Research of WSN-based safety monitoring system for middle-and-small
sized dams
1
Tang Wenliang
,
2
Yan Liping
*1,Corresponding Author
Software School, East China Jiaotong University, China
twl_ecjtu@163.com
2.
Software School, East China Jiaotong University, China
ylp_park@163.com
Abstract
Aiming at the serious safety hidden danger of middle-and-small size dams in our country, the paper
designs a WSN-based dam safety monitoring system. Firstly it proposes a kind of multi-hop, clustering
structure of WSN, which greatly reduce the consumption of the network and improve the reliability of it.
Then based on the thought of saving energy this paper proposes a routing rule based on genetic
algorithm, which is self-organized, self-adaptive and self-studied, and as a result the reduction and
equilibrium of energy consumption is produced in the data transmission process and the survival cycle
of the network is greatly lengthened. Error exists in the data collected in multi-sensor network to some
degree and influences the reliability of dam safety monitoring. In order to improve the accuracy and
reliability of collected data from several sensors this paper finally applies data fusion algorithm in the
dam safety monitoring system. Through calculating the arithmetic average values and using batch
estimation algorithm, we fuse the measurement values from different sensors and get the data more
approaching the true value so that the correct judgment and decision can be made in time to ensure the
dam safety.
Keywords: Wireless sensor networks; Data fusion; Safety monitoring
1. Introduction
Constructing dams and reservoirs is an important world-wide act to comprehensively utilize water
resource. Our country has constructed about 90 thousand dams, the most of which were built in fiftieth
to seventieth of the twentieth century. Due to kinds of reasons many dams are inferior and some have
been developed into ill ones, in which there are hidden dangers. Few dams have been broken and
caused great loss. Therefore it is necessary to monitor the dam safety in time and maintain it in the long
time.
At present the already existing dam safety monitoring systems are mainly based on cable sensors,
which are accurate in collecting signals, good at resisting disturbance and serialized in production [1-4].
But the number of cables is great, the maintenance expense is high and it is even impossible to lay out
cables in some structures. With the development of sensor technology, wireless communication
technology and MEMS (Micro-Electro-Mechanism System), wireless sensor technology has begun to
be applied in civil engineering monitoring field and has become a research spot [5-8]. Thus wireless
sensor networks emerge as the times require. In the wireless sensor network, the sensor data is
transmitted to the base station for centralized processing. Collecting a great deal of initial data leads to
excessive consumption and shortened network cycle. In addition sensors of low costs are confined to
definite bandwidth so that it is difficult to satisfy the timeliness [9-11].
Combined with the above problems and on the basis of the characteristics and design demand of the
safety monitoring system for middle-and-small size dams, this paper firstly proposes a kind of
multi-hop, clustering structure of Wireless Sensor Networks (WSN), which greatly reduce the
consumption of the network and improve the reliability of it[12, 13]. Saving energy is the core demand
of the WSN for the dam safety monitoring. Thus this paper proposes a routing rule based on genetic
algorithm, which is self-organized, self-adaptive and self-studied, and as a result the reduction and
equilibrium of energy consumption is produced in the data transmission process and the survival cycle
of the network is greatly lengthened[14, 15].
In the dam safety monitoring process many kinds of sensors are needed to be monitored, such as
water level sensor, stress sensor, shift sensor and leakage sensor. Because of the inaccuracy of sensors
themselves and other factors such as monitoring environment and human disturbance the monitoring