A Probabilistic Convolution-Based Model for the Equivalent Reliability of the Distribution System

In traditional reliability assessment of power transmission systems, the distribution network is often equivalenced with a constant load, which is no longer able to meet the reliability evaluation requirements of modern power systems. Hence, this paper suggests a precise alternative model for evaluating the reliability of power transmission systems' distribution networks, employing probabilistic convolution. The continuous load curve of the distribution network undergoes discretization, transforming into a discrete probability model of load demand. Furthermore, using the sequential Monte Carlo approach, a discrete probability distribution model is developed for the distribution network's failure intensity. By utilizing the theory of probabilistic convolution, the problem of solving the precise equivalent model of the distribution network is transformed into determining the joint probability density distribution and expectation of independent distributed two-dimensional discrete random variables. This approach accurately reflects the impact of variations in both the distribution network's load level and reliability level on the power transmission system. As a result, it achieves the precise equivalent of the distribution network's load level in reliability assessment. The efficacy of the suggested approach is showcased by evaluating and analyzing the RBTS test system, particularly its distribution system at BUS4.