Dynamic stochastic optimal power flow of wind power and the electric vehicle integrated power system considering temporal-spatial characteristics

Abstract

The intermittent volatility of wind power integrated into the grid poses a great threat to the stable operation of power systems on the supply side. Conversely, large-scale charging of electric vehicles (EVs) also brings new challenges to dispatch on the demand side. In response, the randomness and temporal-spatial correlations of stochastic wind power generation are considered in this paper. Additionally, the EV charging infrastructure is studied. A dynamic stochastic optimal power flow (DSOPF) for wind farms and EVs integrated power system based on the chance-constrained programming model is proposed. An optimal dispatch scheme is obtained by solving the dynamic optimal power flow. After that, dynamic probabilistic power flow based on cumulants is performed under the scheme to obtain the probability distribution of state variables. The upper and lower bounds of chance constraints are adjusted according to the probability distribution function until they are all satisfied. Illustrative examples demonstrate the effectiveness of DSOPF for firming the variable wind energy, and EV charging is performed on the Institute of Electrical and Electronics Engineers systems. On this basis, different EV charging modes and the temporal-spatial correlations are specifically discussed.

Multi-stage Stochastic Alternating Current Optimal Power Flow with Storage: Bounding the Relaxation Gap

“…Non-anticipativity is guaranteed in [20] using affine-linear policies but with a (linear) DC OPF model. Non-anticipativity of the decisions is also guaranteed in [21] which considers an iterative procedure to optimize a decision policy. By comparison, the approach by scenario trees developed here accounts both for non-anticipativity constraints and for the nonlinearity of the stochastic OPF problem, whereas it benefits from theoretical convergence guarantees.…”

“…The inequalities on S and S Lin arise from the constraint (3) which implies that I is non-negative component-wise, from passivity of the network and from constraints (12) and (20). The inequalities on s 0 and s Lin 0 can then be deduced by the inequality between S and S Lin and constraints ( 13) and (21). Comparing ( 14) and ( 22), using the passivity of the network and the inequalities between S and S Lin , one gets for all −−→ (i, j) in E:…”

Stochastic Security Assessment for Power Systems With High Renewable Energy Penetration Considering Frequency Regulation

“…In [16], a non-parametric approach is used to simulate the forecast error sequence of solar power based on the historical data. Moreover, the methodology in [17] can be used to compute the correlation coefficient matrix [18] to generate the correlated forecast error samples, when given the spatial correlation between the renewable power injections at adjacent nodes.…”

Random-Fuzzy Chance-Constrained Programming Optimal Power Flow of Wind Integrated Power Considering Voltage Stability

“…Currently, research on SOPF mostly concentrates on a single time section. In this direction, reference [19] from the aspect of day-ahead time horizon and depicting WG output forecast via auto regression moving average, has researched chanceconstrained DSOPF which considers randomness of forecast error and temporal-spatial correlation. In [20], the scenario tree model is applied to approximate the stochastic nature of WG in the 24-h operation horizon optimization problem.…”