Jiawei Zhang

The increasing penetration of renewable energy leads to the frequent occurrences of zero-marginal-cost phenomenon in the spot market and causes missing money problems for renewable power producers (RPPs). Engagement in the forward market is a viable solution for RPPs to address this issue. This paper establishes a distributionally robust meanvariance model to optimize the RPPs' forward trading strategies. The model is further transformed into a tractable second-ordercone-programming one. Based on this model, RPPs are able to ensure their revenue sufficiency and mitigate revenue risks through the engagement in the forward market. We conduct case studies on RPPs' forward trading strategies under three typical market scenarios. In addition, we study the impacts of delivery mechanisms and the price of forward market on RPPs' trading strategies.

Poorly damped converter-driven oscillations are becoming increasingly prevalent as conventional synchronous generation (SG) is progressively replaced by inverter-based generation (IBG). Future power systems should strategically optimize the investment decisions of IBG and SG to prevent unstable dynamic interactions. However, different combinations of candidate generators would introduce varying oscillation modes, complicating the task of reserving sufficient oscillation damping for various critical modes during generation expansion. To address this challenge, this paper formulates a constraint-learning-based generation expansion planning framework. First, a cost-dominant sampling method is proposed to generate diverse generation mix scenarios, encompassing various on-off states and outputs of candidate generators. Next, stability constraints are learned for each generator and then grouped into an ensemble through the proposed physical-ensemble constraint learning method. These constraints are dynamically integrated into generation expansion and activated based on the generator's on-off status to maintain converter-driven stability across various generation mix scenarios. The feasibility and effectiveness of the proposed framework are validated on a modified IEEE standard test case and a real-world power grid with high IBG penetration. Case studies demonstrate that the proposed method outperforms state-of-the-art approaches by reducing computational burden while improving the proportion of stable states to over 95% in the operation simulations of the planned power systems.

The frequency responses at different buses following a disturbance differ from that of the center-of-inertia (COI). Buses exhibiting much lower frequency nadirs than the COI nadir may pose stability concerns. In this letter, we demonstrate that nodal frequency responses are composed of global and local components. We propose a natural analytical decomposition method to separate these two components. The decomposition provides an insight into the overlap of angle and frequency dynamics. Based on this decomposition, we also derive an indicator that identifies buses susceptible to frequency stability issues. We believe that our findings are fundamental for deepening the understanding of nodal frequency dynamics, with potential applications in related studies. Our work is supported by both numerical simulations and rigorous theoretical analysis.

High voltage direct current (HVDC) power injections decrease power systems' short-circuit ratio (SCR) and the grid strength. They may cause voltage stability problems, especially in systems with a high share of renewable. The synchronous condenser (SynCon) is a promising solution for increasing the SCR. However, planning SynCons to meet the SCR requirements is challenging since the SCR requirement is a set of nonlinear eigenvalue inequality constraints that are not friendly to solvers. Based on generalized spectral decomposition, this paper proposes a Rayleigh cut method to equivalently transform eigenvalue inequality constraints into a set of mixed- integer linear constraints. Then, a mixed- integer linear programming model is built to solve the SCR-constrained SynCon placement problem. The proposed approach's convergence, complexity, and error analysis are all analyzed theoretically. Also, numerical comparisons with the other seven models suggest the proposed model is effective and efficient, outperforming the other models in terms of solution quality and computation time.

Increasing renewable energy penetrations bring complex feasibility and stability problems. Data-driven methods are applied in extracting and embedding these feasibility and stability rules in power system operations and planning. This paper presents a method of alternate support vector machine decision trees for rule extraction problems. The method significantly improves the classical decision-tree-based algorithms' efficiency, stability, and versatility. Finally, we apply the method to several power and energy system scenarios to show its effectiveness.

Renewable energy resources, such as wind and solar energy, have become the primary components of power systems. However, the uncertainty and fluctuations associated with these resources increase the difficulty to follow renewable fluctuations using conventional generators. Energy storage systems are one of the best choices for improving the mechanical performance limitations of conventional units. In this paper, we analyze the dynamic performance of the conventional-storage frequency regulation model and provide parameter and capacity setting rules for storage. Furthermore, we allocate the storage capacity to buses under different operation modes to minimize transmission-line fluctuations. We use a possible planning power systems in Jiangsu Province, China, in 2030 to analyze and verify our proposal. The system has more than 1000 buses, 2000 branches and 700 generators, and 220, 500 and 1000 kV voltage levels. The results show that using a small amount of storage is feasible for improving regulation performances. Additionally, the optimal energy storage placement effectively reduces the fluctuations and optimizes the power flows in the transmission systems.

Increasing renewable energy penetrations bring complex feasibility and stability problems. Data-driven methods are applied in extracting and embedding these feasibility and stability rules in power system operations and planning. This paper presents a method of alternate support vector machine decision trees for rule extraction problems. The method has significant improvements to the classical decision-tree-based algorithms in terms of efficiency, stability and versatility. Finally, we apply the method to several power and energy system scenarios to show its effectiveness.

The increasing integration of variable renewable energy increases the demand for a power system’s frequency regulation resources. The decreasing share of controllable power generators makes the frequency regulation resources insufficient. Simultaneously, distributed energy resources (DER) on the distribution network side, such as storage devices, electric vehicles, and controllable building energy systems, have excellent potential to provide frequency support. However, power system dispatchers cannot control these resources in the same manner as conventional power sources. These sources have power output uncertainties and cannot directly access power system communication networks. This paper proposes a cloud-edge collaboration and wireless communication coordination framework to facilitate DER frequency regulations. The cloud balances the power uncertainty of controllability in edges together with computing burden, wire, and wireless communication loads. Also, the edges adjust storage’s power outputs to achieve frequency regulations and covering PV power fluctuations. A numerical case study demonstrates the effectiveness of the proposed method. The solution holds 95.5% frequency regulation adjustable power capacity of theoretical maximum on average. Meanwhile, the computing costs reduce by 36.3%, and wire communication only operates in 0.35% of a day as a supplement (reduces by 99.65%).