[1]聂磊,蔡文涛,黄一凡,等.基于时间序列的微藻生长环境参数的预测模型[J].江苏农业学报,2021,(05):1183-1189.[doi:doi:10.3969/j.issn.1000-4440.2021.05.013]
　NIE Lei,CAI Wen-tao,HUANG Yi-fan,et al.Prediction model of microalgae growth environment parameters based on time series[J].,2021,(05):1183-1189.[doi:doi:10.3969/j.issn.1000-4440.2021.05.013]
点击复制

基于时间序列的微藻生长环境参数的预测模型()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2021年05期
页码:
1183-1189
栏目:
耕作栽培·资源环境
出版日期:
2021-10-30

文章信息/Info

Title:
Prediction model of microalgae growth environment parameters based on time series
作者:
聂磊; 蔡文涛; 黄一凡; 董正琼
（湖北工业大学机械工程学院/湖北省现代制造质量工程重点实验室，湖北武汉430068）
Author(s):
NIE Lei; CAI Wen-tao; HUANG Yi-fan; DONG Zheng-qiong
（College of Mechanical Engineering, Hubei University of Technology/Hubei Key Laboratory of Manufacture Quality Engineering, WuHan 430068, China）
关键词:
钝顶螺旋藻; 自回归滑动平均模型; 自回归滑动平均-卡尔曼滤波模型; 环境参数
Keywords:
Spirulina platensis; autoregressive moving average model; autoregressive moving average-Kalman filter model; environmental parameters
分类号:
Q949.2
DOI:
doi:10.3969/j.issn.1000-4440.2021.05.013
文献标志码:
A
摘要:
掌握微藻生长环境中的温度、硝酸盐浓度、氧气浓度等参数的变化规律有利于提高生物产出量。本研究利用钝顶螺旋藻全生长周期中的大量环境参数，引入时间序列预测方法，通过极大似然估计，得到线性回归模型，同时引入状态空间预测方法，将低维的线性模型映射到高维空间，以消除低维空间下模型的不准确性。为评估模型的有效性，采用了杜宾-瓦特森检验法、图像法、均方根误差、平均绝对误差和最大误差等评估方法和指标进行评估和验证。预测结果显示，自回归滑动平均模型和自回归滑动平均-卡尔曼滤波模型均可用于预测微藻生长的环境参数；与前者相比，后者预测误差更小，模型拟合更精确，能更好地揭示钝顶螺旋藻生长过程中环境参数的内在规律。
Abstract:
Mastering the variation law of parameters such as temperature, nitrate concentration and oxygen concentration in the growth environment of microalgae is beneficial to improve the biological output. In this study, a large number of environmental parameters in the full growth cycle of Spirulina platensis were used, and the time series prediction method was introduced to obtain a linear regression model by maximum likelihood estimation, while a state space prediction method was introduced to map the low-dimensional linear model to the high-dimensional space to eliminate the inaccuracy of the model in the low-dimensional space. Evaluation methods and indicators such as the Durbin-Watson test, image method, root mean square error, mean absolute error and maximum error were used to assess the validity of the model. The prediction results showed that both the autoregressive moving average model and the autoregressive moving average-Kalman filter model could be used to predict the environmental parameters of microalgae growth. Compared with the autoregressive moving average, the autoregressive moving average-Kalman filter model has smaller prediction error and more accurate model fitting, which can better reveal the intrinsic patterns of environmental parameters during the growth of Spirulina platensis.

参考文献/References:

[1]ZHAO X L, KATRIN S, GERO C, et al. Hydrothermal carbonization of Spirulina platensis and Chlorella vulgaris combined with protein isolation and struvite production[J]. Bioresource Technology Reports, 2019, 1(6):159-167.
[2]IMAR D S, REGIN M Q, ALHADJ D M, et al. Vitamin a status in healthy women eating traditionally prepared spirulina ( Dihé ) in the Chad Lake area[J]. PLoS One, 2018, 13(1): e0191887.
[3]闫春宇,胡冰涛,王素英. 常压室温等离子体诱变对螺旋藻中氨基酸成分的影响[J]. 食品与发酵工业, 2017, 43(1): 60-65.
[4]CHEN G S, CAI Y, SU Y Y, et al. Effects of Spirulin a algae as a feed supplement on nutritional value and flavour components of silkie hens eggs[J]. Journal of Animal Physiology and Animal Nutrition, 2019, 103(5):1408-1417.
[5]罗光宏,雷婕,张喜峰,等. 碳酸二甲酯三相分配体系萃取分离螺旋藻多糖[J]. 食品与发酵工业, 2020, 46(17): 196-203.
[6]朱孝晨. 钝顶螺旋藻藻蓝蛋白与多糖的制备及生物活性研究[D]. 烟台：烟台大学, 2018.
[7]王雪飞. 光生物反应器中市政污水培养钝顶螺旋藻的条件优化[D]. 无锡：江南大学, 2017.
[8]CARVALHO A P, MEIRELES L A, MALCATA F X. Microalgal reactors: a review of enclosed system designs and performances[J]. Biotechnology Progress, 2006, 22(6):1490-1506.
[9]NORSKER N H, BARBOSA M J, VERMUE M H, et al. Microalgal production--a close look at the economics[J]. Biotechnology Advances, 2011, 29(1):24-27.
[10]余灿. 城市污水户外跑道池培养微藻的生长促进方法研究[D]. 哈尔滨：哈尔滨工业大学,2015.
[11]BAKY H E, BAROTY G E, MOSTAFA E M . Optimization growth of spirulina (Arthrospira) platensis in photobioreactor under varied nitrogen concentration for maximized biomass, carotenoids and lipid contents.[J]. Recent Patents on Food, 2020, 11(1):40-48.
[12]LI S H, SONG C F, LI M D, et al. Effect of different nitrogen ratio on the performance of CO2 absorption and microalgae conversion (CAMC) hybrid system[J]. Bioresource Technology, 2020, 306: 123126.
[13]李姿,任洪艳,周骞骞,等. 光强和CO2体积分数对不同生长时期的钝顶螺旋藻生长和固碳的交互影响[J]. 食品与生物技术学报, 2014, 33(8): 827-836.
[14]赵嶝科,刘慧,张少斌,等. Ni2+对钝顶螺旋藻生长及藻胆蛋白含量的影响[J]. 水产科学, 2020, 39(5):1-5.
[15]李日清.在线pH仪在污水处理系统中的使用及维护[J].山西电子技术,2011,4(5):35-36.
[16]李桐，董维红，张琦琛，等.基于时间序列模型的黑龙江省粮食水足迹分析与预测[J].排灌机械工程学报，2020，38(11）：1152-1159.
[17]甄晓菊，张雪红，吴国明，等．基于Sentinel-2A NDVI时间序列数据的冬小麦识别[J].江苏农业科学，2019，47(16)：239-245.
[18]江显群,陈武奋,邵金龙,等. 大坝变形预报模型应用[J]. 排灌机械工程学报，2019, 37(10)：870-874，920．
[19]吴婕. 数据驱动的雷达健康预测技术研究[D].北京：中国电子科技集团公司电子科学研究院, 2019.
[20]李波,林聪,刘清蝉,等. 基于时序建模的光纤电流互感器随机噪声卡尔曼滤波方法[J]. 电机与控制学报, 2017, 21(4): 83-88，.
[21]张瑞国,李春雨,丁志宏,等. 基于时间序列模型的雷达数据随机误差建模与补偿[J]. 现代雷达, 2016, 38(11): 49-52.
[22]唐俊,赵成萍,周新志. 基于EVI-RBF的玉米长势监测及产量预测[J].江苏农业学报,2020,36(3):577-583.
[23]孙振华. 基于动态数据驱动的生物氧化槽进气量预测研究[D]. 乌鲁木齐：新疆大学, 2018.
[24]修春波,任晓,李艳晴,等. 基于卡尔曼滤波的风速序列短期预测方法[J]. 电工技术学报, 2014, 29(2): 253-259.

备注/Memo

备注/Memo:
收稿日期：2021-03-17基金项目：国家自然科学基金面上项目(51975191)；湖北省教育厅重点项目(D20201401)作者简介：聂磊(1978-),男,湖北武汉人，博士，教授，主要从事电子制造工艺可靠性研究。（E-mail）leinie@ hust.edu.cn通讯作者：董正琼，（E-mail）dongzhq@hbut.edu.cn
更新日期/Last Update: 2021-11-09