Research On Natural Gas Pipeline Leak Detection Based On Collaborative Prediction And Recursive Filtering

The importance of pipeline leak detection technology in energy security management is self-evident.With the continuous improvement of smart grid construction,wireless sensors are applied to more and more pipeline transportation engineering in terms of monitoring and management.The frequent scrapping of instruments and meters caused by the uninterrupted data transmission of a large number of sensors makes it necessary to develop energy-saving pipeline leak detection technology,which is of great significance for achieving sustainable development of energy.In order to save energy,pulse-width modulation transmission mechanism is widely used in many pipeline systems,and the working status of sensor nodes is cyclically alternated between active and dormant.Therefore,studying the leak detection technology based on pulse-width modulation transmission mechanism can help save energy,which has great practical significance.In this paper,natural gas pipelines are taken as the research object,and a pipeline leak detection scheme based on recursive filtering is proposed.This paper intends to explore the linear model and nonlinear model of pipeline leak detection schemes under the pulse-width modulation transmission mechanism,combined with cooperative prediction algorithms.The main work is as follows:1)A state space expression and a discrete model were established for the natural gas pipeline flow system.Firstly,based on the principles of mass conservation and momentum conservation,the pipeline system was modeled,and the dynamic equations of parameters such as internal pressure,flow and density of the pipeline were established according to the segmental characteristics of the pipeline.Secondly,the continuous time model was discretized to obtain a discrete system model.Finally,by setting appropriate boundary conditions,the state space model of the natural gas pipeline flow system was obtained,providing a solid theoretical basis for the natural gas pipeline leakage detection scheme proposed in this paper.2)In this paper,a pipeline leakage detection method based on Kalman filter was proposed for the state space model of natural gas pipeline,and its effectiveness was verified by natural gas pipeline leakage detection experiment.Firstly,based on the state estimation theory of Kalman filter,the estimated values of the natural gas pipeline pressure and flow were obtained,and then the residual was obtained by subtracting the filtering estimated observation value from the pressure and flow observation value.Secondly,the residual threshold was determined under normal operating conditions,and then the pipeline leakage was judged by whether the observation residual exceeded the normal threshold.Finally,the effectiveness of the detection method based on Kalman filter was verified in the natural gas pipeline leakage detection experiment.3)For a class of nonlinear time-varying systems under the duty cycle transmission mechanism,a recursive filtering scheme combining a cooperative prediction method is proposed.The purpose of this paper is to design a practical recursive filtering scheme for linear time-varying systems affected by duty cycle scheduling under energy constraints.First,the duty cycle transmission mechanism is modeled according to the corresponding scheduling rules.Secondly,the cooperative prediction algorithm is used to obtain the unmeasured output of the sleeping sensor,and a recursive filtering scheme is proposed on this basis.In addition,the filter gain is calculated by minimizing the trace of the error covariance.Subsequently,the boundedness of the designed filtering algorithm error covariance was analyzed.Finally,through simulation examples and natural gas pipeline leakage detection experiments,the effectiveness of the algorithm designed in this paper in nonlinear systems4)For nonlinear systems with saturated sensors,a nonlinear recursive filtering scheme combining the cooperative prediction method is proposed.Taking the nonlinear case of sensor saturation as an example,a nonlinear recursive filtering algorithm is proposed under the duty cycle transmission mechanism,so that the designed pipeline leakage detection scheme can still achieve satisfactory filtering performance when applied to actual systems.To prove the effectiveness of the algorithm designed in this paper in nonlinear systems,first,a set of matrix differential equations are solved to obtain the upper bound of the filtering error covariance.Secondly,the filter gain matrix that minimizes this upper bound is calculated,and the boundedness of the filtering error covariance upper bound is analyzed.Finally,through simulation examples and natural gas pipeline leakage detection experiments,the effectiveness of the algorithm designed in this paper in nonlinear systems with sensor saturation is verified.