Probabilistic Reliability Calculation Method Of Distribution Networks Based On Improved Three-Point Estimation Method And Edgeworth Series

The massive access to distributed power sources has transformed the distribution network into an active distribution network with full interconnection of multiple sources and users.which has increased the complexity of distribution network reliability assessment.The expansion of the distribution network also leads to an exponential increase in the amount of data.and the performance bottleneck of traditional reliability assessment algorithms gradually emerges.The problem of further improving the computational speed of distribution network reliability while ensuring the computational accuracy has become an urgent issue to be solved.Since the expected value of the traditional reliability index does not fully reflect the change information of the index value,the concept of probabilistic reliability of the distribution network is introduced.Probabilistic reliability is defined as the probability distribution of reliability indicators,which can describe the fluctuation range and the probability of occurrence of different calculated values of indicators,reflecting the risk level of system operation in terms of probability,helping to identify system states that have a low probability of occurrence,and is particularly important for detecting incidents that do not occur very frequently in the system but have a serious impact.Current research into the probabilistic reliability of distribution networks has few methods that can combine both accuracy and speed,so an efficient and accurate method for calculating probabilistic reliability of distribution networks needs to be investigated in depth.In addition,there is a correlation between distributed power sources or loads in close geographical proximity due to conditions such as light and temperature.The analysis and calculation of the reliability of the active distribution network needs to reveal the influence of the load and distributed power supply power fluctuation in the system and the correlation between them on the reliability of the distribution network,so the correlation between the load and distributed power supply cannot be ignored in the probabilistic reliability calculation process.This paper proposes a fast and accurate calculation method for the probabilistic reliability of distribution networks to address the above issues.To address the problem of large number of samples required by the traditional Monte Carlo method.this paper uses an improved three-point estimation method and a third-order polynomial normal transformation to generate the distribution network probabilistic reliability calculation samples,which effectively reduces the size of the input sample points of the Monte Carlo method while retaining the correlation of random variables.Then the sequential Monte Carlo method is applied to the reduced sample points to obtain the values of the reliability indicators corresponding to each sample point,and the probability density function of each reliability indicator is obtained using the Edgeworth series,and the probability reliability of the indicators is further obtained by segmental integration,i.e.the probability distribution of the reliability indicators.Finally,an arithmetic analysis is carried out based on the improved IEEE-RBTS Bus6 for the F4 feeder.The results show that there is only a maximum deviation of 2.192%between the calculation results of the distribution system reliability index of the proposed method and the traditional Monte Carlo method,and the calculation time of the proposed method is only 1.05%of that of the traditional Monte Carlo method,which verifies the efficiency and accuracy of the proposed method in the probabilistic reliability calculation of the distribution network.