Research Article
Application of Principal Component Analysis Algorithm in Abnormal Diagnosis of Electricity Bill Reading and Receiving Data
@INPROCEEDINGS{10.4108/eai.8-12-2023.2344734, author={Weihua Zhao}, title={Application of Principal Component Analysis Algorithm in Abnormal Diagnosis of Electricity Bill Reading and Receiving Data}, proceedings={Proceedings of the 5th Management Science Informatization and Economic Innovation Development Conference, MSIEID 2023, December 8--10, 2023, Guangzhou, China}, publisher={EAI}, proceedings_a={MSIEID}, year={2024}, month={4}, keywords={wavelet transform; principal component analysis algorithm; electricity fee collection data; abnormal diagnosis}, doi={10.4108/eai.8-12-2023.2344734} }- Weihua Zhao
Year: 2024
Application of Principal Component Analysis Algorithm in Abnormal Diagnosis of Electricity Bill Reading and Receiving Data
MSIEID
EAI
DOI: 10.4108/eai.8-12-2023.2344734
Abstract
In order to understand the abnormal diagnosis of electricity bill reading and receiving data, an application research based on principal component analysis algorithm in the abnormal diagnosis of electricity bill reading and receiving data is put forward. In this paper, firstly, an abnormal diagnosis method of electricity bill reading and receiving data based on principal component analysis algorithm is proposed. The collected data of electricity bill copying and checking is taken as the research sample data, and it is denoised and preprocessed by wavelet transform. On this basis, the correlation coefficient matrix is used to solve the principal components of data, and the number of principal components of data contribution rate is determined. Secondly, the principal component analysis expression is used to establish a data anomaly diagnosis model, and the data anomaly diagnosis of electricity bill copying and receiving is realized through the model. Finally, in order to verify the comprehensive effectiveness of the proposed method based on the principal component analysis algorithm, a simulation experiment was carried out on Win7 system, CPU i55600U@2.6GHz and memory 16GB3200MHz. The experimental results show that by denoising the data, the accuracy of the abnormal diagnosis result of the proposed method is effectively improved and the diagnosis delay is obviously reduced.