Pipe microleakage detection method based on deep neural network

Abstract

The invention provides a pipe microleakage detection method based on a deep neural network. By means of the pipe microleakage detection method based on the deep neural network, the pipe leakage monitoring scope is widened, a detection point can be selected flexibly, and secondary damage of an original pipe cannot be caused. The pipe microleakage detection method comprises the following steps of 1,collecting the pipe leakage sound frequency; 2, conducting preprocessing on data; 3, generating a pipe microleakage detection model based on a convolution neural network; and 4, conducting training on the pipe microleakage detection model, and conducting pipe microleakage detection through the pipe microleakage detection model obtained after training is complete. The pipe microleakage detection method has the beneficial effects that by means of the microleakage detection model based on the convolution neural network, a non-contact type detection method can be provided for pipe leakage detection; defects, such as pipe body damage, of an existing method are overcome; and the data foundation is laid for scientific research on relevant problems.

Description

technical field [0001] The invention relates to the field of pipeline tightness detection and detection, and in particular, a method for detecting pipeline micro-leakage based on a deep neural network. Background technique [0002] Pipeline leak detection refers to the tightness detection of water supply, oil, gas and other transportation pipelines in operation. It is of great significance to discover the leakage of dangerous goods in time and prevent the occurrence of major accidents. At present, the common method for detecting pipeline leakage is the pipeline leakage detection method based on hardware, such as the negative pressure wave method, which uses the sudden drop in pressure at the leakage point of the pipeline, the pressure difference of the surrounding fluid, and the fluid flows from the high pressure area to the low pressure area, forming negative pressure fluctuations The principle is to detect the leakage of the pipeline and locate the leakage point of the pi...

AI Technical Summary

Problems solved by technology

However, there are still some deficiencies in the above pipeline leakage detection methods. Taking oil and gas transportation pipelines as an example, conventional detection methods need to install fluid pressure gauges, flow meters and other equipment in the pipeline memory, which destroys the integrity of the pipeline and increases the manufacturing cost of the pipeline. , the risk of leakage of the pipeline itself also increases

Images