Study On Construction Safety Decision Based On Knowledge Reduction

Underground space building's security is important since it often built on geological complex, road narrow, underground pipeline dense, busy traffic center downtown. Whenever the underground space projects construction or the operation period, it should be taken a strict monitoring to the data for the stability of the main structure and the safety of environment. The project is a long-term work; its characteristics include the various items, the long distance, too many points to monitor, the period frequency and large amount of data in the monitoring period. Facing lots of data which may be incomplete,inaccurate and uncertain, is necessary to establish the mathematical model for efficient synthesis utilization, to exert fully the aided decision-making role of the underground space construction safety monitoring system.Z.Paw lak proposes rough sets theory which is a kind of depicting incomplete and uncertain mathematical tool. It can analyze the inaccurate, uncertain, incomplete information effectively. Therefore, for relevant information about underground space project, this paper utilizes related equipment for data monitoring collection and uses the rough sets theory to analyze the collected data. To solve the problems of construction safety decision, this paper introduces a construction safety decision method based on knowledge reduction. By analyzing the classic knowledge reduction algorithm, it proposed a new method for complete and incomplete knowledge reduction. The main jobs in the paper are as follows:1. For the complete knowledge system, the complex reduction by constructing new knowledge partition and relative conditions partition are proposed. And an applicable and complete knowledge reduction algorithm for the consistent and inconsistent decision table is proposed. On the other hand, in order to reduce huge dynamic knowledge and satisfy the need to build knowledge base, a quick and dynamic knowledge reduction algorithm for the consistent and inconsistent decision table based on hash table is studied.2. For the incomplete knowledge system, since it is unreasonable for asymmetric similarity relation to classify obviously similar relation, dynamic valued asymmetric similarity relation is defined and a kind of valued asymmetric similarity relation model with auto-selecting threshold is built. This model makes data selection more flexible and reasonable. 3. The consistent classes and maximal consistent blocks in the incomplete knowledge system are researched and the rapid reduction algorithm based on the consistent classes is designed in this paper.4. The algorithms proposed in this paper have been used to the data reduction and security decision-making during construction project and has a good effect.