| Smart Grid massive accident happens in our lives which bring serious consequences, many experts and scholars study and find that many software reliability model which is applied to deal with software reliability evaluation and fault diagnosis can not get a good effect, therefore how to let the software reliability model adapt to the control system of smart grid become the core issue of this thesis to study.The goal of this thesis is to build highly reliable operation mode using software reliability model based on formal method. Starting with the software reliability research of the existing smart grid control system, we analyze the characteristics of smart grid and it’s control system, hold the basic elements of software reliability model, use the rigor and consistency of formal method to solve the ambiguity of software reliability, use Bayesian network to solve reliability measurement problem of smart grid control system software modular, calculate the reliability of the interaction between modules based on Bayesian method; Using Petri network deal of parallel issues in the processing software of fault diagnosis,using the basic structure simplification and transient state level compression solve the problem of Petri net state space explosion. in this thesis, integrating the characteristics of Bayesian and Petri net we propose a BNP model which is suitable for smart grid control system, in the last draw the validity and reliability of this model with specific examples.This thesis mainly to the research content:(1) Analysis of the control characteristics of smart grid control systems, the hierarchical control technology;(2) Analyze software module partition and its management, use the Bayesian method to process module reliable measurement of smart grid control system, and propose Bayesian networks model and its inference;(3) Using the P/T model partitioning subnet process, complete parallel processing of process based on Petri net, put forward feasible suggestions for the problem of space explosion phenomenon in software of large-scale smart grid state;(4) Comprehensive the characteristics of Bayesian and Petri net, proposed the BNP model, using examples to prove the validity and feasibility of the model.As can be seen by an example, the BNP model can well describe the dynamic process of software systems; As much as possible to consider the influencing factors of software reliability; Reliability probability calculation of fuzzy sample also can get good results; The results provide a good basic for analyzing thereliability of the system operation after a certain time, also can be directly used to guide the development process. |
PDF Full Text Request