This paper proposes an infeasible interior-point algorithm with full-Newton step for linear complementarity problem,which is an extension of Roos about linear optimization. The main iteration of the algorithm consists...This paper proposes an infeasible interior-point algorithm with full-Newton step for linear complementarity problem,which is an extension of Roos about linear optimization. The main iteration of the algorithm consists of a feasibility step and several centrality steps. At last,we prove that the algorithm has O(nlog n/ε) polynomial complexity,which coincides with the best known one for the infeasible interior-point algorithm at present.展开更多
为实现综合能源系统中能源的高效利用,同时考虑到不同能源系统属于不同能源供应商,通常只进行部分信息交互,提出一种基于近似牛顿方向(approximate Newton directions,AND)的电热气综合能源系统最优能流解耦算法。将整体的优化问题根据...为实现综合能源系统中能源的高效利用,同时考虑到不同能源系统属于不同能源供应商,通常只进行部分信息交互,提出一种基于近似牛顿方向(approximate Newton directions,AND)的电热气综合能源系统最优能流解耦算法。将整体的优化问题根据能源主体解耦,分解为电力系统、热力系统、天然气系统优化子问题,通过部分信息的传递,对子问题交替迭代,进行分布式求解,最终达到整体优化的效果。该算法信息传输量少,结构简单,计算速度快,求解过程中,子问题不需要求最优解,只需要迭代1次,提高了求解效率。针对具体算例,将AND算法与集中式内点法(centralized interior point method,CIPM)的计算结果进行对比分析,验证了所提算法的可行性和有效性。展开更多
基金Supported by the National Natural Science Fund Finances Projects(71071119)
文摘This paper proposes an infeasible interior-point algorithm with full-Newton step for linear complementarity problem,which is an extension of Roos about linear optimization. The main iteration of the algorithm consists of a feasibility step and several centrality steps. At last,we prove that the algorithm has O(nlog n/ε) polynomial complexity,which coincides with the best known one for the infeasible interior-point algorithm at present.
基金The author gratefully acknowledges the partial supports of the Ph.D.Foundation Grant(0527003)of Chinese Education Ministry and the Science Foundation Grant (06DZ037,05DZ11) of Shanghai Education Committee.
文摘为实现综合能源系统中能源的高效利用,同时考虑到不同能源系统属于不同能源供应商,通常只进行部分信息交互,提出一种基于近似牛顿方向(approximate Newton directions,AND)的电热气综合能源系统最优能流解耦算法。将整体的优化问题根据能源主体解耦,分解为电力系统、热力系统、天然气系统优化子问题,通过部分信息的传递,对子问题交替迭代,进行分布式求解,最终达到整体优化的效果。该算法信息传输量少,结构简单,计算速度快,求解过程中,子问题不需要求最优解,只需要迭代1次,提高了求解效率。针对具体算例,将AND算法与集中式内点法(centralized interior point method,CIPM)的计算结果进行对比分析,验证了所提算法的可行性和有效性。
