Research And Design Of Tungsten Tailings Pond Safety Monitoring And Warning System Based On Wireless Sensor Networks

Tungsten tailings pond is the necessary facility to maintain the normal production of tungstenmines, and it is also the major hazards of tungsten tailings pond. The security monitoring andwarning of tungsten tailings pond becomes one of the hot spots of people’s attention.The traditionaltungsten tailings pond safety monitoring systems used artificial observation and wired transmissionmode, and it has many flaws such as long measuring cycles, low efficiency and poor real-time dataperformance. The cable transmission also has flaws such as high cost of installation andmaintenance, complexion wiring, limited transmission distance. In recent years, the technology ofwireless sensor network due to its strong survival ability, high accuracy and reliability, dynamicway of ad hoc network, strong scalability and low power consumption, etc., which can well solvetheexistingproblemsoftungstentailings pond safetymonitoring.In this thesis,according to the research contents of National Natural Science Foundationproject, proposed the safety monitoring and warning system of tungsten tailings pond based onwireless sensor network, focusing on the system overall scheme design, the improvement ofcluster-tree network routing algorithm, the sensor node hardware and software system design anddevelopment ofPCmanagementsystem.(1) First, according to the current status, characteristics and requirements of tungsten tailingspond, to research the main form of tailings pond harm and risk, then determine the main indexparameters that need to be monitored of the safety monitoring system. According to the actualsituation of the tailings pond to arrange the monitoring points reasonably, then to complete theoverallschemeofthesystem.(2)According to the actual demands of tungsten tailings pond monitoring and warning systemto improve the cluster-tree routing algorithm from optimization of the network path, balancing thenetwork energy and increasing the network data transmission rate, enhanced the data transmissioncapacity andstability oftheZigBeewireless network.(3) Choosing the low-power consumption SoC chip CC2530as MCU and the RF transceiverto complete the hardware design of sensor nodes. Aiming at the limited wireless sensor networknode energy to design a hybrid way of sensor node power supply based on the solar battery, supercapacitorandlithiumbattery. (4) Completing the design of wireless networking, data transmission, serial communicationsoftware of three types of nodes in ZigBee2007protocol stack integration environment.Developing the PC management system by LabVIEW software, and dispiays real-time monitoringdata in the form of chart and form, achieves sound and light warning when the collected dataexceeds the present safety threshold. Background database using SQL Server2005databasemanagement system to design the database table structure, and realize the monitoring data storage,reportsquerying andprinting, etc.(5) Researching the use of support vector machine regression (SVR) algorithm to set up theforecast model of the monitoring parameter data.The LabVIEW back panel calls Matlab softwareas the operation layer to creat corresponding SVR algorithm modular for specific monitoringparameters. The mean error of the SVR regression model is0.1%, the error range is less than theBP neural network,and the prediction accuracy is higher,conforms the requirements for the safeoperation ofthetungstentailings pond.