| In recent years,microgrid technology has been widely concerned and rapidly developed by the power industry at home and abroad by effectively improving the grid’s ability to accept distributed generation and improving the power supply quality and reliability of the network.However,the uncertainty of a large number of distributed generation in microgrid will have a negative impact on the power quality of users;On the other hand,when the upper power grid fails and the microgrid is in off grid operation mode,its internal load demand depends on the power output of the micro power supply,which is likely to cause the load to be forced to cut off,directly affecting the power reliability of users.Therefore,it is of great significance to carry out microgrid reliability assessment for power grid planning and design and power quality of users.In the reliability evaluation of microgrid,the sequential Monte Carlo algorithm is used to simulate the distributed generation such as photovoltaic and wind power,and the time-series polymorphism uncertainty of load will lead to slow convergence of calculation,and the contradiction between calculation accuracy and calculation cost is prominent.In order to solve the above problems,a data-driven algorithm for microgrid reliability evaluation is proposed.Firstly,the traditional reliability model is optimized,and the time series probability simulation model,photovoltaic intermittent equivalent random model and time-varying load fluctuation equivalent random model are established;Secondly,by using the continuous random variables defined in the model,the reliability index is transformed into an integral problem on the multidimensional hypercube,and then the multidimensional integral model of microgrid reliability evaluation is established;At the same time,the Lagrange multiplier method is used to solve the optimal density function of random variables with the minimum variance as the objective,and the Monte Carlo integration method is used to estimate the numerical solution of the analytical formula to form a specific method suitable for microgrid data-driven,so as to achieve the goal of reducing the computational cost;Finally,the effectiveness and applicability of the proposed model and algorithm are verified from the two aspects of adequacy and security of microgrid,taking the calculation accuracy,variance and variance coefficient as evaluation criteria.The research results show that the reliability evaluation theory based on data-driven algorithm can realize the real evaluation of the reliability level of the independent operation microgrid,effectively solve the contradiction between the calculation cost and the calculation accuracy,greatly save the calculation cost of the evaluation on the basis of ensuring the calculation accuracy,and provide quantitative analysis for the decisive factors affecting the reliability index,It also provides targeted guidance for capacity optimization in microgrid reliability evaluation.In addition,the reliability evaluation algorithm of microgrid in grid connected mode is studied.The equivalent generator model of distribution network is adopted to simplify the distribution network system and reflect the supporting effect of distribution network on the reliability of microgrid.The reliability parameters of the equivalent generator model are defined and the probability simulation is realized.This paper analyzes the influence of the component failure of the superior distribution network on the operation state of the equivalent generator and static switch at point of common coupling,and further considers the output mode of the external network and microgrid,puts forward the microgrid load balancing strategy with equivalent generator.A case study is carried out to verify the effectiveness of the above evaluation method.The results show that the reliability level of microgrid in the grid connected state is better than that in the island mode,and the reliability level is also different under different grid connected schemes. |
PDF Full Text Request