Intelligent adaptive system and method for monitoring leakage of oil pipeline networks based on big data

Patent metadata

Field	Value
Publication number	US-9797799-B2
Application number	US-201514692502-A
Country	US
Kind code	B2
Filing date	Apr 21, 2015
Priority date	Apr 28, 2014
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

The present invention relates to an intelligent adaptive system and method for monitoring leakage of oil pipeline networks based on big data. The present invention effectively analyzes a large amount of data collected on site within a reasonable time period and obtains a state of a pipeline network by an intelligent adaptive method, thereby obtaining a topological structure of a pipeline network. The present invention specifically adopts a flow balance method in combination with information conformance theory to analyze whether the pipeline network has leakage; small amount of leakage and slow leakage can be perfectly and accurately alarmed upon detection; as a generalized regression neural network is adopted to locate a leakage of the pipeline network, an accuracy of a result is increased. Therefore, the present invention adopts a policy and intelligent adaptive method based on big data to solve problems of detecting and locating leakage of the pipeline network.

First claim

What is claimed is:

1. An intelligent adaptive system for monitoring leakage of oil pipeline networks based on big data, comprising an upper computer and a lower computer; the lower computer comprises a data collector, a filter circuit, an amplifying circuit and a PLC central processing unit,
wherein the data collector is used for collecting millisecond pressure, millisecond flow rate, temperature and density at an inlet and outlet of a pipeline, conducting multi-source consistent processing to all kinds of collected signals, converting the signals into standard uniform data and sending them to the filter circuit,
wherein the filter circuit is used for conducting noise filtering to the collected signals and sending the signals after filtering to the amplifying circuit,
wherein the amplifying circuit is used for amplifying the collected signals and sending the signals after amplification to the PLC central processing unit,
wherein the PLC central processing unit is used for conducting analogue-to-digital conversion to the collected signals, conducting time correction to the data at the inlet and outlet of each segment of the pipeline, and sending the signals after time correction to the upper computer,
wherein the upper computer is used for obtaining an initial state of a pipeline network in accordance with states of valves and pumps, thereby obtaining a topological structure of the whole pipeline network,
wherein the upper computer is used for determining a priority detection range of a pipeline network area by judging whether the difference between collected inlet flow rate and outlet flow rate exceeds a threshold;
wherein the upper computer is used for determining whether a change of collected signals is human-induced by inspecting whether the states of all valves and pumps are the same as the initial states and judging whether the difference between flow rates is equal to an artificial defueling amount or an artificial refueling amount,
wherein the upper computer is used for conducting conformance testing to the collected flow rates and determining whether the collected flow rates are effective, and
wherein the upper computer is used for determining a theoretical time difference between a pressure wave generated by a theoretical preset leakage point arriving at an upstream sensor and a downstream sensor, taking a sequence of the theoretical time difference and a sequence of a length of each segment of the pipeline as the input of a generalized regression neural network, taking a sequence of the distance from each preset leakage point to the inlet of the pipeline where the preset leakage point is located as an expected output, conducting training to obtain a nonlinear model, and putting an actual time difference between a pressure wave generated by a leakage point arriving at the upstream sensor and the downstream sensor into the nonlinear model obtained by training to obtain the location of an actual leakage point.

2. An intelligent adaptive method for monitoring leakage of oil pipeline networks based on big data of claim 1 , comprising the following steps:
Step 1: Determining the initial state, namely an opening state or a closing state, of each valve and pump in the pipeline network, and obtaining the initial state of the pipeline network in accordance with the states of valves and pumps, thereby obtaining the topological structure of the whole pipeline network;
Step 2: Using the data collector to collect millisecond pressure, millisecond flow rate, temperature and density at the inlet and outlet of the pipeline, and conducting multi-source consistent processing to all kinds of collected signals, converting the signals into standard uniform data and sending them to the filter circuit;
Step 3: Using the PLC central processing unit to conduct analogue-to-digital conversion to the collected signals, conducting time correction to the data at the inlet and outlet of each segment of the pipeline, and sending the data to the upper computer for storage;
Step 4: Judging whether the difference between collected inlet flow rate and outlet flow rate exceeds the threshold; if so, investigating a historical working condition of the pipeline network and determining a monitoring station of the leakage point corresponding to a historical flow rate nearest to this flow rate, and taking the pipeline network area monitored by this monitoring station as the priority detection range; otherwise, returning to execute Step 2;
Step 5: Inspecting whether the states of all valves and pumps are the same as the initial states; if so, executing Step 7; otherwise, determining the artificial defueling amount or the artificial refueling amount of each valve and pump having state changes, determining the difference between the inlet and outlet flow rates of the pipeline network, and executing Step 6;
Step 6: Judging whether the difference between the inlet and outlet flow rates is equal to the artificial defueling amount or the artificial refueling amount; if not, executing Step 7; otherwise, returning to execute Step 2;
Step 7: Conducting conformance testing to the collected flow rates and determining whether the collected flow rates are effective as follows:
Judging whether

F
>

M
+
2

2
⁢
R

is true; if true, executing Step 7; otherwise, returning to execute Step 2,
wherein M is a total number of sensors; R is a number of flow sensors installed at each monitoring station; F is a number of the flow sensors collecting changed data;
Step 8: Determining the actual time difference between the pressure waves generated by the leakage point arriving at the upstream sensor and the downstream sensor, wherein the formula for calculating a cross correlation coefficient between a pressure signal value collected at an upstream end and the pressure signal value collected at a downstream end is as follows:

R

x

i

y
j

⁢

=

lim

N
->
∞

⁢

1
N

⁢

∑

t
=
1

N

⁢

x
⁡

Assignees

• Univ Northeastern

Inventors

• Zhang Huaguang
• Ma Dazhong
• Feng Jian
• Liu Jinhai
• Wang Gang
• Wu Zhenning
• Sun Qiuye
• Li Xiaoyu

Classifications

• G01M3/2807 Primary

  for pipes (G01M3/2892, G01M3/30 take precedence) · CPC title

• G01M3/243

  for pipes · CPC title