A method for fusion mapping of multivariate monitoring data in intelligent integrated platform

Abstract

The present invention discloses a method for fusion mapping of multivariate monitoring data of an intelligent integrated platform, comprising the following steps: the intelligent integrated platform monitors and collects real-time data of mechanical, kinematic and dynamic information, and performs screening, cleaning and preprocessing; establishes a BP neural network model of the state information of the monitoring object according to the monitoring and collection data of the intelligent integrated platform; establishes a numerical simulation model for monitoring the object of the intelligent integrated platform, and obtains the information data of the monitoring object through simulation calculation; compares the real-time data collected by the intelligent integrated platform monitoring with the simulation data information of the monitoring object to obtain difference data; realizes iterative evolution of the simulation model; uses the simulation data of the iteratively evolved simulation model as training data to correct the model parameters of the BP neural network model; and displays data of the three-dimensional digital model of the monitoring object. The method of the present invention realizes the all-round display of the time domain and air domain of the monitoring object, and improves the ductility of the monitoring data in the time domain and air domain.

Description

Technical Field

[0001] The present invention relates to intelligent monitoring technology, and in particular to a multivariate monitoring data fusion mapping method for an intelligent integrated platform. Background Art

[0002] The intelligent integrated platform needs to monitor the mechanical, kinematic and dynamic parameters of the equipment, and there are many types and quantities of monitoring parameters. Due to the limitations of factors such as the installation space and cost of monitoring sensors, it is impossible to display the status information of each part of the monitored object. At the same time, the monitoring data can only display historical storage data. The uncollected equipment limit state data and unmonitored future data cannot be displayed, and the expression of predicted data cannot be provided to users. By using the intelligent integrated platform monitoring and simulation multivariate data fusion mapping method, the intelligent integrated platform can display the parameters of each part of the monitored object under any working condition and at any time. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for fusion mapping of multivariate monitoring data of an intelligent integrated platform in view of the defects in the prior art.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a method for fusion mapping of multivariate monitoring data of an intelligent integrated platform, comprising:

[0005] 1) The intelligent integrated platform monitors and collects real-time data of mechanical, kinematic and dynamic information for screening, cleaning and preprocessing;

[0006] 2) Establish a BP neural network model of the monitoring object status information based on the monitoring and collection data of the intelligent integration platform;

[0007] 3) Establishing a numerical simulation model for monitoring the intelligent integrated platform object, wherein the numerical simulation model includes a mechanical model, a kinematic model and a dynamic model;

[0008] Perform mechanical model simulation based on the load and boundary condition information obtained from monitoring, and calculate the mechanical state simulation information of the monitored object;

[0009] The kinematic and dynamic information data of the monitored object are obtained through simulation calculation;

[0010] 4) Compare the real-time data collected by the intelligent integration platform with the simulated data information of the monitored object to obtain differential data;

[0011] Use the difference data as correction data to reverse the simulation model, correct the simulation model parameters, and realize the iterative evolution of the simulation model;

[0012] 5) Using the simulation data of the iteratively evolved simulation model as training data, the model parameters of the BP neural network model are corrected;

[0013] 6) On the three-dimensional digital model of the monitored object, for locations with monitoring points, real-time data acquired by the monitoring points are mapped to the three-dimensional digital model of the monitored object; for locations without monitoring points, mechanical simulation data calculated by the mechanical simulation model are mapped to locations where there is no monitoring data on the three-dimensional digital model of the monitored object, and kinematic and dynamic information data predicted by the BP neural network model are mapped to locations where there is no monitoring data on the three-dimensional digital model of the monitored object;

[0014] 7) Perform consistency analysis and processing on the three-dimensional digital model of the monitored object mapped with the monitoring data and the predicted data, and use filtering algorithms to clean and eliminate error data to achieve a multi-dimensional panoramic display of the multivariate data of the monitored object.

[0015] The beneficial effects produced by the present invention are:

[0016] The present invention adopts a fusion display method of real-time monitoring data, model simulation data and prediction data. Real-time monitoring data displays real-time status data collected by sensors, model simulation data displays simulation data calculated by mechanical simulation models, and prediction data displays prediction data obtained by kinematic and dynamic neural network models of monitored objects. Through the fusion display of the three types of data, the display of real-time and future data of monitored objects is realized, and the parts without sensors are also observed, thereby realizing the all-round and multi-angle display of the time domain and air domain of the monitored objects, expanding the scope of monitoring, and improving the ductility of monitoring data in the time domain. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0018] Figure 1 It is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0020] like Figure 1 As shown, a method for fusion mapping of multivariate monitoring data of an intelligent integrated platform includes:

[0021] 1) The intelligent integrated platform monitors and collects real-time data of mechanical, kinematic and dynamic information for screening, cleaning and preprocessing;

[0022] 2) Establish a BP neural network model of the monitoring object status information based on the monitoring and collection data of the intelligent integration platform;

[0023] 3) Establish a numerical simulation model for monitoring the intelligent integrated platform object, perform simulation based on the load and boundary condition information obtained from monitoring, and obtain the mechanical, kinematic and dynamic information data of the monitored object through simulation calculation;

[0024] Numerical simulation model: Use the principles of mechanics to build the mechanical models of the derrick, hook, hydraulic cylinder and the connection position with the deck; use kinematics and dynamics methods to build the kinematics and dynamics models of the key components of the drilling and production system;

[0025] Input the load and boundary condition constraints into the mechanical model of the derrick, the big hook, the hydraulic cylinder and the connection position with the clamping plate; obtain the state simulation information of the monitored object through the simulation calculation of the above mechanical model;

[0026] 4) Compare the real-time data collected by the intelligent integration platform with the simulated data information of the monitored object to obtain differential data;

[0027] Compare the stress, strain, position, speed, load and constraint information of the operating system obtained through real-time monitoring with the simulation information of the monitored object to obtain difference data;

[0028] Use the difference data as correction data to reverse the simulation model, correct the simulation model parameters, and realize the iterative evolution of the simulation model;

[0029] 5) Using the simulation data of the iteratively evolved simulation model as training data, the model parameters of the BP neural network model are corrected;

[0030] 6) On the three-dimensional digital model of the monitored object, for locations with monitoring points, real-time data acquired by the monitoring points are mapped to the three-dimensional digital model of the monitored object; for locations without monitoring points, mechanical simulation data calculated by the mechanical simulation model are mapped to locations where there is no monitoring data on the three-dimensional digital model of the monitored object, and kinematic and dynamic information data predicted by the BP neural network model are mapped to locations where there is no monitoring data on the three-dimensional digital model of the monitored object;

[0031] 7) Perform consistency analysis and processing on the three-dimensional digital model of the monitored object mapped with the monitoring data and the predicted data, and use filtering algorithms to clean and eliminate error data to achieve a multi-dimensional panoramic display of the multivariate data of the monitored object. Using the constructed three-dimensional digital model of the monitored object, the currently collected status information is displayed in the form of text, charts, and cloud maps; the predicted display and visual expression of the status information of any working condition and any time of each part of the monitored object of the operating system is achieved.

[0032] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for fusion mapping of multivariate monitoring data of an intelligent integrated platform, characterized in that: The following steps are involved: 1) The intelligent integrated platform monitors the collected real-time data of mechanical, kinematic and dynamic information, and performs screening, cleaning and preprocessing; 2) Establish a BP neural network model of the monitoring object status information based on the monitoring and collection data of the intelligent integration platform; 3) Establishing a numerical simulation model for monitoring the intelligent integrated platform object, wherein the numerical simulation model includes a mechanical model, a kinematic model and a dynamic model; Perform mechanical model simulation based on the load and boundary condition information obtained from monitoring, and calculate the mechanical state simulation information of the monitored object; The kinematic and dynamic information data of the monitored object are obtained through simulation calculation; 4) Compare the real-time data collected by the intelligent integration platform with the simulated data information of the monitored object to obtain differential data; Use the difference data as correction data to reverse the simulation model, correct the simulation model parameters, and realize the iterative evolution of the simulation model; 5) Using the simulation data of the iteratively evolved simulation model as training data, the model parameters of the BP neural network model are corrected; 6) On the three-dimensional digital model of the monitored object, for locations with monitoring points, mapping the real-time data acquired by the monitoring points onto the three-dimensional digital model of the monitored object; For positions without monitoring points, the mechanical simulation data calculated by the mechanical simulation model are mapped to the positions where the three-dimensional digital model of the monitored object has no monitoring data, and the kinematic and dynamic information data predicted by the BP neural network model are mapped to the positions where the three-dimensional digital model of the monitored object has no monitoring data; 7) Performing data consistency processing on the three-dimensional digital model of the monitored object mapped with the monitoring data and the prediction data to achieve a multi-dimensional panoramic display of the multivariate data of the monitored object.

Images