Numerical modelling of coastal morphology is a complex and sometimes unrewarding exercise and often not yielding tangible results. Typically, the underlying drivers of morphology are not properly accounted for in nume...Numerical modelling of coastal morphology is a complex and sometimes unrewarding exercise and often not yielding tangible results. Typically, the underlying drivers of morphology are not properly accounted for in numerical models. Such inaccuracies combined with a paucity of validation data create a difficulty for coastal planners/engineers who are required to interpret such morphological models to develop coastal management strategies. This study develops an approach to long term morphological modelling of a barrier beach system that includes the findings of over 10 years of coastal monitoring on a dynamic coastal system. The novel approach to predicting the long term evolution of the area combines a mix of short term hydrodynamic monitoring and long term morphological modelling to predict future changes in a breached barrier system. A coupled wave, wind, hydrodynamic and sediment transport numerical model was used to predict the coastal evolution in the dynamic barrier beach system of Inner Dingle Bay, Co. Kerry, Ireland. The modelling approach utilizes the schematisation of inputs to reflect observed trends. The approach is subject to two stages of validation both quantitative and qualitative. The study highlights the importance of considering all the parameters responsible for driving coastal evolution and the necessity to have long term monitoring results for trend based validation.展开更多
Many of the important questions facing farming systems in the world today require long-term studies to provide meaningful information and answers. A long-term agronomic experiment (LTAE) should (1) have long-term obje...Many of the important questions facing farming systems in the world today require long-term studies to provide meaningful information and answers. A long-term agronomic experiment (LTAE) should (1) have long-term objectives; (2) study important soil processes or ecological processes; and (3) be related to the productivity and sustainability of systems. A well established LTAE can provide both insights into how the system operates and foresight into where the system goes. The prerequisites for setting up a LTAE are the secured land, continuous funding and dedicated scientists. A number of principles must be considered carefully when establishing a LTAE, (1) the site must be representative of large areas; (2) the treatments should be simple, but focusing on the big questions; (3) the plots should be large enough to allow subsequent modification of the experiment if this becomes necessary; (4) crop rotations should minimise, wherever possible, the risk of build-up of pests and diseases, and rotational phase should be considered in a rotational experiment; (5) a clearly defined experimental protocol should be developed to ensure data collected is scientifically valid and statistically analysable, but with flexibility to allow essential changes; (6) soil samples, possibly plant samples, should be achieved to provide better answer to the original questions when new, perhaps more accurate analytical techniques are developed, or answer new research questions that were not considered in the original design. The MASTER experiment in Australia was used as a case study to demonstrate how these principles are implemented in practice.展开更多
脱硫过程是具有高度动态非线性和较大延迟时间的复杂工业过程,为了解决烟气脱硫过程的建模问题,设计了注意力机制下的深度长短期记忆(attention mechanism-based long short-term memory,AttLSTM)网络,并基于该网络设计自动编码器,完成...脱硫过程是具有高度动态非线性和较大延迟时间的复杂工业过程,为了解决烟气脱硫过程的建模问题,设计了注意力机制下的深度长短期记忆(attention mechanism-based long short-term memory,AttLSTM)网络,并基于该网络设计自动编码器,完成脱硫过程异常点的检测。该文首次提出使用AttLSTM网络自编码器对脱硫过程进行离群点检测,并且该网络模型同样首次应用于脱硫过程的辨识任务中。从更深的意义上讲,该文尝试使用深度学习模型对复杂系统进行辨识,所建立的AttLSTM网络之前未出现在系统辨识领域,该网络的出现可以丰富辨识模型的选择,同时为人工智能技术在系统辨识领域和控制领域的应用与推广提供参考。实验结果表明,相比于之前文献出现的脱硫过程建模方法,所提方法在不同性能指标上均具有更好的表现,由此可以证明深度AttLSTM网络在脱硫场景下的有效性。展开更多
由于电池荷电状态(state of charge,SOC)无法直接测量,且传统的SOC估算方法精度低。为了提升锂离子电池SOC估算精度,对比了不同深度学习网络模型应用于SOC估算的效果,并提出了一种基于DRSN-CW-LSTM网络的锂离子电池SOC估算方法。该方法...由于电池荷电状态(state of charge,SOC)无法直接测量,且传统的SOC估算方法精度低。为了提升锂离子电池SOC估算精度,对比了不同深度学习网络模型应用于SOC估算的效果,并提出了一种基于DRSN-CW-LSTM网络的锂离子电池SOC估算方法。该方法基于长短期记忆网络(long-short-term memory,LSTM)和逐通道不同阈值的深度残差收缩网络(deep residual shrinkage networks with channel-wise thresholds,DRSN-CW),利用锂离子电池电压、电流、温度、容量等数据信息在深度残差收缩网路中进行特征提取,通过LSTM进一步拟合时间序列数据趋势,实现锂离子电池在使用周期内SOC的预测。在DRSN-CW网络的残差收缩模块中可以实现自适应噪声数据处理功能,消除锂离子电池数据流质量对SOC预测的负面影响。利用锂电池公共数据集训练所提出的网络,对比了3种神经网络模型在该两组数据集上的预测效果。实验结果表明,所提出的深度学习模型在两组公开数据集上的MAE和RMSE均值都控制在5%以内,相比其他3种深度学习模型有更好的抗噪性能和预测性能,且估算精度高。展开更多
文摘Numerical modelling of coastal morphology is a complex and sometimes unrewarding exercise and often not yielding tangible results. Typically, the underlying drivers of morphology are not properly accounted for in numerical models. Such inaccuracies combined with a paucity of validation data create a difficulty for coastal planners/engineers who are required to interpret such morphological models to develop coastal management strategies. This study develops an approach to long term morphological modelling of a barrier beach system that includes the findings of over 10 years of coastal monitoring on a dynamic coastal system. The novel approach to predicting the long term evolution of the area combines a mix of short term hydrodynamic monitoring and long term morphological modelling to predict future changes in a breached barrier system. A coupled wave, wind, hydrodynamic and sediment transport numerical model was used to predict the coastal evolution in the dynamic barrier beach system of Inner Dingle Bay, Co. Kerry, Ireland. The modelling approach utilizes the schematisation of inputs to reflect observed trends. The approach is subject to two stages of validation both quantitative and qualitative. The study highlights the importance of considering all the parameters responsible for driving coastal evolution and the necessity to have long term monitoring results for trend based validation.
基金supported by the CAS/SAFEA International Partnership Program for Creative Research Teams (CXTD-Z2005-2-4)
文摘Many of the important questions facing farming systems in the world today require long-term studies to provide meaningful information and answers. A long-term agronomic experiment (LTAE) should (1) have long-term objectives; (2) study important soil processes or ecological processes; and (3) be related to the productivity and sustainability of systems. A well established LTAE can provide both insights into how the system operates and foresight into where the system goes. The prerequisites for setting up a LTAE are the secured land, continuous funding and dedicated scientists. A number of principles must be considered carefully when establishing a LTAE, (1) the site must be representative of large areas; (2) the treatments should be simple, but focusing on the big questions; (3) the plots should be large enough to allow subsequent modification of the experiment if this becomes necessary; (4) crop rotations should minimise, wherever possible, the risk of build-up of pests and diseases, and rotational phase should be considered in a rotational experiment; (5) a clearly defined experimental protocol should be developed to ensure data collected is scientifically valid and statistically analysable, but with flexibility to allow essential changes; (6) soil samples, possibly plant samples, should be achieved to provide better answer to the original questions when new, perhaps more accurate analytical techniques are developed, or answer new research questions that were not considered in the original design. The MASTER experiment in Australia was used as a case study to demonstrate how these principles are implemented in practice.
文摘脱硫过程是具有高度动态非线性和较大延迟时间的复杂工业过程,为了解决烟气脱硫过程的建模问题,设计了注意力机制下的深度长短期记忆(attention mechanism-based long short-term memory,AttLSTM)网络,并基于该网络设计自动编码器,完成脱硫过程异常点的检测。该文首次提出使用AttLSTM网络自编码器对脱硫过程进行离群点检测,并且该网络模型同样首次应用于脱硫过程的辨识任务中。从更深的意义上讲,该文尝试使用深度学习模型对复杂系统进行辨识,所建立的AttLSTM网络之前未出现在系统辨识领域,该网络的出现可以丰富辨识模型的选择,同时为人工智能技术在系统辨识领域和控制领域的应用与推广提供参考。实验结果表明,相比于之前文献出现的脱硫过程建模方法,所提方法在不同性能指标上均具有更好的表现,由此可以证明深度AttLSTM网络在脱硫场景下的有效性。
文摘由于电池荷电状态(state of charge,SOC)无法直接测量,且传统的SOC估算方法精度低。为了提升锂离子电池SOC估算精度,对比了不同深度学习网络模型应用于SOC估算的效果,并提出了一种基于DRSN-CW-LSTM网络的锂离子电池SOC估算方法。该方法基于长短期记忆网络(long-short-term memory,LSTM)和逐通道不同阈值的深度残差收缩网络(deep residual shrinkage networks with channel-wise thresholds,DRSN-CW),利用锂离子电池电压、电流、温度、容量等数据信息在深度残差收缩网路中进行特征提取,通过LSTM进一步拟合时间序列数据趋势,实现锂离子电池在使用周期内SOC的预测。在DRSN-CW网络的残差收缩模块中可以实现自适应噪声数据处理功能,消除锂离子电池数据流质量对SOC预测的负面影响。利用锂电池公共数据集训练所提出的网络,对比了3种神经网络模型在该两组数据集上的预测效果。实验结果表明,所提出的深度学习模型在两组公开数据集上的MAE和RMSE均值都控制在5%以内,相比其他3种深度学习模型有更好的抗噪性能和预测性能,且估算精度高。