Power Quality Probabilistic Evaluation On Distribution System With Considering Distributed Generators

With the exploitation and application of renewable energy resources, it exits a number of new problems with the high penetration of distributed generators (DGs) in the current distribution network of the power system, especially within the issue of power qualities. It is significanlty serious performed by the intermittent output generation such as wind power and photovotalic (PV) arrays. Therefore, utilizing more scientific and rational methodology to evaluate the power quality indices effectively becomes the essetial subject to be carefully studied.In this paper, the probabilistic evaluation model for power qulities via considering the characteristic of stochastic power output by DGs. The content in very detail will be including the following aspects:First, the stochastic power output by DGs and variation of load power makes the concept of probabilistic load flow (PLF) need to be introduced in inflecting the reletive statistical performance data for nodal votage. Basd on the emphasis and analysis of2n point estimate method for PLF solutions, the more rational and exaustive solution-the modified2n point estimate method is proposed to obtain the each moment value of nodal voltages. Combining with the Cornish-Fisher series expansion, the probabilisty desity function (PDF) of nodal voltages will be derived. The case analysis will be studied to verify the rationality and effect of the proposed evaluation algorithm.Second, poiting at the regular three-phase unbalance and the impact by the connection of DGs in the distribution system, the evaluation of distribution system three phase voltage unbalance factor (VUF) will be emphasized, based on the analysis of the conventional distribution system state estimation (DSSE) method. In this paper, the probabilistic model-based DSSE is proposed. Meanwhile, the feasible model calculation will be derived via the reaserch of object function, concept and contruction of pseudo-measurement, and the correlative probabilistic distribution. Besides, the detailed solution of DSSE is deducted, combining with the case study. The results implys the fitness of the VUF evaluation and the conclusion is also significant for the guidness of the connection of DGs into distribution systems.Third, based on the reaserching area of the harmonic current evaluation, the algorithm primarily applied in the ecnomic model for the paramter estimation will be used. The constrution of Markov Chain Monte Carlo (MCMC) is introduced into the probabilistic evalution models. Moreover, the reaserch of sampling, solving, and processing approach will be studied in order to derive the relative model of the mathematical expression for the harmonic issue evalution. Meanwhile, the model will be verified by applying the typical5th and7th harmonic data with considering the influence of proporation of linear and nonlinear loads in different periods. The result is more comprehensive and close to the real distribution system.In the end, by analyzing the concept and method of comprehensive evaluation and the relative courses, a new method named sample set centralized constraint and entropy weight is proposed in this paper. By utilizing the projection pursuit to solve the complicated data format integration issue and the entropy and entropy weight principle, the method is applied to modify the weight of sample set. Comparing with the current evaluation method, this approach is more simple and convinient to use. Besides, the evaluation process for the practise issue is reaserched and verified with the case study. After that, by applying the eigenvalue weighted method, the power quality grade is classified, which is also sophisticated to connection of renewable energy-based DGs to distribution system, providing with suggestions and advises.