A digital twin system for identification and control of large-tonnage loaders

By integrating operation environment identification, data expansion and twin model control systems on large tonnage loaders, the difficulties of performance prediction and trajectory planning in complex environments are solved, and the adaptability and efficient management of the system are realized, energy consumption is reduced and improved Overall performance of the loader.

CN114444816BActive Publication Date: 2024-11-15HENAN POLYTECHNIC UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210232924.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-09
Publication Date
2024-11-15
Estimated Expiration
2042-03-09

AI Technical Summary

Technical Problem

It is difficult for large-tonnage loaders to achieve rapid and accurate performance prediction and operation trajectory planning in complex operating environments, resulting in high energy consumption and low efficiency of the system, and lack of technical support for intelligent and unmanned management.

Method used

A digital twin system including an operating environment identification system, a data expansion system, a twin model control system and a data storage system was designed. The identification environment is used for laser rangefinder, image sensor, position sensor, etc., and through data fusion and group intelligence optimization algorithms Compute mechanical properties, plan the best loading trajectory, and store historical data to support subsequent optimization.

Benefits of technology

The large-tonnage loader has realized the identification, perception and adaptive mechanical performance calculation of the operating environment, planned the best loading trajectory, and provided scientific based design, monitoring and full life cycle management, reducing energy consumption and improving efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114444816B_ABST
    Figure CN114444816B_ABST
Patent Text Reader

Abstract

The present invention discloses a digital twin system for identifying and controlling a large-tonnage loader, which involves machine learning and big data mining technologies, and digital twin technologies for complex electromechanical intelligent equipment, including an operating environment identification system, a data expansion system, a twin model control system, and a data storage system. The operating environment identification system mainly utilizes functional devices such as laser rangefinders, image sensors, and position sensors to identify the operating environment of large-tonnage devices, and screens, filters, and classifies the data obtained by identification to obtain useful operating environment data. The data expansion system is mainly used to fuse the currently obtained time series big data, operating object morphology big data, equipment morphology small sample data, and historical operating data to form an initial data set that fuses multi-source information. The present invention can provide a scientific basis for the design, monitoring, performance prediction, and full life cycle management of large-tonnage loaders.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin intelligent monitoring of electromechanical equipment, and specifically to an identification and control digital twin system for a large-tonnage loader. Background Art

[0002] Large-tonnage loaders are widely used in heavy-duty loading occasions such as mines. They have the advantages of fast speed, good maneuverability and easy operation. With the vigorous development of artificial intelligence technology, digital twin technology that uses digital methods to display the state of the physical world has emerged. Digital twin technology mainly presents and analyzes the performance of physical equipment in a virtual environment. It has the advantages of low R&D cost and short cycle. This technology has been developed and applied in many industries. Large-tonnage loaders usually have complex operating environments. Accurately and quickly predicting the performance of large-tonnage loaders and planning their operating trajectories in real time have become urgent issues to be solved. The use of digital twin technology to perform "identification-control" simulation on large-tonnage loaders can effectively reduce the system's operating energy consumption and improve its efficiency, which is of great significance to the development of intelligent and unmanned loaders. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the existing defects and provide a digital twin system for identifying and controlling large-tonnage loaders, which can realize the identification and perception of the surrounding environment of the large-tonnage loader, and can also enable the large-tonnage loader to adaptively calculate its own mechanical properties according to the current working environment, and plan the optimal loading trajectory under the current working conditions. It provides a scientific basis for the design, monitoring, performance prediction and full life cycle management of large-tonnage loaders, and can effectively solve the problems in the background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a digital twin system for identifying and controlling a large-tonnage loader, comprising an operating environment identification system, a data expansion system, a twin model control system and a data storage system, wherein the operating environment identification system mainly utilizes functional devices such as laser rangefinders, image sensors, and position sensors to identify the operating environment of large-tonnage devices, and screens, filters, and classifies the data obtained by identification to obtain useful operating environment data; the data expansion system is mainly used to fuse the currently obtained time series big data, operating object morphology big data, equipment morphology small sample data and historical operating data to form an initial data set that integrates multi-source information; the twin model control system uses high-fidelity technology to perform three-dimensional virtual mapping of the physical model of the large-tonnage loader to obtain a digital model consistent with the physical state performance; based on the initial data set that integrates multi-source information, an advanced swarm intelligent optimization algorithm is used to calculate and evaluate the physical and mechanical properties of the large-tonnage loader, and at the same time, the loading trajectory under the current environment is planned; the data storage system stores the obtained data and provides data basis for predicting and optimizing the twin model of subsequent large-tonnage loading as important historical data.

[0005] Furthermore, the working environment recognition system includes the sensitivity characteristics of laser rangefinder, image sensor, position sensor, lidar, and inertial navigation sensor unit, establishes a transmission model of measurement errors in complex environments such as strong vibration and strong wind and sand, filters and cleans the obtained data, and realizes rapid reconstruction of the complex environment of large-tonnage loaders based on the optimal spatial reasonable layout method and the superposition and fusion method of large-scale multi-view point cloud data.

[0006] Furthermore, the data expansion system adopts a logical judgment method to transform the data to address the problem that the data structures and types from different sources are significantly different, and divides the time series data into identification segments of different environments and working conditions to obtain a pre-processed initial data set with obvious characteristics and effectiveness.

[0007] Furthermore, the twin model control system includes the mapping construction of the twin model of a large-tonnage loader and the planning of the loading trajectory by an advanced swarm intelligence algorithm. The mapping construction of the twin model of a large-tonnage loader adopts an equivalent characterization model of non-uniform virtual materials with geometric or physical properties, and dynamically sets the properties of virtual material units based on measured data to establish a digital twin model of a large-tonnage loader that is dynamically consistent with the measured data. The advanced swarm intelligence algorithm plans the loading trajectory by using advanced swarm intelligence optimization methods to evaluate its mechanical properties, and obtains future data based on historical analysis data through machine learning algorithms such as long short-term memory neural networks, and optimizes it by building a digital model to obtain the optimal loading trajectory of the loader in the current state.

[0008] Furthermore, the storage system saves historical operation data, mechanical performance data of the current loader, and operation trajectory data, and provides data basis for predicting and optimizing the subsequent twin model of large-tonnage loading as important historical data.

[0009] Compared with the prior art, the present invention has the following beneficial effects:

[0010] The present invention can not only realize the recognition and perception of the working environment by the large-tonnage loader, but also enable the large-tonnage loader to adaptively calculate its own mechanical properties according to the current working environment. On this basis, the large-tonnage loader fully considers the characteristics of the working conditions and plans the optimal loading trajectory under the current working conditions, providing a scientific basis for the design, monitoring, performance prediction and full life cycle management of the large-tonnage loader. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a technical flow chart of the present invention;

[0012] Figure 2 This is the “identification-control” block diagram of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] Embodiment 1

[0015] See also Figure 1-2The present invention provides a technical solution: a large-tonnage loader identification and control digital twin system, including an operating environment identification system, a data expansion system, a twin model control system and a data storage system, wherein the operating environment identification system mainly uses functional devices such as laser rangefinders, image sensors, and position sensors to identify the operating environment of large-tonnage devices, and screens, filters, and classifies the data obtained by identification to obtain useful operating environment data. The operating environment identification system includes the sensitivity characteristics of laser rangefinders, image sensors, position sensors, laser radars, and inertial navigation sensor units, and establishes the transmission of measurement errors in complex environments such as strong vibrations and strong winds and sand. The model filters and cleans the acquired data. Based on the optimal spatial layout and the superposition and fusion method of large-scale multi-view point cloud data, the rapid reconstruction of the complex environment of large-tonnage loaders is achieved. The data expansion system is mainly used to fuse the currently acquired time series big data, operation object morphology big data, equipment morphology small sample data and historical operation data to form an initial data set that integrates multi-source information. The data expansion system uses logical judgment to transform the data for problems with significantly different data structures and types from different sources, and divides the time series data into identification and segmentation of different environments and working conditions to obtain a pre-processed initial data set with obvious characteristics and effective, twin model The control system uses high-fidelity technology to perform three-dimensional virtual mapping of the physical model of the large-tonnage loader to obtain a digital model consistent with the physical state performance. Based on the initial data set that integrates multi-source information, it uses advanced swarm intelligence optimization algorithms to calculate and evaluate the physical and mechanical properties of the large-tonnage loader, and at the same time plans the loading trajectory under the current environment. The twin model control system includes the mapping construction of the large-tonnage loader twin model and the planning of the loading trajectory by advanced swarm intelligence algorithms. The mapping construction of the large-tonnage loader twin model adopts an equivalent characterization model of non-uniform virtual materials with geometric or physical properties, and dynamically sets the properties of virtual material units based on measured data to establish The digital twin model of a large-tonnage loader is dynamically consistent with the measured data. The advanced swarm intelligence algorithm plans the loading trajectory by using advanced swarm intelligence optimization methods to evaluate its mechanical properties. Future data is obtained based on historical analysis data through machine learning algorithms such as long short-term memory neural networks. The digital model is constructed to optimize it and obtain the optimal loading trajectory of the loader in the current state. The data storage system stores the obtained data. The data storage system saves historical operation data, the mechanical performance data of the current loader, and the operation trajectory data, and provides data basis for the prediction and optimization of the subsequent large-tonnage loading twin model as important historical data.

[0016] like Figure 1The figure shows the operating process of the digital twin system of a large-tonnage loader, in which the physical layer is the identification and data collection of the operating conditions of the large-tonnage loader, and the acquired data is filtered and cleaned to obtain data useful for subsequent loading operation trajectory planning; the analysis layer is mainly for the fusion of the acquired data, and on this basis, the relevant mechanical performance analysis, calculation and prediction of the large-tonnage loader are performed; the virtual layer is to digitally equivalently model the physical model to obtain a twin model consistent with the state performance of the physical model, and combine the fused data to plan the loading trajectory for the current operating condition to obtain the optimal operating trajectory parameters.

[0017] Real-time control process based on image recognition technology Figure 2 As shown, a variety of different sensor devices are used to detect the operations around the large-tonnage loader, obtain the surrounding environment data of the large-tonnage loader, and clean and reduce the noise of the obtained data; the current data is expanded in combination with the historical data of the loader to realize the calculation and predictive analysis of the mechanical properties of the loader; based on the expanded data, a digital twin model consistent with the physical state information of the loader is constructed to achieve the intrinsic structure performance twin mapping consistent with the physical equipment entity model, and on this basis, real-time loading trajectory planning and performance evaluation of the large-tonnage loader are carried out, and the obtained data is archived to provide data for subsequent twin performance analysis and model correction.

[0018] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital twin system for identification and control of a large-tonnage loader, including an operating environment identification system, a data expansion system, a twin model control system, and a data storage system. Features: The working environment recognition system uses laser rangefinders, image sensors, and position sensor functional devices to identify the working environment of large-tonnage equipment, screen, filter, and classify the data obtained from the recognition to obtain useful working environment data. The data expansion system is used to fuse the currently obtained time series big data, working object morphology big data, equipment morphology small sample data, and historical working data to form an initial data set that integrates multi-source information. The twin model control system uses high-fidelity technology to perform three-dimensional virtual mapping of the physical model of the large-tonnage loader to obtain a digital model consistent with the physical state performance. Based on the initial data set that integrates multi-source information, the advanced swarm intelligence optimization algorithm is used to calculate and evaluate the physical and mechanical properties of the large-tonnage loader, and at the same time plan the loading trajectory under the current environment. The data storage system stores the obtained data and uses it as important historical data to provide data basis for predicting and optimizing the twin model of subsequent large-tonnage loading; the working environment recognition system includes laser rangefinders, image sensors, and position sensors. The sensitivity characteristics of sensors, lidars, and inertial navigation sensor units are used to establish a transmission model of measurement errors in complex environments with strong vibrations and strong winds and sandstorms. The obtained data is filtered and cleaned, and based on the optimal spatial reasonable layout method and the superposition and fusion method of large-scale multi-view point cloud data, the complex environment of large-tonnage loaders can be quickly reconstructed; the twin model control system includes the mapping construction of the twin model of large-tonnage loader and the planning of loading trajectories by advanced swarm intelligence algorithms. The mapping construction of the twin model of large-tonnage loader adopts an equivalent characterization model of non-uniform virtual materials with geometric or physical properties, and dynamically sets the properties of virtual material units based on measured data to establish a large-tonnage loader digital twin model that is dynamically consistent with the measured data. The advanced swarm intelligence algorithm plans the loading trajectory by using advanced swarm intelligence optimization methods to evaluate its mechanical properties, and obtains future data based on historical analysis data through a long short-term memory neural network machine learning algorithm, and optimizes it by building a digital model to obtain the optimal loading trajectory of the loader in the current state.

2. According to claim 1, a large-tonnage loader identification and control digital twin system, Features: The data expansion system adopts a logical judgment method to transform the data to solve the problem that the data structures and types from different sources are significantly different, and divides the time series data into identification segments of different environments and working conditions to obtain a pre-processed initial data set with obvious characteristics and effectiveness.

3. According to claim 1, a large-tonnage loader identification and control digital twin system, Features: The storage system saves historical operation data, mechanical performance data of the current loader, and operation trajectory data, and provides data basis for prediction and optimization of the subsequent twin model of large-tonnage loading as important historical data.