scholarly journals Research on Runoff Simulations Using Deep-Learning Methods

2021 ◽  
Vol 13 (3) ◽  
pp. 1336
Author(s):  
Yan Liu ◽  
Ting Zhang ◽  
Aiqing Kang ◽  
Jianzhu Li ◽  
Xiaohui Lei
Keyword(s):  
Research Area ◽  
Data Driven ◽  
Distributed Model ◽  
Model Parameters ◽  
Optimal Parameters ◽  
Output Data ◽  
Runoff Modeling ◽  
Model Component ◽  
Hanjiang River ◽  
Lstm Network

Runoff simulations are of great significance to the planning management of water resources. Here, we discussed the influence of the model component, model parameters and model input on runoff modeling, taking Hanjiang River Basin as the research area. Convolution kernel and attention mechanism were introduced into an LSTM network, and a new data-driven model Conv-TALSTM was developed. The model parameters were analyzed based on the Conv-TALSTM, and the results suggested that the optimal parameters were greatly affected by the correlation between the input data and output data. We compared the performance of Conv-TALSTM and variant models (TALSTM, Conv-LSTM, LSTM), and found that Conv-TALSTM can reproduce high flow more accurately. Moreover, the results were comparable when the model was trained with meteorological or hydrological variables, whereas the peak values with hydrological data were closer to the observations. When the two datasets were combined, the performance of the model was better. Additionally, Conv-TALSTM was also compared with an ANN (artificial neural network) and Wetspa (a distributed model for Water and Energy Transfer between Soil, Plants and Atmosphere), which verified the advantages of Conv-TALSTM in peak simulations. This study provides a direction for improving the accuracy, simplifying model structure and shortening calculation time in runoff simulations.

scholarly journals Improving WetSpa model to predict streamflows for gaged and ungaged catchments

2013 ◽  
Vol 16 (4) ◽  
pp. 758-771 ◽  
Author(s):  
Alireza Safari ◽  
F. De Smedt
Keyword(s):  
Model Performance ◽  
Flow Equation ◽  
Distributed Model ◽  
Model Parameters ◽  
Second Phase ◽  
Modified Model ◽  
Wetspa Model ◽  
Runoff Modeling ◽  
Intercomparison Project ◽  
High Flows

In the second phase of the Distributed Model Intercomparison Project (DMIP2), the WetSpa model is applied to simulate flows at basin and subbasin scales. Parent basins and their nested subbasins are modeled as gaged and ungaged basins, respectively. Available observations in the subbasins were only used to validate the model predictions. Gaged basins simulation results show that the predictions and observations are in good agreement. However, major peaks are underestimated, as is often the case in runoff modeling. Underestimation of high flows and in particular peak flows indicates that the prediction of actual runoff coefficients in the current WetSpa model needs to be improved. Also, sensitivity analysis of the model parameters reveals that the baseflow recession coefficient is the most sensitive parameter and care should be taken when modeling ungaged basins. Hence, by estimating this parameter for each subbasin separately, the model performance for the subbasins can be improved. To do this, a Boussinesq groundwater flow equation is used to improve the prediction of baseflow recession coefficients in the subbasins. Comparison between the original and the modified WetSpa models shows that the modified model yields relatively higher performances for the subbasins, creating a more accurate model for predicting ungaged subbasins.

A Review and evaluation of Machine Translation methods for Lumasaaba

2020 ◽  
pp. 3-17
Author(s):  
Peter Nabende
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Machine Translation ◽  
Language Processing ◽  
Research Area ◽  
Data Driven ◽  
East African ◽  
Data Set ◽  
African Languages ◽  
Translation Methods

Natural Language Processing for under-resourced languages is now a mainstream research area. However, there are limited studies on Natural Language Processing applications for many indigenous East African languages. As a contribution to covering the current gap of knowledge, this paper focuses on evaluating the application of well-established machine translation methods for one heavily under-resourced indigenous East African language called Lumasaaba. Specifically, we review the most common machine translation methods in the context of Lumasaaba including both rule-based and data-driven methods. Then we apply a state of the art data-driven machine translation method to learn models for automating translation between Lumasaaba and English using a very limited data set of parallel sentences. Automatic evaluation results show that a transformer-based Neural Machine Translation model architecture leads to consistently better BLEU scores than the recurrent neural network-based models. Moreover, the automatically generated translations can be comprehended to a reasonable extent and are usually associated with the source language input.

Chaotic motion control of a vibro-impact system based on data-driven control method

2021 ◽  
pp. 107754632110433
Author(s):  
Xiao-juan Wei ◽  
Ning-zhou Li ◽  
Wang-cai Ding
Keyword(s):  
Motion Control ◽  
Data Model ◽  
Chaotic Motion ◽  
Control Method ◽  
Data Driven ◽  
Damping Coefficient ◽  
Input Output ◽  
Output Data ◽  
Controlled System ◽  
Impact System

For the chaotic motion control of a vibro-impact system with clearance, the parameter feedback chaos control strategy based on the data-driven control method is presented in this article. The pseudo-partial-derivative is estimated on-line by using the input/output data of the controlled system so that the compact form dynamic linearization (CFDL) data model of the controlled system can be established. And then, the chaos controller is designed based on the CFDL data model of the controlled system. And the distance between two adjacent points on the Poincaré section is used as the judgment basis to guide the controller to output a small perturbation to adjust the damping coefficient of the controlled system, so the chaotic motion can be controlled to a periodic motion by dynamically and slightly adjusting the damping coefficient of the controlled system. In this method, the design of the controller is independent of the order of the controlled system and the structure of the mathematical model. Only the input/output data of the controlled system can be used to complete the design of the controller. In the simulation experiment, the effectiveness and feasibility of the proposed control method in this article are verified by simulation results.

scholarly journals Conditioning rainfall-runoff model parameters for ungauged catchments and land management impacts analysis

2009 ◽  
Vol 13 (6) ◽  
pp. 893-904 ◽  
Author(s):  
N. Bulygina ◽  
N. McIntyre ◽  
H. Wheater
Keyword(s):  
Parameter Space ◽  
Land Management ◽  
Base Flow ◽  
Distributed Model ◽  
Flow Index ◽  
Model Parameters ◽  
Ungauged Catchments ◽  
Management Interventions ◽  
Simulated Event ◽  
Base Flow Index

Abstract. Data scarcity and model over-parameterisation, leading to model equifinality and large prediction uncertainty, are common barriers to effective hydrological modelling. The problem can be alleviated by constraining the prior parameter space using parameter regionalisation. A common basis for regionalisation in the UK is the HOST database which provides estimates of hydrological indices for different soil classifications. In our study, Base Flow Index is estimated from the HOST database and the power of this index for constraining the parameter space is explored. The method is applied to a highly discretised distributed model of a 12.5 km2 upland catchment in Wales. To assess probabilistic predictions against flow observations, a probabilistic version of the Nash-Sutcliffe efficiency is derived. For six flow gauges with reliable data, this efficiency ranged between 0.70 and 0.81, and inspection of the results shows that the model explains the data well. Knowledge of how Base Flow Index and interception losses may change under future land use management interventions was then used to further condition the model. Two interventions are considered: afforestation of grazed areas, and soil degradation associated with increased grazing intensity. Afforestation leads to median reduction in modelled runoff volume of 24% over the simulated 3 month period; and a median peak flow reduction ranging from 12 to 15% over the six gauges for the largest simulated event. Uncertainty in all results is low compared to prior uncertainty and it is concluded that using Base Flow Index estimated from HOST is a simple and potentially powerful method of conditioning the parameter space under current and future land management.

Dynamic Data-Driven Spatiotemporal System Behavior Prediction With Simulations and Sensor Measurement Data

2018 ◽  
Author(s):  
Xiangxue Zhao ◽  
Shapour Azarm ◽  
Balakumar Balachandran
Keyword(s):  
Dynamical System ◽  
Real Time ◽  
Measurement Data ◽  
Data Driven ◽  
Sensor Data ◽  
Model Parameters ◽  
High Fidelity ◽  
System Behavior ◽  
Sensor Measurement

Online prediction of dynamical system behavior based on a combination of simulation data and sensor measurement data has numerous applications. Examples include predicting safe flight configurations, forecasting storms and wildfire spread, estimating railway track and pipeline health conditions. In such applications, high-fidelity simulations may be used to accurately predict a system’s dynamical behavior offline (“non-real time”). However, due to the computational expense, these simulations have limited usage for online (“real-time”) prediction of a system’s behavior. To remedy this, one possible approach is to allocate a significant portion of the computational effort to obtain data through offline simulations. The obtained offline data can then be combined with online sensor measurements for online estimation of the system’s behavior with comparable accuracy as the off-line, high-fidelity simulation. The main contribution of this paper is in the construction of a fast data-driven spatiotemporal prediction framework that can be used to estimate general parametric dynamical system behavior. This is achieved through three steps. First, high-order singular value decomposition is applied to map high-dimensional offline simulation datasets into a subspace. Second, Gaussian processes are constructed to approximate model parameters in the subspace. Finally, reduced-order particle filtering is used to assimilate sparsely located sensor data to further improve the prediction. The effectiveness of the proposed approach is demonstrated through a case study. In this case study, aeroelastic response data obtained for an aircraft through simulations is integrated with measurement data obtained from a few sparsely located sensors. Through this case study, the authors show that along with dynamic enhancement of the state estimates, one can also realize a reduction in uncertainty of the estimates.

scholarly journals ONLINE PARAMETER IDENTIFICATION OF RICE TRANSPLANTER MODEL BASED ON IPSO-EKF ALGORITHM

2020 ◽  
Vol 61 (2) ◽  
pp. 25-34 ◽  
Author(s):  
Yibo Li ◽  
Hang Li ◽  
Xiaonan Guo
Keyword(s):  
Parameter Identification ◽  
Measured Data ◽  
Model Parameters ◽  
Identification Algorithm ◽  
Output Data ◽  
Swarm Optimization ◽  
Model Based ◽  
Simulation Results ◽  
Filtering Effect ◽  
Ipso Algorithm

In order to improve the accuracy of rice transplanter model parameters, an online parameter identification algorithm for the rice transplanter model based on improved particle swarm optimization (IPSO) algorithm and extended Kalman filter (EKF) algorithm was proposed. The dynamic model of the rice transplanter was established to determine the model parameters of the rice transplanter. Aiming at the problem that the noise matrices in EKF algorithm were difficult to select and affected the best filtering effect, the proposed algorithm used the IPSO algorithm to optimize the noise matrices of the EKF algorithm in offline state. According to the actual vehicle tests, the IPSO-EKF was used to identify the cornering stiffness of the front and rear tires online, and the identified cornering stiffness value was substituted into the model to calculate the output data and was compared with the measured data. The simulation results showed that the accuracy of parameter identification for the rice transplanter model based on the IPSO-EKF algorithm was improved, and established an accurate rice transplanter model.

scholarly journals An integrated framework for building trustworthy data-driven epidemiological models: Application to the COVID-19 outbreak in New York City

2021 ◽  
Author(s):  
Sheng Zhang ◽  
Joan Ponce ◽  
Zhen Zhang ◽  
Guang Lin ◽  
George Karniadakis
Keyword(s):  
New York ◽  
New York City ◽  
York City ◽  
Robustness Analysis ◽  
Data Driven ◽  
Model Parameters ◽  
Calibration Model ◽  
Epidemiological Models ◽  
Analysis Model ◽  
Identifiability Analysis

AbstractEpidemiological models can provide the dynamic evolution of a pandemic but they are based on many assumptions and parameters that have to be adjusted over the time when the pandemic lasts. However, often the available data are not sufficient to identify the model parameters and hence infer the unobserved dynamics. Here, we develop a general framework for building a trustworthy data-driven epidemiological model, consisting of a workflow that integrates data acquisition and event timeline, model development, identifiability analysis, sensitivity analysis, model calibration, model robustness analysis, and forecasting with uncertainties in different scenarios. In particular, we apply this framework to propose a modified susceptible–exposed–infectious–recovered (SEIR) model, including new compartments and model vaccination in order to forecast the transmission dynamics of COVID-19 in New York City (NYC). We find that we can uniquely estimate the model parameters and accurately predict the daily new infection cases, hospitalizations, and deaths, in agreement with the available data from NYC’s government’s website. In addition, we employ the calibrated data-driven model to study the effects of vaccination and timing of reopening indoor dining in NYC.

scholarly journals NetPyNE, a tool for data-driven multiscale modeling of brain circuits

eLife ◽  
2019 ◽  
Vol 8 ◽  
Author(s):  
Salvador Dura-Bernal ◽  
Benjamin A Suter ◽  
Padraig Gleeson ◽  
Matteo Cantarelli ◽  
Adrian Quintana ◽  
...  
Keyword(s):  
Multiple Scales ◽  
Network Models ◽  
Brain Regions ◽  
Data Driven ◽  
Model Parameters ◽  
Biophysical Modeling ◽  
Neuron Network ◽  
Graphical Interfaces ◽  
Information Theoretic Measures ◽  
High Level

Biophysical modeling of neuronal networks helps to integrate and interpret rapidly growing and disparate experimental datasets at multiple scales. The NetPyNE tool (www.netpyne.org) provides both programmatic and graphical interfaces to develop data-driven multiscale network models in NEURON. NetPyNE clearly separates model parameters from implementation code. Users provide specifications at a high level via a standardized declarative language, for example connectivity rules, to create millions of cell-to-cell connections. NetPyNE then enables users to generate the NEURON network, run efficiently parallelized simulations, optimize and explore network parameters through automated batch runs, and use built-in functions for visualization and analysis – connectivity matrices, voltage traces, spike raster plots, local field potentials, and information theoretic measures. NetPyNE also facilitates model sharing by exporting and importing standardized formats (NeuroML and SONATA). NetPyNE is already being used to teach computational neuroscience students and by modelers to investigate brain regions and phenomena.

scholarly journals Aircraft Engines Remaining Useful Life Prediction with an Improved Online Sequential Extreme Learning Machine

2020 ◽  
Vol 10 (3) ◽  
pp. 1062 ◽  
Author(s):  
Tarek Berghout ◽  
Leïla-Hayet Mouss ◽  
Ouahab Kadri ◽  
Lotfi Saïdi ◽  
Mohamed Benbouzid
Keyword(s):  
Extreme Learning Machine ◽  
Life Prediction ◽  
Remaining Useful Life ◽  
Data Driven ◽  
Selection Strategy ◽  
Mapping Technique ◽  
Model Parameters ◽  
Aircraft Engines ◽  
Useful Life ◽  
Learning Machine

The efficient data investigation for fast and accurate remaining useful life prediction of aircraft engines can be considered as a very important task for maintenance operations. In this context, the key issue is how an appropriate investigation can be conducted for the extraction of important information from data-driven sequences in high dimensional space in order to guarantee a reliable conclusion. In this paper, a new data-driven learning scheme based on an online sequential extreme learning machine algorithm is proposed for remaining useful life prediction. Firstly, a new feature mapping technique based on stacked autoencoders is proposed to enhance features representations through an accurate reconstruction. In addition, to attempt into addressing dynamic programming based on environmental feedback, a new dynamic forgetting function based on the temporal difference of recursive learning is introduced to enhance dynamic tracking ability of newly coming data. Moreover, a new updated selection strategy was developed in order to discard the unwanted data sequences and to ensure the convergence of the training model parameters to their appropriate values. The proposed approach is validated on the C-MAPSS dataset where experimental results confirm that it yields satisfactory accuracy and efficiency of the prediction model compared to other existing methods.

scholarly journals A Data-Driven Scheduling Approach for Hydrogen Penetrated Energy System Using LSTM Network

2019 ◽  
Vol 11 (23) ◽  
pp. 6784
Author(s):  
Suyang Zhou ◽  
Di He ◽  
Zhiyang Zhang ◽  
Zhi Wu ◽  
Wei Gu ◽  
...  
Keyword(s):  
Value At Risk ◽  
High Speed ◽  
Energy System ◽  
Data Driven ◽  
Mixed Integer ◽  
Conditional Value At Risk ◽  
Data Driven Approach ◽  
Lstm Network ◽  
Rolling Optimization ◽  
Optimization Mechanism

Intra-day control and scheduling of energy systems require high-speed computation and strong robustness. Conventional mathematical driven approaches usually require high computation resources and have difficulty handling system uncertainties. This paper proposes two data-driven scheduling approaches for hydrogen penetrated energy system (HPES) operational scheduling. The two data-driven approaches learn the historical optimization results calculated out using the mixed integer linear programing (MILP) and conditional value at risk (CVaR), respectively. The intra-day rolling optimization mechanism is introduced to evaluate the proposed data-driven scheduling approaches, MILP data-driven approach and CVaR data-driven approach, along with the forecasted renewable generation and load demands. Results show that the two data-driven approaches have lower intra-day operational costs compared with the MILP based method by 1.17% and 0.93%. In addition, the combined cooling and heating plant (CCHP) has a lower frequency of changing the operational states and power output when using the MILP data-driven approach compared with the mathematical driven approaches.

Sign in / Sign up

Export Citation Format

Share Document