Lost in Optimization of Water Distribution Systems: Better Call Bayes

The main goal of this paper is to show that Bayesian optimization could be regarded as a general framework for the data driven modelling and solution of problems arising in water distribution systems. Hydraulic simulation, both scenario based, and Monte Carlo is a key tool in modelling in water distribution systems. The related optimization problems fall in a simulation/optimization framework in which objectives and constraints are often black-box. Bayesian Optimization (BO) is characterized by a surrogate model, usually a Gaussian process, but also a random forest and increasingly neural networks and an acquisition function which drives the search for new evaluation points. These modelling options make BO nonparametric, robust, flexible and sample efficient particularly suitable for simulation/optimization problems. A defining characteristic of BO is its versatility and flexibility, given for instance by different probabilistic models, in particular different kernels, different acquisition functions. These characteristics of the Bayesian optimization approach are exemplified by the two problems: cost/energy optimization in pump scheduling and optimal sensor placement for early detection on contaminant intrusion. Different surrogate models have been used both in explicit and implicit control schemes. Showing that BO can drive the process of learning control rules directly from operational data. BO can also be extended to multi-objective optimization. Two algorithms have been proposed for multi-objective detection problem using two different acquisition functions.

Rehabilitation of operation regimes in aged irrigation schemes based on hydraulic simulation

The selection and employment of proper methods in water distribution causes increasing in water productivity and the level of satisfaction of water users. It is faced with more difficulties in aged irrigation projects due to temporal changes such as changes in the crop patterns, development of the command area and destruction of canals and hydraulic structures. The plan of operation methods have some hydraulic and social complexities and therefore is usually simplified or implemented experimentally. This research investigates the best options for water distribution to the paddy fields in a subunit of Sefidroud irrigation scheme based on field survey, recording real data and hydraulic simulation with employing SOBEK hydrodynamic model. Different operation scenarios were defined and then simulated in the current physical state of the scheme through replacing the exhausted intake structures with sluice gates. Finally, the better operation scenarios during the irrigation season were suggested based on the distribution indices. The results show that in spite of the current situation, water loss could reach the minimum by employing modification scenarios and indices of adequacy and equity of water distribution improve.

A Sensor Data Processing Algorithm for Wind Turbine Hydraulic Pitch System Diagnosis

Modern wind turbines depend on their blade pitch systems for start-ups, shutdowns, and power control. Pitch system failures have, therefore, a considerable impact on their operation and integrity. Hydraulic pitch systems are very common, due to their flexibility, maintainability, and cost; hence, the relevance of diagnostic algorithms specifically targeted at them. We propose one such algorithm based on sensor data available to the vast majority of turbine controllers, which we process to fit a model of the hydraulic pitch system to obtain significant indicators of the presence of the critical failure modes. This algorithm differs from state-of-the-art, model-based algorithms in that it does not numerically time-integrate the model equations in parallel with the physical turbine, which is demanding in terms of in situ computation (or, alternatively, data transmission) and is highly susceptible to drift. Our algorithm requires only a modest amount of local sensor data processing, which can be asynchronous and intermittent, to produce negligible quantities of data to be transmitted for remote storage and analysis. In order to validate our algorithm, we use synthetic data generated with state-of-the-art aeroelastic and hydraulic simulation software. The results suggest that a diagnosis of the critical wind turbine hydraulic pitch system failure modes based on our algorithm is viable.

Thermo-hydraulic Simulation of AP1000 Nuclear Reactor Fuel Assembly

One of the challenges of future nuclear power is the development of safer and more efficient nuclear reactor designs. The AP1000 reactor based on the PWR concept of generation III + has several advantages, which can be summarized as: a modular construction, which facilitates its manufacture in series reducing the total construction time, simplification of the different systems, reduction of the initial capital investment and improvement of safety through the implementation of passive emergency systems. Being a novel design it is important to study the thermohydraulic behavior of the core applying the most modern tools. To determine the thermohydraulic behavior of a typical AP1000 fuel assembly, a computational model based on CFD was developed. A coupled neutronic-thermohydraulic calculation was performed, allowing to obtain the axial power distribution in the typical fuel assembly. The geometric model built used the certified dimensions for this type of installation that appear in the corresponding manuals. The thermohydraulic study used the CFD-based program ANSYS-CFX, considering an eighth of the fuel assembly. The neutronic calculation was performed with the program MCNPX version 2.6e. The work shows the results that illustrate the behavior of the temperature and the heat transfer in different zones of the fuel assembly. The results obtained agree with the data reported in the literature, which allowed the verification of the consistency of the developed model.

Hydrodynamic modelling of historical flood event using one dimensional HEC-RAS in Kelantan basin, Malaysia

Hydraulic simulation models are critical tools for analysing the hydraulic properties of a river’s system flow. The work focuses on the simulation of a river flow in a Kelantan basin using the one-dimensional (1D) Hydrologic Engineering Center - River Analysis System (HECRAS). In the present study, cross-sections from survey data were utilised into the RAS Mapper provided in HEC-RAS 5.0 to simulate the river flow in the region. This study highlights the modelling methodology with a focus on data collection and its importance during the calibration and validation process. The model was used to discover the expected peak flood levels based on historical flood events. Simulated flows were utilised to examine the potential of the model during the model development procedure. The simulation outcomes reveal that the simulated flows are in excellent agreement during the model development as the obtained R2 value was between 0.95 to 1.0 during both model calibration and validation. This demonstrates the applicability of the HEC-RAS 1D model in simulating precise river flow, especially for flood events.

Flow hydraulic simulation through two sand traps, using Ansys fluent

Computational fluid dynamics is a tool that allows to simulate and observe the behavior of any fluid, based on a physical, hydraulic, and hydrodynamic analysis. This research analyses the behavior of the flow in a sand trap, which is a structure used to remove sand particles with a minimum size of 0.10 mm, prior to treatment in a drinking-water plant. The objective of this study is to determine the highest efficiency between two sand traps, one with a double smooth screen and the other with a double perforated screen (with diffusers), based on the simulation and analysis behavior of the flow inside each sand trap. The methodology used includes the traditional design of each unit based on Hazen’s model and Stokes viscosity law, to later carry out the numerical model simulation from Ansys Fluent (pre-processing, processing, and post-processing). The result shows that perforated double screen sand trap generates a removal efficiency of 78%, while the smooth double screen 28%. In addition, other four units of interleaved screens are proposed, in these cases efficiencies of up to 50% are observed and it is shown that it is necessary to implement at least two perforated screens (with diffusers) to guarantee an efficiency greater than 70%. Hydraulic simulation has a broad impact on infrastructure works and consulting.

Assessment and modeling of sewer network development utilizing Arc GIS and SewerGEMS in Kabul city of Afghanistan

The absence of a wastewater collection, management, and disposal scheme is one of Kabul’s most serious environmental issues. This has resulted in both health and ecological problems. This research used Arc GIS and SewerGEMS tools to assess the viability of a decentralized sewerage collection model in the research area. The research area was chosen as the city’s 5th district. Land-use and land-cover, Digital Elevation Model (DEM), and Satellite data were used to construct the network’s geometry in the Arc map environment. SewerGEMS software was used to perform hydraulic simulation and modeling. The variables were regulated based on the results of the study using conventional wastewater topology guidelines. Based on the outputs of hydraulic analysis, it is concluded that the decentralized wastewater collection system would be the best option for the area. It can be deduced from hydraulic design findings that the hydraulic model was successfully developed and built. The methodology can be applied for the development of future wastewater master plans of the city.

Theoretical Analysis and Semi-Analytical Formulation for Efficient Thermal-Hydraulic-Mechanical Reservoir Simulation

The research of multiphysical thermal-hydraulic-mechanical (THM) simulation has achieved significant progress in the past decade. Currently, two general approaches for poromechanical simulation co-exist in the reservoir simulation community, namely the stress approach with stress as the primary variable for the mechanical governing equations and the displacement approach with displacement as the primary variable. In this work, we aim to provide a theoretical foundation and a practical semi-analytical solution for the stress approach based on the Navier-Beltrami-Michell Equations. Moreover, we will clarify the relationship (and equivalence) between the two approaches. We have firstly proven the existence and uniqueness of the stress solution of Navier-Beltrami-Michell equation with given pressure and temperature field. Moreover, we have demonstrated the equivalence of the stress formulation to the displacement formulation. Based on Fourier's expansion, we have developed a general semi-analytical solution for thermal-hydraulic-mechanical process. The semi-analytical solution takes the pressure solution from the hydraulic simulation module (or a commercial reservoir simulator) and directly predicts the stress tensor of the multiphysical system. As such, the solution can be programmed fully coupled with the hydraulic simulation module to predict the stress field with varying pressure and temperature of homogeneous poroelastic rocks under given stress boundary conditions. From the work above, we have laid a theoretical foundation for the stress approach. The derived semi-analytical solution of the stress field shows excellent accuracy. The solution has been used to predict the transient stress field of a dual-porosity system during primary depletion. This paper is arguably the first trial to clarify the relationship between the stress approach and the displacement approach. Moreover, the derived semi-analytical solution provides a convenient yet precise way to obtain the stress field without time-consuming numerical simulation.