Font Size: a A A

Research And Implementation Of Reliability Assessment Method For Regional Distribution Network Operation

Posted on:2022-06-05Degree:MasterType:Thesis
Country:ChinaCandidate:J X ZhuFull Text:PDF
GTID:2512306530479874Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
With the improvement of my country's comprehensive strength and the implementation of the strategy of "power transmission from west to east,national interconnection",the national basic energy industry has developed tremendously,the coverage area of the power grid is gradually increasing,the structure is becoming more and more complex,and the degree of interconnection of the power grids in various regions is continuously improved.,Put forward higher requirements for the reliability evaluation of power grid operation.Therefore,it has become a more urgent topic to study theories and evaluation methods related to the reliability of power grid operation and realize the reliability evaluation system of power grid operation.Power grid operation reliability evaluation methods have higher requirements for the accuracy of calculation results and calculation efficiency.Traditional analytical methods need to involve a large number of components when calculating large-scale power grids,which leads to lower calculation efficiency.In terms of method,it requires continuous sampling and spends a lot of time to repeat complex power flow calculations,which is difficult to meet the requirements of calculation efficiency.In recent years,with the rise of machine learning algorithms,machine learning algorithms can be added to traditional evaluation methods.In order to realize the fast and accurate assessment of the reliability of the power system operation.Short-term load forecasting is an important part of power grid operation reliability evaluation.Accurately predicting electric load and its changing trend through reliable and high-quality load forecasting models can improve the basic data support for subsequent grid operation reliability evaluation.In short-term load forecasting Not only should the time series correlation of the data be considered,but also the influence of external factors should be considered,and a suitable model should be built to improve the accuracy of prediction.To this end,this article takes the regional distribution network as the object,and conducts research on short-term component reliability modeling,short-term load forecasting methods,regional distribution network operation reliability evaluation methods and system implementation,and provides theories for subsequent power grid reliability research.According to the specific research content as follows:1.Improve the reliability model of traditional transmission line components.Because the transmission line is outdoors for a long time,its health is affected by the environment.In addition,it also considers its own life span and the impact of the power grid operation status through the above three influencing factors Establish reliability models of short-term transmission line components to make operational reliability assessment more accurate and practical.2.Taking into account the timing correlation of the Prophet model and the faster running speed of the XGBoost(e Xtreme gradient boosting)algorithm,a short-term load forecasting method based on the Prophet-XGBoost hybrid model is proposed.First,the Prophet model and the XGBoost model are used for short-term Load forecast,and then integrate the two forecast results through the reciprocal error method to obtain the final forecast value.The accuracy of the method is verified by the electrical load and meteorological data of a certain area in the southwest.3.Combine the traditional Monte Carlo simulation method(MCS)with the support vector machine algorithm(SVM),and use the optimal SVM model obtained by training to directly predict the optimal load shedding amount,avoiding the complex optimal load shedding amount Calculation,so as to achieve the purpose of improving the calculation efficiency under the premise of ensuring the accuracy of the calculation results.Finally,the feasibility of the method is verified by the RTS79 system example.4.At present,the research on the reliability evaluation system of the distribution network is not yet mature.Based on the above research,taking a certain area in the southwest as an example,construct a distribution network operation reliability evaluation system,which has operational reliability evaluation,etc.The function is helpful to help grid companies accurately grasp the development and operation of the power grid in the current economic development situation in the region,grasp the utilization efficiency level of the existing power grid,and understand the safety and reliability level of the power grid.
Keywords/Search Tags:Regional distribution network, Operational reliability evaluation system, Machine learning, Monte Carlo simulation method, Short-term load forecasting
PDF Full Text Request
Related items