Study On Reservoir Water Remote Measurement And Control System

With our country paying more and more attention to the shortage of water resources and environment pollution problem, monitoring precisely and using of water resource becomes more and more important. There are many water conservancy projects, such as water diversion, transfer and so on in china. In these projects, most of the reservoir water are transported to the using area through the pipeline of large aperture and most of time the water is not full of the flowing tube. But many of the traditional measurement method under the condition of large aperture, non-full of pipe can not meet the measurement accuracy requirements. Multi-path ultrasonic measurement can solve the problem of large aperture measurement, but poor precision will come in the non-full pipe; electromagnetic flowmeter of non-full pipe and liquidometer can measure the water in the non-full pipes accurately, but as the diameter increase, costs are also increasing rapidly. Also at this stage most of the water management system is using the B/S architecture or C/S architecture, but for some small reservoirs of small size and monitoring points, a low solution is needed. So a comprehensive solution of remote measurement system under the condition of large aperture, non-full pipe and designing a suitable and small scale application has important significance.For the problem of accuracy under the condition of large aperture, non-full pipe using the ultrasonic flowmeter, a kind of high-speed-neural network algorithm is put forward based on the existing theory of algorithm. The speed means that the pipe is divided into several corresponding acoustic path by the multi-path ultrasonic flowmeter to measure each acoustic path velocity; height means that the measured water level by the liquidometer in the non-full pipe; neural network algorithm means that using the establishment of a three layer BP neural network to calculate the current flow by the measured velocity and height information. At the same time, using the Gauss-Jacoby numerical algorithm to determine the numbers of paths and height.This paper design an embedded remote water measurement system, in the light of actual needs of small size and low cost water metering system. The system overall used start type of network structure, composed by a center node and multiple collection node composition, Center node and collection node communicate with each other through GPRS network, No communication between collection nodes. The STM32processor is used as Collection Nodes for multi-channel data acquisition and processing, and sends the collected data to the central through GPRS network. S5PV210embedded processor is used for the central node, equipped with Linux operating system, receive GPRS data via Ethernet, storing data to a SQlite Database, running the embedded WEB server. Thus user can achieve the aims of data via the browser.Verified by the experiment, the measuring accuracy of the algorithm has improved as to original, the system runs stably, better meet the actual needs. The system has low cost, strong scalability, con venient monitoring, for small water metering system has the certain reference value.