Metering method and system for flow of water meter based on improved Elman neural network

Abstract

The invention discloses a metering method and system for flow of a water meter based on an improved Elman neural network. The method comprises acquiring time difference data of each flow point ultrasonic wave in a pipeline; performing random selection on the preprocessed time difference data to obtain a training sample set and a test sample set; constructing the improved Elman neural network model, training the improved Elman neural network model according to the training sample set and using an adaptive error inverse propagation algorithm, and performing simulation verification on the improved Elman neural network model through the test sample set to determine a flow metering model; and acquiring the time difference data of the current ultrasonic wave of a to-be-detected ultrasonic flowmeter by calling the flow metering model, and determining the instantaneous flow rate of the to-be-detected ultrasonic flowmeter according to the time difference data. According to the metering method and system for flow of the water meter based on the improved Elman neural network, the adaptive error inverse propagation algorithm is adopted for training, so that the specific convergence speed is high, thereby avoiding the problem that complex parameter correction and correction compensation are involved to reduce the metering error, and the measurement precision is high.

Description

technical field [0001] The invention relates to the technical field of water meter flow measurement, in particular to a water meter flow measurement method and system based on an improved Elman neural network. Background technique [0002] A water meter is an instrument for measuring water flow, which can be divided into two types: mechanical metering and electronic flow metering. Mechanical metering water meters occupy an absolute proportion in the current market due to their mature technology, low cost, and convenient installation and debugging. However, the water meter with mechanical metering has the disadvantages of low flowmeter measurement accuracy, large pressure loss, single application range and difficult meter reading. In terms of electronic flow measurement: At present, flow formulas derived from measurement principles are generally used, which involve complex corrections or empirical estimates of different parameters, and for some water meters that are easily a...

AI Technical Summary

Problems solved by technology

However, the water meter with mechanical metering has the disadvantages of low flowmeter measurement accuracy, large pressure loss, single application range and difficult meter reading.

In terms of electronic flow measurement: At present, flow formulas derived from measurement principles are generally used, which involve complex corrections or empirical estimates of different parameters, and for some water meters that are easily affected by working conditions, real-time corresponding Correction and compensation to reduce measurement errors, the process is complex and has certain errors

At the same time, there are still errors in the installation and sampling of sensors in the actual flow measurement.

Therefore, the corresponding law between the data collected by the electronic water meter and the water flow has certain complexity, nonlinearity and uncertainty.

Images