Yiyan Li

Prof. Yiyan Li


Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
16 January 2023 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
YLYiyan Li
Yiyan Li
LSLidong Song
Lidong Song
SZSi Zhang
Si Zhang
LKLaura Kraus
Laura Kraus
TATaylor Adcox
Taylor Adcox
NLNing Lu
Ning Lu
This paper presents a Temporal Convolutional Network (TCN) based hybrid PV forecasting framework for enhancing hours-ahead utility-scale PV forecasting. The hybrid framework consists of two forecasting models: a physics-based trend forecasting (TF) model and a data-driven fluctuation forecasting (FF) model. Three TCNs are integrated in the framework for: i) blending the inputs from different Numerical Weather Prediction sources for the TF model to achieve superior performance on forecasting hourly PV profiles, ii) capturing spatial-temporal correlations between detector sites and the target site in the FF model to achieve more accurate forecast of intrahour PV power drops, and iii) reconciling TF and FF results to obtain coherent hours-ahead PV forecast with both hourly trends and intra-hour fluctuations well preserved. To automatically identify the most contributive neighboring sites for forming a detector network, a scenario-based correlation analysis method is developed, which significantly improves the capability of the FF model on capturing large power fluctuations caused by cloud movements. The framework is developed, tested, and validated using actual PV data collected from 95 PV farms in North Carolina. Simulation results show that the performance of 6 hours ahead PV power forecasting is improved by 20% -30% compared with state-of-the-art methods.

7 July 2023 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
YLYiyan Li
Yiyan Li
LSLidong Song
Lidong Song
YHYi Hu
Yi Hu
HLHanpyo Lee
Hanpyo Lee
DWDi Wu
Di Wu
NLNing Lu
Ning Lu
This paper introduces a Generative Adversarial Nets (GAN) based, Load Profile Inpainting Network (Load-PIN) for restoring missing load data segments and estimating the baseline for a demand response event. The inputs are time series load data before and after the inpainting period together with explanatory variables (e.g., weather data). We propose a Generator structure consisting of a coarse network and a fine-tuning network. The coarse network provides an initial estimation of the data segment in the inpainting period. The fine-tuning network consists of self-attention blocks and gated convolution layers for adjusting the initial estimations. Loss functions are specially designed for the fine-tuning and the discriminator networks to enhance both the point-to-point accuracy and realisticness of the results. We test the Load-PIN on three real-world data sets for two applications: patching missing data and deriving baselines of conservation voltage reduction (CVR) events. We benchmark the performance of Load-PIN with five existing deep-learning methods. Our simulation results show that, compared with the state-of-the-art methods, Load-PIN can handle varying-length missing data events and achieve 15%-30% accuracy improvement.

9 March 2022 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
LSLidong Song
Lidong Song
YLYiyan Li
Yiyan Li
NLNing Lu
Ning Lu
This paper presents a novel two-stage load profile super-resolution (LPSR) framework, ProfileSR-GAN, to upsample the low-resolution load profiles (LRLPs) to high-resolution load profiles (HRLPs). The LPSR problem is formulated as a Maximum-a-Posteriori problem. In the first-stage, a GAN-based model is adopted to restore high-frequency components from the LRLPs. To reflect the load-weather dependency, aside from the LRLPs, the weather data is added as an input to the GAN-based model. In the second-stage, a polishing network guided by outline loss and switching loss is novelly introduced to remove the unrealistic power fluctuations in the generated HRLPs and improve the point-to-point matching accuracy. To evaluate the realisticness of the generated HRLPs, a new set of load shape evaluation metrics is developed. Simulation results show that: i) ProfileSR-GAN outperforms the state-of-the-art methods in all shape-based metrics and can achieve comparable performance with those methods in point-to-point matching accuracy, and ii) after applying ProfileSR-GAN to convert LRLPs to HRLPs, the performance of a downstream task, non-intrusive load monitoring, can be significantly improved. This demonstrates that ProfileSR-GAN is an effective new mechanism for restoring high-frequency components in downsampled time-series data sets and improves the performance of downstream tasks that require HR load profiles as inputs.

4 August 2023 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
YHYi Hu
Yi Hu
YLYiyan Li
Yiyan Li
LSLidong Song
Lidong Song
HLHan Pyo Lee
Han Pyo Lee
PRP. J. Rehm
P. J. Rehm
NLNing Lu
Ning Lu
This paper presents a deep-learning framework, Multi-load Generative Adversarial Network (MultiLoad-GAN), for generating a group of synthetic load profiles (SLPs) simultaneously. The main contribution of MultiLoad-GAN is the capture of spatial-temporal correlations among a group of loads that are served by the same distribution transformer. This enables the generation of a large amount of correlated SLPs required for microgrid and distribution system studies. The novelty and uniqueness of the MultiLoad-GAN framework are three-fold. First, to the best of our knowledge, this is the first method for generating a group of load profiles bearing realistic spatial-temporal correlations simultaneously. Second, two complementary realisticness metrics for evaluating generated load profiles are developed: computing statistics based on domain knowledge and comparing high-level features via a deep-learning classifier. Third, to tackle data scarcity, a novel iterative data augmentation mechanism is developed to generate training samples for enhancing the training of both the classifier and the MultiLoad-GAN model. Simulation results show that MultiLoad-GAN can generate more realistic load profiles than existing approaches, especially in group level characteristics. With little finetuning, MultiLoad-GAN can be readily extended to generate a group of load or PV profiles for a feeder or a service area.

17 May 2017 in Journal of Modern Power Systems and Clean Energy
Verified
Journal Article
Open access
PDF file
YLYiyan Li
Yiyan Li
DHDong Han
Dong Han
ZYZheng Yan
Zheng Yan
In this paper, a data-driven linear clustering (DLC) method is proposed to solve the long-term system load forecasting problem caused by load fluctuation in some developed cities. A large substation load dataset with annual interval is utilized and firstly preprocessed by the proposed linear clustering method to prepare for modelling. Then optimal autoregressive integrated moving average (ARIMA) models are constructed for the sum series of each obtained cluster to forecast their respective future load. Finally, the system load forecasting result is obtained by summing up all the ARIMA forecasts. From error analysis and application results, it is both theoretically and practically proved that the proposed DLC method can reduce random forecasting errors while guaranteeing modelling accuracy, so that a more stable and precise system load forecasting result can be obtained.

7 October 2021 in IEEE Transactions on Cloud Computing
Verified
Journal Article
Closed access
Qiangang JiaQiangang Jia
Qiangang Jia
SCSijie Chen
Sijie Chen
ZYZheng Yan
Zheng Yan
YLYiyan Li
Yiyan Li
In the residential demand response area, currently the incentive-based method (e.g., direct load control, DLC) may impair users’ comfort and autonomy, while the price-based method can hardly guarantee users’ engagements. This paper proposes an edge-cloud integrated demand response framework to achieve an effect-predictable residential demand response without harming users’ benefits. First, we combine the cloud-computing resource (cloud) and the home-installed smart thermostats (edges) to formulate an efficient, cost-effective, and data-secured infrastructure to implement the demand response program. Then, we model the demand response problem between the load aggregator and its served residential users as a bi-level optimization problem, and the key is for the load aggregator to find the optimal incentive strategy. To solve this problem, we introduce an RL algorithm, i.e., Continuous Action Reinforcement Learning Automata, to quickly obtain the optimal incentive strategy under an incomplete information scenario. Simulation results based on 136 real-world residential users in Austin area demonstrate that the proposed CEI-DR framework can increase the social welfare by about $8.6/h compared to the traditional DLC method during a normal DR event.