Interpretable and explainable hybrid model for daily streamflow prediction based on multi-factor drivers.

Streamflow time series data typically exhibit nonlinear and nonstationary characteristics that complicate precise estimation. Recently, multifactorial machine learning (ML) models have been developed to enhance the performance of streamflow predictions. However, the lack of interpretability within these ML models raises concerns about their inner workings and reliability. This paper introduces an innovative hybrid architecture, the TCN-LSTM-Multihead-Attention model, which combines two layers of temporal convolutional networks (TCN) followed by one layer of long short-term memory (LSTM) units, integrated with a Multihead-Attention mechanism for predicting streamflow with streamflow causation-driven prediction samples (RCDP), employing local and global interpretability studies through Shapley values and partial dependency analysis. The find_peaks method was used to identify peak flow events in the test dataset, validating the model's generality and uncovering the physical causative patterns of streamflow. The results show that (1) compared to the LSTM model with the same hyperparameter settings, the proposed TCN-LSTM-Multihead-Attention hybrid model increased the R2 by 52.9%, 2.5%, 43.1%, and 10.7% respectively at four stations in the test set predictions using RCDP samples. Moreover, comparing the prediction results of the hybrid model under different samples in Hengshan station, the R2 for RCDP increased by 5.06% and 1.22% compared to streamflow autoregressive prediction samples (RAP) and meteorological-soil volumetric water content coupled autoregressive prediction samples (MCSAP) respectively. (2) Historical streamflow data from the preceding 3 days predominantly influences predictions due to strong autocorrelation, with flow quantity (Q) typically emerging as the most significant feature alongside precipitation (P), surface soil moisture (SSM), and adjacent station flow data. (3) During periods of low and normal flow, historical data remains the most crucial factor; however, during flood periods, the roles of upstream inflow and precipitation become significantly more pronounced. This model facilitates the identification and quantification of various hydrodynamic impacts on flow predictions, including upstream flood propagation, precipitation, and soil moisture conditions. It also elucidates the model's nonlinear relationships and threshold responses, thereby enhancing the interpretability and reliability of streamflow predictions.

Numerical Study of Viscoelastic Microfluidic Particle Manipulation in a Microchannel with Asymmetrical Expansions.

Microfluidic microparticle manipulation is currently widely used in environmental, bio-chemical, and medical applications. Previously we proposed a straight microchannel with additional triangular cavity arrays to manipulate microparticles with inertial microfluidic forces, and experimentally explored the performances within different viscoelastic fluids. However, the mechanism remained poorly understood, which limited the exploration of the optimal design and standard operation strategies. In this study, we built a simple but robust numerical model to reveal the mechanisms of microparticle lateral migration in such microchannels. The numerical model was validated by our experimental results with good agreement. Furthermore, the force fields under different viscoelastic fluids and flow rates were carried out for quantitative analysis. The mechanism of microparticle lateral migration was revealed and is discussed regarding the dominant microfluidic forces, including drag force, inertial lift force, and elastic force. The findings of this study can help to better understand the different performances of microparticle migration under different fluid environments and complex boundary conditions.

Mechanism and Effects of Cellular Creep in a Microfluidic Filter.

Biomicroparticles such as proteins, bacterium, and cells are known to be viscoelastic, which significantly affects their performance in microfluidic applications. However, the exact effects and the quantitative study of cellular viscoelastic creep within different applications remain unclear. In this study, the cellular-deforming evolution within a filter unit was studied using a multiphysics numerical model. A general cellular creep deformation process of viscoelastic particle trapping in pores was revealed. Two featured variables, namely, the maximum surface displacement and the volumetric strain, were identified and determined to quantitatively describe the evolution. The effects of flow conditions and physical characteristics of the microparticles were studied. Furthermore, a Giardia concentration experiment was conducted using an integrated hydraulic filtration system with a porous membrane. The experimental results agreed well with the numerical analysis, indicating that, compared to pure elastic particles, it is more difficult to release cellular material matters including cells, chemical synthetic particles, and microbes from trapping due to their time-accumulated creep deformation.