Research On Power System Load Flow Simulation Based On Pandapower And Speech Recognition Technology

Power flow simulation is an important means to analysis steady-state characteristics of power systems.It plays a pivotal role at the planning,design and operation stages of power grids.Recently,many researchers have been conducted in this area.Considerable efforts have been paid to numerical calculation method,software development and visual display so far.However,there are still dificiencies in existing researches.In terms of simulation data,there is a problem of data incompatibility between different simulation tools.It is necessary to conduct data conversion research to improve the efficiency of utilizating simulation data.In terms of human-computer interaction,conventional interactive devices such as keyboard and mouse have been used in power flow simulation.Literature searches indicate that an integration of speech recognition technology and power flow simulation has not been studied so far.Therefore,it is necessary to improve the intelligent level of human-computer interaction.In terms of drawing power flow diagram,manual operation mode has always been used in traditional simulation tools,whereas more effective automatic drawing methods are required to deal with large-scale power system.In this thesis,new methods of data conversion,human-computer interaction and automatic drawing of power flow diagram are investigated systematically on the basis of a power flow simulation tool which named Pandapower and speech recognition technology so as to improve the intelligent level of power flow simulation.A new conversion method from PSD-BPA data to Pandapower is presented.Firstly,the power flow input data of PSD-BPA and Pandapower are analyzed.Secondly,the process and algorithm of data automatic conversion are constructed,and the conversion methods of various component models in power system are discussed.Finally,the example result demonstrates the correctness of the method.A novel method for integrating speech recognition technology into power flow simulation is presented.Based on convolutional neural network and self-attention mechanism,the design and implementation of speech interaction system for power flow simulation is discussed from the view of speech acoustic model,speech language model,speech power flow instruction and speech response module.The test results shows the validity of proposed method.An automatic generation method of power flow diagram suitable for large-scale power systems is also presented.The topological structure of power system is analyzed first and then the power grid topology is divided into inter-station topology and intra-station topology by using the depth-first algorithm.Based on the topological layering and a visualization tool which named Pyecharts,the layout between stations and the internal layout of stations are developed by using the the force-oriented layout algorithm,followed by the generation of the power flow diagram.Tests of practical examples indicate that the proposed method is effective.