Dataset underlying the PhD thesis 'Unlocking flexibility: risk-aware operational water and energy management'.

This thesis explores risk-aware operational decision-making methods to support the integration of Renewable Energy Sources (RES) into the energy system by enhancing energy flexibility under operational uncertainty. Amidst the urgent global shift towards RES to combat climate change, this work identifies and addresses the challenges posed by the intermittent and uncertain nature of renewable energies, such as wind and solar power, to grid stability and energy reliability.<br>Central to this objective is the use of Demand Response (DR) strategies to balance energy supply and demand dynamically using electricity spot markets, mitigating the risks associated with the variability and uncertainty of renewable energy sources. A significant contribution of this work lies in the examination of the water-energy nexus through a case study of the Noordzeekanaal–Amsterdam-Rijnkanaal in the Netherlands.<br>This dataset contains the underlying code for:Exploring the potential benefits of applying DR to the NZK-ARK, a critical piece of the Dutch water defence system and a large energy consumer. By formulating a new MPC problem, in which we propose a multi-market strategy, DAM and IDM prices are used to optimize energy-cost optimal pump schedules.Several methodologies applied to model operational uncertainty, detailing how neural networks are trained and optimized for forecasting purposes, what architecture is used to quantify uncertainty and generate probabilistic forecasts, a method for multi-distribution sampling with fixed correlation between variables for the purpose of time series sampling, and a scenario-reduction method is to condense a large uncertainty representation into an optimal subset and sparse scenario tree representation.A stochastic MPC framework where we apply risk-aware constraint formulations for a computationally efficient and pragmatic trade-off between energy cost savings and water level violations.RayCast, a spatially distributed Quantile Regression irradiance nowcast, that proficiently predicts irradiance quantiles using satellite data.

本论文探究了风险感知型运行决策方法，旨在通过提升运行不确定性场景下的能源灵活性，推动可再生能源（Renewable Energy Sources, RES）融入能源系统。在全球为应对气候变化而加速向可再生能源转型的紧迫背景下，本研究识别并解决了风能、太阳能等可再生能源的间歇性与不确定性特性对电网稳定性及能源可靠性构成的挑战。
本研究的核心在于采用需求响应（Demand Response, DR）策略，通过电力现货市场动态平衡能源供需，缓解可再生能源波动性与不确定性所引发的相关风险。本研究的一项重要贡献在于，以荷兰北泽运河-阿姆斯特丹-莱茵运河（Noordzeekanaal–Amsterdam-Rijnkanaal）为案例开展水-能源关联（water-energy nexus）的相关研究。
本数据集包含以下研究内容的实现代码：
1. 探究将需求响应应用于NZK-ARK（荷兰水防护系统的关键组成部分，同时也是大型能源消费主体）所能带来的潜在效益。通过构建新型模型预测控制（Model Predictive Control, MPC）问题，本文提出了多市场策略，利用日前市场（Day-Ahead Market, DAM）与日内市场（Intra-Day Market, IDM）电价优化得到能源成本最优的泵调度方案。
2. 多种用于建模运行不确定性的方法：详述了神经网络的训练与优化流程以支撑预测任务、用于量化不确定性并生成概率预测结果的架构、一种面向时间序列采样且可实现变量间固定相关性的多分布采样方法，以及一种将大规模不确定性表征压缩为最优子集与稀疏场景树表征的场景缩减方法。
3. 一种随机模型预测控制框架，其中应用了风险感知约束公式，以在能源成本节约与水位违规风险之间实现计算高效且务实的权衡。
4. RayCast：一种空间分布式分位数回归（Quantile Regression）辐照度临近预报模型，可借助卫星数据精准预测辐照度分位数。