Distribution Network Line Loss Calculation

The line losses calculation in distribution network is to calculate electric energy waste of all the electric power system elements. The distribution network is the end of the electric power network. It connects to the user directly and has much equipment. It brings a big waste because there was a lot loss in the transmission of electric energy. So the line losses calculation in distribution network is economically significant for power-supplied corporation. As the development of power distribution construction, the calculation work directs the rationality of the topology design and the compare of the project's effect of measures which want to decrease loss. So it is necessary to research the line losses calculation in distribution network.The great issue of the calculation work is that the calculate methods and measures restrict the evolvement of the decrease losses work. The methods for electric power network are no longer suitable for distribution network, because of the particularity and complexity of distribution structure. The current methods also have limitation in the computer implementation, so the method which is applicable for the distribution structure and computer implement should be brought forward. On the other hand, because the number of elements in the distribution network is large and the running data is random, it is difficult to collect the complete running data and the workload is great if an exact calculation is made. So in some cases, the convenient estimate method which can get an accurate result relatively is needed.Firstly, the paper introduces the electric loss calculation of the distributed loads and the current calculation methods for line losses. The characteristics of all the methods referred to are also analyzed.Then the forward/backward sweep method is researched and the sequence layer storage structure based on the index table is brought forward. This structure can cost less memory cost and eliminate the dependency of node number. Efficiency is also improved under this structure. On the other hand, the method of using XML file to organize the distribution data is proposed. This organization can figure the distribution structure intuitively and the distribution branches can be modified flexibly under this circumstance. Based on the technology above, the node objects are established and the program design and implementation are completed. The validity of this improved method is verified by instances.The application of Artificial Neural Networks in line losses estimation of distribution network is studied. The estimate and realization process of line losses calculation are discussed in detail using Back-Propagation network arithmetic. Then the method is used to forecast the line losses of a region and the result is all right.Finally, the design of prototype system's architecture is discussed in the thesis. The whole system design is presented and two core modules are introduced in detail.