Yanan Zhang

Dr. Yanan Zhang

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
16 August 2024 in IEEE Transactions on Sustainable Energy
Verified
Journal Article
Closed access
YZYanan Zhang
Yanan Zhang
GZGan Zhou
Gan Zhou
ZLZhan Liu
Zhan Liu
LHLi Huang
Li Huang
YRYucheng Ren
Yucheng Ren
Heating, ventilation, and air conditioning (HVAC) systems constitute a large proportion of building energy consumption and provide considerable potential for power grid regulation. While the HVAC power consumption forecasting task is generally straightforward with sufficient historical data, it becomes challenging when dealing with scarce data. Such situation is common in cases of intermittent data collection or early system implementations, where precise forecasting is required despite limited data available. Considering accessible datasets from nearby or similar HVAC systems through energy management systems, this paper proposes an adaptive transfer learning framework to tackle this issue. Specifically, the framework leverages diverse source domains, employing model-level regularizers to quantify domain discrepancies and an adaptive parameter regulation mechanism to dynamically align source domains with the target domain. Embedded within the framework, a unique deep learning architecture with attention mechanisms is proposed, capable of identifying complex temporal patterns and hierarchical features in HVAC systems. Experiments on public HVAC datasets demonstrate the generalization, accuracy and robustness of our methodology under diverse data-scarce scenarios.

1 March 2026 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
YZYanan Zhang
Yanan Zhang
GZGan Zhou
Gan Zhou
HMHuan Mao
Huan Mao
WGWei Gu
Wei Gu
YFYanjun Feng
Yanjun Feng
RBRui Bo
Rui Bo
Low-voltage distribution networks often suffer from incomplete or outdated network records, making it challenging to obtain the topology and line parameters under actual operating conditions. To address this issue, a joint identification method is proposed based on the propagation of load transient characteristics. First, the principle of load characteristic propagation is elaborated, and the concept of coupling impedance is introduced. Second, a set of linear regression equations is established based on the changes in current and voltage of the terminal measurements before and after load switching, and then these equations are solved using the least squares method to form the coupling impedance matrix. Finally, a node recursive algorithm is applied to iteratively derive the network topology and calculate line parameters in a bottom-up manner. Case studies show that by effectively extracting the information contained in the transient measurement dataset, the proposed method can correctly identify the network topology and accurately estimate line parameters even under three-phase imbalanced conditions, providing high practical value for engineering applications.

10 February 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
QGQuanxi Guo
Quanxi Guo
YZYanan Zhang
Yanan Zhang
FGFrancesco Grimaccia
Francesco Grimaccia
Alessandro NiccolaiAlessandro Niccolai
Alessandro Niccolai
Data-driven research on battery state is growing rapidly. However, obtaining high-quality battery data often requires extensive and costly experiments, which undoubtedly limits the development of this field. To address this issue, this paper proposes an Increment Capacity (IC) based conditional diffusion model framework for generating high-quality multidimensional battery time series data. This method first uses a variational autoencoder (VAE) to reduce the dimensionality of the IC of early cycles. Subsequently, a convolutional long short-term memory network is used to predict the reduced IC to obtain the IC of future cycles. Finally, the conditional information composed of the IC is input into the Denoising Diffusion Probabilistic Model (DDPM) using a feature-wise linear modulation method. Experimental results show that the proposed method can generate voltage, current, and SOC data for subsequent battery cycles using only early battery data and significantly outperforms traditional diffusion models in terms of accuracy and stability

16 September 2026 in Energies
Verified
Journal Article
Open access
PDF file
ZZZiyao Zhou
Ziyao Zhou
YWYujue Wang
Yujue Wang
YZYanan Zhang
Yanan Zhang
GZGan Zhou
Gan Zhou
FYFeng Yanjun
Feng Yanjun
Within the same low-voltage distribution network (LVDN), the short electrical distances between customers lead to highly similar steady-state voltage curves. This reduces the discriminative capability of traditional similarity metrics and limits the accuracy of customer–meter box relationship identification. To address this problem, this paper proposes a two-stage framework for identifying customer–meter box relationships based on dynamic responses to load switching. In the first stage, the consistency of the dynamic voltage responses of same-phase customers within the same meter box is used to calculate the similarity between customers voltage event sequences. The customers are then divided into single-phase clusters by phase. In the second stage, using the current events generated by load switching as the driving quantity and the cross-phase voltage events as the response quantity, a cross-phase cluster matching model is constructed. Combined with a voting mechanism and an optimized allocation strategy, this approach enables accurate matching of single-phase clusters across different phases and complete reconstruction of the meter box topology. A case study using high-frequency measurement data from an actual LVDN in Nanjing shows that the proposed method achieves an identification accuracy of 100%, demonstrating its effectiveness and superiority in identifying the meter boxes assignments of customers that are electrically close to one another.

1 January 2025 in IET Smart Grid
Verified
Journal Article
Open access
YZYanan Zhang
Yanan Zhang
GZGan Zhou
Gan Zhou
YFYanjun Feng
Yanjun Feng
ZLZhan Liu
Zhan Liu
LHLi Huang
Li Huang
RBRui Bo
Rui Bo
Although deep learning is increasingly promising in the field of Non‐Intrusive Load Monitoring (NILM) these days, the high costs of data recording and labelling represent a significant challenge for the training of supervised models. To address this, a cost‐effective sequence‐to‐points NILM solution is proposed, integrating three‐point labelling with non‐causal convolution techniques. The approach introduces a semi‐automatic labelling framework for obtaining NILM three‐point data, which provides a low‐cost data collection and labelling solution for large‐scale applications. Then, a novel loss function combining coordinate loss and confidence loss is developed to address the positional misalignment and negative sample confusion in sequence‐to‐points scenario in NILM. Furthermore, an advanced neural network architecture based on multi‐scale non‐causal temporal convolution techniques is designed to capture unique features and operational modes of different appliances. Experimental results on the UK‐DALE dataset show that the proposed mixed loss function has an advantage over plain Mean Absolute Error (MAE) on the sequence‐to‐points occasion, and the novel network outperforms on all of the appliances, demonstrating its potential for practical NILM applications.

20 August 2025 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
YZYanan Zhang
Yanan Zhang
GZGan Zhou
Gan Zhou
QGQuanxi Guo
Quanxi Guo
Marco MerloMarco Merlo
Marco Merlo
To address the unavailability of indoor temperature data and limited model generalization in dynamic response modeling of building air conditioning systems (ACs), a framework combining event-driven pseudo-feature generation and an improved generative adversarial networks (IGANs) architecture is proposed. By utilizing data from a data-sufficient source domain, an event-driven algorithm is applied to generate pseudo indoor features and improve the inference accuracy of indoor information. The dynamic response modeling of the target user’s AC system is then reformulated into a data-driven, multi-step time series prediction problem, where a reconstructed dataset consisting of the generated pseudo features, serves as the model input. Then, an IGANs architecture, specifically designed for regression tasks, is introduced, along with the corresponding training strategy that balances spatial alignment and discrepancy capture, improving model performance. Comparative experiments on real-world datasets validate the effectiveness of the proposed method, and sensitivity analysis further reveals the impact of potential factors on the results.