CN-116468925-B - Heavy oil storage tank water cutting control method, water cutting system, electronic equipment and storage medium

CN116468925BCN 116468925 BCN116468925 BCN 116468925BCN-116468925-B

Abstract

The invention discloses a water cutting control method for a heavy oil storage tank, which comprises the following steps of S110 collecting top view images and side view images of fluid in a water cutting pipeline, S151 adopting an ICM algorithm based on a Markov random field to divide the top view images and the side view images, S152 obtaining a heavy oil proportion P 1 of the top view images and a heavy oil proportion P 2 of the side view images according to division results, S153 calculating a heavy oil proportion n=mu 1 P 1 +μ 2 P 2 of the fluid in the water cutting pipeline through weighted summation, wherein mu 1 is a weight of the top view images, mu 2 is a weight of the side view images, mu 1 +μ 2 =1, S154 compares the calculated heavy oil proportion n of the fluid in the water cutting pipeline with a preset closing threshold value n 0 , and when n is larger than or equal to n 0 , the water cutting state of the water cutting pipeline is kept and the water cutting pipeline returns to the step S110. The invention also discloses a heavy oil storage tank water cutting system, electronic equipment and a storage medium. According to the invention, the water cutting pipeline image is collected in two directions and the image is segmented, so that the accurate control of water cutting can be realized.

Inventors

HUANG ZIJIAN
SHI LEI
ZHAO YATONG
YANG XIAO

Assignees

中国石油化工股份有限公司
中国石油化工股份有限公司大连石油化工研究院

Dates

Publication Date: 20260505
Application Date: 20220111

Claims (12)

1. The water cutting control method for the heavy oil storage tank is characterized by comprising the following steps of: S110, acquiring a top view image and a side view image of fluid in a cut water pipeline; S151, performing image segmentation on the top view image and the side view image by adopting an ICM algorithm based on a Markov random field; S152, obtaining heavy oil duty ratio of top view image according to the segmentation result And heavy oil duty ratio of side view image ; S153 calculating the heavy oil duty ratio of the fluid in the cut water pipe by weighted summation In which, in the process, For the weights of the top-down image, For the weight of the side-view image, Weights of top-view images And weighting of side view images Weight of top view image determined according to heavy oil density In the middle of Is the ratio of the density of heavy oil to the density of water; S154 comparing the calculated heavy oil duty ratio of the fluid in the cut water pipe And preset closing threshold When (when) Closing the water-cutting pipeline when When the water cut state of the water cut pipe is maintained and the process returns to step S110.
2. The heavy oil tank water cut control method according to claim 1, wherein the preset closing threshold value 5% -8%.
3. The heavy oil tank water cut control method according to claim 1, further comprising, before step S151: S120, performing image recognition on the overlook image and the side view image by using a convolutional neural network to obtain a recognition result; S130, closing the water cutting pipeline according to the identification result, returning to step 110 or proceeding to step S151.
4. The heavy oil tank water cut control method according to claim 3, wherein step S120 includes: S121, acquiring a historical overlooking image and a historical side view image of fluid in a water cutting pipeline, and establishing a training sample set; s122, setting labels for each historical top view image and historical side view image in the training sample set, and marking as cut water and shut off; S123, inputting the training sample set and the corresponding set label into a convolutional neural network for training, and obtaining a trained convolutional neural network model; s124, identifying the acquired overlook image and side view image by using the trained convolutional neural network model to obtain an identification result.
5. The heavy oil tank water cut control method according to claim 4, wherein step S130 includes: when the recognition results of the overlook image and the side view image at the same moment are both 'closed', closing the water cutting pipeline; When the recognition results of the top view image and the side view image at the same time are both "cut water", maintaining the cut water state of the cut water pipe and returning to the step S110, and When the recognition results of the top view image and the side view image at the same time are different, the process advances to step S151.
6. The heavy oil tank water cut control method according to claim 4, wherein step S122 includes: image segmentation is carried out on the historical overlooking image and the historical side view image by adopting an ICM algorithm based on a Markov random field; Obtaining heavy oil duty ratio of historical overlook image according to the segmentation result And heavy oil duty cycle of historical side view images ; When (when) When the label of the historical overlook image is closed, the label is closed The label of the history overlook image is cut water; When (when) When the label of the history side view image is closed, the label is closed The label of the history side view image is "cut water".
7. A heavy oil storage tank water cutting system, comprising: an image acquisition unit for acquiring a top view image and a side view image of the fluid in the cut water pipe; The recognition unit is used for carrying out image recognition on the overlook image and the side view image by utilizing the convolutional neural network to obtain a recognition result; A calculation unit for dividing the top view image and the side view image by using ICM algorithm based on Markov random field, and obtaining heavy oil duty ratio of the top view image according to the division result And heavy oil duty ratio of side view image Calculating the heavy oil ratio of fluid in the water cutting pipeline through weighted summation In which, in the process, For the weights of the top-down image, For the weight of the side-view image, Weights of top-view images And weighting of side view images Weight of top view image determined according to heavy oil density In the following Is the ratio of the density of heavy oil to the density of water; and a control unit for controlling the water cut pipe according to the identification result obtained by the identification unit and the heavy oil ratio of the fluid in the water cut pipe calculated by the calculation unit.
8. The heavy oil tank water cutting system of claim 7, wherein the identification unit comprises: the model acquisition submodule is used for establishing a training sample set based on the historical overlook image and the historical side view image of the fluid in the water cutting pipeline, setting labels for each historical overlook image and each historical side view image in the training sample set, marking as water cutting and closing, inputting the training sample set and the corresponding set labels into the convolutional neural network for training to obtain a trained convolutional neural network model, and And the image recognition sub-module is used for recognizing the acquired overlook image and side view image by utilizing the trained convolutional neural network model, so as to obtain a recognition result.
9. The heavy oil tank water cutting system of claim 7, wherein the image acquisition unit comprises: A transparent pipe section which is a section of a cut water pipe; two cameras for respectively collecting top view image and side view image of fluid in the water cutting pipeline through the transparent pipe section, and And the heat preservation piece is used for keeping the working temperature of the transparent pipe section at 45-55 ℃.
10. The heavy oil tank water cutting system of claim 7, further comprising: and the flushing unit is used for flushing the closed water cutting pipeline.
11. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to cause the at least one processor to perform the heavy oil tank water cut control method according to any one of claims 1 to 6.
12. A non-transitory computer-readable storage medium storing computer-executable instructions for causing a computer to perform the heavy oil tank water cut control method according to any one of claims 1 to 6.

Description

Heavy oil storage tank water cutting control method, water cutting system, electronic equipment and storage medium Technical Field The invention relates to the technical field of storage tanks in petrochemical industry, in particular to a heavy oil storage tank water cutting control method, a water cutting system, electronic equipment and a storage medium. Background The water cutting device is a product developed and designed according to the need of dewatering a storage tank in the petrochemical industry, is arranged at the lower part of an oil tank, is used in the petroleum industry and the chemical industry, replaces an oil tank manual water cutting mode, realizes automatic water cutting, greatly reduces the labor intensity of operation and the phenomenon of hydrogen sulfide poisoning during manual water cutting, reduces the waste of oil products and environmental pollution, and meets the standard requirements of drainage and pipelining of a petrochemical system. The traditional water cutting device is divided into a mechanical water cutting device and an electronic water cutting device, the mechanical water cutting device utilizes the mechanical force generated by buoyancy to automatically open and drain water, and the water draining device can be automatically closed after the water draining is completed, so that manual intervention is not needed, energy is not consumed, and the explosion-proof problem does not exist. However, most mechanical water cutting devices commonly have the problem of malfunctioning when being applied to heavy oil storage tanks, mainly because the heavy oil density such as crude oil is similar to that of water, the phenomenon of water-in-oil, water-in-water oil and the like is prominent, the traditional mechanical water cutting devices work principle is that the buoyancy generated by a floating ball is amplified through a lever and then the combined action of mechanical structures such as springs and the like is matched, so that the control of the valve of the water cutting device by the floating ball is realized, and when the buoyancy generated by the floating ball in the heavy oil is similar to that in water, the valve of the water cutting device cannot be accurately controlled. The electronic water cutter identifies the medium in the current pipeline through the sensor, and controls the valve according to the detection result of the sensor. For example, two sensors are installed at different positions of a pipeline by adopting a liquid column resonance principle, when a mixed medium passes through the pipeline, the sensors can send detection results to a controller, and the controller can judge whether to close a water cutting valve according to the oil content of the medium due to the difference of the viscosity and the dielectric constant of the medium. The electronic water cutter has the characteristic of high sensitivity, and can be used in a heavy oil storage tank, but generally after a few months of use, a sensor probe can be stuck by heavy oil to cause sensor failure, so that the maintenance period is too short, and the electronic water cutter cannot be suitable for high-strength operation storage tanks such as refineries and the like. The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art. Disclosure of Invention The invention aims to provide a heavy oil storage tank water cutting control method, a water cutting system, electronic equipment and a storage medium, so that the problem that the traditional mechanical water cutting device has an unsatisfactory application effect on a heavy oil storage tank is solved. Another object of the present invention is to provide a method for controlling water cut of a heavy oil storage tank, a water cut system, an electronic device, and a storage medium, thereby improving stability and reliability of water cut of the heavy oil storage tank in high-intensity operation. To achieve the above object, according to a first aspect of the present invention, there is provided a heavy oil storage tank water cut control method comprising the steps of: S110, acquiring a top view image and a side view image of fluid in a cut water pipeline; S151, performing image segmentation on the top view image and the side view image by adopting an ICM algorithm based on a Markov random field; S152, obtaining a heavy oil duty ratio P 1 of the top view image and a heavy oil duty ratio P 2 of the side view image according to the segmentation result; S153 calculates the heavy oil ratio n=μ 1P1+μ2P2 of the fluid in the cut-off pipeline by weighted summation, wherein μ 1 is the weight of the top view image, μ 2 is the weight of the side view image, μ 1+μ2 =1; S154, comparing the calculated heavy oil ratio n of the