An Efficient And Practical Three-Phase Analytical Optimization Algorithm For Reactive Power Planning Of Distribution Networks

The purpose of reactive power planning optimization for a distribution network is to achieve the optimal allocation of reactive power resources to reduce the electric energy loss, maintain system voltage levels in a reasonable range.For the reactive power planning optimization of a practical distribution system, since hundreds of medium feeders need to be calculated with a lot of parameter adjustment and planning scheme comparison being required, it’s necessary to propose an efficient and practical optimization algorithm for the reactive power planning of a distribution network in order to solve the problems of unstable calculation and long execution time for existing algorithms.The prices of 0.4k V and 10 k V voltage levels shunt capacitors are surveyed, and a linear function is selected to fit the cost models of shunt capacitors. Based on the linear model, the selection of capacitors’ different capacity investment for 0.4k V and 10 k V voltage levels are discussed and the related proposal is given.A reactive power planning optimization model is employed, which takes into account both the compensation benefit and voltage constraint, based on the economic net benefit and voltage constraint, a practical three-phase analytical optimization algorithm for reactive power planning of distribution networks is proposed, in which medium/low voltage shunt capacitors, main transformer taps, DGs and weakly meshed networks are taken into account. The proposed algorithm is particularly suitable for the reactive power planning optimization in a large-scale medium/low voltage system, because no upper limit of capacitor size and the total number of candidate nodes are necessary beforehand with the satisfactory calculation accuracy and much less execution time.Based on a node number optimization and an approximate power flow analysis, a first-phase optimization is used to select the compensation node with the largest net benefit and no voltage constraint in the whole network as a compensation node; with the compensation influence of the selected capacitors being considered, the compensation siting and sizing process is repeated until no further compensation is needed because of a technology and/or economy constraint, resulting in the initial compensation siting, sizing and ordering.Based on the initial solution of first-phase optimization, a second-phase optimization takes into account the mutual influence of selected capacitors’ sizes and locations to make an iterative correction calculation: based on the node compensation ordering of first-phase optimization, the location and size of only one node is optimized every time with the locations and sizes of the other nodes being fixed, until the location and size of any node cannot be further improved.Because of the approximate power flow analysis of first-phase and second-phase optimization, a third-phase optimization is used to further optimize the results of second-optimization with the ac power flow analysis being employed.The propose algorithm and some other ones are applied to the an IEEE33 node test system, an IEEE33 node test system with distribution transformers being added, an actual 133 nodes feeder system, an IEEE33 node test system with distribution transformers, OLTC and DG being added, and a test system of integrated functions. The results show the effectiveness, stability and practicality of the proposed algorithm.