论文部分内容阅读
自动需求响应是智能电网与用户实现信息和能量互动的重要实现手段。为解决基于实时电价的自动需求响应技术应用中包含多类负荷用户的电能综合规划问题，建立优化问题的数学模型。针对该模型提出一种基于动态抗体记忆库的免疫优化算法。设计依据二重亲和度阈值检测的抗体记忆库更新机制，在优化结束后能够为用户提供多个备选可行解。采用先验知识疫苗接种的方法，提高算法的求解精度。通过抗体种群的优值马尔可夫链的转移概率分析，证明了算法的收敛性；利用实际算例验证了所提算法的有效性。对比分析的结果表明，所提算法比其他算法具有更好的全局优化能力和搜索效率。“,”Automatic demand response is an important realization means of achieving information and energy interaction between smart grid and users. In application of spot price based automatic demand response, to solve the problem of integrated power planning for a user with various loads, a mathematical model of optimization was established. According to this model, a flexible antibody memory based immune optimization algorithm was proposed. To provide the user several alternative feasible solutions after the optimization process, a refresh mechanism according to double affinity threshold detection was designed. The accuracy of this algorithm was improved by use of prior knowledge vaccines inoculation. The convergence of this algorithm was proved by the transition probability analysis of antibody population best values’ Markov chain. Then the feasibility of proposed algorithm was verified by the optimization result of a practical example. Comparison result shows that the proposed algorithm has a better performance of global optimization and searching efficiency than other algorithms.