Research On Black Box Test Technology Of Smart Meter Software

Smart meter is an important power metering equipment,which is related to the national economy and the people’s livelihood.According to statistics,the fault rate of field operation of the smart meter is 0.004%,which caused by the software fault as high as 20%.Therefore,software has become the key factor which affecting the stability and reliability of smart meter.In this paper,the software structure,fault modes and reasons of the smart meter are analyzed.The important fault repeat method is studied,and some test cases with depth and breadth are designed based on the fault information.The validity of the designed test cases are verified on the test platform.The main work of this paper is as follows:Firstly,the common embedded software structure is analyzed and the characteristics of various structures are discussed.As one of special embedded software,the working principle of smart meter software is analyzed.Combining with the modular design of the hardware,the typical software structure of the smart meter is designed and the rationality of the system structure is analyzed.Secondly,based on FMEA(Failure Mode and Effects Analysis)method,software faults and reasons of the smart meter are deeply analyzed,which occurred in field and laboratory testing process.The failure reason is studied.In order to quickly query the fault information and locate the fault pattern accurately,the fault pattern recognition based on neural network is designed and the best classification result is achieved,which can help to design the test cases.Then,based on the black box test method and principle of test cases,the test cases with depth and breadth are designed for faulty modules.Combined with a specific phenomenon,the fault recovery and detection method are studied by using the design principle of fault injection.The designed test cases are verified on the test platform.The result shows that the designed test cases can stimulate the fault of smart meter and detect the potential software fault.The validity of the test cases are verified.It has a great significance for the testing of smart meters.Finally,the main content is summarized and future work is presented.