Methods For Defending False Data In Power System State Estimation

False data,mamaliciosly injected by hackers to tamp with power system state estimation,are stealthy bad data.Considering the fact that false data can efficiently bypass the traditional bad data detection and identification algorithms,hackers can delebrately manipulate these state estimation outputs by specially constructing false data,then the distorted dispatching orders would further pose great threats to the safe and reliable operations of power systems.Therefore,it makes urgent sense that more attention should be paid to investigating the data security vulnerabilities of real power systems and formulating the corresponding defense mechanisms in the aspect of ensuring the future smart grids a better service for national economy development.According to their defense performance,these current proposed defense mechanisms can be classified into into 3 types:detection,identification and containment.The detection mechanisms can only determine the existence ranther than their specific locations of false data.However,the identification mechanisms can further realize the precise position of false data.Since the containment mechanisms focus on fundamentally stopping hackers from constructing false data,they have been considered as the most effective defense way for securing the reliability of power system state estimation.Measurements physical protection strategies,dynamic adjustments of network parameters or topology,and electric power information fusion defense methods are the mian 4 contaiment mechanisms presented currently,but their applications are somewhat limited by huge investments,restricted global network transmission capability,high computational complexity,et al.In order to overcome the defects of theses defense algorithms mentioned above,defense mechanisms,with improving the inherent data security vulnerabilities in bad data detection and identification algorithm as the core for containing false data,are proposed.The work and main research achievements of this paper are as follows:1.The research status of defending against false data in power system state estimation is systematically reviewed.Bsed on a summary of the defects in these current methods,the core of defending false data is pointed out,saying that the inherent data security vulnerabilities in bad data detection and identification algorithms should be improved.2.A method for defending false data based on state vector mining is proposed.Hidden Markov model is adopted to model the historical running state database,and the outputs of generators which are difficult to be manipulated by hackers is regarded as the observation state sequences,then the system state sequences are decoded with the aid of Viterbi algorithm.3.A method for defending false data based on state variable modification is proposed.State consistency test is proposed to locate the state variables that are suspected of being distorted by false data,and then state variable correction vector is constructed to corrupt the stealthiness of false data,which guarantees that false data can not avoid bad data detection and identification,further more the false data can be identified and eliminated.4.The IEEE-14 bus and IEEE-118 bus standard test systems are adopted to validate the effectiveness of these proposed methods in containing false data.