Research On Marine Salinity Sensing Technology Based On Miniaturized Devices

Marine environment monitoring technology is an important support for the development and utilization of marine resources,among which marine salinity sensor is a high-end core equipment to obtain the physical characteristics of the ocean and achieve ocean hydrographic observation,and its development level and innovation ability has become an important standard to measure a country’s marine science and technology indicators.With the increase in demand for marine exploration,salinity parameter monitoring towards the water surface high-density,miniaturization,low-cost marine Internet of Things direction,the traditional salinity sensor has high observation costs,low network density and other deficiencies,it is difficult to meet the salinity sensor mass,high-density deployment needs.The miniaturization is an important development trend of marine salinity sensors and an important way to achieve low-power design and lowcost batch production of sensors.In this thesis,we design and implement a marine salinity measurement system based on miniaturized devices.The miniaturized and planar design of the sensing probe is carried out by using micro and nano fabrication methods,combined with multi-physics field simulation,to realize the functional replacement of the traditional temperature and conductivity sensing probes,and the preparation,processing and packaging of the gridtype high-impedance platinum resistor and the strip-shaped seven-electrode planar electrode on the same substrate to realize the integrated temperature and conductivity measurement.A miniaturized salinity measurement module was designed,including a sensing probe module,a master control module,an A/D conversion module,a communication module and a power supply module.In addition,a two-strategy improved seagull optimization algorithm is used to improve the initial weights and thresholds of the BP neural network,and an improved SOA-BP neural network based on the back propagation neural network is established.A model of conductivity data compensation algorithm based on improved SOA-BP neural network,which effectively eliminates the influence of non-target parameters and improves the accuracy of conductivity parameter measurement.In order to verify the overall performance of the measurement system,the performance testing experiments of the miniaturized salinity measurement system were carried out in the laboratory and on the shore of the marine station,and the measurement results were analyzed.The laboratory test results show that the actual accuracy of temperature measurement reaches 0.056℃,the actual accuracy of electrical conductivity measurement reaches 0.059 mS/cm,and it also has good repeatability and stability,and the maximum measurement error rate of salinity is 0.31%.In the shore experimental test stage of the ocean station,compared with the Alec temperature and salinity meter,the average error of electrical conductivity measurement is 0.03 mS/cm,the average error of temperature measurement is 0.04℃,and the average error of salinity measurement is0.068 psu.Therefore,the miniaturized salinity measurement system meets the application requirements at the present stage,can give full play to the advantages of miniaturization in volume,power consumption and cost,and effectively serve the construction of marine Internet of things and marine three-dimensional monitoring system.