Explainable heat demand forecasting for the novel control strategies of district heating systems

Zdravković, Milan; Ćirić, Ivan; Ignjatović, Marko

doi:10.1016/j.arcontrol.2022.03.009

Cited by 26 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…9(b). It is worth noting that the thermal demand profiles are assumed to be known to the NMPC regulator, given that the development of suitable forecasting algorithms is beyond the scope of this work and different related techniques are available in the literature (see [30], [52]).…”

Section: B Control Resultsmentioning

confidence: 99%

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

de Giuli,

La Bella,

Scattolini

2024

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

This article addresses the data-based modeling and optimal control of district heating systems (DHSs). Physical models of such large-scale networked systems are governed by complex nonlinear equations that require a large amount of parameters, leading to potential computational issues in optimizing their operation. A novel methodology is hence proposed, exploiting operational data and available physical knowledge to attain accurate and computationally efficient DHSs dynamic models. The proposed idea consists in leveraging multiple recurrent neural networks (RNNs) and in embedding the physical topology of the DHS network in their interconnections. With respect to standard RNN approaches, the resulting modeling methodology, denoted as physics-informed RNN (PI-RNN), enables to achieve faster training procedures and higher modeling accuracy, even when reduced-dimension models are exploited. The developed PI-RNN modeling technique paves the way for the design of a nonlinear model predictive control (NMPC) regulation strategy, enabling, with limited computational time, to minimize production costs, to increase system efficiency and to respect operational constraints over the whole DHS network. The proposed methods are tested in simulation on a DHS benchmark referenced in the literature, showing promising results from the modeling and control perspective.

show abstract

Section: B Control Resultsmentioning

confidence: 99%

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

de Giuli,

La Bella,

Scattolini

2024

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

show abstract

“…To facilitate the integration of LIME with our Hybrid 1D CNN-GRU model, we leverage the "RecurrentTabularExplainer" package, specifically tailored for handling multidimensional inputs, such as sequential data processed by deep learning models [51,52]. As shown in Fig.…”

Section: B Utilizing Lime For Explanationsmentioning

confidence: 99%

“…Signal data, including ECG signals, possess unique characteristics and temporal dependencies that demand specialized treatment during the explanation process [54]. The traditional approach of LIME, which is designed primarily for tabular data, may not fully capture the intricate patterns and dynamics present in signal data [52]. This limitation becomes evident from the outcomes depicted in Fig.…”

Section: ) Investigating the Understandability Of Lime Explanationsmentioning

confidence: 99%

Sig-Lime: A Signal-Based Enhancement of Lime Explanation Technique

Abdullah,

Zahid,

Turki

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Interpreting machine learning models is facilitated by the widely employed locally interpretable model-agnostic explanation (LIME) technique. However, when extending LIME to signal data, its credibility falters due to perturbation techniques used to generate local datasets. These techniques disrupt temporal dependencies among features, leading to unrealistic data points and potentially misleading explanations. Additionally, LIME faces instability and local fidelity issues, limiting its suitability for real-world applications. The absence of a dedicated LIME package tailored for interpreting signal data further diminishes comprehensibility, especially when applied to models trained on such data. In this paper, we introduce Signalbased LIME (Sig-LIME) to address these limitations. Sig-LIME leverages a novel data generation technique that captures temporal dependence among features, enhancing credibility and stability. It combines a random forest model and heatmaps to provide illuminating explanations for predictions drawn from electrocardiogram (ECG) signals, improving model transparency. Empirical findings underscore the enhanced interpretability and comprehension of model predictions attained by Sig-LIME compared to baseline LIME. Our quantitative evaluation based on an analysis of variance (ANOVA) framework, reveals a notable improvement in stability with Sig-LIME, evidenced by an f-statistic of 0.0 and p-values of 1, indicating a complete absence of variation between multiple runs. Regarding local fidelity, Sig-LIME surpasses the baseline LIME, exhibiting a lower average Euclidean distance of 0.49 compared to 17.24. Sig-LIME excels in generating data more akin to the original, achieving remarkable stability and significantly enhancing credibility and local fidelity in the explanations it generates.

show abstract

“…An uncertainty map is then produced to visualize and interpret the prediction. The authors in [29,30] use Local Interpretable Model-agnostic Explanations (LIME) in a weather forecast context to interpret the decisions from their model. LIME approximates the deep learning network with a simple model as a linear one to understand the relationships in the weather features.…”

Section: Related Workmentioning

confidence: 99%

SAR-UNet: Small Attention Residual UNet for Explainable Nowcasting Tasks

Renault

Mehrkanoon

2023

2023 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

The accuracy and explainability of data-driven nowcasting models are of great importance in many socio-economic sectors reliant on weather-dependent decision making. This paper proposes a novel architecture called Small Attention Residual UNet (SAR-UNet) for precipitation and cloud cover nowcasting. Here, SmaAt-UNet is used as a core model and is further equipped with residual connections, parallel to the depthwise separable convolutions. The proposed SAR-UNet model is evaluated on two datasets, i.e., Dutch precipitation maps ranging from 2016 to 2019 and French cloud cover binary images from 2017 to 2018. The obtained results show that SAR-UNet outperforms other examined models in precipitation nowcasting from 30 to 180 minutes in the future as well as cloud cover nowcasting in the next 90 minutes. Furthermore, we provide additional insights on the nowcasts made by our proposed model using Grad-CAM, a visual explanation technique, which is employed on different levels of the encoder and decoder paths of the SAR-UNet model and produces heatmaps highlighting the critical regions in the input image as well as intermediate representations to the precipitation. The heatmaps generated by Grad-CAM reveal the interactions between the residual connections and the depthwise separable convolutions inside of the multiple depthwise separable blocks placed throughout the network architecture.

show abstract

Explainable heat demand forecasting for the novel control strategies of district heating systems

Cited by 26 publications

References 20 publications

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

Physics-Informed Neural Network Modeling and Predictive Control of District Heating Systems

Sig-Lime: A Signal-Based Enhancement of Lime Explanation Technique

SAR-UNet: Small Attention Residual UNet for Explainable Nowcasting Tasks

Contact Info

Product

Resources

About