Multiple Time Steps Research Articles

The Lehtinen–Pirjola method modified for efficient modelling of geomagnetically induced currents in multiple voltage levels of a power network

Abstract. The need for accurate assessment of the geomagnetic hazard to power systems is driving a requirement to model geomagnetically induced currents (GIC) in multiple voltage levels of a power network. The Lehtinen–Pirjola method for modelling GIC is widely used but was developed when the main aim was to model GIC in only the highest voltage level of a power network. Here we present a modification to the Lehtinen–Pirjola (LP) method designed to provide an efficient method for modelling GIC in multiple voltage levels. The LP method calculates the GIC flow to ground from each node. However, with a network involving multiple voltage levels, many of the nodes are ungrounded, i.e. have infinite resistance to ground, which is numerically inconvenient. The new modified Lehtinen–Pirjola (LPm) method replaces the earthing impedance matrix [Ze] with the corresponding earthing admittance matrix [Ye] in which the ungrounded nodes have zero admittance to ground. This is combined with the network admittance matrix [Yn] to give a combined matrix ([Yn] + [Ye]), which is a sparse symmetric positive definite matrix allowing efficient techniques, such as Cholesky decomposition, to be used to provide the nodal voltages. The nodal voltages are then used to calculate the GIC in the transformer windings and the transmission lines of the power network. The LPm method with Cholesky decomposition also provides an efficient method for calculating GIC at multiple time steps. Finally, the paper shows how software for the LP method can be easily converted to the LPm method and provides examples of calculations using the LPm method.

A system for electric vehicle’s energy-aware routing in a transportation network through real-time prediction of energy consumption

To tackle the problem of range anxiety of a driver of an electric vehicle (EV), it is necessary to accurately estimate the power/energy consumption of EVs in real time, so that drivers can get real-time information about the vehicle’s remaining range. In addition, it can be used for energy-aware routing, i.e., the driver can be provided with information that on which route less energy consumption will take place. In this paper, an integrated system has been proposed which can provide reliable and real-time estimate of the energy consumption for an EV. The approach uses Deep Auto-Encoders (DAE), cross-connected using latent space mapping, which consider historical traffic speed to predict the traffic speed at multiple time steps in future. The predicted traffic speed is used to calculate the future vehicle speed. The vehicle speed, acceleration along with wind speed, road elevation, temperature, battery’s SOC, and auxiliary loads are used as input to a multi-channel Convolutional Neural Network (CNN) to predict the energy consumption. The prediction is further fine-tuned using a Bagged Decision Tree (BDT). Unlike other existing techniques, the proposed system can be easily generalized for other vehicles as it is independent of internal vehicle parameters. Comparison with other benchmark techniques shows that the proposed system performs better and has a least mean absolute percentage error of 1.57%.

Prediction of gas concentration evolution with evolutionary attention-based temporal graph convolutional network

Accurate prediction of gas concentration is of great importance in many safe-based systems and applications. However, prediction accuracy of gas concentration is limited by not only the temporal evolution of gas concentration but also the spatial characteristics of gas dispersion. To capture the spatial and temporal dependences simultaneously, an evolutionary attention-based temporal graph convolutional network (EAT-GCN) is proposed, which has three outstanding features: (1) graph convolutional network (GCN) is used to capture spatial dependence by learning topological structures of a gas sensor network; (2) gated recurrent unit (GRU) is adopted to retain temporal dependence by learning dynamic changes of gas concentration, and (3) evolutionary attention is introduced to improve the ability of GRU to pay different degrees of attention to the sub-window features within multiple time steps. Finally, kernel extrapolation distribution mapping algorithm is employed to visualize the predicted results of gas concentration and update the gas distribution map. Compared with CNN, GCN, GRU, T-GCN, A3T-GCN and EA-GRU models, the proposed EAT-GCN model improves the prediction accuracy by 13.46%, 124.21%, 33.92%, 23.39%, 46.63%, and 23.97%, respectively. Experiments demonstrate that the designed model captures spatiotemporal correlation from gas concentration data and achieves better prediction accuracy than state-of-the-art baseline methods.

Molecular arrangement and fringe identification and analysis from molecular dynamics (MAFIA-MD): A tool for analyzing the molecular structures formed during reactive molecular dynamics simulation of hydrocarbons

Reactive molecular dynamics (RMD) simulations are becoming popular, with the recent developments in high-performance and efficient computing architecture, for investigating fundamental physico-chemical behaviors of reacting species. Due to the complexity of hydrocarbon systems, characterization of molecules formed during RMD simulations of hydrocarbons can be very challenging for simulations involving a large number of molecules. The novel post-processing utility “MAFIA-MD” – an acronym for “Molecular Arrangement and Fringe Identification and Analysis from Molecular Dynamics” – presented in this manuscript can analyze and perform molecular characterization of a large number of RMD trajectory files (.XYZ files) simultaneously. The utility will be useful for analysis and chemical characterization of trajectories without the large bond information files. A graphical user interface (GUI) is developed for easy operation. The utility analyzes the existing cyclic structures in the domain and generates statistics of alicyclic and aromatic hydrocarbons based on the number of carbon atoms. Alongside the statistical analysis, the program also separates the constituent molecules and extracts their chemical information in terms of Simplified Molecular-Input Line-Entry System (SMILES) and Spatial Data File (.SDF)”. A methodology for calculating fringe spacing is also implemented in the code for validation of RMD simulation with High-Resolution Transmission Electron Microscopy (HRTEM) images. Program summaryProgram title:MAFIA-MDCPC Library link to program files:https://doi.org/10.17632/s7dsk553fh.1Developer's repository link:https://github.com/kmmukut/MAFIA-MD.gitLicensing provisions: MITProgramming language:Python 3.7Supplementary material: Bash script for splitting long continuous trajectory files with multiple timestep into trajectory files for individual timestep and some example input trajectory files for testing.Nature of problem: Reactive molecular dynamics (RMD) simulations of hydrocarbons allow chemical reactions between different hydrocarbon molecules through bond breaking and new bond formation. The information about the chemical changes is stored in the bond information files and trajectory of individual atoms are stored in trajectory files. While the bond files can become very large for large simulations, trajectory files remain quite small. MAFIA-MD will be useful for practitioners interested in analyzing the chemical structures of emerging molecular clusters from small individual trajectory files without requiring the large, memory-intensive bond information files. Identification of chemical characteristics, particularly aromatic and alicyclic ring structures and molecular fringes, are important for gaining insights from RMD simulations of hydrocarbons, especially in exploration of soot formation during combustion. The capability of isolating the cyclic and non-cyclic molecules from trajectory files is the novel part of MAFIA-MD.Solution method: The coordinates of individual atoms are taken as the input for the program. The coordinates are then analyzed to create an adjacency matrix [1] based on the bond distance between carbon atoms. This adjacency matrix is then converted into a directed graph and the cyclic structures are extracted from the graph by implementing an efficient depth-first-search algorithm [2] developed by Johnson [3]. Apart from the ability to isolate cyclic and non-cyclic components by chemical structural analysis, a methodology to calculate fringe spacing is also implemented by calculating the distances between cyclic structures. The utility uses a Python interface for easy management of the input and output parameters. The detailed solution methodology has been discussed in the manuscript.Additional comments including restrictions and unusual features: In the current version, the utility can analyze only carbon and hydrogen atoms. The future versions will implement the addition of other common atoms found in aromatic molecules, e.g. nitrogen, oxygen, sulfur, etc. MAFIA-MD can be run on Linux, macOS, and Windows operating systems.

Power analysis for detecting the effects of best management practices on reducing nitrogen and phosphorus fluxes to the Chesapeake Bay Watershed, USA

In 2010 the U.S. Environmental Protection Agency established the Total Maximum Daily Load (TMDL) which is a “pollution diet” that aims to reduce the amount of nitrogen and phosphorus entering the Chesapeake Bay, the largest estuary in the United States, by 25 and 24% percent, respectively. To achieve this goal the TMDL requires the implementation of Best Management Practices (BMPs), which are accepted land management practices for reducing pollutant runoff to nearby bodies of water. While the TMDL requires that the necessary management actions be in place by 2025 to eventually reach targeted nutrient loads, the ability to detect an effect of BMPs while assuming that one has occurred (i.e. statistical power) is still not well understood. The goal of this study was to investigate the power and required timelines to detect nutrient reductions in streams and rivers as the result of BMP implementation at the Chesapeake Watershed scale. Power estimates were produced using SPAtially Referenced Regression On Watershed attributes (SPARROW) models, which offer a flexible statistical framework and were recently extended to allow for modeling multiple time steps. Nitrogen and phosphorus focused models were calibrated to estimate the power to detect reductions in flux from numerous constituent sources. To confidently detect a decrease in constituent flux reaching the Chesapeake Bay’s tidal waters from a specific constituent source, reductions ranging from 30–60% were required for the nitrogen model. In contrast, reductions of up to 80% were not detectable under the phosphorus model. The timelines necessary to detect reductions in nitrogen flux ranged from 11 to several hundred years under different rates-of-change and management scenarios. The approach proposed here can help better understand the ability to detect the effects of BMPs on a regional scale and help guide future management actions and monitoring programs.

Large-scale simulation of graphene and structural superlubricity with improved smoothed molecular dynamics method

Smoothed molecular dynamics (SMD) method is an efficient atomic simulation method by introducing mapping strategy between atoms and one set of background mesh into molecular dynamics (MD) procedure. Much larger feasible time step can be adopted in SMD method because of the smoothing effect for high-frequency modes of motion. A dual-mesh SMD method is proposed to improve the accuracy of graphene simulation. The multiple-time-step (MTS) integration scheme is adopted for MD–SMD coupling and an adaptive scheme for identifying MD region is developed. An alternating load balancing strategy is proposed to ensure high parallel efficiency when the MTS coupling method is executed on parallel computers. The proposed method is validated with several typical examples of single- and multiple-layer graphene as well as the graphene structural superlubricity problems.

Improving the scaling and performance of multiple time stepping‐based molecular dynamics with hybrid density functionals

Density functionals at the level of the generalized gradient approximation (GGA) and a plane-wave basis set are widely used today to perform ab initio molecular dynamics (AIMD) simulations. Going up in the ladder of accuracy of density functionals from GGA (second rung) to hybrid density functionals (fourth rung) is much desired pertaining to the accuracy of the latter in describing structure, dynamics, and energetics of molecular and condensed matter systems. On the other hand, hybrid density functional based AIMD simulations are about two orders of magnitude slower than GGA based AIMD for systems containing ~100 atoms using ~100 compute cores. Two methods, namely MTACE and s-MTACE, based on a multiple time step integrator and adaptively compressed exchange operator formalism are able to provide a speed-up of about 7-9 in performing hybrid density functional based AIMD. In this work, we report an implementation of these methods using a task-group based parallelization within the CPMD program package, with the intention to take advantage of the large number of compute cores available on modern high-performance computing platforms. We present here the boost in performance achieved through this algorithm. This work also identifies the computational bottleneck in the s-MTACE method and proposes a way to overcome it.

An Efficient Gaussian-Accelerated Molecular Dynamics (GaMD) Multilevel Enhanced Sampling Strategy: Application to Polarizable Force Fields Simulations of Large Biological Systems.

We introduce a novel multilevel enhanced sampling strategy grounded on Gaussian-accelerated Molecular Dynamics (GaMD). First, we propose a GaMD multi-GPUs-accelerated implementation within the Tinker-HP molecular dynamics package. We introduce the new "dual-water" mode and its use with the flexible AMOEBA polarizable force field. By adding harmonic boosts to the water stretching and bonding terms, it accelerates the solvent-solute interactions while enabling speedups, thanks to the use of fast multiple-time step integrators. To further reduce the time-to-solution, we couple GaMD to Umbrella Sampling (US). The GaMD─US/dual-water approach is tested on the 1D Potential of Mean Force (PMF) of the solvated CD2-CD58 system (168 000 atoms), allowing the AMOEBA PMF to converge within 1 kcal/mol of the experimental value. Finally, Adaptive Sampling (AS) is added, enabling AS-GaMD capabilities but also the introduction of the new Adaptive Sampling-US-GaMD (ASUS-GaMD) scheme. The highly parallel ASUS-GaMD setup decreases time to convergence by, respectively, 10 and 20 times, compared to GaMD-US and US. Overall, beside the acceleration of PMF computations, Tinker-HP now allows for the simultaneous use of Adaptive Sampling and GaMD-"dual water" enhanced sampling approaches increasing the applicability of polarizable force fields to large-scale simulations of biological systems.

WEB.BM – a web-based river basin management model with multiple time-step optimization and the SSARR channel routing options

ABSTRACT This paper presents the WEB.BM model for multipurpose multireservoir river basin management that uses Mixed Integer Linear Programming (MILP) solver to optimize basin operation over single or multiple time steps, with flexible time step length and optional use of both reservoir and hydrological channel routing. The principal novelty is the capability of the model to solve multiple time step optimization while simultaneously properly accounting for hydrological channel routing as constraints to optimization, which is demonstrated on two numerical tests presented in this paper, which can be used as benchmark by other researchers. There are no other known optimization-based river basin management models with this capability. The model can include multiple reservoirs, hydropower plants, environmental flow targets, return flows from irrigation, diversion volume licences, apportionment agreements between neighbouring states and net evaporation. Operating priorities among users are represented by a selection of user-defined weight factors. The model can be accessed at www.riverbasinmanagement.com.

A multiple time step algorithm for trajectory surface hopping simulations.

A multiple time step (MTS) algorithm for trajectory surface hopping molecular dynamics has been developed, implemented, and tested. The MTS scheme is an extension of the ab initio implementation for Born-Oppenheimer molecular dynamics presented in the work of Liberatore et al. [J. Chem. Theory Comput. 14, 2834 (2018)]. In particular, the MTS algorithm has been modified to enable the simulation of non-adiabatic processes with the trajectory surface hopping (TSH) method and Tully's fewest switches algorithm. The specificities of the implementation lie in the combination of Landau-Zener and Tully's transition probabilities during the inner MTS time steps. The new MTS-TSH method is applied successfully to the photorelaxation of protonated formaldimine, showing that the important characteristics of the process are recovered by the MTS algorithm. A computational speed-up between 1.5 and 3 has been obtained compared to standard TSH simulations, which is close to the ideal values that could be obtained with the computational settings considered.

Multimodal Spatio-Temporal Prediction with Stochastic Adversarial Networks

Spatio-temporal (ST) data is a collection of multiple time series data with different spatial locations and is inherently stochastic and unpredictable. An accurate prediction over such data is an important building block for several urban applications, such as taxi demand prediction, traffic flow prediction, and so on. Existing deep learning based approaches assume that outcome is deterministic and there is only one plausible future; therefore, cannot capture the multimodal nature of future contents and dynamics. In addition, existing approaches learn spatial and temporal data separately as they assume weak correlation between them. To handle these issues, in this article, we propose a stochastic spatio-temporal generative model (named D-GAN) which adopts Generative Adversarial Networks (GANs)-based structure for more accurate ST prediction in multiple time steps. D-GAN consists of two components: (1) spatio-temporal correlation network which models spatio-temporal joint distribution of pixels and supports a stochastic sampling of latent variables for multiple plausible futures; (2) a stochastic adversarial network to jointly learn generation and variational inference of data through implicit distribution modeling. D-GAN also supports fusion of external factors through explicit objective to improve the model learning. Extensive experiments performed on two real-world datasets show that D-GAN achieves significant improvements and outperforms baseline models.

A Real-Time Safety-Based Optimal Velocity Model

Modeling safety-critical driver behavior at signalized intersections needs to account for the driver’s planned decision process, where a driver executes a plan to avoid collision in multiple time steps. Such a process can be embedded in the Optimal Velocity Model (OVM) that traditionally assumes that drivers base their “mental intention” on a distance gap only. We propose and evaluate a data-driven OVM based on real-time inference of roadside traffic video data. First, we extract vehicle trajectory data from roadside traffic footage through our advanced video processing algorithm (VT-Lane) for a study site in Blacksburg, VA, USA. Vehicles engaged in car-following episodes are then identified within the extracted vehicle trajectories database, and the real-time time-to-collision (TTC) is calculated for all car-following instances. Then, we analyze the driver behavior to predict the shape of the underlying TTC-based desired velocity function. A clustering approach is used to assess car-following behavior heterogeneity and understand the reasons behind outlying driving behaviors at the intersection to design our model accordingly. The results of this assessment show that the calibrated TTC-based OVM can replicate the observed driving behavior by capturing the acceleration pattern with an error 20% lower than the gap distance-based OVM.

Towards a Monte Carlo simulation of a pebble bed type high temperature gas cooled reactor using Geant4

This paper introduces the stochastic Monte Carlo (MC) modelling of a nuclear reactor core using the Geant4 framework. The simulation is exercised in the context of a High Temperature Gas Cooled Reactor (HTGCR) that uses helium (a noble gas) as a coolant and graphite as a neutron moderator. The study presents the results from the implementation of basic neutronics, scalability, geometrical discretisation for studying the spatial variation of physical parameters, time slicing and adaptation of Geant4 for correct intra-slice persistence, a scheme of integration with thermal hydraulics by workflow scheduling, validation of the thermal macroscopic cross-section behaviour, the process of fission, burn, decay, and differential energy depositions for the various physics processes, validation of the Xenon effects on the neutronics, criticality and core follow over multiple time steps.

Comparing multi-step ahead building cooling load prediction using shallow machine learning and deep learning models

Accurate building cooling load prediction is beneficial in managing optimal operation to conserve energy user and operational cost. Several physics-based and data-driven models proposed to forecast building cooling load focus on one-step-ahead prediction. Deep learning-based Long-term memory (LSTM) models are shown to perform well for one-step short-term (1Hour Ahead) building cooling load prediction. However, no prior studies have examined the prediction performance of shallow machine learning methods over deep learning models to forecast building cooling demand over multi-steps across diverse real-world datasets. A multi-step model learns a single parametric function from input time series and forecasts an array of building cooling load values (multi-step) simultaneously. A comprehensive study has been carried out to evaluate the performance of six data-driven models (2 shallow learning, 3 deep sequential learning, and 1 heuristic method) to predict multi-step long-term (1Day Ahead) building cooling load. Our results demonstrate variant of the LSTM model, the Recurrent Neural Network Multi-Input Multi-Output (RNN-MIMO) network architecture, performs consistently well compared to its deep learning counterparts and shallow machine learning techniques, both tree boosting and support vector regression. Notable conclusions from results obtained are twofold: Firstly, Long short-term memory (LSTM) based RNN-MIMO architecture performs well in both short-term (1Hour Ahead) and long-term (1Day Ahead) multi-step forecast horizon. RNN-MIMO is up to 33% more accurate, on average, in terms of mean absolute error over existing, state-of-the-art shallow machine learning models both Support Vector Regression (SVR) and tree boosting techniques (XGBoost). Our findings have significant implications for practice. Notably, machine learning models trained on one-step-ahead predictions cannot be deployed readily to predict multiple time steps into the future since longer prediction horizons impose additional training and fine-tuning efforts over each of the multiple steps. RNN-MIMO model’s ability to predict multiple time steps simultaneously eliminates the need for manual fine-tuning of individual models for each required forecast horizon.

Automated Temporal Tracking of Coherently Evolving Density Fronts in Numerical Models

AbstractOceanic density fronts can evolve, be advected, or propagate as gravity currents. Frontal evolution studies require methods to temporally track evolving density fronts. We present an automated method to temporally track these fronts from numerical model solutions. First, at all time steps contiguous density fronts are detected using an edge detection algorithm. A front event, defined as a set of sequential-in-time fronts representing a single time-evolving front, is then identified. At time stepi, a front is compared to each front at time stepi+ 1 to determine if the two fronts are matched. Anifront grid point is trackable if the minimum distance to thei+ 1 front falls within a range. Theifront is forward matched to thei+ 1 front when a sufficient number of grid points are trackable and the front moves onshore. A front event is obtained by forward tracking a front for multiple time steps. Within an event, the times that a grid point can be tracked is its connectivity and a pruning algorithm using a connectivity cutoff is applied to extract only the coherently evolving components. This tracking method is applied to a realistic 3-month San Diego Bight model solution yielding 81 front events with duration ≥ 7 h, allowing analyses of front event properties including occurrence frequency and propagation velocity. Sensitivity tests for the method’s parameters support that this method can be straightforwardly adapted to track evolving fronts of many types in other regions from both models and observations.

Soft clustering based probabilistic power flow with correlated inter temporal events

There is a growing demand to deploy renewable energy sources (RES) and plug-in electric vehicle (PEV) charging station (EVCS) in distribution networks. This, however, will change power grid landscapes and potentially impose uncertainties in short and long term grid planning. The solution to quantify randomness of intermittent events is to implement probabilistic power flow (PPF) analysis. The computation of PPF due to myriad sources of unknown uncertainties is very expensive. This computation analysis becomes more complex when inter-period temporal-spatial correlation and multivariate uncertainties are considered. This paper addresses the issue by resorting to a novel hybrid nonparametric algorithm for short-term PPF analysis in radial distribution networks. The proposed algorithm deals with probabilistic nonlinear loads, unit generators including PV panels, Wind farms, and EVCS charging pattern. Temporal correlation characterized by memory coefficient is used to model inter-event temporal correlation (IETC) over multiple time steps. Detailed uncertainty modeling of PEVs driving and their queues at EVCSs are included. The proposed method is superior over other existing numerical methods as it accounts for the multivariate inter temporal uncertainty events of RES and PEVs charging. Simulation results in a spatio-temporally correlated IEEE 34 node radial distribution system revealed that using IETC coefficients leads to improved prediction accuracy with average error less than 2.5% and improves the computation burden. The proposed method is applied to IEEE 123 node to test the performance of our proposed method. A comparative analysis is done with other well-known analytical/numerical methods. The PPF Computation time of our proposed method is shown to be faster than other methods when IETC is implemented. The results show insensitivity of our proposed method to the size of the system.

Multistep traffic speed prediction: A deep learning based approach using latent space mapping considering spatio-temporal dependencies

Traffic management in a city has become a major problem due to the increasing number of vehicles on roads. Intelligent Transportation System (ITS) can help the city traffic managers to tackle the problem by providing accurate traffic forecasts. For this, ITS requires a reliable traffic prediction algorithm that can provide accurate traffic prediction at multiple time steps based on past and current traffic data. In recent years, a number of different methods for traffic prediction have been proposed which have proved their effectiveness in terms of accuracy. However, most of these methods have either considered spatial information or temporal information only and overlooked the effect of other. In this paper, to address the above problem a deep learning based approach has been developed using both the spatial and temporal dependencies. To consider spatio-temporal dependencies, nearby road sensors at a particular instant are selected based on the attributes like traffic similarity and distance. Two pre-trained deep auto-encoders were cross-connected using the concept of latent space mapping and the resultant model was trained using the traffic data from the selected nearby sensors as input. The proposed deep learning based approach was trained using the real-world traffic data collected from loop detector sensors installed on different highways of Los Angeles and Bay Area. The traffic data is freely available from the web portal of the California Department of Transportation Performance Measurement System (PeMS). The effectiveness of the proposed approach was verified by comparing it with a number of machine/deep learning approaches. It has been found that the proposed approach provides accurate traffic prediction results even for 60-min ahead prediction with least error than other techniques.

Artificial Neural Networks as Propagators in Quantum Dynamics.

The utilization of artificial neural networks (ANNs) provides strategies for accelerating molecular simulations. Herein, ANNs are implemented as propagators of the time-dependent Schrödinger equation to simulate the quantum dynamics of systems with time-dependent potentials. These ANN propagators are trained to map nonstationary wavepackets from a given time to a future time within the discrete variable representation. Each propagator is trained for a specified time step, and iterative application of the propagator enables the propagation of wavepackets over long time scales. Such ANN propagators are developed and applied to one- and two-dimensional proton transfer systems, which exhibit nuclear quantum effects such as hydrogen tunneling. These ANN propagators are trained for either a specific time-independent potential or general potentials that can be time-dependent. Hierarchical, multiple time step algorithms enable parallelization, and the extension to higher dimensions is straightforward. This strategy is applicable to quantum dynamical simulations of diverse chemical and biological processes.

Field scale spatial wheat yield forecasting system under limited field data availability by integrating crop simulation model with weather forecast and satellite remote sensing

CONTEXTAn accurate crop yield forecast with sufficient lead time is critical for various applications, such as crop management, resources mobilization, agri-commodity trading, crop insurance, etc. Accurate yield forecasting well ahead of harvest at field scale with minimal field input data remains a challenge. OBJECTIVEThis study aimed to develop a novel prototype wheat yield forecasting system by assimilating remote sensing (RS) derived crop parameters and weather forecast into InfoCrop-Wheat crop simulation model (CSM), using minimum field measurements. METHODSThe CSM was calibrated and validated at both research farm and farmers' fields. The crop LAI was retrieved through inversion of the PROSAIL radiative transfer model from Sentinel-2A and Landsat-8 images and validated using in-situ LAI measurements. The CSM was modified to test assimilation of RS derived LAI through “Ensemble Kalman Filtering” (EnKF) and “Forcing” strategies at multiple time-steps. The RS derived LAI was not only used to correct/replace model simulated LAI but other model state variables were also adjusted accordingly. A major challenge of adjusting crop phenology based on RS derived LAI was also attempted. The WRF weather forecast was bias-corrected and incorporated into the modified model-LAI assimilation framework. Generic crop management inputs were specified to the model. Finally, the study demonstrated a workable prototype of a field scale wheat growth and yield forecasting system under limited field data availability. RESULTS AND CONCLUSIONSThe inversion of PROSAIL showed an RMSE of 0.56 m2/m2 in LAI retrievals. Model validation with measured inputs showed normalized error (NE) of 6‐–8% in grain yield. The proposed framework showed only 2%, 5%, 3% and 1% higher NE in simulating days to anthesis, days to physiological maturity, dry matter and grain yield, respectively, than with measured inputs. The “EnkF” outperformed “Forcing” for predicting crop yield as well as phenology and growth of wheat using generic management inputs. The system showed acceptable accuracy in forecasting phenology, dry matter and yield of wheat at field scale when weighted adaptive bias-correction of weather forecast was incorporated with a 15 days lead time. SIGNIFICANCEThe prototype can be scaled-up for wheat and other crops for predicting real-time crop condition and yield losses at farmers' field for a range of applications, notably, crop-insurance, resources allocation, targeted agro-advisories and triggering contingency plans. It offers considerable potential for objective assessment of crops in the marginal and smallholder systems supporting the smart farming paradigm.

Real-time burst detection based on multiple features of pressure data

Abstract Pipe bursts are an essential issue for water loss in water distribution systems. This study proposes a real-time burst detection method that combines multiple data features of multiple time steps. The method sets burst thresholds in three dimensions according to different moments at a specific monitoring point, and achieves burst identification based on a classification model. First, three data features, namely, absolute pressure value, predicted deviation value obtained by pressure variation value, of historical pressure at each time step are scored based on the Western Electric Company rules. The scores represent different abnormalities. Then, the scores corresponding to the three features are used as input of the decision tree classification model. The trained model is used for detecting burst events. Results show that this method achieves 99.56% detection accuracy, indicating that it is effective for burst detection. The proposed method outperformed the single-feature-based method and provides good results in water distribution systems.