Dirk Schwanenberg

Short-Term Reservoir Optimization for Flood Mitigation under Meteorological and Hydrological Forecast Uncertainty

State-of-the-art applications of short-term reservoir management integrate several advanced components, namely hydrological modelling and data assimilation techniques for predicting streamflow, optimization-based techniques for decision-making on the reservoir operation and the technical framework for integrating these components with data feeds from gauging networks, remote sensing data and meteorological weather predictions. In this paper, we present such a framework for the short-term management of reservoirs operated by the Companhia Energética de Minas Gerais S.A. (CEMIG) in the Brazilian state of Minas Gerais. Our focus is the Três Marias hydropower reservoir in the São Francisco River with a drainage area of approximately 55,000 km and its operation for flood mitigation. Basis for the anticipatory short-term management of the reservoir over a forecast horizon of up to 15 days are streamflow predictions of the MGB hydrological model. The semi-distributed model is well suited to represent the watershed and shows a Nash-Sutcliffe model performance in the order of 0.83-0.90 for most streamflow gauges of the data-sparse basin. A lead time performance assessment of the deterministic and probabilistic ECMWF forecasts as model forcing indicate the superiority of the probabilistic model. The novel short-term optimization approach consists of the reduction of the ensemble forecasts into scenario trees as an input of a multi-stage stochastic optimization. We show that this approach has several advantages over commonly used deterministic methods which neglect forecast uncertainty in the short-term decision-making. First, the probabilistic forecasts have longer forecast horizons that allow an earlier and therefore better anticipation of critical flood events. Second, the stochastic optimization leads to more robust decisions than deterministic procedures which consider only a single future trajectory. Third, the stochastic optimization permits to introduce advanced chance constraints for refining the system operation.

Short-term management of hydropower assets of the Federal Columbia River Power System

We focus on the short-term optimization of large-scale hydropower systems with a mixture of storage reservoirs and run-of-the-river projects. If sufficient operational flexibility is available within the operational constraints, the system is capable of balancing the transmission network by compensating for load fluctuations and power production of other renewables such as solar and wind resources. The proposed optimization model is based on a nonlinear system representation of the hydropower system for forecasting state trajectories over a forecast horizon in combination with nonlinear programming for computing optimal release trajectories for specific hydropower projects. We present the application of a deterministic version of the approach to the short-term management of the Federal Columbia River Power System in the Pacific Northwest of the USA during the chum spawning season. An assessment of the computational performance of the approach for different optimization algorithms shows a superior performance of the Interior Point OPTimizer (IPOPT) in combination with the HSL/MA27 linear equation solver. In particular, the scaling properties are promising and will enable an extension of the deterministic approach towards a multi-stage stochastic optimization for taking into account the forecast uncertainty.

Comparison Between Two Hydrodynamic Models for Flooding Simulations at River Lima Basin

According to EU flood risks directive, flood hazard maps should include information on hydraulic characteristics of vulnerable locations, i.e. the inundated areas, water depths and velocities. These features can be assessed by the use of advanced hydraulic modelling tools which are presented in this paper based on a case study in the river Lima basin, Portugal. This river includes several flood-prone areas. Ponte Lima town is one of the places of higher flood risk. The upstream dams can lower the flood risks if part of its storage capacity is allocated for mitigating flood events. However, proper management of dam releases and the evaluation of downstream river flows should be considered for preventing flood damages. A hydrological and a one-dimensional hydrodynamic model were implemented, and at a particular flood-prone town, inundation was assessed using a two-dimensional model. The hydrological model is based on the well known Sacramento model. For this purpose, two different modelling implementations were analysed: a model based on a finite element mesh and a model based on rectangular grids. The computational performance of the two modelling implementations is evaluated. Historical flood events were used for model calibration serving as a basis for the establishment of different potential flood scenarios. Intense precipitation events in the river’s basin and operational dam releases are determinant for the occurrence of floods at vulnerable downstream locations. The inundation model based on the unstructured mesh reveals to be more computationally efficient if high spatial resolution is required. A new combination of software tools for floods simulation is presented including an efficient alternative for simulation of 2-D inundation using a finite element mesh instead of a grid.

Simplified hydraulic modelling in model predictive control of flood mitigation measures along rivers

Model predictive control (MPC) is an advanced technique for controlling water resources systems over short-term prediction horizons by optimizing control trajectories of hydraulic structures. Future system states are forecasted by an internal model that simulates the system response to a set of external disturbances and control inputs. This paper deals with the design and numerical implementation of dedicated simplified hydraulic models for representing flood routing processes in MPC. We present alternatives for the numerical implementation in simulation mode as well as first-order sensitivity mode and discuss the trade-off between model accuracy and computational performance in sequential and simultaneous MPC set-ups in application to two test cases: The first one is a fictitious estuary case with little potential for simplifying the full dynamic model. The second case is an application of MPC to the control of flood detention measures along the Rhine River in Germany. Both cases show the high potential ...

Performance of Deterministic and Probabilistic Hydrological Forecasts for the Short-Term Optimization of a Tropical Hydropower Reservoir

Hydropower is the most important source of electricity in Brazil. It is subject to the natural variability of water yield. One building block of the proper management of hydropower assets is the short-term forecast of reservoir inflows as input for an online, event-based optimization of its release strategy. While deterministic forecasts and optimization schemes are the established techniques for short-term reservoir management, the use of probabilistic ensemble forecasts and multi-stage stochastic optimization techniques is receiving growing attention. The present work introduces a novel, mass conservative scenario tree reduction in combination with a detailed hindcasting and closed-loop control experiments for a multi-purpose hydropower reservoir in a tropical region in Brazil. The case study is the hydropower project Três Marias, which is operated with two main objectives: (i) hydroelectricity generation and (ii) flood control downstream. In the experiments, precipitation forecasts based on observed data, deterministic and probabilistic forecasts are used to generate streamflow forecasts in a hydrological model over a period of 2 years. Results for a perfect forecast show the potential benefit of the online optimization and indicate a desired forecast lead time of 30 days. In comparison, the use of actual forecasts of up to 15 days shows the practical benefit of operational forecasts, where stochastic optimization (15 days lead time) outperforms the deterministic version (10 days lead time) significantly. The range of the energy production rate between the different approaches is relatively small, between 78% and 80%, suggesting that the use of stochastic optimization combined with ensemble forecasts leads to a significantly higher level of flood protection without compromising the energy production.

Parameter estimation for eutrophication models in reservoirs

Excessive eutrophication is a major water quality issue in lakes and reservoirs worldwide. This complex biological process can lead to serious water quality problems. Although it can be adequately addressed by applying sophisticated mathematical models, the application of these tools in a reservoir management context requires significant amounts of data and large computation times. This work presents a simple primary production model and a calibration procedure that can efficiently be used in operational reservoir management frameworks. It considers four state variables: herbivorous zooplankton, algae (measured as chlorophyll-a pigment), phosphorous and nitrogen. The model was applied to a set of Portuguese reservoirs. We apply the model to 23 Portuguese reservoirs in two different calibration settings. This research work presents the results of the estimation of model parameters.

Short Term Optimal Operation of Water Supply Reservoir under Flood Control Stress using Model Predictive Control

Reservoir operations require enhanced operating procedures for water systems under stress attributed to growing water demand and consequences of changing hydro-climatic conditions. This study focuses on the management of the Yuvacik Dam Reservoir for water supply and flood mitigation in the Marmara Region of Turkey. We present an improved operating technique for fulfilling the conflicting water supply and flood mitigation objectives. This is accomplished by incorporating the long term water supply objectives into a Guide Curve (GC) whereas the extreme floods are attenuated by means of short-term optimization based on Model Predictive Control (MPC). The reference case implements operating rules with a constant GC at maximum forebay elevation targeting the fulfillment of the water supply objective. We compare the reference with a new time-dependent GC, derived using an Implicit Stochastic Optimization (ISO) approach. This new curve shows nearly the same performance regarding the water supply objectives, but significantly reduces the flooding risk downstream of the dam. Possible flood events observed at the end of the wet season, when the reservoir is at the maximum level to enable water supply for the dry season, can be eliminated by the application of an additional short-term optimization by MPC. The robustness of the approach is demonstrated via hindcasting experiments.

Multi-step Flow Routing Using Artificial Neural Networks for Decision Support

The use of surrogate and data-driven models has the potential to decrease the computational effort of streamflow predictions in time-critical model applications such as Data Assimilation (DA) or Model Predictive Control (MPC). In the present work, it was evaluated the use of Artificial Neural Network (ANN) as replacement of a physical model in a MPC implementation for the multi-objective optimization of a reservoir system. The presented application covers the flow routing of a reservoir release from Tres Marias dam in Brazil and downstream tributaries to Pirapora gauge for lead times between 1 and 360 h (15 days). The ANNs were trained using Levenberg–Marquadt algorithm, and three different transfer functions were evaluated. It was also tested two output correction techniques, namely an ARX model for error prediction and bias correction for lead time until 15 days ahead. The ANNs model shows good capability of handling with minor disturbances. Best performance was found using the purelin transfer function. For the error correction, the ARX models showed better error reduction when compared with the bias correction technique, which could not reduce the error at the end of lead time.

Short-Term Hydropower Optimization and Assessment of Operational Flexibility

AbstractHydroelectric power systems are largely characterized by variability and uncertainty in water resource obligations. Market volatility and the growing number of operational obligations for flood control, navigation, environmental obligations, and ancillary services (including load-balancing requirements for renewable resources) further the need to quantify sources of uncertainty. The variations caused by these factors require the hydropower system to have enough upward and downward flexibility for control technologies, such as dynamic optimal control load-following, unit commitment, or automatic generation, to be effective. Therefore, it is increasingly important to identify measures of operational flexibility to better manage uncertainty and operational obligations. The objective of this paper is to present and discuss approaches for assessment of operational flexibility as a function of dynamic states and control input and how the available operational flexibility can be used by hydropower produc...

Short-term management of hydropower reservoirs under meteorological uncertainty by means of multi-stage optimization

Abstract Uncertainty in meteorology, market volatility and balancing requirements for introducing renewable energy resources into the power grid, environmental obligations require robust management of non-intermittent energy sources such as hydropower. In this paper, a probalistic management system is shown and its performance is discussed in relation to the deterministic one. In the system, scenario trees enable to setup a multi-stage stochastic optimization approach as the mathematical formulation of the short-term system management. The Federal Columbia River Power System (FCRPS), managed by the Bonneville Power Administration, the US Army Corps of Engineers and the Bureau of Reclamation, serves as a large-scale test case for the application of the management system and proves that the stochastic approach is feasible and verify the operational applicability within a real-time environment.