arXiv subjects

Philipp Althaus

Publications and source records attributed to Philipp Althaus.

4 recordsLinked to original sources

arXiv2026

Systematic Evaluation of TabPFN-TS for Zero-Shot Probabilistic Heat Load Forecasting in District Heating Networks

District heating energy hubs require reliable heat load forecasts for efficient operational scheduling. Conventional forecasting workflows train system-specific models on historical data, which can become burdensome when networks change through new consumers, retrofits, or changing operating regimes. Zero-shot time-series foundation models and in-context forecasting offer a promising alternative: they can adapt at inference time from recent observations rather than by repeated retraining. This study systematically evaluates TabPFN-TS against time-series foundation models and trained machine-learning baselines for probabilistic heat load forecasting in district heating networks. Unlike foundation models pretrained on large collections of real time series, TabPFN-TS relies on synthetic pretraining data, which avoids direct pretraining-test overlap but raises the question of whether the learned prior captures district heating dynamics. We analyze covariate choice, context length, temporal resolution, and prediction horizon on representative operating weeks, validate the selected configuration over a full year, and test transferability on a second network. The results identify hourly 24-hour forecasting with a 12-week rolling context and ambient temperature as a parsimonious high-performing configuration; longer context windows do not improve accuracy. TabPFN-TS remains close to Chronos-2 in deterministic accuracy, reaching CVRMSE values of 13.06% versus 12.48% on the main dataset, and lies within the critical-difference threshold in the daily-rank comparison. Although Chronos-2 achieves the lowest aggregate full-year error, TabPFN-TS shows better empirical calibration. Finally, the diagnostic findings motivate a Multi-Resolution Residual-Correction Forecaster that combines a low-frequency Base Forecaster with a short-horizon Residual Forecaster to improve longer-horizon planning accuracy.

cs.LG↗

arXiv2023

Method development for lowering supply temperatures in existing buildings using minimal building information and demand measurement data

Regarding climate change, the need to reduce greenhouse gas emissions is well-known. As building heating contributes to a high share of total energy consumption, which relies mainly on fossil energy sources, improving heating efficiency is promising to consider. Lowering supply temperatures of the heating systems in buildings offers a huge potential for efficiency improvements since different heat supply technologies, such as heat pumps or district heating, benefit from low supply temperatures. However, most estimations of possible temperature reductions in existing buildings are based on available measurement data on room level or detailed building information about the building's physics to develop simulation models. To reveal the potential of temperature reduction for several buildings and strive for a wide applicability, the presented method focuses on estimations for temperature reduction in existing buildings with limited input data. By evaluating historic heat demand data on the building level, outdoor temperatures and information about installed heaters, the minimal actual necessary supply temperature is calculated for each heater in the building using the LMTD approach. Based on the calculated required supply temperatures for each room at different outdoor temperatures, the overall necessary supply temperatures to be provided to the building are chosen. Thus, the minimal heatcurve possible for an existing building is deduced. The method described is applied to multiple existing office buildings at the campus of Forschungszentrum Juelich, Germany, demonstrating the fast application for several buildings with limited expenditure. Furthermore, a developed adapted heatcurve is implemented in one real building and evaluated in relation to the previously applied heatcurve of the heating system.

eess.SY↗

arXiv2022

Demand Response for Flat Nonlinear MIMO Processes using Dynamic Ramping Constraints

Volatile electricity prices make demand response (DR) attractive for processes that can modulate their production rate. However, if nonlinear dynamic processes must be scheduled simultaneously with their local multi-energy system, the resulting scheduling optimization problems often cannot be solved in real time. For single-input single-output processes, the problem can be simplified without sacrificing feasibility by dynamic ramping constraints that define a derivative of the production rate as the ramping degree of freedom. In this work, we extend dynamic ramping constraints to flat multi-input multi-output processes by a coordinate transformation that gives the true nonlinear ramping limits. Approximating these ramping limits by piecewise affine functions gives a mixed-integer linear formulation that guarantees feasible operation. As a case study, dynamic ramping constraints are derived for a heated reactor-separator process that is subsequently scheduled simultaneously with its multi-energy system. The dynamic ramping formulation bridges the gap between rigorous process models and simplified process representations for real-time scheduling.

math.OC↗

arXiv2021

Dynamic Ramping for Demand Response of Processes and Energy Systems based on Exact Linearization

The increasing share of volatile renewable electricity production motivates demand response. Substantial potential for demand response is offered by flexible processes and their local multi-energy supply systems. Simultaneous optimization of their schedules can exploit the demand response potential, but leads to numerically challenging problems for nonlinear dynamic processes. In this paper, we propose to capture process dynamics using dynamic ramping constraints. In contrast to traditional static ramping constraints, dynamic ramping constraints are a function of the process state and can capture high-order dynamics. We derive dynamic ramping constraints rigorously for the case of single-input single-output processes that are exactly input-state linearizable. The resulting scheduling problem can be efficiently solved as a mixed-integer linear program. In a case study, we study two flexible reactors and a multi-energy system. The proper representation of process dynamics by dynamic ramping allows for faster transitions compared to static ramping constraints and thus higher economic benefits of demand response. The proposed dynamic ramping approach is sufficiently fast for application in online optimization.

math.OC↗