A novel decentralized method based on the system engineering concept for reliability‐security constraint unit commitment in restructured power environment

Abstract

In the deregulated power environment, including Central operator (CO) and Micro Grids (MGs), different parts of the network are dedicated to the private sector, and each of them seeks to increase their profits independently. The CO and MGs should cooperate and collaborate in terms of operating, security and reliability in the whole power system. This article proposes a new method based on a System of System (SoS) concept for the secure and economic hourly generation scheduling to find optimal operational point. The main methodology includes three steps. In the first step, the power system is divided into several subsystems by using a spectral clustering partitioning technique to reduce converge time by decentralizes methods. And also load forecasting based on a Gaussian probability distribution function to avoid conventional calculation and considering uncertainty of the loads has been presented. To find a similar scenario, and reduction scenario with low probability, the probabilistic method has been addressed. The main contribution of this method is removing scenarios with low value of probabilities and scenarios which are similar to each other. In fact, the reduced set must include scenarios which are scattered appropriately in the uncertain space while holding high probabilities. In order to estimate the similarity (distance) between two scenarios the Kantorovich distance is implemented. In the second step, the hierarchical Bi-level optimization approach is used to execute the decentralized decision making to solve the Security Constraints Unit Commitment (SCUC) problem between CO and MGs. Regarding all physical relations and shared data among CO and MGs, the SoS concept and Bi-level optimization are presented to find the optimal operating point of autonomous systems. In the third step, a random number of generators will be select. Hence, the initial iteration number is set. In this step, sampling from state space to classifying reliability object achieved (The expected energy not supplied and loss of load probability are the reliability criterion). The presented method is evaluated using a 6-bus, the RTS 24-bus,

An optimal probabilistic spinning reserve quantification scheme considering frequency dynamic response in smart power environment

“…The probability distribution function is categorized into five segments and the amount of the probability for each segment is calculated to produce the required scenarios. [32][33][34][35] 3.4 | Demand side management modeling (load shifting program)…”

“…Most methods of reliability analysis are including analytical, simulation, and recently population-based intelligent search methods. 34 Reliability analysis contains simple levels such as state selection and evaluation, and index calculation. In this paper, an intelligent search technique is utilized for reliability evaluation, which is proposed in Reference 34.…”

Enhancement in reliability‐constrained unit commitment considering state‐transition‐process and uncertain resources

“…After constructing the regression function , a cross‐verification is employed to check the validity of the regression function. Especially, the leave‐one‐out method, introduced in [23], is used for model evaluation. Each sampled data is considered and assessed by using it, which ensures the obtained function can effectively explain the mapping between rescheduling and the normal‐state scheduling result.…”

Decentralized approach for security enhancement of wind‐integrated energy systems coordinated with energy storages

“…In this manner, decomposition algorithms are presented as proper solutions to meet the difficulties associated with this problem. Therefore, decomposition algorithms are employed in References 16‐21 to tackle the SCOPF problem in energy systems. In Reference 16, a new approach for security‐constrained economic dispatch is presented that uses an iterative decomposition technique to reduce the computational burden and reach a lower operating cost.…”