Refine
Document Type
- ZIB-Report (11)
- Article (8)
- In Proceedings (7)
- Other (2)
Keywords
- CO2 Transport (1)
- Energiewende (1)
- Hydrogen Infrastructure (1)
- Hydrogen Transport (1)
- Investment planning (1)
- Mixed Integer Programming (1)
- Multi-energy systems (1)
- Network Design (1)
- Network Flows (1)
- Pipeline Sizing (1)
- Time series (1)
- Unit commitment (1)
Deep learning for spatio-temporal supply and demand forecasting in natural gas transmission networks
(2021)
Germany is the largest market for natural gas in the European Union, with an annual consumption of approx. 95 billion cubic meters. Germany's high-pressure gas pipeline network is roughly 40,000 km long, which enables highly fluctuating quantities of gas to be transported safely over long distances. Considering that similar amounts of gas are also transshipped through Germany to other EU states, it is clear that Germany's gas transport system is essential to the European energy supply. Since the average velocity of gas in a pipeline is only 25km/h, an adequate high-precision, high-frequency forecasting of supply and demand is crucial for efficient control and operation of such a transmission network. We propose a deep learning model based on spatio-temporal convolutional neural networks (DLST) to tackle the problem of gas flow forecasting in a complex high-pressure transmission network. Experiments show that our model effectively captures comprehensive spatio-temporal correlations through modeling gas networks and consistently outperforms state-of-the-art benchmarks on real-world data sets by at least 21$\%$.
The results demonstrate that the proposed model can deal with complex nonlinear gas network flow forecasting with high accuracy and effectiveness.
Deep learning for spatio-temporal supply anddemand forecasting in natural gas transmission networks
(2021)
Germany is the largest market for natural gas in the European Union, with an annual consumption of approx. 95 billion cubic meters. Germany's high-pressure gas pipeline network is roughly 40,000 km long, which enables highly fluctuating quantities of gas to be transported safely over long distances. Considering that similar amounts of gas are also transshipped through Germany to other EU states, it is clear that Germany's gas transport system is essential to the European energy supply. Since the average velocity of gas in a pipeline is only 25km/h, an adequate high-precision, high-frequency forecasting of supply and demand is crucial for efficient control and operation of such a transmission network. We propose a deep learning model based on spatio-temporal convolutional neural networks (DLST) to tackle the problem of gas flow forecasting in a complex high-pressure transmission network. Experiments show that our model effectively captures comprehensive spatio-temporal correlations through modeling gas networks and consistently outperforms state-of-the-art benchmarks on real-world data sets by at least 21%.
The results demonstrate that the proposed model can deal with complex nonlinear gas network flow forecasting with high accuracy and effectiveness.
In the transition towards a pure hydrogen infrastructure, utilizing the existing natural gas infrastructure is a necessity. In this study, the maximal technically feasible injection of hydrogen into the existing German natural gas transmission network is analysed with respect to regulatory limits regarding the gas quality. We propose a transient tracking model based on the general pooling problem including linepack. The analysis is conducted using real-world hourly gas flow data on a network of about 10,000 km length.
As a result of the legislation for gas markets introduced by the European Union in 2005, separate independent companies have to conduct the transport and trading of natural gas. The current gas market of Germany, which has a market value of more than 54 billion USD, consists of Transmission System Operators (TSO), network users, and traders. Traders can nominate a certain amount of gas anytime and anywhere in the network. Such unrestricted access for the traders, on the other hand, increase the uncertainty in the gas supply management. Some customers’ behaviors may cause abrupt structural changes in gas flow time series. In particular, it is a challenging task for the TSO operators to predict gas nominations 6 to 10 hours ahead. In our study, we aim to investigate the regime changes in the time series of nominations to predict the 6 to 10 hours ahead of gas nominations.
With the emergence of ”Big Data” the analysis of large data sets of high-dimensional energy time series in network structures have become feasible. However, building large-scale data-driven and computationally efficient models to accurately capture the underlying spatial and temporal dynamics and forecast the multivariate time series data remains a great challenge. Additional constraints make the problem more challenging to solve with conventional methods. For example, to ensure the security of supply, energy networks require the demand and supply to be balanced.
This paper introduces a novel large-scale Hierarchical Network Regression model with Relaxed Balance constraint (HNR-RB) to investigate the network dynamics and predict multistep-ahead flows in the natural gas transmission network, where the total in- and out-flows of the network have to be balanced over a period of time. We concurrently address three main challenges: high dimensionality of networks with more than 100 nodes, unknown network dynamics, and constraint of balanced supply and demand in the network. The effectiveness of the proposed model is demonstrated through a real-world case study of forecasting demand and supply in a large-scale natural gas transmission network. The results demonstrate that HNR-RB outperforms alternative models for short- and mid-term horizons.
The imperative to decarbonize energy systems has intensified the need for efficient transformations within the heating sector, with a particular focus on district heating networks. This study addresses this challenge by proposing a comprehensive optimization approach evaluated on the district heating
network of the Märkisches Viertel of Berlin. Our objective is to simultaneously optimize heat production with three targets: minimizing costs, minimizing CO2-emissions, and maximizing heat generation from Combined Heat and Power (CHP) plants for enhanced efficiency.
To tackle this optimization problem, we employed a Mixed-Integer Linear Program (MILP) that encompasses the conversion of various fuels into heat and power, integration with relevant markets, and considerations for technical constraints on power plant operation. These constraints include startup
and minimum downtime, activation costs, and storage limits. The ultimate goal is to delineate the Pareto front, representing the optimal trade-offs between the three targets. We evaluate variants of the 𝜖-constraint algorithm for their effectiveness in coordinating these objectives, with a simultaneous focus on the quality of the estimated Pareto front and computational efficiency. One algorithm explores solutions on an evenly spaced grid in the objective space, while another dynamically adjusts the grid based on identified solutions. Initial findings highlight the strengths and limitations of each algorithm, providing guidance on algorithm selection depending on desired outcomes and computational constraints.
Our study emphasizes that the optimal choice of algorithm hinges on the density and distribution of solutions in the feasible space. Whether solutions are clustered or evenly distributed significantly influences algorithm performance. These insights contribute to a nuanced understanding of algorithm selection for multi-objective multi-energy system optimization, offering valuable guidance for future research and practical applications for planning sustainable district heating networks.
The investigation of energy transition paths toward a sustainable and decarbonized future under uncertainty is a critical aspect of contemporary energy planning and policy development. There are numerous methods for analysing uncertainties and sensitivities and many studies on sustainable transformation paths, but there is a lack of combined application to relevant use-cases.
In this study, we investigate the sensitivity of energy transition paths to uncertainties in operational and investment costs of power plants in the metropolitan area of Berlin and its rural surroundings.
By employing the linear programming energy system model oemof-B3, we extensively focus on the system's energy technologies, such as wind turbines, photovoltaics, hydro and combustion plants, and energy storages. Greenhouse gas reduction and electrification rates per commodity are realized by selected constraints.
Our research aims to discern how investments in energy production capacities are influenced by uncertainties of other energy technologies' investment and operational costs in the system. We apply a quantitative approach to investigate such interdependencies of cost variations and their impact on long-term energy planning. Thus, the analysis sheds light on the robustness of energy transition paths in the face of these uncertainties.
The region Berlin-Brandenburg serves as a case study and thus reflects on the present space conflicts to meet energy demands in urban and suburban areas and their rural surroundings. An electricity-intensive scenario is selected that assumes a 100 % reduction in greenhouse gas emissions by 2050. With the results of the case study, we show how our approach enables rural and metropolitan decision-makers to collaborate in achieving sustainable energy.
Decision-making in long-term energy planning can be made more robust and flexible by acknowledging the identified sensitivities and enable such regions better to navigate challenges and uncertainties associated with sustainable energy planning.