CN-121982191-A - Self-adaptive material taking method and system for stacker-reclaimer based on dynamic boundary model

CN121982191ACN 121982191 ACN121982191 ACN 121982191ACN-121982191-A

Abstract

The invention discloses a stacker-reclaimer self-adaptive material taking method and system based on a dynamic boundary model. The method comprises the steps of S1, dynamic boundary modeling, S2, safety track planning, S3, load self-adaptive adjustment, wherein the steps S2 and S3 are executed in parallel and are mutually coupled to form a double-constraint control loop. The system comprises a sensing module, a control module and an execution module. The invention realizes the spanning from static preset to dynamic perception and from passive response to active safety, effectively solves the problems of uneven material taking, high collision risk and low efficiency, and remarkably improves the operation safety, efficiency and intelligentization level.

Inventors

WANG ZHENXIANG
SHI JING
JIA LANHUI
FAN PAN
WANG HAILAN
WANG JIANQING

Assignees

中交机电工程局有限公司

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (8)

1. The self-adaptive material taking method of the stacker-reclaimer based on the dynamic boundary model is characterized by comprising the following steps of: S1, modeling a dynamic boundary: Three-dimensional point cloud data of a material taking working face are obtained in real time through three-dimensional scanning equipment arranged at the front end of a cantilever of a stacker-reclaimer; Carrying out point cloud registration and three-dimensional differential calculation on the three-dimensional point cloud data and a pre-stored reference model, identifying a difference region between the actual contour of the material pile and the model contour, and updating according to the difference region to generate a dynamic effective material taking boundary model reflecting the current actual working condition; S2, safety track planning: Based on an effective material taking boundary model, calculating a dynamic safety distance between a material taking head and a material pile boundary by combining prestored material repose angle parameters, and planning or adjusting a movement track of the material taking head in real time to ensure that material taking operation is always positioned in the safety boundary; s3, load self-adaptive adjustment: Monitoring the load current of the material taking driving motor in real time; the load current is used as feedback, and the pitching angle of the material taking head and/or the advancing speed of the whole machine are dynamically adjusted through the self-adaptive PID controller, so that the material taking power is stabilized within a preset target range; Wherein, the steps S2 and S3 are executed in parallel and are mutually coupled to form a double-constraint control loop, and the control logic of the loop is as follows: A first priority that when the step S2 judges that collision or collapse risk exists, the output track correction instruction will cover the load adjustment instruction of the step S3; And the second priority is that when the risk is not existed, the expected track output in the step S2 and the optimized load parameter output in the step S3 are coupled and resolved by a central processing unit to generate a final control instruction to drive the actuating mechanism to act.
2. The adaptive material taking method of stacker-reclaimer based on dynamic boundary model of claim 1, further comprising S4, iterative learning and prediction: Recording a dynamic boundary model sequence, load current data and corresponding operation parameters in a plurality of continuous operation periods; Training a boundary evolution prediction model through a recurrent neural network based on the recorded data, and predicting the evolution trend of the material pile boundary in the next working period; and taking the predicted boundary trend as priori information, and applying the priori information to the initial path planning of the next working period to realize prospective control.
3. The adaptive material taking method of a stacker-reclaimer based on a dynamic boundary model according to claim 2, wherein in S4, a boundary evolution prediction model is trained by a recurrent neural network, specifically, a long-short-term memory network is utilized to perform time sequence modeling on a dynamic boundary model sequence so as to predict a material pile profile of a specific time step in the future.
4. The adaptive material taking method of a stacker-reclaimer based on a dynamic boundary model of claim 1, wherein in S1, the three-dimensional scanning device is a laser radar and a three-dimensional stereo camera.
5. The adaptive reclaimer method of claim 1, wherein the reference model is a three-dimensional model of a stockpile built from an initial scan or a morphological model of a stockpile predicted from historical operational data.
6. The adaptive material taking method of a stacker-reclaimer based on a dynamic boundary model as set forth in claim 1, wherein in S2, the dynamic safety distance is calculated by calculating a minimum safety operation distance for preventing collapse of a material pile according to a slope angle of an effective material taking boundary model and a repose angle of the material, and using the distance as a rigid constraint condition for track planning.
7. The adaptive material taking method of a stacker-reclaimer based on a dynamic boundary model as defined in claim 1, wherein in S3, the operation parameters of the material taking head are dynamically adjusted, specifically including decoupling control of a pitching mechanism, a slewing mechanism and a complete machine running mechanism of the material taking arm; the load adaptive regulating loop is configured to independently and cooperatively adjust pitch angle, swing speed, and travel speed according to changes in load current so that the reclaimer system maintains stability of movement of each mechanism while maintaining constant or optimal power.
8. A stacker-reclaimer adaptive reclaiming system based on a dynamic boundary model for executing the stacker-reclaimer adaptive reclaiming method based on the dynamic boundary model as set forth in any one of claims 1-7, comprising: A perception module, comprising: the three-dimensional scanning unit comprises a laser radar and a three-dimensional camera which are arranged at one end of a stacker-reclaimer cantilever, which is close to a material taking head, and is used for scanning the material taking working surface to obtain complete material pile contour point cloud data; the load detection unit comprises a current sensor and a power transmitter which are arranged in a power loop of the material taking driving motor and is used for directly measuring the real-time load of the material taking driving motor; A pose sensing unit comprising: the inclination angle sensor is arranged on a cantilever of the stacker-reclaimer and is used for directly measuring the pitching angle of the cantilever; The rotary encoder is arranged in the rotation driving mechanism of the cantilever and is used for detecting the rotation angle of the cantilever; the high-precision positioning module is used for detecting the position of the whole machine and comprises a Gray bus paved along the whole course of a stacker-reclaimer track and an address detector arranged on a traveling mechanism of the stacker-reclaimer; the control module is connected with each unit of the sensing module through an industrial bus in a signal way, and the hardware carrier of the control module is an industrial computer and comprises: The core processing unit is used as a system operation core; The dynamic boundary modeling unit is operated on the core processing unit and is configured to execute point cloud registration and three-dimensional differential calculation so as to complete the dynamic boundary modeling step; the safety track planning unit is operated on the core processing unit and is configured to execute dynamic safety distance calculation and track planning so as to complete the safety track planning step; The load self-adaptive adjusting unit is operated on the core processing unit and is configured to operate a self-adaptive PID control algorithm so as to complete the load self-adaptive adjusting step; The decision fusion unit is operated on the core processing unit and is configured to receive the outputs of the safety track planning unit and the load self-adaptive adjusting unit, and carry out priority judgment and instruction coupling calculation according to the logic of the dual constraint control loop to generate a final control instruction; The digital twin unit is operated in the industrial computer or an upper computer communicated with the industrial computer, and is used for constructing a mirror model synchronous with the states of the physical stock ground and the equipment in a virtual space based on real-time data, and carrying out simulation verification and optimization on a control instruction; the execution module is in signal connection with the control module and is used for receiving and executing control instructions, and the execution module comprises: The pitching driving mechanism is used for controlling the pitching angle of the cantilever; The rotation driving mechanism is used for controlling the rotation angle of the cantilever; The walking driving mechanism is used for controlling the whole machine to travel along the track; And the material taking driving mechanism is used for driving the material taking head to rotate.

Description

Self-adaptive material taking method and system for stacker-reclaimer based on dynamic boundary model Technical Field The invention relates to the technical field of intelligent control of bulk cargo loading and unloading equipment, in particular to a self-adaptive material taking method and system of a stacker-reclaimer based on a dynamic boundary model. Background Stacker-reclaimers are key devices for stacking and reclaiming in large bulk yards. The automation and intelligence levels are directly related to the efficiency and safety of the whole bulk material treatment system. The existing flow control method of stacker-reclaimers, such as "a flow control method and device of stacker-reclaimers and stacker-reclaimers (CN 110980318B)" of the chinese patent, is to adjust the rotation speed of the bucket wheel or the conveying arm by comparing the obtained real-time flow value with a fixed preset value. Although the method realizes stable control of the flow to a certain extent, the method still has obvious defects: firstly, the control is based on a fixed preset flow value, so that the sudden change of the material taking resistance caused by material pile collapse, landslide, material granularity and humidity change in the material taking process cannot be adapted, and the control strategy is stiff; Secondly, the control is passive response type control, the control is only carried out after the flow deviation occurs, response lag exists, and system oscillation or overshoot is easy to cause; Finally, the sensing dimension is single, only parameters such as flow or motor current are relied on, and real-time sensing and understanding of the three-dimensional form of the working object, namely the material pile, are lacked, so that the control system cannot see real environment change, and the safety and efficiency problems such as material pile gnawing (collapse caused by impacting the root of the material pile) or empty digging (material taking head does not contact materials) cannot be fundamentally solved. In addition, another type of automated reclaiming scheme based on a preset fixed path (e.g., trapezoidal, layered) relies entirely on an initial, static stockpile model. Once the actual stockpile shape is inconsistent with the model due to natural slump or repeated material taking operation, the automatic process fails, manual intervention is still required, and the degree of automation and the reliability are not ideal. Therefore, there is a strong need in the art for an intelligent material taking scheme capable of "sensing-modeling-decision-making" closed-loop linkage, which is characterized in that the dynamic change of the material pile morphology can be sensed and predicted in real time, and the material taking strategy can be adjusted autonomously, safely and efficiently accordingly. Disclosure of Invention The invention aims to overcome the defects in the prior art and provides a stacker-reclaimer self-adaptive material taking method and system based on a dynamic boundary model. According to the invention, the dynamic boundary model is built in real time, and double constraint control of safety track planning and load self-adaptive adjustment is realized on the basis, so that the purposes of safety, high efficiency and self-adaptive material taking are finally achieved. The invention adopts the following technical scheme to realize the aim: the adaptive material taking method of the stacker-reclaimer based on the dynamic boundary model comprises the following steps: S1, modeling a dynamic boundary: Three-dimensional point cloud data of a material taking working face are obtained in real time through three-dimensional scanning equipment arranged at the front end of a cantilever of a stacker-reclaimer; Carrying out point cloud registration and three-dimensional differential calculation on the three-dimensional point cloud data and a pre-stored reference model, identifying a difference region between the actual contour of the material pile and the model contour, and updating according to the difference region to generate a dynamic effective material taking boundary model reflecting the current actual working condition; S2, safety track planning: Based on an effective material taking boundary model, calculating a dynamic safety distance between a material taking head and a material pile boundary by combining prestored material repose angle parameters, and planning or adjusting a movement track of the material taking head in real time to ensure that material taking operation is always positioned in the safety boundary; s3, load self-adaptive adjustment: Monitoring the load current of the material taking driving motor in real time; the load current is used as feedback, and the pitching angle of the material taking head and/or the advancing speed of the whole machine are dynamically adjusted through the self-adaptive PID controller, so that the material taking power is stabilized within a preset targe