An integrated air-space-ground-sea submarine pipeline inspection and early warning method based on computing power application

By creating a network computing satellite cluster in the air and using Beidou satellite to transmit subsea pipeline data for real-time analysis, the problem of low efficiency of AI data back-passing ground analysis during subsea pipeline inspection is solved, and the efficiency of subsea pipeline detection and early warning is improved, and the efficiency of subsea pipeline detection and early warning is achieved. Commercial services that integrate space, earth and sea.

CN115865958BActive Publication Date: 2024-11-15CHINA TELECOM DIGITAL INTELLIGENCE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211360150.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-02
Publication Date
2024-11-15
Estimated Expiration
2042-11-02

AI Technical Summary

Technical Problem

The existing technology has a technical bottleneck in the ground analysis of AI data back-passing ground in the inspection of subsea pipelines, resulting in inefficient inspections.

Method used

By creating a network computing satellite cluster in the air, using Beidou satellite to transmit subsea pipeline detection data for real-time analysis and early warning, avoiding data back to ground calculation, using critical external pressure threshold calculation and signal enhancement model, computing node scheduling model, and subsea pipeline The inspection model performs data processing.

Benefits of technology

It improves the efficiency of submarine pipeline detection and inspection, realizes real-time prediction and early warning, solves the bottleneck of AI data back-passing ground analysis, and enhances the ability of marine pipeline detection and rescue.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115865958B_ABST
    Figure CN115865958B_ABST
Patent Text Reader

Abstract

The present invention discloses an integrated submarine pipeline inspection and early warning method based on computing power application, including: step 1, collecting submarine pipeline detection data, calculating the critical external pressure threshold, combining other inspection data as deep-sea data of the submarine pipeline area, and transmitting it to the air base station through Beidou satellite; step 2, the air base station creates a network computing power satellite group, and performs ocean pipeline detection analysis and early warning based on Beidou satellite RDSS data. The computing power satellite solves the technical bottleneck that all AI data needs to be transmitted back to the ground for analysis, and plays its role in the air to complete the prediction of real-time inspection data in the air without transmitting it back to the ground for calculation, thereby greatly improving the efficiency of submarine pipeline detection and inspection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of ocean detection technology, and specifically relates to an air-space-ground-sea integrated submarine pipeline inspection and early warning method based on computing power application. Background Art

[0002] In 2022, the new generation of Beidou will provide more accurate and reliable services to global users, and realize star-to-star networking and interconnection through intersatellite links. In the same year, China Telecom's "Tiantong-1" world's first and largest 5G SA commercial network also provided support for China Telecom's precise positioning of aerial base station satellites. With the commercialization of aerial base stations and Beidou satellite navigation, the generation of aerial satellite commercial data is increasingly closely related to the ground and the ocean. The aerial satellite commercial service model is diverse. The ground facilities combined with the aerial base station and Beidou interact with the ground application platform. The relevant technical solutions calculate the early warning of the submarine pipeline based on the signal propagation time and related data measured by the ground control center. Use the real-time location data and related maps of navigation users to manage submarine pipelines. At the same time, the application platform technology architecture has put forward higher requirements for pipeline disaster early warning in complex scenarios to realize the integration of air, land, sea and big data to support artificial intelligence. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an integrated air-space-ground-sea submarine pipeline inspection and early warning method based on computing power application in view of the deficiencies of the above-mentioned prior art. The computing power satellite is used to solve the technical bottleneck that all AI data need to be transmitted back to the ground for analysis, and its role in the air is played to complete the prediction of real-time inspection data in the air without transmitting it back to the ground for calculation, thereby greatly improving the efficiency of submarine pipeline detection and inspection.

[0004] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0005] An air-space-ground-sea integrated submarine pipeline inspection and early warning method based on computing power application, comprising:

[0006] Step 1: Collect submarine pipeline detection data, calculate the critical external pressure threshold, combine it with other inspection data as deep-sea data of the submarine pipeline area, and transmit it to the air base station via Beidou satellite;

[0007] Step 2: The aerial base station creates a network computing satellite cluster to conduct marine pipeline detection, analysis and early warning based on Beidou satellite RDSS data.

[0008] To optimize the above technical solutions, the specific measures taken also include:

[0009] In the above step 1, the detector installed on the submarine pipeline collects data, calculates the critical external pressure threshold and performs external pressure verification to obtain the optimal critical external pressure threshold, where:

[0010] The calculation formula of critical external pressure threshold is:

[0011]

[0012] Per is the critical external pressure threshold;

[0013] σF is the specified minimum yield strength;

[0014] D is the outer diameter of the steel pipe;

[0015] t is the minimum wall thickness;

[0016] C is the critical crushing pressure of an absolutely smooth circular tube.

[0017] The external pressure verification formula used in the above step 1 is:

[0018]

[0019] Among them, η xp is the axial stress σ x = Allowable use coefficient when 0;

[0020] η yp is the hoop stress σ x = Allowable use coefficient when 0;

[0021] σ xcr is the axial critical stress;

[0022] σ ycr is the critical hoop stress.

[0023] The above step 1 uses BeiDou satellite to pull the BeiDou receiver of the drilling or fishing vessel to collect deep-sea data of the submarine pipeline area and transmit it to the aerial base station.

[0024] The above step 2 uniformly transmits the RDSS short messages of the Beidou satellite to the core satellite, which then transfers them to the network computing satellite cluster to realize air network AI.

[0025] In the above-mentioned network computing satellite cluster, the network computing satellite and the core satellite use NR exclusive channels to transmit data messages.

[0026] The above-mentioned network computing satellite cluster deployment signal enhancement model, computing node scheduling model, and submarine pipeline detection model;

[0027] Among them, the signal enhancement model uses the signal-to-noise ratio as the strength of the signal received by the Beidou receiver on the sea surface. When the signal strength is confirmed to be 3db or above, it indicates that the data is true and valid;

[0028] The computing power node scheduling model obtains the optimal computing power node in the air by calculating the quality of computing power nodes;

[0029] The submarine pipeline inspection model uses the critical external pressure threshold as the characteristic value of the monitoring indicator to judge the indicator abnormality.

[0030] The above-mentioned computing power node quality = average latency + average energy consumption;

[0031] The average delay formula is:

[0032]

[0033] Among them, S k (t)) is the time when the computing task is offloaded to the server-side computing queue of the cloud data center.

[0034] Q k (t)The task queue existing on the edge computing node.

[0035] t represents the tth decision time;

[0036] k, K represent the kth edge node and the total number of edge nodes, respectively;

[0037] The average energy consumption formula is:

[0038]

[0039] In the formula, w k 、w N+1 They represent the energy consumption weights of edge computing nodes and cloud data center servers respectively;

[0040] T are edge computing cycles.

[0041] The above submarine pipeline inspection model is:

[0042] According to the conditional probability formula, under the condition that the characteristic value T is satisfied, the probability that a suspected anomaly is judged as an anomaly R1 is

[0043]

[0044] Among them, P(T|R1) represents the probability of the eigenvalue T appearing in the abnormal subset of the sample set;

[0045] P(R1) represents the proportion of abnormal subsets in the entire sample set.

[0046] The probability of a suspected abnormality being judged as normal R2 under the condition of the eigenvalue T is expressed as

[0047]

[0048] Among them, P(T|R2) represents the probability of the eigenvalue T appearing in the normal subset of the sample set, and P(R2) represents the proportion of the normal subset in the entire sample set.

[0049] If P(R1|T)>P(R2|T), the monitored indicator is considered abnormal; otherwise, it is normal.

[0050] The present invention has the following beneficial effects:

[0051] First, the detector installed on the submarine pipeline collects data to obtain the critical external pressure threshold. A secondary external pressure calibration is performed to obtain the optimal critical external pressure threshold. The data is then transmitted to fishing vessels, detection wells and other sea surface facilities.

[0052] Secondly, BeiDou satellites pull deep-sea data of submarine pipeline areas collected by BeiDou receivers on drilling or fishing vessels and transmit them to the aerial base station network computing node satellites.

[0053] Then, an aerial base station and a network computing satellite are built, and after the Beidou satellite RDSS data is transmitted to the aerial base station network computing satellite, the marine pipeline detection analysis and early warning are analyzed.

[0054] This will help the aerial base station to provide commercial services for marine rescue, survey and other emergency marine scenarios, and ultimately complete the commercial services of air, land, and sea integration.

[0055] 1. Use artificial intelligence models to conduct predictive analysis for submarine pipeline inspection, early warning and other application scenarios.

[0056] 2. Computing satellites can be used to resolve the technical bottleneck of all AI data needing to be transmitted back to the ground for analysis. Computing satellites can be used to predict real-time inspection data in the air without transmitting it back to the ground for calculation, thereby greatly improving the efficiency of submarine pipeline detection and inspection.

[0057] 3. The network computing satellite of the present invention is different from ordinary satellites in that the chip material is mainly silicon photonic chips; the network computing satellite and the core satellite use NR exclusive channels to transmit data messages; the computing node scheduling model is used to analyze and obtain the optimal computing node in the air; the signal enhancement program uses the signal-to-noise ratio as the strength of the signal received by the Beidou receiver on the sea surface. When the signal strength is confirmed to be 3db or above, it proves that the AI ​​computing data of the submarine pipeline inspection model for sea fishing boats, detection wells, etc. is true and valid. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is the principle diagram of the method in the present invention;

[0059] Figure 2 There are various factors that cause submarine pipeline failures. DETAILED DESCRIPTION

[0060] The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0061] like Figure 1 As shown, a method for early warning of submarine pipeline inspection based on computing power application for air, land, and sea integration includes:

[0062] Step 1: Collect submarine pipeline detection data, calculate the critical external pressure threshold, combine it with other inspection data as deep-sea data of the submarine pipeline area, and transmit it to the air base station via Beidou satellite;

[0063] In the embodiment, the critical external pressure threshold is obtained by collecting data through detectors installed on the submarine pipeline; the Beidou satellite pulls the deep sea data of the submarine pipeline area collected by the Beidou receiver of the drilling or fishing boat; and transmits it to the aerial base station network computing node satellite; specifically, the sonar sensing system technology is combined with the pipeline transponder appearance image data collected by the underwater robot perception system, the critical external pressure threshold related data obtained by the collection data of the pipeline area where the submarine transponder is located, and the robot inspection trajectory data are four parts of data.

[0064] (1) The detectors installed on the submarine pipeline collect data, calculate the critical external pressure threshold, and perform external pressure verification to obtain the optimal critical external pressure threshold;

[0065] Under the effect of external overpressure on the seabed, calculate the critical external pressure:

[0066]

[0067] Per is the critical external pressure, MPa;

[0068] σF is the specified minimum yield strength, MPa;

[0069] D is the outer diameter of the steel pipe, mm;

[0070] t is the minimum wall thickness, mm;

[0071] C is the critical crushing pressure of an absolutely smooth circular tube, MPa;

[0072] When the submarine pipeline is subjected to the combined effects of bending moment, axial force and annular force, the critical external pressure can be checked according to the following formula:

[0073]

[0074] η xp is the axial stress σ x = Allowable use coefficient when 0;

[0075] η yp is the hoop stress σ x = Allowable use coefficient when 0;

[0076] σx is the axial stress, MPa;

[0077] σ x is the hoop stress, MPa;

[0078] σ xcr is the axial critical stress;

[0079] σ ycr is the critical hoop stress.

[0080] (2) BeiDou satellite pulls the BeiDou receiver on the drilling or fishing vessel to collect deep-sea data of the submarine pipeline area and transmit it to the airborne base station.

[0081] Step 2: The aerial base station creates a network computing satellite cluster to conduct marine pipeline detection, analysis and early warning based on Beidou satellite RDSS data.

[0082] In the embodiment, an aerial base station and a network computing satellite are constructed, and after the Beidou satellite RDSS data is transmitted to the aerial base station network computing satellite, marine pipeline detection analysis and early warning are performed. Specifically, the Beidou two-way RDSS short message is sent to the aerial base station (China Telecom "Tiantong No. 1"). The aerial base station satellite performs real-time tracking and corresponding early warning of the entire process of the optical cable and oil pipeline of the submarine underwater robot through data sorting and analysis. The status is abnormal and normal. At the same time, the data is transmitted to the ground IAB base station through the satellite relay point closest to the ground of the aerial base station for data storage and archiving, truly completing the whole process of integrated air-space-ground-sea submarine pipeline detection combined with underwater robot application.

[0083] S1, create a network computing satellite group at the air base station, and transmit the seabed data collected by the sea surface pipeline detection receiver to the 6G air base station core satellite (for example: China Telecom's "Tiantong-1") through the Beidou signal receiver deployed by itself through the Beidou satellite's RDSS short message method, and then transfer it to the network computing satellite group by the core satellite. Realize air network AI.

[0084] The difference between network computing satellites and ordinary satellites:

[0085] 1. The chip material is mainly silicon photonic chip

[0086] 2. NR exclusive channels are used to transmit data messages between the network computing satellite and the core satellite.

[0087] Secondly, the computing power node scheduling model is used to analyze and obtain the optimal computing power node in the air.

[0088] Then, the signal enhancement program uses the signal-to-noise ratio as the strength of the signal received by the Beidou receiver on the sea surface. When the signal strength is confirmed to be 3db or above, it proves that the AI ​​calculation data of the submarine pipeline inspection model for sea fishing boats, detection wells, etc. is true and effective.

[0089] S2, the network computing satellite cluster mainly deploys three parts: signal enhancement model, computing node scheduling model, and submarine pipeline detection model.

[0090] 1. Signal Enhancement Program

[0091] Signal Strength and Carrier-to-Noise Ratio

[0092] In fact, the strength of the received signal power cannot fully describe the clarity or quality of the signal. Generally speaking, it is also necessary to know the strength of the signal relative to the noise. The signal-to-noise ratio is usually used to describe the strength of the signal received by the receiver, which is defined as the ratio of the signal power PR to the noise power N.

[0093] The signal-to-noise ratio is usually expressed in decibels, without units. The higher the signal-to-noise ratio, the better the signal quality. A low signal-to-noise ratio affects the receiver's acquisition and tracking performance.

[0094] Noise power N is equal to noise temperature Tn: N = kT n B n

[0095] Carrier-to-noise ratio expression:

[0096] Relationship between signal-to-noise ratio and carrier-to-noise ratio: C / N0 = SNR × B n

[0097] The formula for calculating the noise temperature Tn is called the Fris formula, and the specific expression is

[0098]

[0099] 2. Computing Node Scheduling Model:

[0100] Latency is the most important indicator in the communication system. Therefore, through the reasonable allocation of network resources and computing resources, we can ensure that performance indicators such as computing task latency and system energy consumption are optimized under the constraints of limited resources and task priority.

[0101] Average delay formula

[0102] S k (t)) is the time when the computing task is offloaded to the server-side computing queue of the cloud data center.

[0103] Q k (t) The task queue existing on the edge computing node.

[0104] k, K represent the kth edge node and the total number of edge nodes, respectively;

[0105] t represents the tth decision time.

[0106] Average energy consumption formula

[0107] In the formula, w k 、w N+1 They represent the energy consumption weights of edge computing nodes and cloud data center servers respectively.

[0108] T are edge computing cycles.

[0109] Computing node quality = average latency + average energy consumption

[0110] 3. Submarine pipeline inspection model

[0111] This method has only two states, one is abnormal R1 and the other is normal R2. The extracted monitoring index characteristic value is expressed as

[0112] T={t1,t2,...,t 12}(For example: Critical external pressure threshold: MPa)

[0113] According to the conditional probability formula, under the condition that the characteristic value T is met, the probability that the suspected anomaly is judged as R1 is

[0114] Among them, P(T|R1) represents the probability of the feature value T appearing in the abnormal subset in the sample set, and P(R1) represents the proportion of the abnormal subset in the entire sample set.

[0115] Similarly, the probability that a suspected anomaly is judged as R2 under the condition that the eigenvalue T appears is expressed as

[0116]

[0117] Among them, P(T|R2) represents the probability of the eigenvalue T appearing in the normal subset of the sample set, and P(R2) represents the proportion of the normal subset in the entire sample set.

[0118] Due to Bayesian thinking, each eigenvalue t i They are independent of each other. After the probability of suspected anomalies being judged as R1 and R2 is calculated based on the probability of relatively independent events, the two probabilities are compared. If P(R1|T)>P(R2|T), the monitored indicator is considered abnormal; otherwise, it is normal.

[0119] Various factors that cause submarine pipeline failures such as Figure 2 shown.

[0120] Advantages of doing this:

[0121] 1. Use artificial intelligence models to conduct predictive analysis for submarine pipeline inspection, early warning and other application scenarios.

[0122] 2. Computing satellites can be used to resolve the technical bottleneck of all AI data needing to be transmitted back to the ground for analysis. Computing satellites can be used to predict real-time inspection data in the air without transmitting it back to the ground for calculation, thereby greatly improving the efficiency of submarine pipeline detection and inspection.

[0123] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A method for early warning of submarine pipeline inspection based on computing power application for air, land, and sea, characterized in that: include: Step 1: Collect submarine pipeline detection data, calculate the critical external pressure threshold, combine it with other inspection data as deep-sea data of the submarine pipeline area, and transmit it to the air base station via Beidou satellite; Step 2: The aerial base station creates a network computing satellite group to conduct marine pipeline detection, analysis and early warning based on Beidou satellite RDSS data; In the step 1, the detector installed on the submarine pipeline first collects data to obtain the critical external pressure threshold, performs external pressure verification, obtains the optimal critical external pressure threshold, and transmits the data to the sea surface facilities, including the Beidou receiver of the drilling or fishing vessel; secondly, the Beidou satellite pulls the deep sea data of the submarine pipeline area collected by the Beidou receiver of the drilling or fishing vessel, and transmits it to the air base station; The calculation formula of critical external pressure threshold is: p er is the critical external pressure threshold; σ F is the specified minimum yield strength; D is the outer diameter of the steel pipe; t is the minimum wall thickness; C is the critical crushing pressure of an absolutely smooth circular tube; The network computing power satellite group deploys a signal enhancement model, a computing power node scheduling model, and a submarine pipeline detection model; Among them, the signal enhancement model uses the signal-to-noise ratio as the strength of the signal received by the Beidou receiver on the sea surface. When the signal strength is confirmed to be 3db or above, it indicates that the data is true and valid; The computing node scheduling model obtains the optimal computing node in the air by calculating the computing node quality. The computing node quality = average delay + average energy consumption. The submarine pipeline inspection model uses the critical external pressure threshold as the characteristic value of the monitoring indicator to judge the indicator abnormality.

2. According to claim 1, a method for early warning of submarine pipeline inspection based on computing power application for air, land, and sea integration is characterized in that: The external pressure verification formula is: Among them, η xp is the axial stress σ x = Allowable use coefficient when 0; η yp is the hoop stress σ y = Allowable use coefficient when 0; σ xcr is the axial critical stress; σ ycr is the critical hoop stress.

3. The method for early warning of submarine pipeline inspection based on computing power application based on air-space-ground-sea integration according to claim 1 is characterized in that: The step 2 uniformly transmits the RDSS short messages of the Beidou satellite to the core satellite, which then transfers them to the network computing satellite group to realize air network AI.

4. The method for early warning of submarine pipeline inspection based on computing power application based on air-space-ground-sea integration according to claim 3 is characterized in that: In the network computing satellite cluster, the network computing satellite and the core satellite use an NR exclusive channel to transmit data messages.

5. The method for early warning of submarine pipeline inspection based on computing power application based on air-space-ground-sea integration according to claim 1 is characterized in that: The average delay formula is: Among them, S k (t) is the time when the computing task is offloaded to the server-side computing queue of the cloud data center; Q k (t) The task queue existing on the edge computing node; t represents the tth decision time; k, K represent the kth edge node and the total number of edge nodes, respectively; The average energy consumption formula is: In the formula, w k 、w N+1 They represent the energy consumption weights of edge computing nodes and cloud data center servers respectively; T are edge computing cycles.

6. The method for early warning of submarine pipeline inspection based on computing power application based on air-space-ground-sea integration according to claim 1 is characterized in that: The submarine pipeline inspection model is: According to the conditional probability formula, under the condition that the characteristic value T is satisfied, the probability that a suspected anomaly is judged as an anomaly R1 is Among them, P(T|R1) represents the probability of the eigenvalue T appearing in the abnormal subset of the sample set; P(R1) represents the proportion of abnormal subsets in the entire sample set; The probability of a suspected abnormality being judged as normal R2 under the condition of the eigenvalue T is expressed as Among them, P(T|R2) represents the probability of the feature value T appearing in the normal subset of the sample set, and P(R2) represents the proportion of the normal subset in the entire sample set; If P(R1|T)>P(R2|T), the monitored indicator is considered abnormal; otherwise, it is normal.