1,721,226 research outputs found
Integrating Inverse Reinforcement Learning and Direct Policy Search for Modeling Multipurpose Water Reservoir Systems
Full text linkSystem identification and optimal control have always contributed to water resources systems planning and management. Although water control problems are commonly formulated as multi-objective Markov Decision Processes, accurately modeling reservoir systems controlled by human operators remains challenging due to the absence of a formal definition of the objective function guiding their behavior. In this letter, we introduce a mixed Reinforcement Learning approach to model the dynamics of multipurpose reservoir systems. Specifically, our method first uses Inverse Reinforcement Learning to extract the tradeoff among competing objectives from historical observations of the reservoir system dynamics. The identified objective function is then used in the formulation of an optimal control problem returning a closed-loop policy which allows the simulation of the observed dynamics of the reservoir system. We demonstrate the potential of the proposed method in a real-world application involving the multipurpose regulation of Lake Como in northern Italy. Results show that our approach effectively infers the tradeoff between flood control and water supply adopted in the observed system's operation, and yields a control policy that closely approximates the observed system dynamics
Exploring the Climate Puzzle: A Surprising Twist in Fighting Climate Change
Full text linkSometimes, when scientists try to help people, they can end up with a surprise ending in which things do not work out as expected. Their “help” might even accidentally make the situation worse for some people. We wanted to know if this could be true for a strategy to slow down climate change: charging countries a fee when they cut down forests to create farmland. We used computers to predict what might happen if countries were charged different fees, to keep things fair. Specifically, countries with less money would only have to pay low (or no) fees, while rich countries would pay higher fees. However, our computer model showed that this plan could have unexpected negative consequences for water availability in some places that pay low fees, like certain regions in Africa. This tells us that, as we fight climate change, we must keep our eyes open for unintended consequences that could result from our attempts to help the planet
Reducing the Intrusiveness of Energy and Water End-use Disaggregation via Social Media and Users Interactions
No full textWater end-use disaggregation from smart-metered water consumption data is key to design water demand management strategies. Yet, despite large investments for smart metering networks, few water utilities are actually exploiting the potential of end-use-based consumer profiling. This is mostly due to the intrusiveness of existing disaggregation algorithms, which require intensive and time consuming human interaction for extracting appliance-specific information and estimating the end-use patterns. We propose to explore the potential for social media in facilitating the interactions between water utilities and water users in order to collect consumption diaries. The evaluation of the value of this user-generated information is performed by using the collected diaries as input for a disaggregation algorithm, which is shown to attain acceptable performance in reproducing the end-use patterns
Assessing the value of cooperation and information exchange in large water re- sources systems by multi-agent optimization: the Zambezi River case.
No full textFlexible and adaptive water systems operations through more informed and dynamic decisions
No full textTimely adapting the operations of water systems to be resilient against rapid changes in both hydroclimatic and socioeconomic forcing is generally recommended as a part of planning and managing water resources under uncertain futures. A great opportunity to make the operations more flexible and adaptive is offered by the unprecedented amount of information that is becoming available to water system operators, providing a wide range of data at increasingly higher temporal and spatial resolution. Yet, many water systems are still operated using very simple information systems, typically based on basic statistical analysis and the operator’s experience. In this work, we discuss the potential offered by incorporating improved information to enhance water systems operation and increase their ability of adapting to different external conditions and resolving potential conflicts across sectors. In particular, we focus on the use of different variables associated to different dynamics of the system (slow and fast) diversely impacting the operating objectives on the short-, medium- and long-term. The multi-purpose operations of the Hoa Binh reservoir in the Red River Basin (Vietnam) is used to demonstrate our approach. Numerical results show that our procedure is able to automatically select the most valuable information for improving the Hoa Binh operations and mitigating the conflict between short-term objectives, i.e. hydropower production and flood control. Moreover, we also successfully identify low-frequency climate information associated to El-Nino Southern Oscillation for improving the performance in terms of long-term objectives, i.e. water supply. Finally, we assess the value of better informing operational decisions for adapting the system operations to changing conditions by considering different climate change projections
Modeling the behavior of water reservoir operators via eigenbehavior analysis
No full textThe large number of dammed rivers worldwide emphasizes the need to couple models of natural processes with models describing human behaviors. However, such behavioral models are often simplistic and lack proper validation against observational data. In this work, we contribute a new approach to infer the typical operations of water reservoirs from historical observations, using data-driven behavioral modeling based on eigenbehavior analysis. The approach is demonstrated using monthly storage data from 172 reservoirs in California, USA. Results show that the proposed method identifies four typical behavioral profiles, which are strongly linked to key features of the reservoirs. Moreover, we show how the identified models can be used for discovering behavioral profiles, and associated reservoir characteristics, that are vulnerable to drought conditions
Exploring the potential of desalination and aquaponics in the integrated management of arid river basins: the case of the Nile River basin
No full textAssessing the Value of Post-processed State-of-the-art Long-term Weather Forecast Ensembles within An Integrated Agronomic Modelling Framework
No full textOver recent years, long-term climate forecast from global circulation models (GCMs) has been demonstrated to show increasing skills over the climatology, thanks to the advances in the modelling of coupled ocean-atmosphere dynamics. Improved information from long-term forecast is supposed to be a valuable support to farmers in optimizing farming operations (e.g. crop choice, cropping time) and for more effectively coping with the adverse impacts of climate variability. Yet, evaluating how valuable this information can be is not straightforward and farmers' response must be taken into consideration. Indeed, while long-range forecast are traditionally evaluated in terms of accuracy by comparison of hindcast and observed values, in the context of agricultural systems, potentially useful forecast information should alter the stakeholders' expectation, modify their decisions and ultimately have an impact on their annual benefit. Therefore, it is more desirable to assess the value of those long-term forecasts via decision-making models so as to extract direct indication of probable decision outcomes from farmers, i.e. from an end-to-end perspective. In this work, we evaluate the operational value of thirteen state-of-the-art long-range forecast ensembles against climatology forecast and subjective prediction (i.e. past year climate and historical average) within an integrated agronomic modeling framework embedding an implicit model of farmers' behavior. Collected ensemble datasets are bias-corrected and downscaled using a stochastic weather generator, in order to address the mismatch of the spatio-temporal scale between forecast data from GCMs and distributed crop simulation model. The agronomic model is first simulated using the forecast information (ex-ante), followed by a second run with actual climate (ex-post). Multi-year simulations are performed to account for climate variability and the value of the different climate forecast is evaluated against the perfect foresight scenario based on the expected crop productivity as well as the land-use decisions. Our results show that not all the products generate beneficial effects to farmers and that the forecast errors might be amplified by the farmers decisions
Dynamic emulation modelling for the optimal operation of water systems: an overview
No full textDespite sustained increase in computing power over recent decades, computational limitations remain a major barrier to the effective and systematic use of large-scale, process-based simulation models in rational environmental decision-making. Whereas complex models may provide clear advantages when the goal of the modelling exercise is to enhance our understanding of the natural processes, they introduce problems of model identifiability caused by over-parameterization and suffer from high computational burden when used in management and planning problems. As a result, increasing attention is now being devoted to emulation modelling (or model reduction) as a way of overcoming these limitations. An emulation model, or emulator, is a low-order approximation of the process-based model that can be substituted for it in order to solve high resource-demanding problems. In this talk, an overview of emulation modelling within the context of the optimal operation of water systems will be provided. Particular emphasis will be given to Dynamic Emulation Modelling (DEMo), a special type of model complexity reduction in which the dynamic nature of the original process-based model is preserved, with consequent advantages in a wide range of problems, particularly feedback control problems. This will be contrasted with traditional non-dynamic emulators (e.g. response surface and surrogate models) that have been studied extensively in recent years and are mainly used for planning purposes. A number of real world numerical experiences will be used to support the discussion ranging from multi-outlet water quality control in water reservoir through erosion/sedimentation rebalancing in the operation of run-off-river power plants to salinity control in lake and reservoirs
- …
