Intelligent control system of anode carbon block automatic stereoscopic warehouse

Abstract

The invention belongs to the technical field of stereoscopic warehouses, and particularly relates to an intelligent control system of an automatic stereoscopic warehouse for anode carbon blocks, which comprises the following steps: a machine vision perception module for acquiring storage location state, carbon block position, transportation equipment position and personnel position information in real time through an industrial camera, and generating three-dimensional space coordinate data; the multi-sensor fusion positioning module integrates a laser range finder, an encoder and an RFID tag and performs millimeter-level positioning on a stacking machine, a conveyor and a carbon block, and the error range is smaller than or equal to + / -2mm; the intelligent scheduling engine is based on a multi-device and multi-task coordination model; according to the method, high-density point cloud data are generated by adopting Intel RealSense D455 on the top surface, two Baser blaze-101 are deployed in a side-looking mode, the top surface data are fused through an ICP algorithm, the complete contour of the carbon block is constructed, the position of a person is monitored in combination with a wide-angle camera, millisecond-level accurate recognition of the position and posture of the carbon block in the stacking process is achieved, stacking failure caused by inclination is avoided, and meanwhile the safety of the person is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of automated warehouse technology, and in particular to an intelligent control system for an automated warehouse for anode carbon blocks. Background Technology

[0002] In the electrolytic aluminum industry, anode carbon blocks are a core production raw material, and their storage management directly impacts production continuity and cost control. Traditional anode carbon block warehouses commonly suffer from the following technical problems, severely hindering the industry's intelligent upgrading: Traditional stacker cranes rely on manual positioning and operation. In scenarios with racks over 12 meters high, the stacking error of carbon blocks often exceeds ±5mm, leading to uneven load distribution on the racks and a risk of collapse. The empty-run rate is as high as 45%, with a daily throughput of less than 200 tons. Furthermore, manual scheduling is susceptible to fatigue and lack of experience, making it impossible to meet the 24-hour continuous feeding requirements of electrolytic cells. A single layer of carbon blocks can bear a load of up to 5 tons; improper manual clamping force control easily causes breakage, with a breakage rate as high as 3%, resulting in direct economic losses exceeding one million yuan annually. Electrolytic aluminum companies need to invest a large amount of manpower in repetitive handling operations, and the efficiency of manual operation fluctuates greatly, making it difficult to achieve large-scale, standardized management.

[0003] Traditional warehouses lack multi-level safety protection, relying only on basic anti-collision strips and limit switches, with a response time exceeding 1 second, making them unable to cope with sudden collisions when stacker cranes are running at high speeds (≥120m / min). Key components (such as drive chains and motors) lack condition monitoring, and fault prediction relies on regular inspections. Sudden failures can lead to unplanned downtime of up to 8 hours per incident, severely impacting production rhythm. The safety level only meets the ISO 13849-1 PLC standard, which cannot effectively prevent equipment from falling or goods from tipping over under extreme conditions such as fork overload or chain breakage. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes an intelligent control system for an automated three-dimensional warehouse for anode carbon blocks, which more accurately solves the problems mentioned in the background art.

[0005] This invention is achieved through the following technical solution: This invention proposes an intelligent control system for an automated three-dimensional warehouse for anode carbon blocks, comprising the following steps: a machine vision perception module: real-time acquisition of warehouse location status, carbon block position, transport equipment position, and personnel position information via industrial cameras, generating three-dimensional spatial coordinate data; a multi-sensor fusion positioning module: integrating a laser rangefinder, encoder, and RFID tag to perform millimeter-level positioning of the stacker crane, conveyor, and carbon blocks, with an error range ≤ ±2mm; an intelligent scheduling engine: based on a multi-device, multi-task coordination model, dynamically allocating work tasks to the stacker crane, AGV, and conveyor system, optimizing path planning, and reducing equipment idle time by ≥30%; and three-dimensional modeling. Model and Quality Inspection Module: Constructs a 3D model of the charcoal block using at least three sets of line lasers and area array cameras, compares it with a standard model, and automatically identifies defects such as cracks and deformations with an accuracy rate of ≥99.5%; Adaptive Stacking Control Module: Dynamically adjusts the fork clamping force and stacking height according to the size of the charcoal block, supporting a maximum stacking height of 12m and a single-layer load capacity of ≥5 tons; Safety Collision Avoidance System: Monitors the equipment operating area in real time using ultrasonic sensors and lidar, triggering an emergency braking response time of ≤0.2 seconds; Dynamic Inventory Management Module: Updates the charcoal block entry and exit time, location, and batch information in real time, generates visual reports, and supports inventory turnover rate analysis; 5G wireless communication module: Enables low-latency data interaction between machine vision system, material handling equipment and scheduling management system, with communication latency ≤10ms; Fault self-diagnosis module: Monitors equipment operating parameters, predicts the lifespan of key components such as motors and reducers, and provides early warning of potential faults up to 72 hours in advance; Remote operation and maintenance module: Enables centralized monitoring of multiple warehouses through cloud platform, supports remote parameter adjustment and firmware upgrade, and reduces on-site maintenance frequency by ≥50%.

[0006] Preferably, the machine vision perception module includes a top depth camera and a side depth camera, which are installed above and to the side of the stacker crane, respectively, to simultaneously acquire image data of the top and side surfaces of the carbon block and construct a complete three-dimensional contour.

[0007] Preferably, the intelligent scheduling engine adopts a multi-objective optimization model based on genetic algorithms, with operation efficiency, energy consumption and equipment wear as optimization objectives, to generate the optimal task sequence.

[0008] Preferably, in the 3D modeling and quality inspection module, the line laser scanning frequency is ≥10kHz, the area array camera resolution is ≥12 million pixels, and it supports real-time modeling of dynamic objects.

[0009] Preferably, the adaptive stacking control module includes a cylinder lifting adjustment mechanism and an electric push rod lateral adjustment mechanism, with the cylinder stroke range being 0-500mm and the electric push rod displacement accuracy being ±0.1mm.

[0010] Preferably, the safety anti-collision system further includes terminal limit protection, chain breakage protection, and cargo rope breakage protection devices, and the power source of the equipment is immediately cut off when any of the protection devices is triggered.

[0011] Preferably, the inventory dynamic management module supports seamless integration with the ERP system, automatically synchronizing production plans and purchase orders, and realizing early warning of upper and lower inventory limits and automatic replenishment.

[0012] Preferably, the 5G wireless communication module adopts the Time-Sensitive Networking (TSN) protocol to ensure deterministic transmission of critical control commands.

[0013] Preferably, the fault self-diagnosis module collects equipment operation data through vibration sensors, temperature sensors, and current sensors, and predicts the remaining service life based on an LSTM neural network model.

[0014] Preferably, the remote operation and maintenance module supports VR / AR remote assistance, allowing experts to mark equipment fault points through a virtual interface and guide on-site personnel to complete maintenance operations.

[0015] Compared with the prior art, the present invention provides an intelligent control system for an automated three-dimensional warehouse for anode carbon blocks, which has the following beneficial effects: The intelligent control system of this automated storage and retrieval system for anode carbon blocks generates high-density point cloud data on the top surface using Intel RealSense D455, and deploys two Basler blaze-101 cameras on the side to fuse the top surface data using the ICP algorithm to construct the complete outline of the carbon blocks. Combined with wide-angle cameras to monitor personnel positions, it achieves millisecond-level accurate identification of the position and posture of the carbon blocks during the stacking process, avoiding stacking failures caused by tilting and ensuring personnel safety.

[0016] The intelligent control system of this automated storage and retrieval system for anode carbon blocks encodes task sequences into integer arrays and constructs a fitness function with energy consumption (E), time (T), and equipment wear index (W) as constraints. This reduces the idle rate of equipment and the energy consumption per unit, while also increasing the daily workload. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the intelligent control system for an automated three-dimensional warehouse for anode carbon blocks proposed in this invention. Detailed Implementation

[0018] To more clearly and completely illustrate the technical solution of the present invention, the present invention will be further described below with reference to the accompanying drawings. Example

[0019] like Figure 1As shown in the figure, an embodiment of the present invention proposes an intelligent control system for an automated three-dimensional warehouse for anode carbon blocks. The system includes a machine vision perception module: an industrial camera (resolution ≥ 5 megapixels) is installed on top of the stacker crane, generating 3D coordinate data of the storage location using a binocular vision algorithm, with a positioning accuracy of ±1mm; a wide-angle camera is installed on the side to monitor personnel positions, triggering a safety zone alarm with a response time ≤ 0.5 seconds. A multi-sensor fusion positioning module is also included: a laser rangefinder (accuracy ±0.1mm) is installed on the stacker crane track; a fork encoder with a resolution of 0.01° is used; and RFID tags are embedded in the carbon block pallets, achieving a 3D positioning error ≤ ±1.5mm. An intelligent scheduling engine is also included: based on a genetic algorithm model, with constraints of operating efficiency (target value ≥ 95%), energy consumption (target value ≤ 8kWh / ton), and equipment wear rate (target value ≤ 5%), the generated task sequence reduces the equipment idle rate by 35%. 3D Modeling and Quality Inspection Module: Employing 3 sets of line lasers (12kHz scanning frequency) and a 12-megapixel area array camera, the carbon block model is constructed in ≤2 seconds, with a crack detection sensitivity of 0.1mm and a defect identification accuracy of 99.8%. Adaptive Stacking Control Module: A cylinder lifting mechanism (500mm stroke) automatically adjusts according to the carbon block height, and an electric push rod (±0.05mm accuracy) dynamically adjusts the clamping force, achieving a 100% pass rate in the 12m stacking stability test. Safety Collision Avoidance System: Combined monitoring by ultrasonic sensors (detection distance 0.1-5m) and lidar (angular resolution 0.1°), emergency braking distance is ≤0.3m, reducing collision risk by 90%. Inventory Dynamic Management Module: Inbound / outbound data update delay is ≤0.1 seconds, the error rate of generated inventory turnover rate reports is ≤1%, and the success rate of data synchronization with the SAP ERP system is ≥99.9%. 5G Wireless Communication Module: Employing the TSN protocol, the data transmission latency between the machine vision and scheduling systems is ≤8ms, with a critical command packet loss rate of 0.005%, ensuring stacker crane synchronization control accuracy of ±2mm. Fault Self-Diagnosis Module: Data is collected through vibration sensors (sampling frequency 10kHz), temperature sensors (accuracy ±0.5℃), and current sensors (accuracy ±0.1A). The LSTM model predicts motor lifespan with an error of ≤5%, and provides a 92% accuracy rate for 72-hour early warning. Remote Maintenance Module: The cloud platform supports simultaneous access from 100 warehouses, VR remote assistance latency is ≤200ms, expert-annotated fault location accuracy is ±10cm, and on-site repair efficiency is improved by 40%.

[0020] In this invention, the camera layout optimization method for the machine vision perception module includes: a top-view depth camera (Intel RealSense D455, installed in the center of the stacker crane beam, covering the entire shelf layer, generating point cloud data with a density ≥100 points / cm²); and two side-view depth cameras (two Basler blaze-101, positioned at 45° angles on either side of the fork, simultaneously acquiring side images, fusing the top-view data using the ICP algorithm to construct a complete charcoal block outline with an error ≤±0.8mm). This achieves a charcoal block posture recognition accuracy of 99.9%, avoiding stacking failures caused by tilting.

[0021] In this invention, the genetic algorithm optimization method of the intelligent scheduling engine includes chromosome encoding: encoding the task sequence as an integer array (e.g., [3,1,2] represents the execution order of task 3→1→2); fitness function:

[0022] Where E is energy consumption, T is time, and W is the equipment wear index; after 100 iterations, it converges, and the optimal task sequence increases the daily workload by 28% and reduces energy consumption by 19%.

[0023] In this invention, the hardware configuration of the 3D modeling and quality inspection module includes: a line laser: a Keyence LK-H052 with an output power of 50mW and a scanning line width of 10mm, which, together with a high-speed galvanometer, achieves a scanning frequency of 12kHz; an area scan camera: a FLIRBlackfly S BFS-U3-12S4C with a global shutter and a frame rate of 120fps, covering a 1m×1m area at a working distance of 0.5m; and a dynamic modeling frame rate of ≥15fps, meeting the real-time inspection requirements when the conveyor speed is 2m / s.

[0024] In this invention, the adaptive stacking control module's adjustment mechanism design includes a cylinder lifting mechanism: SMCCQ2B20-20DM, with a stroke of 500mm and a load capacity of 2000N, achieving stepless speed regulation via a proportional valve; and an electric push rod: LINAKLA37, with a displacement accuracy of ±0.05mm, a maximum thrust of 5000N, and integrated force sensor feedback of clamping force. This reduces the carbon block breakage rate from 3% to 0.2%, and the stacking layer error is ≤1 layer.

[0025] In this invention, the safety anti-collision system employs multiple protection mechanisms: terminal limit protection: Omron E6B2-CWZ6C encoders are installed at both ends of the stacker crane track, triggering a PLC emergency stop command when the travel exceeds the limit; chain breakage protection: Hall sensors are deployed next to the transmission chain to monitor changes in chain pitch, cutting off the motor power within 0.1 seconds when a chain breaks; and cargo rope breakage protection: tension sensors are installed at the four corners of the charcoal block lifting device, triggering the anti-fall device when the single rope tension decreases by 30%. The system has operated continuously for 1000 hours without any collision accidents, meeting ISO 13849-1 PLd safety level.

[0026] In this invention, the ERP integration method for the dynamic inventory management module includes the following data interface: communication with the SAP ERP system via the OPC UA protocol, transmitting inbound and outbound records in JSON format (fields include material number, batch, quantity, and time); automatic replenishment logic: when the inventory level is below the safety threshold, the system generates a purchase order and pushes it to the ERP approval process; inventory data synchronization delay is ≤0.3 seconds, and the number of stockouts is reduced by 85%.

[0027] In this invention, the TSN configuration of the 5G wireless communication module includes time synchronization: the IEEE 802.1AS protocol is used to achieve inter-device clock synchronization with a deviation of ≤1μs; traffic scheduling: dedicated time slots are allocated for critical control commands (such as reserving a 2ms transmission window every 10ms) to ensure deterministic delay; the stacker crane synchronization control error is reduced from ±10mm to ±1.8mm, meeting the requirements for high-precision stacking.

[0028] In this invention, the LSTM model training for the fault self-diagnosis module involves data acquisition: deploying a three-dimensional vibration sensor (acceleration range ±50g) on ​​the motor bearing housing, sampling at a frequency of 10kHz, and continuously collecting operational data for 3 months; feature engineering: extracting time-domain features (root mean square, peak value) and frequency-domain features (spectral centroid, frequency band energy); model training: constructing a two-layer LSTM network (64 hidden layer dimensions) using PyTorch, and training it for 200 epochs on an NVIDIA Tesla T4 GPU; the remaining service life prediction error is ≤8%, and the fault warning accuracy is 94%.

[0029] In this invention, the VR / AR assistance process of the remote operation and maintenance module includes: equipment modeling: constructing a digital twin model of the warehouse using Unity 3D with an accuracy of ≤2cm; AR annotation: experts use HoloLens 2 devices to annotate fault points on the virtual model via voice commands (such as "motor X-axis vibration exceeds the standard"); on-site synchronization: on-site personnel receive the annotation information via iPad Pro and use AR navigation to locate the faulty equipment; the average troubleshooting time is reduced from 120 minutes to 45 minutes, and the maintenance error rate is reduced by 70%.

[0030] Finally, it should be noted that the basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification, and therefore remain within the spirit and scope of the exemplary embodiments of this specification. Furthermore, this specification uses specific terms to describe embodiments of this specification. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined. Moreover, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods of this specification.

[0031] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent control system for an automated three-dimensional warehouse for anode carbon blocks, comprising the following steps: Machine vision perception module: Real-time data collection of storage location status, charcoal block location, transportation equipment location, and personnel location information using industrial cameras, generating three-dimensional spatial coordinate data; Multi-sensor fusion positioning module: integrates laser rangefinder, encoder and RFID tag to perform millimeter-level positioning of stacker crane, conveyor and charcoal block, with an error range of ≤±2mm; Intelligent scheduling engine: Based on a multi-device, multi-task coordination model, it dynamically allocates work tasks to stacker cranes, AGVs, and conveyor systems, optimizes path planning, and reduces equipment idle time by ≥30%; 3D modeling and quality inspection module: Utilizes at least three sets of line lasers and area array cameras to construct a 3D model of the carbon block, compares it with a standard model, and automatically identifies defects such as cracks and deformations, with an accuracy rate of ≥99.5%. Adaptive stacking control module: dynamically adjusts the fork clamping force and stacking height according to the size of the charcoal blocks, supports a maximum stacking height of 12m, and a single-layer load capacity of ≥5 tons; Safety collision avoidance system: Real-time monitoring of the equipment's operating area via ultrasonic sensors and lidar, triggering emergency braking response time ≤0.2 seconds; The inventory dynamic management module updates the time, location, and batch information of charcoal blocks entering and leaving the warehouse in real time, generates visual reports, and supports inventory turnover rate analysis. 5G wireless communication module: Enables low-latency data interaction for machine vision systems, material handling equipment, and scheduling management systems, with a communication latency of ≤10ms; Fault self-diagnosis module: Monitors equipment operating parameters, predicts the lifespan of key components such as motors and reducers, and provides early warning of potential faults up to 72 hours in advance; Remote operation and maintenance module: Enables centralized monitoring of multiple warehouses through the cloud platform, supports remote parameter adjustment and firmware upgrade, and reduces the frequency of on-site maintenance by ≥50%.

2. The intelligent control system according to claim 1, characterized in that, The machine vision perception module includes a top depth camera and a side depth camera, which are installed above and to the side of the stacker crane, respectively, to simultaneously acquire image data of the top and side surfaces of the carbon blocks and construct a complete three-dimensional contour.

3. The intelligent control system according to claim 1, characterized in that, The intelligent scheduling engine adopts a multi-objective optimization model based on genetic algorithms, with operation efficiency, energy consumption and equipment wear as optimization objectives, to generate the optimal task sequence.

4. The intelligent control system according to claim 1, characterized in that, In the 3D modeling and quality inspection module, the line laser scanning frequency is ≥10kHz, the area array camera resolution is ≥12 million pixels, and it supports real-time modeling of dynamic objects.

5. The intelligent control system according to claim 1, characterized in that, The adaptive stacking control module includes a cylinder lifting adjustment mechanism and an electric push rod lateral adjustment mechanism. The cylinder stroke range is 0-500mm, and the electric push rod displacement accuracy is ±0.1mm.

6. The intelligent control system according to claim 1, characterized in that, The safety collision avoidance system also includes terminal limit protection, chain breakage protection, and cargo rope breakage protection devices. When any of these protection devices is triggered, the equipment power source is immediately cut off.

7. The intelligent control system according to claim 1, characterized in that, The dynamic inventory management module supports seamless integration with the ERP system, automatically synchronizing production plans and purchase orders, and enabling early warning of upper and lower inventory limits and automatic replenishment.

8. The intelligent control system according to claim 1, characterized in that, The 5G wireless communication module adopts the Time-Sensitive Networking (TSN) protocol to ensure deterministic transmission of critical control commands.

9. The intelligent control system according to claim 1, characterized in that, The fault self-diagnosis module collects equipment operation data through vibration sensors, temperature sensors, and current sensors, and predicts the remaining service life based on an LSTM neural network model.

10. The intelligent control system according to claim 1, characterized in that, The remote operation and maintenance module supports VR / AR remote assistance, allowing experts to mark equipment fault points through a virtual interface and guide on-site personnel to complete maintenance operations.

Images