feasibility of response surface design experiments applied to synergistically induce high yields in microalgae under a magnetic field and light quality is explored,indicating that response surface methodology is not suitable for optimization of the two culture conditions.2.Collecting the VIS/NIRS of C.pyrenoidosa during the growth time,and with the treatment of spectral pretreatment,feature band selection,chemometrics modeling,con-structed quantitative discriminant models between hyperspectral data and biomass,pho-tosynthetic pigments(Chl-a,Chl-b,Car),carbohydrates,and proteins,and the model performed well.The biomass prediction model used Autosacling pretreatment com-bined with the CARS-MLR,R_p~2and RPD were 0.979 and 7.650;Chl-a,Chl-b and Car used SNV pretreatment combined with SPA-MLR,R_p~2were 0.967,0.960 and 0.961,RPD were 5.740,5.348 and 5.206;carbohydrates used WT pretreatment combined with IRF-RFR,R_p~2and RPD were 0.979 and 7.650;proteins used S-G pretreatment with SA-RFR,R_p~2and RPD were 0.968 and 12.993.Finally,HSI was used to visualize the spatial distribution and abundance of components during the growth of C.pyrenoidosa,it per-formed well.3.The VIS/NIRS of T.obliquus during the growth time were used with the treat-ment of spectral pretreatment,feature band selection,chemometrics modeling,and con-structed quantitative discriminant models between hyperspectral data and biomass,pho-tosynthetic pigments(Chl-a,Chl-b,Car),carbohydrates,and proteins,the model per-formed well.The biomass prediction model used SNV pretreatment combined with the CARS-MLR,R_p~2and RPD were 0.967 and 6.212;Chl-a used MAF pretreatment com-bined with SPA-MLR R_p~2and PRD were 0.968 and 5.902;Chl-b used WT pretreatment combined with SPA-MLR R_p~2and PRD were 0.947 and 4.613;Car used MAF pretreat-ment combined with SPA-MLR R_p~2and PRD were 0.960 and 8.005;carbohydrates used raw spectrum combined with IRF-RFR,R_p~2and RPD were 0.996 and 36.156;proteins used WT pretreatment with SA-RFR,R_p~2and RPD were 0.909 and 10.116.Finally,HSI was used to visualize the spatial distribution and abundance of components during the growth of T.obliquus,got good results. |