| In order to guarantee the security, serviceability and durability etc. of significant engineering structures, an effective way of structural health monitoring (SHM) should be adopted for a lot of established significant engineering structures and infrastructural facilities to appraise theirs safe situation. Traditionally, the wired way is adopted generally, which takes a great deal of work to lay circuit, maintain the structures difficultly and can't locate nodes nimbly. The wireless sensor network will play an important role in SHM for its high density, low-power consumption, less assembling time and locating nodes nimbly.At present, the research of wireless network sensor focuses on routing algorithm of sensor node, energy algorithm, organization algorithm and management scheme on wireless sensor networks. There have less research about the data fusion and clock synchronism algorithm. The existing algorithm and system are related to appointed application, which have certain limitation. Be aimed at concrete application background, the data fusion and clock synchronism algorithm require a step further research.In this paper, wireless sensors data fusion, time synchronization and their network system are developed for the purpose of SHM on the basis of sensing, information disposal and wireless communication technique. The main contents include:(1) An data fusion algorithm combined with routing based on multicast tree by analyzing the existing achievement for data fusion of WSN is presented, which is composed of three parts. One is constructing a multicast tree and the second one is flooding the interest. The third one is fusing data. The new algorithm constructs a multicast tree rooted sink. When the sink node flooding its interest in the second phase, the interest is flooded according to the optimizing directed diffusion routing based on passive clustering. An improved consensus algorithm in data fusing is adopted, which has different confidence distance for different measuring precision. And the supporting matrix is fuzzed, which can avoid the subjective error in determining the threshold value. The effectiveness of this method is confirmed through theoretic analysis and simulation. It has low delay and reduces the energy consumption effectively. It can be concluded that this method could make full use of the data from WSN.(2) By summarizing the application background of WSN time synchronization, the algorithms which have been already proposed are analyzed and compared .In order to reduce the overall energy dissipated in the network, an event-triggered time synchronization algorithm based on clock precision difference is presented. This algorithm is composed of two phases: calculations of clock precision difference and clock synchronization. In this algorithm, nodes' clocks run by their own rates, when an event occurs, firstly, compares the energy difference between post-damage waveform and pre-damage waveform received by each sensor node in structural damage detection model to use time series, then estimates round-trip delay and synchronization difference by exchanging synchronization signals. The simulation results indicate that the algorithm can save energy efficiently, prolong the life of networks for the event occurs in low frequency in SHM and has better clock precision than other synchronization algorithm.(3) Be aimed at the organization of wireless sensor network for SHM, the three layer wireless sensor network topological structure consisting of node—cluster head—gateway is given on the base of the existing sensor network topological modules. And the wireless sensor network can be visited in long-distance while the network system is connected with Internet. And an energy-efficiency network model of wireless sensor network based on passive clustering strategy is put forward .The construction and maintenance of network model is explained at length.
|
PDF Full Text Request