西北太平洋柔鱼BP神经网络渔场预报模型比较研究

Comparative study on the forecasting models of squid fishing ground in the northwest Pacific Ocean based on BP artificial neural network

柔鱼是西北太平洋的重要经济种类。研究根据1995-2001年7-11月采集的鱿钓生产数据以及相对应的海洋环境因子数据,包括经纬度、表温(SST)和海平面高度距平(SSHA),分别以单位捕捞努力量渔获量(CPUE)和捕捞努力量作为中心渔场指标,采用BP神经网络方法,以经纬度、海洋环境因子作为输入因子,分别以CPUE和捕捞努力量作为输出因子,采用4-3-1和4-2-1两种模型,共4种方案对西北太平洋柔鱼渔场进行预报,并以拟合残差最小的模型作为最优预报模型。分析结果显示,7-11月各月中心渔场预报模型均以4-3-1模型为最优,但7、8月最优预报模型以捕捞努力量为输出的4-3-1模型,9、10、11月最优预报模型以CPUE为输出的4-3-1模型,总体平均误差以捕捞努力量为输出的4-3-1模型为最小。研究认为,CPUE和捕捞努力量作为中心渔场预报指标有差异,以捕捞努力量为输出的4-3-1模型较合适作为柔鱼渔场预报模型。

Squid is one of the important economic species in the northwestern Pacific. Using Catch per Unit Effort and V% as the target of central fishing ground and adopting BP artificial neural network, we forecast fishing ground in the northwest Pacific Ocean. The study was based on the data of squid fishing and relevant marine environment factors, including longitude, latitude, SST and SSHA from July to November from 1995 to 2001. The input factor is marine environment factor, the output factors are CPUE and V% and 4-3-1 and 4-2- 1 model total 4 kinds models were used to compare which is the best suitable model for fishery forecast. The minimum fitting residual of model is the best one. Result shows that 4-3-1 is the best suitable model for each month, but the best suitable model for July and August is 4-3-1 with output V% and best suitable model for September, October and November is 4-3-1 with output CPUE, the minimum overall average error is 4-3-1 model output V%. Research suggests that there are differences as a center of fishery forecast targets by CPUE and V%, and the 4-3-1 model output V% can be used as forecasting model of squid fishing ground.