MENG Wei1,2*, LIU Yang3, LI Jian-wen1, LIU Xiao-lin1, GUO Xian-jiu4,CHENG Zhi-juan1
1. Suzhou Nuclear Power Research Institute Company Limited, Suzhou 215004,China;2. College of Information Science and Technology, Dalian Maritime University, Dalian 116026, China;3. College of Electronic and Information Engineering, Dalian Jiaotong University, Dalian 116028, China;4. College of Information Engineering, Dalian Ocean University,Dalian 116023,China
Abstract: An early warning model of marine organism detection is established to deal with the problem of marine organism invasion in a nuclear power plant cold source system based on multi-source information fusion. The effects of marine biological density, relative current velocity, relative wind speed, salinity and temperature on marine organism invasion were evaluated and influence coefficients and influence functions of five factors were calculated to establish the invasion organism detection model. Based on linear regression model and fuzzy neural network prediction method, the sampling data of multi-sensor were analyzed and calculated, and the early model of marine organism invasion was obtained. The results showed that the value of marine biological invasion intensity was obtained, and then the possibility of water intake blockage in nuclear power plant was judged by fusing the data characteristics of five factors. The measured data showed that the estimated error percentage of the two algorithms were within [-0.2,0.2], and the Euclidean distances were 0.016 8 and 0.007 8, respectively. The proposed marine organism detection and early warning model here is characterized by high precision, detection and early warning of marine organism in nuclear power plants to ensure the safety operation in the nuclear power cold source system.