Research On Key Technology Of Intelligent Control For Autonomous Subsea In-pipe Robot

Subsea high pressure pipeline is the important equipment to achieve the gathering and transportation for offshore oil and gas. After its putting into use, the subsea pipeline suffers the affection from dielectric corrosion, stress variation and etc, so flaws can appear and grow on the walls of pipelines. As a result there exist potential pipeline leakages. Any failure of these offshore pipelines will not only affect productivity negatively but also cause tremendous environmental hazards.An effective way to ensure safe and efficient operation for the long-distance subsea oil pipeline is to perform periodic inspects for them by in-pipe robot, evaluate the detected flaws, then locate and repair them. For the metallic electromagnetic shielding, after being introduced into the pipeline, the in-pipe robot will run autonomously according to given programme and be no longer under the control of us. The inner environment of in-service oil pipelines is extremely bad and it induces many disadvantages, such as that the reliability and accuracy for locating defects by the automomous in-pipe robot can not meet the engineering requirements, and the in-pipe robot might encounter the impassability inside the pipeline for unclear reason and accordingly the pipeline repair tusk is defeated. The problems severely affected the in-pipe robot's usability and security. The above is the project background for this paper, and the main contents of the paper are as follows.(1) Design and realization for the in-pipe robot intelligent controller: according to control mission requirements for the in-pipe robot, the intelligent controller was developed with an embedded PC104 industrial computer as the hardware platform and embedded Linux+Rt-Linux real-time extension as the system software platform. We presented in detail the fault treatment strategy during the in-pipe robot's operation, application layer protocol for CAN communication and intelligent controller software structure. Based on RS232-CAN protocol converter, the simulation platform was construced to test the function for the intelligent controller, which also facilitated the design and validation of control program.(2) Autonomous localization technique of in-pipe robot based on multi-sensor data fusion: based on the structural characteristics of subsea oil pipelines, a novel localization method was brought forward, which firstly achieves rough localization using eddy current sensor to detect girth welds in pipelines, then carries out precise localization between adjacent girth welds by multi-odometer. A production rule was presented to achieve fault-tolerance processing for undetected girth welds, which are induced by in-pipe robot's unstable crawl speed, using movement distance information. Multi-odometer is applied to get redundant movement distance information and the consensus data fusion algorithm is adopted to process them, so in-pipe robot's localization precision is improved as well. The tests on experimental pipeline system and active duty beach oil pipeline show the novel in-pipeline localization method is reliable and practical. According to autonomous localization characteristics of in-pipe robot, we introduced Classical Probability Theory to analyze the reliability for in-pipe robot's rough localization and Distributed Detection Fusion Theory based on Bayesian Principle to analyze the Bayes risk for in-pipe robot's precise localization. When making pipeline maintenance plan, we can evaluate the reliability for the in-pipe robot's autonomous localization process, and therefore could reduce activity risk from autonomous localization failure. The tests on experimental pipeline system show the novel in-pipeline localization method is reliable and practical.(3) Reactive self-rescue control for autonomous in-pipe robot based on Reinforcement Learning: a control technique to achieve self-rescue for autonomous in-pipe robot from obstacle environment based on reinforcement learning was proposed. Motion control strategy of self-rescue was got on line through the interaction between the in-pipe robot and the obstacle environment to avoid loss of rescue activity and task failure. Aprior knowledge of working environment was applied to direct the design of heuristic reward function for the reinforcement learning system, which guaranteed the correct direction of searching and learning control strategy. The tests on the experimental pipeline system showed that it was feasible to achieve self-rescue by self-learning control strategy when the in-pipe robot being in environment obstacle.(4) Experimental verification for the in-pipe robot's intelligent controller: based on the constructed simulation platform, we verified the fundamental function for the intelligent controller, such as CAN bus communication, process flow implemented by control software, fault disposal strategy, Data Collection and Preservation and et al. On the experimental pipeline system and active duty beach oil pipeline, we verified the precision and reliability for the intelligent controller to execute autonomous localization control, and the effect for self-rescure control.The maintenance method with autonomous in-pipe robot for long-distance subsea pipelines is first adopted by internal and overseas, which has the advantages that it has low-cost implementation and is easy for operation, besides, there are no special requirements for maintained pipelines themselves, so it adapts to the point reparation of most welded offshore pipelines. Nevertheless, to adopt the maintenance method sets an extremely high requirement for the mechanism design, driving force and control system's reliability and adaptability of the autonomous in-pipe robot. The research purpose for this paper is to improve the adaptability to the awful pipeline environment by the autonomous in-pipe robot based on intelligent control techniques. The paper research achievements will promote the subsea in-pipe robot into practical application early. Correspondingly, the in-pipe robot will guarantee the safe run of the existing offshore oil pipelines of our country and therefore have a tremendous economic value and social significance.