Outline
- Abstract
- Keywords
- Introduction
- Problem Formulation
- Decision Variables
- Objective Function
- Mt Cost
- Wt Probabilistic Model
- Battery Cost
- Network Cost
- Loss Cost
- Switching Cost
- System Constraints
- Unit Constraints
- System Topology
- Uncertainty Model
- Proposed Algorithm
- Simulation Results
- Conclusion
- References
رئوس مطالب
- چکیده
- کلید واژه ها
- مقدمه
- فرمول بندی مسئله
- متغیرهای تصمیم گیری
- تابع هدف
- هزینه MT
- مدل آماری WT
- هزینه باتری
- هزینه شبکه
- هزینه سوئیچینگ
- محدودیت‌های سیستم
- محدودیت‌های واحد
- تکنولوژی سیستم
- مدل عدم قطعیت
- الگوریتم پیشنهادی
- نتایج شبیه سازی
- نتیجه گیری
Abstract
A stochastic model for day-ahead Micro-Grid (MG) management is proposed in this paper. The presented model uses probabilistic reconfiguration and Unit Commitment (UC) simultaneously to achieve the optimal set points of the MG’s units besides the MG optimal topology for day-ahead power market. The proposed operation method is employed to maximize MG’s benefit considering load demand and wind power generation uncertainty. MG’s day-ahead benefit is considered as the Objective Function (OF) and Particle Swarm Optimization (PSO) algorithm is used to solve the problem. For modeling uncertainties, some scenarios are generated according to Monte Carlo Simulation (MCS), and MG optimal operation is analyzed under these scenarios. The case study is a typical 10-bus MG, including Wind Turbine (WT), battery, Micro-Turbines (MTs), vital and non-vital loads. This MG is connected to the upstream network in one bus. Finally, the optimal set points of dispatchable units and best topology of MG are determined by scenario aggregation, and these amounts are proposed for the day-ahead operation. In fact, the proposed model is able to minimize the undesirable impact of uncertainties on MG’s benefit by creating different scenarios.
Keywords: Micro-grid - reconfiguration - Uncertainty - Unit commitment - Wind power generationConclusions
A stochastic method for simultaneous MG reconfiguration and UC is proposed for each hour of the day-ahead. MG operation and performance are analyzed with different input scenarios. The results show more benefit of this algorithm than only economic dispatch. The proposed algorithm is able to approach to the optimal MG benefit, the units’ set points and MG’s topology for each hour of day-head. WT generation and load demand are considered as uncertain inputs. Considering enough scenarios, optimal three MTs generation, battery charge or discharge power exchange with upstream network and the most repeated topology are determined for each hour of the next day. The average of continuous variables and the most repeated topology for each hour are assessed by scenarios as a suggestion for the next day. This method was applied to a typical MG. The benefit coefficient of variation becomes converges after approximately 50 iterations.
The future work in the field of this paper should take substantial problems which MG manager faces into consideration. Some of the examples include demand response, photovoltaic generation, plugin hybrid electric vehicles, and uncertainty of prices.