[论文解读] Physically Consistent Neural Networks for building thermal modeling: theory and analysis
本文提出物理一致神经网络(PCNNs),一种新颖的神经网络架构,通过在深度神经网络并行的线性模块中整合先验物理知识,确保建筑热力预测的热力学一致性。PCNNs 在 3 天预测时域内相比经典 RC 模型误差降低 40%,同时保持泛化鲁棒性,并在验证数据上优于标准 LSTM 模型。
Due to their high energy intensity, buildings play a major role in the current worldwide energy transition. Building models are ubiquitous since they are needed at each stage of the life of buildings, i.e. for design, retrofitting, and control operations. Classical white-box models, based on physical equations, are bound to follow the laws of physics but the specific design of their underlying structure might hinder their expressiveness and hence their accuracy. On the other hand, black-box models are better suited to capture nonlinear building dynamics and thus can often achieve better accuracy, but they require a lot of data and might not follow the laws of physics, a problem that is particularly common for neural network (NN) models. To counter this known generalization issue, physics-informed NNs have recently been introduced, where researchers introduce prior knowledge in the structure of NNs to ground them in known underlying physical laws and avoid classical NN generalization issues. In this work, we present a novel physics-informed NN architecture, dubbed Physically Consistent NN (PCNN), which only requires past operational data and no engineering overhead, including prior knowledge in a linear module running in parallel to a classical NN. We formally prove that such networks are physically consistent - by design and even on unseen data - with respect to different control inputs and temperatures outside and in neighboring zones. We demonstrate their performance on a case study, where the PCNN attains an accuracy up to 40% better than a classical physics-based resistance-capacitance model on 3-day long prediction horizons. Furthermore, despite their constrained structure, PCNNs attain similar performance to classical NNs on the validation data, overfitting the training data less and retaining high expressiveness to tackle the generalization issue.
研究动机与目标
- 解决经典神经网络在建筑热力建模中因违反热力学等物理定律而导致的泛化失败问题。
- 通过将物理约束直接嵌入网络架构中,克服纯数据驱动模型的局限性,且无需工程开销。
- 开发一种在所有控制输入和外部温度条件下均保持物理一致性的模型,即使在长时预测中亦然。
- 在保留深度学习表达能力的同时,提升对经典物理模型(如 RC 模型)的预测精度。
- 使神经网络在控制导向应用(如强化学习)中可被可靠使用,其中物理合理性至关重要。
提出的方法
- 设计一种混合架构,其中标准深度神经网络处理一般输入,而独立的线性模块则利用已知参数(如传热系数)编码物理定律。
- 通过关键输入(如室外温度、邻近区域温度、太阳得热和控制输入)的线性变换,将物理约束整合到网络输出预测中。
- 正式证明:无论控制输入和外部温度如何组合,网络输出在训练和推理过程中均保持物理一致性。
- 在黑箱神经网络与物理信息模块之间使用残差连接,使网络能够学习理想物理行为的偏差,同时保持一致性。
- 仅使用历史运行数据进行模型训练,无需工程专业知识或对物理参数的手动调优。
- 以先验知识(如缩放参数 a, b, c, d)初始化物理模块,并在训练过程中允许其微调,同时保持物理合理性。
实验结果
研究问题
- RQ1能否设计一种神经网络架构,在无需工程专业知识的前提下,保证建筑热力预测的物理一致性?
- RQ2将物理定律直接嵌入网络结构是否能提升相比标准神经网络的泛化性能?
- RQ3数据驱动模型在保持物理一致性的前提下,能在多大程度上超越经典物理基 RC 模型的长期温度预测性能?
- RQ4在分布偏移情况下,PCNN 与标准深度学习模型(如 LSTM)在训练和验证误差方面的表现如何比较?
- RQ5即使初始值非理想,物理信息模块是否能有效从数据中学习,同时保持物理合理性?
主要发现
- PCNNs 在 3 天预测时域内相比调优后的经典电阻-电容(RC)模型,预测误差降低 40%。
- 尽管结构受限,PCNNs 在验证数据上的表现与标准 LSTM 相当,且在训练过程中表现出更少的过拟合。
- PCNNs 的泛化能力优于经典神经网络:训练时表现较差,但在验证数据上显著更优,表明过拟合减少且鲁棒性增强。
- 该模型在所有测试的控制输入和外部温度条件下均保持物理一致性,无论在训练还是推理阶段,且已正式证明。
- 物理信息模块在整个训练过程中保持物理合理性,如 a, b, c, d 等参数仅围绕初始值缓慢调整,未违反物理约束。
- PCNNs 无需工程开销或物理参数的手动调优,因此具有高度灵活性,可直接部署于实际建筑控制应用中。
更好的研究,从现在开始
从阅读论文到最终审阅,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。