International Core Journal of Engineering 2020-26 | Page 29

VI. C ONCLUSION
R EFERENCES
This paper proposes a method for risk identification of
settlement data based on evidence theory. The multi-
evidence fusion decision-making method is integrated into
the settlement data risk identification. Based on the
calculation results of evidence theory, the risk treatment
method of settlement data is proposed. The following
conclusions were obtained.
[1]
R. Ge, L. Chen, Y. Wang, D. Liu, “Optimization and Design of
China's Power Market Construction Path,” Automation of Electric
Power Systems, vol. 41, no. 24, pp. 10-15, 2017.
[2] M. Nemeth, “Multivariate Statistical Methods: A Primer,”
Technometrics, vol.39, no. 4, pp. 426-427, 1994.
[3] J. Wang, J. Chiang, “A cluster validity measure with outlier detection
for support vector clustering,” IEEE Transactions on Systems Man
and Cybernetics, Part B-Cybernetics, vol.38, no. 1, pp. 78-89, 2008.
[4] S. Hu, G. Li, W. Lu, H. Feng, “Anomaly Data Detection Method for
Wireless Sensor Networks Based on Neural Network,” Computer
Science, vol.41, no. b11, pp. 208-211, 2014.
[5] H. Wu, X. Wang, N. Vocationalamp, “Network Anomaly Data
Classification Method Research Based on Optimization of SVM,”
Bulletin of Science & Technology, 2016.
[6] L. Zadeh, “A simple view of the Dempster-Shafer theory of evidence
and its implication for the rule of combination,” Fuzzy Sets, Fuzzy
Logic, And Fuzzy Systems:Selected Papers by Lotfi A Zadeh, pp.
674-679, 1986.
[7] Y. Liu, Y. Kong, “Power System Transient Stability Assessment
Based on LSSVM and Evidence Theory,” Electric Power Science &
Engineering, vol.26, no. 1, pp. 34-38, 2010.
[8] K. Ghosh, S. Natarajan, R. Srinivasan, “Hierarchically Distributed
Fault Detection and Identification through Dempster–Shafer Evidence
Fusion,” Industrial & Engineering Chemistry Research, vol.50, no. 15,
pp. 9249-9269, 2011.
[9] A. Dempster, “Upper and Lower Probabilities Induced by a
Multivalued Mapping,” Annals of Mathematical Statistics, vol.38, no.
2, pp. 325-339, 1967.
[10] A. Dempster, “A Generalization of Bayesian Inference,” Journal of
the Royal Statistical Society, vol.30, no. 2, pp. 205-247, 1968.
[11] X. Duan, “Evidence Theory and Decision Making, Artificial
Intelligence,” Renmin University of China Press, 1993.
[12] W. Yi, “A Data Smoothing Method: High Order Polynomial
Piecewise Approximation Curve Fitting Method,” Acoustics and
Electronic Engineering, no. 1, pp. 36-39, 1997.
1) The PCA and ICA are used to construct the statistic
as the evidence source, and the evidence data theory is used
to identify the abnormal data. Considering the multi-
evidence fusion of conflict evidence, the reasonable
decision-making method is used to obtain the support level
of each data, and the abnormal transaction data can be
effectively identified. The results of the example show that
the multi-evidence fusion decision can effectively improve
the detection rate of abnormal data, and the processing of
false evidence can also reduce the misrecognition rate of
abnormal data, and improve the accuracy of data risk
identification from various aspects.
2) The reliability can reflect the degree of deviation of
the abnormal data of the settlement, and the abnormality of
the transaction data with the reliability higher than the
reconstruction threshold is within the acceptable range, and
the abnormal data with the reliability lower than the
reconstruction threshold is adopted. Polynomial fitting and
reconstruction methods based on historical correlation. The
results of the example show that the proposed method can
obtain more accurate reconstruction data, reduce the
reconstruction data migration rate, and effectively reduce
the settlement risk without affecting the settlement
efficiency.
7