Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimiz...Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimization routines,which in turn require predictive models that can be queried efficiently over large building clusters.However,conventional physics-based simulators are computationally prohibitive for large-scale control training,and simple data-driven surrogates often lack the generalization needed for heterogeneous clusters.This paper introduces ScaleONet,a deep operator network framework designed for scalable,control-oriented modeling of building-cluster thermal dynamics.ScaleONet leverages the DeepONet paradigm to decouple and share learning across buildings:an LSTM-based branch network encodes outdoor climate and individual HVAC control signals,while a multilayer perceptron(MLP)-based trunk network embeds prediction timestamps,enabling fast predictions for growing clusters with negligible extra cost for each additional building or timestep.To the authors’knowledge,this is the first operator-learning method tailored to indoor air temperature forecasting in heterogeneous building clusters.Validation on thirty Belgian buildings(GenkNet)simulated in Dymola shows that,although a non-operator-learning LSTM baseline slightly outperforms ScaleONet for single-building cases,its error grows monotonically with cluster size.In contrast,ScaleONet’s median per-building-per-day RMSE decreases from 0.59°C at three buildings to 0.53°C at ten and 0.47°C at thirty,compared to 0.95°C for the LSTM at thirty buildings-a 51%reduction in prediction error.Error analysis across envelope heat-loss coefficients(UAbuilding)further reveals that while the LSTM’s RMSE increases for high-𝑈𝐴structures,ScaleONet maintains uniformly low error.With millisecond-scale inference(approximately 4 ms per sample for thirty buildings),ScaleONet is well suited for large-scale reinforcement learning,receding-horizon optimization,and real-time model predictive control.展开更多
Digital twins of lithium-ion batteries are increasingly used to enable predictive monitoring,control,and design at system scale.Increasing their capabilities involves improving their physical fidelity while maintainin...Digital twins of lithium-ion batteries are increasingly used to enable predictive monitoring,control,and design at system scale.Increasing their capabilities involves improving their physical fidelity while maintaining sub-millisecond computational speed.In this work,we introduce machine learning surrogates that learn physical dynamics.Specifically,we benchmark three operator-learning surrogates for the Single Particle Model(SPM):Deep Operator Networks(DeepONets),Fourier Neural Operators(FNOs)and a newly proposed parameter-embedded Fourier Neural Operator(PE-FNO),which conditions each spectral layer on particle radius and solid-phase diffusivity.We extend the comparison to classical machine-learning baselines by including U-Nets.Models are trained on simulated trajectories spanning four current families(constant,triangular,pulse-train,and Gaussian-random-field)and a full range of State-of-Charge(SOC)(0%to 100%).DeepONet accurately replicates constant-current behaviour but struggles with more dynamic loads.The basic FNO maintains mesh invariance and keeps concentration errors below 1%,with voltage mean-absolute errors under 1.7mV across all load types.Introducing parameter embedding marginally increases error but enables generalisation to varying radii and diffusivities.PE-FNO executes approximately 200 times faster than a 16-thread SPM solver.Consequently,PE-FNO’s capabilities in inverse tasks are explored in a parameter estimation task with Bayesian optimisation,recovering anode and cathode diffusivities with 1.14%and 8.4%mean absolute percentage error,respectively,and 0.5918 percentage points higher error in comparison with classical methods.These results pave the way for neural operators to meet the accuracy,speed and parametric flexibility demands of real-time battery management,design-of-experiments and large-scale inference.PE-FNO outperforms conventional neural surrogates,offering a practical path towards high-speed and high-fidelity electrochemical digital twins.展开更多
基金supported by KU Leuven,Belgium through the TECHPED-C2 project(C24M/21/021)which investigates tech-nically feasible and effective solutions for Positive Energy Districts.Additional support was provided by the National University of Singa-pore,Singapore through the Start-Up Grant(A-0009876-00-00)the Ministry of Education,Singapore,under the Academic Research Fund Tier 1(A-8003235-00-00).
文摘Delivering energy flexibility at the district scale entails coordinating control actions across many buildings to shape aggregate demand;this coordination depends on training and deploying control policies and optimization routines,which in turn require predictive models that can be queried efficiently over large building clusters.However,conventional physics-based simulators are computationally prohibitive for large-scale control training,and simple data-driven surrogates often lack the generalization needed for heterogeneous clusters.This paper introduces ScaleONet,a deep operator network framework designed for scalable,control-oriented modeling of building-cluster thermal dynamics.ScaleONet leverages the DeepONet paradigm to decouple and share learning across buildings:an LSTM-based branch network encodes outdoor climate and individual HVAC control signals,while a multilayer perceptron(MLP)-based trunk network embeds prediction timestamps,enabling fast predictions for growing clusters with negligible extra cost for each additional building or timestep.To the authors’knowledge,this is the first operator-learning method tailored to indoor air temperature forecasting in heterogeneous building clusters.Validation on thirty Belgian buildings(GenkNet)simulated in Dymola shows that,although a non-operator-learning LSTM baseline slightly outperforms ScaleONet for single-building cases,its error grows monotonically with cluster size.In contrast,ScaleONet’s median per-building-per-day RMSE decreases from 0.59°C at three buildings to 0.53°C at ten and 0.47°C at thirty,compared to 0.95°C for the LSTM at thirty buildings-a 51%reduction in prediction error.Error analysis across envelope heat-loss coefficients(UAbuilding)further reveals that while the LSTM’s RMSE increases for high-𝑈𝐴structures,ScaleONet maintains uniformly low error.With millisecond-scale inference(approximately 4 ms per sample for thirty buildings),ScaleONet is well suited for large-scale reinforcement learning,receding-horizon optimization,and real-time model predictive control.
基金funding from the project“SPEED”(03XP0585)funded by the German Federal Ministry of ResearchTech-nology and Space(BMFTR)and the project“ADMirABLE”(03ETE053E)funded by the German Federal Ministry for Economic Affairs and Energy(BMWE)support of Shell Research UK Ltd.for the Ph.D.studentship of Amir Ali Panahi and the EPSRC Faraday Institution Multi-Scale Modelling Project(FIRG084).
文摘Digital twins of lithium-ion batteries are increasingly used to enable predictive monitoring,control,and design at system scale.Increasing their capabilities involves improving their physical fidelity while maintaining sub-millisecond computational speed.In this work,we introduce machine learning surrogates that learn physical dynamics.Specifically,we benchmark three operator-learning surrogates for the Single Particle Model(SPM):Deep Operator Networks(DeepONets),Fourier Neural Operators(FNOs)and a newly proposed parameter-embedded Fourier Neural Operator(PE-FNO),which conditions each spectral layer on particle radius and solid-phase diffusivity.We extend the comparison to classical machine-learning baselines by including U-Nets.Models are trained on simulated trajectories spanning four current families(constant,triangular,pulse-train,and Gaussian-random-field)and a full range of State-of-Charge(SOC)(0%to 100%).DeepONet accurately replicates constant-current behaviour but struggles with more dynamic loads.The basic FNO maintains mesh invariance and keeps concentration errors below 1%,with voltage mean-absolute errors under 1.7mV across all load types.Introducing parameter embedding marginally increases error but enables generalisation to varying radii and diffusivities.PE-FNO executes approximately 200 times faster than a 16-thread SPM solver.Consequently,PE-FNO’s capabilities in inverse tasks are explored in a parameter estimation task with Bayesian optimisation,recovering anode and cathode diffusivities with 1.14%and 8.4%mean absolute percentage error,respectively,and 0.5918 percentage points higher error in comparison with classical methods.These results pave the way for neural operators to meet the accuracy,speed and parametric flexibility demands of real-time battery management,design-of-experiments and large-scale inference.PE-FNO outperforms conventional neural surrogates,offering a practical path towards high-speed and high-fidelity electrochemical digital twins.
