TY - JOUR A2 - Morabito, Francesco Carlo AU - Krasnopolsky, Vladimir AU - Nadiga, Sudhir AU - Mehra, Avichal AU - Bayler, Eric PY - 2018 DA - 2018/11/01 TI - adjust Neural Network to a Particular Problem:Neural network - based Empirical Biological Model for叶绿素Concentration in the Upper Ocean SP - 7057363 VL - 2018 AB -神经网络(NN)技术的通用性使其成功地应用于许多科学领域和各种各样的问题。对于每个问题或一类问题，一般的神经网络技术(例如，多层感知器(MLP))通常需要一些调整，这通常是开发一个成功的应用程序的关键。在本文中，我们介绍了一个神经网络应用程序，证明了这种调整的重要性;此外，在这种情况下，应用于一般神经网络技术的调整可以成功地应用于许多其他神经网络应用。我们介绍了一种神经网络技术，将叶绿素“a”(chl-a)的变异(主要由生物过程驱动)与海洋上层的物理过程联系起来，使用基于神经网络的chl-a经验生物模型。在本研究中，利用卫星导出的表层参数场、海表温度(SST)和海表高度(SSH)，以及网格化的盐度和温度剖面(0 - 75m)作为上层海洋动力学的特征。使用了NOAA的可见光成像红外辐射计套件(VIIRS)的叶绿素a场，以及中分辨率成像光谱辐射计(MODIS)和海洋观测宽视场传感器(SeaWiFS)的chl-a浓度。研究了优化神经网络技术的不同方法。使用均方根误差(RMSE)度量和观测海洋颜色(OC)域与神经网络输出之间的相互关系来评估结果。 To reduce the impact of noise in the data and to obtain a stable computation of the NN Jacobian, an ensemble of NN with different weights is constructed. This study demonstrates that the NN technique provides an accurate, computationally cheap method to generate long (up to 10 years) time series of consistent chl-a concentration that are in good agreement with chl-a data observed by different satellite sensors during the relevant period. The presented NN demonstrates a very good ability to generalize in terms of both space and time. Consequently, the NN-based empirical biological model for chl-a can be used in oceanic models, coupled climate prediction systems, and data assimilation systems to dynamically consider biological processes in the upper ocean. SN - 1687-9724 UR - https://doi.org/10.1155/2018/7057363 DO - 10.1155/2018/7057363 JF - Applied Computational Intelligence and Soft Computing PB - Hindawi KW - ER -