Zhichao Shi

Zhichao Shi

Auto-generated profile, based on publicly available information

Zhichao Shi

Department of Electrical and Computer Engineering University of Alberta Edmonton AB Canada

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 November 2017 in IEEE Transactions on Sustainable Energy
Verified
Journal Article
Closed access
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
VDVenkata Dinavahi
Venkata Dinavahi
Interval forecast is an efficient method to quantify the uncertainties in renewable energy production. In this work, the idea of prediction intervals (PIs) is employed to capture the uncertainty of wind power generation in power systems. The recurrent neural network (RNN) model is proposed to construct PIs with the lower upper bound estimation (LUBE) method, which is a powerful nonparametric forecast approach. In addition, a novel comprehensive cost function with a new PI evaluation index is designed with the purpose of enhancing the model training. To tune the parameters of RNN prediction model, the dragonfly algorithm (DA) with a linearly random weight update method is introduced as the optimization tool. The performance of the proposed prediction model is validated by a case study using a real world wind power dataset, and the comparative results show the superiority of the model.

11 January 2018 in IEEE Transactions on Smart Grid
Verified
Journal Article
Closed access
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
SHShengjun Huang
Shengjun Huang
VDVenkata Dinavahi
Venkata Dinavahi
With the development of smart grid, energy management becomes critical for reliable and efficient operation of power systems. In this work, we develop a chance-constrained (CC) energy management model for an islanded microgrid which includes distributed generators (DGs), energy storage system (ESS), and renewable generation such as wind power. The objective function of this model consists of generation cost, emission cost and ESS degradation cost. To capture the uncertainty of renewable generation, a novel ambiguity set is introduced without knowing its probability distribution or exact moment information. Based on the ambiguity set, the chance constraint can be processed with distributionally robust optimization (DRO) method and the energy management problem is reformulated as a tractable second-order conic programming (SOCP) problem. The proposed approach is tested with a case study and simulation results indicate that it is effective and reliable. Moreover, the comparison with the method based on known moment information and some other methods is also conducted to show the performance of the proposed method.

23 October 2018 in IEEE Access
Verified
Journal Article
Open access
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
VDVenkata Dinavahi
Venkata Dinavahi
As a source of clean and pollution-free renewable energy, wind power has attracted much attention and has been increasingly integrated into the existing power system. However, the uncertain and volatile wind speed makes the utilization of wind power a challenging task. Hence, it is essential to design accurate forecast method to deal with the uncertainty caused by wind speed. This work proposes a multiobjective interval prediction method based on wavelet neural network (WNN) for short-term wind speed forecast. This method can generate a set of Pareto optimal solutions which represents a set of prediction models that can directly construct the prediction intervals (PIs). An advanced multiobjective evolutionary algorithm, preference inspired co-evolutionary algorithm using goal vectors (PICEA-g), is investigated to train the WNN model. Two case studies are carried out with real wind speed data of Victoria and Edmonton in Canada to justify the effectiveness of the proposed method. The numerical results also show the superiority of the proposed forecast approach compared with some benchmark methods.

25 April 2019 in IET Generation, Transmission & Distribution
Verified
Journal Article
Open access
PDF file
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
SHShengjun Huang
Shengjun Huang
VDVenkata Dinavahi
Venkata Dinavahi
Microgrids play an important role in modern power systems which can integrate different kinds of distributed energy resources (DERs). To deal with the uncertainty from various factors such as renewable generation, robust energy management for microgrids has become a significant problem. In this work, a novel multistage robust energy management model for grid‐connected microgrids is developed which considers the uncertainty of renewable generation and load demand. The multistage energy management problem is complex and computationally difficult. To solve this problem, a robust version of dual dynamic programming method is proposed which includes a forward pass and a backward pass procedure and has a similar framework with the common stochastic dual dynamic programming (SDDP) method. Based on real datasets, a case study is carried out to validate the effectiveness of the proposed model and solution methodology. Numerical results show that the proposed approach can effectively achieve the robust optimal solution, and the comparison with other methods also testifies the advantage of the proposed multistage robust model.

18 September 2019 in IEEE Access
Verified
Journal Article
Open access
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
VDVenkata Dinavahi
Venkata Dinavahi
The Unit commitment (UC) problem in power systems has been studied for a long time; however, many new challenges have emerged in the UC problem with the increasing penetration of renewable generation which is intermittent and uncertain. Compared with the common uncertainty modeling methods including stochastic programming and robust optimization, in this paper, we develop a data-driven distributionally robust chance-constrained (DDRC) UC model. The proposed two-stage UC model focuses on the commitment decision and dispatch plan in the first stage, and considers the worst-case expected cost for possible power imbalance or re-dispatch in the second stage. To capture the uncertainty of wind power distribution, a distance-based ambiguity set is designed which can be constructed in a data-driven manner. Based on the ambiguity set, the original complicated UC problem is reformulated to a tractable optimization problem which is then solved by the column-and-constraint generation (C&CG) algorithm. The performance of the the proposed approach is validated by case studies with different test systems including the IEEE 6-bus test system, modified IEEE 118-bus system and a practical-scale system, especially the value of data in controlling the conservativeness of the problem.

10 July 2020 in IET Generation, Transmission & Distribution
Verified
Journal Article
Open access
PDF file
ZSZhichao Shi
Zhichao Shi
HLHao Liang
Hao Liang
VDVenkata Dinavahi
Venkata Dinavahi
To improve the overall energy utilisation efficiency, the research of combined cooling, heat, and power (CCHP)‐based microgrids has become prevalent recently. However, the increasing penetration of uncertain renewable generation such as wind power brings new challenges to CCHP‐based microgrids energy management. In this study, the authors propose a two‐stage multi‐period distributionally robust energy management model for CCHP‐based microgrids, and this model considers the non‐anticipativity of uncertainty in dispatch process. A second‐order conic representable ambiguity set is designed to capture the uncertainty of wind power. Based on linear decision rule approximation, the proposed problem is transformed into a tractable mixed‐integer second‐order conic programme problem. Case studies and comparison experiments are conducted in the Matlab environment with real‐world data to validate the performance of the proposed approach. Particularly, the proposed method achieves a less conservative solution and smaller cost compared with a robust optimisation method with the same reliability guarantee. In addition, it is more reliable than the deterministic method which does not consider uncertainty.