Affiliations: [a]
College of Information Engineering, Dalian Ocean University, Dalian, China
| [b] Key Laboratory of Environment Controlled Aquaculture, Ministry of Education, China
| [c] Dalian Key Laboratory of Smart Fisheries, Dalian, China
| [d]
Dalian Xinyulong Marine Biological Seed Technology Co., Ltd., Dalian, China
Correspondence:
[*]
Corresponding author. Donggang He, College of Information Engineering, Dalian Ocean University, Dalian, China. E-mail: [email protected].
Note: [1] This work was supported in part by the General Scientific Research Projects of Liaoning Provincial Department of Education under Grant JYTMS20230489, and in part by Key Laboratory of Environment Controlled Aquaculture (Dalian Ocean University) Ministry of Education Open Project under Grant 202314.
Abstract: The factors of water quality in the intensive marine stichopus japonicus aquaculture process are changing with seasons, so water temperature, salinity, pH value and nitrite were selected as auxiliary variables to measure the concentration of ammonia nitrogen. FCM (Fuzzy C-means) algorithm was adopted to classify them. Based on the EM (Expectation Maximization) algorithm, fuzzy sub-models of ammonia nitrogen concentration were constructed around each operating point, and finally the fuzzy sub-models were combined according to the posterior distribution of the characteristics of the sampling data. Based on the data collected at Xinyulong Marine Biological Seed Technology Co., Ltd, in Dalian China, the ammonia nitrogen concentration prediction model was tested and verified.
Keywords: Water quality, stichopus japonicus, expectation maximization, multi-model
