A Research On Immune Clone Selection Optimization Of Power System Security Engineering Capability Based On Multi-Agent Technology

Power system is a complex and large-scale dynamic system, so it is difficult for common used safety evaluation methods to reflect the power system security engineering capability level as well as the security and stability of power system. The higher the level of security capability of projects, the more reliable, more secure and more stable the projects are. System security engineering capability maturity model(SSE-CMM) is a kind of method to measure the practical ability of safety engineering and using this way to establish power system security engineering capability maturity model. According to the coordination mechanism of Multi agent technology(MAS) and the ability of learning and reasoning, the security capability model of power Multi-agent system is established. Fuzzy modeling and optimization control technology is used to solve the problem of fuzzy modeling and optimization of power system security based on Multi-agent technology. Therefore, it is very important for a power industry to improve the level of security capability and it is meaningful and stable to development of industry and benefits of income.First of all, we need to familiar with characteristics of the power system and the theory of SSE-CMM, mapping, connecting and cutting power production system to constructed power system security engineering process areas and basic practices. Using the fuzzy clustering method to classify the general practice to build public property and general practice project of power system security engineering process. According to the definition principle of capability level, it can be matched to the power system. So this is a power system security engineering capability maturity model.Secondly, we construct a architecture for power security engineering capability model basing on theories and methods of multi-agent technology(MAS). The architecture is mainly composed of five dependent Agents, namely power plant Agent, transformer substation Agent, power transmission sequence Agent, distribution network Agent and coordination Agent. We have designed its internal structure model for these Agents. All these Agents introduced into SSE-CMM, we construct multi-agents of power system security engineering capability model.Then, owing to the uncertainty of parameter of multi agents of power system security engineering capability model, triangular fuzzy parameters will be introduced into this model. Depending on multi-agent system distributed autonomous and multi-agent negotiation mechanism, we design the model for multi-fuzzy goals and multi-fuzzy constraints of multi-agents of power system security engineering capability. We can get membership function and expected value interval of multi-fuzzy goals and multi-fuzzy constraints under different probability levels. According to the relative importance of fuzzy goal and fuzzy constraint, the membership function integration model of fuzzy multiple objectives and multi fuzzy constraints are realized. We use immune clonal selection optimization algorithm to find the optimal security engineering capability.Finally, according to the actual situation of the power system, we apply power system security engineering capability maturity model and multi-agent system security engineering capability maturity model and immune clonal selection optimization method to the power industry. We assess the levels of security engineering capability of this power industry firstly, getting optimal security engineering capability of power multi-agent system under different probability levels. After comparing and analyzing the results, this paper gives some improvement measures which can improve the security engineering capability of this power system as well as verifies the validity of this model and the feasibility of the optimal results.