| Power quality is directly related to national economic benefits and personal property security.With the increasing size of China’s power market,many industries have begun to have higher requirements for good and stable power quality.Poor power quality will affect the performance indicators of electrical equipment and even threaten the safety of the power grid.Compared with developed countries,China’s current power quality management methods are still relatively extensive.Under the energy consumption environment of office buildings,as China’s economy develops toward high quality,the requirements for power quality are becoming higher and higher.Therefore,on the basis of comprehensive and scientific power quality evaluation,the study of improving energy utilization is of great significance.First,the existing power quality evaluation methods are analyzed.The analysis shows that the evaluation results of the existing mainstream power quality evaluation schemes are subjective and uncertain.Therefore,it is of great significance to establish an efficient power quality evaluation model for power quality.Secondly,a power quality evaluation model combining BP neural network and genetic algorithm is proposed.BP neural network can deeply explore the characteristics of the related information between indicators and results,use intelligent algorithms to optimize the electrical energy data itself,can use its predictive function to mark more data,expand the label space,and use genetic algorithms to objectively When calculating the value,the corresponding fitness function can be formulated according to the actual situation of the data.It has strong interpretability and can reflect the index situation that has a key influence on the evaluation result in different environments.Then,using the power quality evaluation model proposed in this thesis,the power quality of office buildings is evaluated.Analyze the characteristics of electrical energy data in office buildings,classify their data anomalies and data missing factors,and formulate corresponding data processing schemes based on the classification.The purpose of data normalization and hierarchical preprocessing is to improve the data mining performance of the subsequent BP neural network and genetic algorithm.Finally,the evaluation results are analyzed and tested.According to the test data,the index weights obtained by BP neural network and genetic algorithm were fitted and tested.Combined with other literature power quality evaluation methods and the combined weights obtained in this article,the results of this example were compared and analyzed. |
