Multi-species coke mixing, stacking and taking system

Abstract

The present application relates to the technical field of coke transportation, and provides a multi-variety coke mixing, stacking and taking system, which comprises a classified stacking module, a stock management module, a stacking and taking execution module and a safe intelligent guarantee module of a transport vehicle. Through the arrangement of the classified stacking module, the stock management module, the stacking and taking execution module and the safe intelligent guarantee module of the transport vehicle, in combination with the stacking area division of the top-charging coke oven coke storage area silo and the stamp-charging coke oven coke storage area silo, and the layered design of the storage space of the upper high-value silo, the middle high-frequency silo and the lower high-substitute silo, multi-dimensional classified stacking of coke based on process source data, quality index data, cost data and production demand data is realized, and the problems of single classification dimension and low space utilization rate in the prior art are solved; the coke dynamic stock map of the stock management module can realize real-time synchronization of the physical warehouse state, so that the operator can intuitively master the position and stock of each batch of coke.

Description

Technical Field

[0001] This invention belongs to the field of coke transportation technology, specifically relating to a multi-variety coke blending, storage, and retrieving system. Background Technology

[0002] The steel production process demands diverse quality requirements for coke. Different ironmaking processes and product quality objectives necessitate coke with varying quality characteristics. Currently, coke producers typically classify coke based on quality indicators such as shatter resistance, abrasion resistance, reactivity, post-reaction strength, ash content, and sulfur content, and store it in designated areas within the stockyard. While some large steel companies have established coke quality databases and employ information management systems to record coke inflows and outflows, these existing technological solutions suffer from the following significant drawbacks:

[0003] (1) Single classification dimension and lack of comprehensive value assessment: The existing classification method is mainly based on a single dimension such as the quality indicators or production process of coke. It fails to comprehensively consider multiple factors such as coke cost data and production demand data (the total amount of coke to be used in the future within the preset period). It is impossible to conduct dynamic value assessment of coke and to achieve precise control and minimization of procurement and usage costs under the premise of meeting production quality requirements. At the same time, due to the lack of prediction of future demand, urgently needed coke may be piled up in a location that is not easy to access (for the access of coke, the higher the storage position, the easier it is to access the material under the action of gravity, and the higher the access efficiency; the lower the storage position, the less the effect of gravity, the greater the difficulty of access, and the longer it takes).

[0004] (2) Extensive storage planning and low space utilization: Existing technologies generally adopt a combination of flat storage and simple zoning, resulting in low efficiency of warehouse space utilization and requiring a huge amount of land resources for the same storage scale.

[0005] (3) Outdated inventory information management: The existing inventory management mainly adopts manual inspection combined with table recording. The information is outdated and seriously out of touch with the real-time status of the physical warehouse. Operators cannot intuitively and quickly grasp the precise location, quantity and status of each batch of coke. The decision on stacking and material retrieval depends on manual experience and on-site search.

[0006] (4) Low material picking and mixing efficiency and lack of intelligent optimization: When production requires the mixing of multiple cokes, the existing technology relies on manual experience to determine the material picking scheme and mixing ratio. It lacks an intelligent mixing strategy based on cost optimization and quality constraints, resulting in high mixing cost, low efficiency, and difficulty in ensuring the stability of coke quality after mixing.

[0007] (5) High transportation safety risks and lack of proactive protection: When transporting coke, the transport vehicle faces a variety of safety risks such as overturning, overheating of the compartment, and dust explosion. Existing technology lacks real-time monitoring and risk warning of the operating status of the transport vehicle, and is unable to carry out adaptive control, resulting in prominent safety hazards. Summary of the Invention

[0008] This invention provides a multi-variety coke blending, storage, and reclaiming system to solve at least one of the aforementioned technical problems. This invention is achieved through the following technical solution:

[0009] A multi-variety coke blending, storage, and reclaiming system includes:

[0010] The classification and storage module is used to classify the batches of coke to be stored based on the acquired process source data, quality index data, cost data and production demand data, and to determine the storage area of ​​the batches of coke to be stored and the target storage space of the corresponding storage area.

[0011] The inventory management module is used to determine the target storage cell for the batch of coke to be stored based on the current dynamic coke inventory map, the storage area corresponding to the batch of coke to be stored, and the target storage space, and to generate the storage planning instruction corresponding to the batch of coke to be stored. It also generates a coke blending strategy based on the externally input production plan, and generates a material picking planning instruction based on the coke blending strategy and the current dynamic coke inventory map.

[0012] The stacking and reclaiming execution module is used to receive and execute stacking planning instructions or reclaiming planning instructions, and control the transport vehicle, storage and reclaiming machine and mixing unit to complete the transportation, mixing and coke conveying operations related to the stacking or reclaiming of coke.

[0013] The intelligent safety assurance module for transport vehicles is used to predict the future risk situation of transport vehicles based on information about the interior of the transport vehicle, vehicle operation information, and road conditions ahead, and to determine the adaptive adjustment strategy of the transport vehicle based on the future risk situation.

[0014] The storage area includes coke storage silos for top-loading coke ovens and coke storage silos for compacted coke ovens. The storage space includes upper high-value silos, middle high-frequency silos, and lower high-substitutability silos. Each storage space is divided into several compartments by partitions.

[0015] Preferably, the categorized storage module includes:

[0016] The coke information acquisition submodule is used to acquire process source data, quality index data, cost data, and production demand data of the batch of coke to be stockpiled.

[0017] The primary classification submodule is used to perform primary classification of the coke batch to be stored based on the process source data of the coke batch to be stored, and to determine the storage area corresponding to the coke batch to be stored.

[0018] The secondary classification submodule is used to perform dynamic value assessment of the batch of coke to be stored based on the quality index data, cost data, and production demand data of the batch of coke to be stored, and to determine the target storage space corresponding to the batch of coke to be stored based on the dynamic value assessment results.

[0019] Preferably, the inventory management module includes:

[0020] The map building and maintenance submodule is used to build a dynamic coke inventory map based on the 3D model of the warehouse, and to receive the stacking or reclaiming operation completion signal from the stacking or reclaiming execution module. Based on the stacking or reclaiming operation completion signal, it updates the inventory and location of the corresponding coke batch in the dynamic coke inventory map in real time, and updates the dynamic coke inventory map.

[0021] The stacking planning submodule is used to determine the target storage cell for the batch of coke to be stacked based on the current dynamic coke inventory map, the stacking area corresponding to the batch of coke to be stacked, and the target storage space through sequential traversal and priority calculation, and to generate the stacking planning instruction corresponding to the batch of coke to be stacked.

[0022] The material requisition planning submodule is used to generate a coke blending strategy based on the production plan input from external sources, and to generate material requisition planning instructions based on the coke blending strategy and the current dynamic coke inventory map.

[0023] Preferably, the heap planning submodule includes a heap feasibility verification unit, a heap path planning unit, and a heap strategy optimization unit.

[0024] Preferably, the material picking planning submodule includes a strategy generation and matching unit, a multi-objective material picking path planning unit, a material picking verification unit, and a final material picking instruction generation unit.

[0025] Preferably, the material sampling verification unit includes a physical simulation verification subunit, a motion interference verification subunit, and a dynamic adjustment subunit.

[0026] Preferably, the stacking and reclaiming execution module includes:

[0027] The transportation submodule is used to receive and control the transportation vehicle to move between the storage area and the material grid based on the stacking planning instructions or the material picking planning instructions, so as to complete the stacking or material picking transportation of coke.

[0028] The material storage and retrieval auxiliary submodule is used to transfer the coke carried by the transport vehicle to the target material cell during the stacking operation, or to transfer the coke in the specific material cell to the transport vehicle during the retrieval operation.

[0029] The mixing and blending execution submodule is used to physically mix different types of coke that are conveyed from specific material compartments to transport vehicles by the storage and retrieval machine during material handling operations, forming a uniform mixed coke.

[0030] Preferably, the intelligent safety protection module for the transport vehicle includes:

[0031] The multi-source information acquisition and processing submodule is used to collect information on the interior of the transport vehicle, vehicle operation information, and road conditions ahead in real time. It processes the collected information to obtain several parameter processing values ​​corresponding to each time point. The processing values ​​of each parameter are arranged in chronological order to obtain several sequences. These sequences include the coke pile unevenness coefficient sequence, the coke average temperature sequence, the coke maximum temperature difference sequence, the average dust concentration in the compartment sequence, the transport vehicle speed sequence, the transport vehicle three-dimensional acceleration sequence, the transport vehicle body tilt angle sequence, and the road condition ruggedness index sequence.

[0032] The risk prediction and matching submodule is used to construct a transportation safety risk prediction matrix based on the sequences obtained by the multi-source information acquisition and processing submodule through time series prediction, extract key risk feature vectors that characterize the future risk situation of the transport vehicle, and match the key risk feature vectors with the corresponding risk pattern vectors in the historical risk pattern library to identify the current risk pattern to be adjusted of the transport vehicle and its risk matching degree.

[0033] The adaptive control submodule determines the adaptive adjustment strategy of the transport vehicle based on the current risk mode to be adjusted and its risk matching degree.

[0034] Preferably, the risk prediction and matching submodule includes a time-series prediction unit, a risk matrix construction unit, a key risk feature vector acquisition unit, and a risk pattern determination unit to be adjusted.

[0035] Preferably, the adaptive control submodule includes:

[0036] The strategy mapping unit pre-stores the correspondence between different risk modes to be adjusted and basic control strategy types, and is used to map the corresponding basic control strategy type based on the current risk mode to be adjusted of the transport vehicle.

[0037] The control parameter calculation unit is used to calculate the specific execution parameters of the basic control strategy type based on the basic control strategy type and risk matching degree of the transport vehicle.

[0038] The control instruction synthesis and issuance unit is used to generate an adaptive adjustment strategy for the transport vehicle based on the basic control strategy type determined by the strategy mapping unit and the specific execution parameters obtained by the control parameter calculation unit, and then issue the strategy to the transport vehicle.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] This invention, by setting up a classification and storage module, an inventory management module, a stacking and reclaiming execution module, and a transport vehicle safety intelligent protection module, combined with the division of storage areas between the coke storage areas of top-loading coke ovens and tamping coke ovens, and the layered design of storage space with upper-level high-value warehouses, middle-level high-frequency warehouses, and lower-level high-substitutability warehouses, achieves multi-dimensional classification and storage of coke based on process source data, quality index data, cost data, and production demand data. This solves the problems of single classification dimensions and low space utilization in existing technologies. The dynamic coke inventory map in the inventory management module can be synchronized with the physical warehouse in real time. The system monitors the status of coke, avoiding information delays and errors caused by manual inspections. It allows operators to intuitively grasp the location and quantity of each batch of coke, while intelligently generating stacking and reclaiming planning instructions. The stacking and reclaiming execution module coordinates transport vehicles, storage and reclaiming machines, and blending devices to achieve automated operations of stacking, transportation, and blending, improving overall operational efficiency. The intelligent safety assurance module for transport vehicles predicts risks and adjusts specific execution parameters based on information inside the vehicle compartment, vehicle operation information, and road conditions ahead, effectively avoiding the risks of overturning, overheating of the compartment, and dust explosions, reducing production interruptions and property losses caused by safety accidents. Attached Figure Description

[0041] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0042] Figure 1 This is a schematic diagram of the composition and structure of a multi-variety coke blending, storage, and reclaiming system provided by the present invention. Detailed Implementation

[0043] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0044] Example 1: This embodiment of the invention provides a multi-variety coke blending, storage, and reclaiming system, such as... Figure 1 As shown, it includes:

[0045] The classification and storage module is used to classify the batches of coke to be stored based on the acquired process source data, quality index data, cost data and production demand data, and to determine the storage area of ​​the batches of coke to be stored and the target storage space of the corresponding storage area.

[0046] The inventory management module is used to determine the target storage cell for the batch of coke to be stored based on the current dynamic coke inventory map, the storage area corresponding to the batch of coke to be stored, and the target storage space, and to generate the storage planning instruction corresponding to the batch of coke to be stored. It also generates a coke blending strategy based on the externally input production plan, and generates a material picking planning instruction based on the coke blending strategy and the current dynamic coke inventory map.

[0047] The stacking and reclaiming execution module is used to receive and execute stacking planning instructions or reclaiming planning instructions, and control the transport vehicle, storage and reclaiming machine and mixing unit to complete the transportation, mixing and coke conveying operations related to the stacking or reclaiming of coke.

[0048] The intelligent safety assurance module for transport vehicles is used to predict the future risk situation of transport vehicles based on information about the interior of the transport vehicle, vehicle operation information, and road conditions ahead, and to determine the adaptive adjustment strategy of the transport vehicle based on the future risk situation.

[0049] The storage area includes coke storage silos for top-loading coke ovens and coke storage silos for compacted coke ovens. The storage space includes upper high-value silos, middle high-frequency silos, and lower high-substitutability silos. Each storage space is divided into several compartments by partitions.

[0050] In this embodiment, the process source data includes the coke oven equipment code that produces coke, which is used to determine whether the coke originates from the top-loading coke oven process or the tamping coke oven process. Specifically, the coke oven equipment code is automatically obtained by scanning the RFID electronic tag or QR code on the coke transport vehicle by scanning the RFID reader or handheld terminal at the warehouse door when the coke is put into storage.

[0051] In this embodiment, the quality index data includes crush resistance (M40), abrasion resistance (M10), reactivity (CRI), post-reaction strength (CSR), ash content, and sulfur content. These quality index data are tested by the quality inspection department before the coke leaves the factory, and the results are directly entered into the coke information management system. The coke information management system is electrically connected to the classification and storage module in this embodiment of the invention, and automatically obtains the complete set of quality index data of the coke when it is put into storage.

[0052] In this embodiment, the cost data includes the unit cost price and the comprehensive transportation cost price. The classification and storage module in this embodiment is electrically connected to the enterprise's procurement cost statistics module. The procurement cost statistics module stores the purchase order number, batch information, unit cost price, and comprehensive transportation cost price for each batch of coke. The unit cost price is divided into two types: one is the quotient of the sum of the raw material purchase cost and all resource and personnel costs in the coke production process (if the enterprise produces the coke itself) and the total weight of the final coke produced; the other is the coke purchase cost price when the enterprise directly purchases coke without producing it itself. Similarly, the comprehensive transportation cost price is also divided into two types: one is the quotient of the sum of the transportation cost of the raw materials and the transportation cost of transporting the coke to the warehouse after production (if the enterprise produces the coke itself) and the total weight of the coke; the other is the quotient of the transportation cost of transporting the coke from the purchase location to the warehouse (if the enterprise does not produce the coke itself) and the total weight of the coke.

[0053] In this embodiment, the production demand data is the total amount of coke required within a preset future period.

[0054] In this embodiment, the storage spaces corresponding to the coke storage areas of the top-loading coke oven and the compacted coke oven both include an upper high-value storage area, a middle high-frequency storage area, and a lower high-substitutability storage area. The upper high-value storage area is used to store coke batches whose comprehensive value assessment value ranks among the top predetermined proportions (typically within the top 15%-20%, e.g., 15%) within the same process source storage area. The middle high-frequency storage area is used to store coke batches whose comprehensive value assessment value ranks among the middle predetermined proportions (typically within the middle 60%-70%, e.g., 70%) within the same process source storage area. The lower high-substitutability storage area is used to store coke batches whose comprehensive value assessment value ranks among the bottom predetermined proportions (typically within the bottom 15%-20%, e.g., 15%) within the same process source storage area.

[0055] In this embodiment, the dynamic coke inventory map is an integrated digital twin visualization system. It not only accurately reproduces the geometric structure of the physical warehouse in three-dimensional space, but more importantly, it uses a data-driven approach to associate and visualize the static attributes (such as quality indicators) and dynamic states (such as real-time inventory levels and locations) of each batch of coke with the specific storage space in the three-dimensional model. Users can view the distribution of coke within the warehouse from any angle through a human-computer interaction interface. Different colors or markers on the dynamic coke inventory map are used to distinguish coke varieties, value grades, or inventory status, including normal, critical, and vacant. In this embodiment, the warehouse refers to the building housing the coke storage silos in the top-loading coke oven area and the coke storage silos in the tamping coke oven area.

[0056] In this embodiment, the stacking planning instructions include stacking strategy instructions and stacking path planning instructions. The material handling planning instructions include coke blending strategy instructions, material handling path planning instructions, and material handling action planning instructions.

[0057] In this embodiment, the transport vehicle is a mobile carrier for coke, used to realize the storage and transportation of coke; the storage and retrieval machine is used to realize the transfer of coke between the storage compartment and the transport vehicle; the mixing device is used to physically mix different types of coke during the material retrieval operation, such as mechanical stirring and turning.

[0058] In this embodiment, the information inside the carriage includes images of the coke pile surface inside the carriage, coke temperature distribution information inside the carriage, and dust concentration information in the air inside the carriage. The images of the coke pile surface inside the carriage are acquired by high-definition industrial cameras deployed at the four corners of the carriage. The coke temperature distribution information inside the carriage is acquired by non-contact infrared temperature sensor arrays deployed on the side walls and bottom of the carriage. The dust concentration information in the air inside the carriage is acquired by laser dust concentration sensors deployed on the top of the carriage.

[0059] In this embodiment, vehicle operating information is obtained through the vehicle CAN bus interface, including vehicle speed, three-dimensional acceleration, and vehicle tilt angle data.

[0060] In this embodiment, road condition information ahead is obtained by scanning the road ahead with a solid-state LiDAR installed at the front of the vehicle to generate three-dimensional point cloud data.

[0061] In this embodiment, the future risk situation of the transport vehicle refers to whether there will be problems such as overturning, overheating of the compartment, and dust explosion in the future.

[0062] In this embodiment, the adaptive adjustment strategy of the transport vehicle is used to adjust the operating parameters of the transport vehicle, including vehicle speed, power of the cooling device, and spray intensity of the dust suppression device. The cooling device is a heat exchanger, and the dust suppression device is a sprayer.

[0063] The beneficial effects of the above technical solution are as follows: This embodiment of the invention, by setting up a classification and storage module, an inventory management module, a stacking and reclaiming execution module, and a transport vehicle safety intelligent protection module, combined with the division of storage areas between the coke storage areas of top-loading coke ovens and tamping coke ovens, and the layered design of storage space with upper-level high-value warehouses, middle-level high-frequency warehouses, and lower-level high-substitutability warehouses, achieves multi-dimensional classification and storage of coke based on process source data, quality indicator data, cost data, and production demand data. This solves the problems of single classification dimensions and low space utilization in existing technologies; the inventory management module provides a dynamic coke inventory... The map storage system can synchronize the physical warehouse status in real time, avoiding information lag and errors caused by manual inspections. It allows operators to intuitively grasp the location and quantity of each batch of coke, and can intelligently generate stacking planning instructions and material retrieving planning instructions. The stacking and retrieving execution module coordinates transport vehicles, storage and retrieving machines, and mixing devices to realize automated operations of stacking, transportation, and mixing, improving overall operational efficiency. The intelligent safety protection module for transport vehicles predicts risks based on information inside the vehicle compartment, vehicle operation information, and road conditions ahead, effectively avoiding the risks of overturning, overheating of the compartment, and dust explosions, reducing production interruptions and property losses caused by safety accidents.

[0064] Example 2: Based on Example 1, the categorized storage module includes:

[0065] The coke information acquisition submodule is used to acquire process source data, quality index data, cost data, and production demand data of the batch of coke to be stockpiled.

[0066] The primary classification submodule is used to perform primary classification of the coke batch to be stored based on the process source data of the coke batch to be stored, and to determine the storage area corresponding to the coke batch to be stored.

[0067] The secondary classification submodule is used to perform dynamic value assessment of the batch of coke to be stored based on the quality index data, cost data, and production demand data of the batch of coke to be stored, and to determine the target storage space corresponding to the batch of coke to be stored based on the dynamic value assessment results.

[0068] Preferably, the secondary classification submodule includes:

[0069] The quality assessment unit is used to obtain the quality assessment value of the batch of coke to be stored based on the quality index data of the batch of coke to be stored.

[0070] The cost assessment unit is used to obtain the cost assessment value of the batch of coke to be stored based on the cost data of the batch of coke to be stored.

[0071] The demand assessment unit is used to obtain the demand assessment value of the batch of coke to be stored based on the production demand data of the batch of coke to be stored.

[0072] The comprehensive value assessment unit is used to determine the comprehensive value assessment value of the batch of coke to be stored based on the quality assessment value, cost assessment value, and demand assessment value of the batch of coke to be stored.

[0073] The storage space determination unit is used to determine the target storage space for storing batches of coke based on the comprehensive value assessment of the batches to be stored.

[0074] In this embodiment, the coke to be stockpiled is classified into primary categories based on process source data to determine the corresponding stockpiling area. Specifically, if the coke to be stockpiled originates from a top-charged coke oven, the coke originating from the top-charged coke oven is allocated to the coke storage area silo of the top-charged coke oven for stockpiling; if the coke to be stockpiled originates from a stamped coke oven, the coke originating from the stamped coke oven is allocated to the coke storage area silo of the stamped coke oven for stockpiling.

[0075] In this embodiment, the dynamic value assessment specifically includes: obtaining a quality assessment value for the batch of coke to be stored based on its quality index data, wherein the better the quality index data, the higher the quality assessment value; obtaining a cost assessment value for the batch of coke to be stored based on its cost data, wherein the lower the cost, the higher the cost assessment value; obtaining a demand assessment value for the batch of coke to be stored based on its production demand data, wherein the higher the urgency of the demand, the higher the demand assessment value; and calculating a weighted sum of the quality assessment value, cost assessment value, and demand assessment value according to preset corresponding weights to obtain a comprehensive value assessment value for the batch of coke to be stored.

[0076] In this embodiment, the formula for calculating the quality assessment value of the batch of coke to be stored, based on the quality index data of the batch to be stored, is as follows:

[0077] ;in, This represents the quality assessment value of the batch of coke to be stockpiled. A higher quality assessment value indicates a better overall quality of the batch of coke. N represents the number of quality indicators included in the quality index data, specifically including crush strength (M40), abrasion resistance (M10), reactivity (CRI), post-reaction strength (CSR), ash content, and sulfur content, i.e., N=6. This represents the standardized score of the j-th quality indicator in the quality indicator data. The value range is 0-100; for positive indicators such as crush resistance (M40) and post-reaction strength (CSR), the larger the value, the better; for negative indicators such as abrasion resistance (M10), reactivity (CRI), ash content and sulfur content, the smaller the value, the better.

[0078] For positive indicators:

[0079] ;

[0080] For negative indicators:

[0081] ;

[0082] in, This represents the actual measured value of the j-th quality indicator in the coke information management system. and These are the preset maximum and minimum benchmarks for the j-th quality indicator within the evaluation system (which can be set according to national standards, industry standards, or enterprise internal control standards). All quality indicators are... and between.

[0083] In this embodiment, the formula for calculating the cost assessment value of the batch of coke to be stored, based on the cost data of the batch to be stored, is as follows:

[0084] ;

[0085] in, The cost assessment value of the batch of coke to be stored. The unit price is the cost benchmark, which can be the average cost of coke purchased by the enterprise over a period of time, or the target cost set based on the market price, and the unit is ton / yuan; and These are the unit cost price and the unit comprehensive transportation cost price in the cost data of the batch of coke to be stored, both in tons / yuan.

[0086] In this embodiment, the formula for calculating the demand assessment value of each batch of coke, based on the production demand data of the batches of coke to be stockpiled, is as follows:

[0087] ;

[0088] in, The demand assessment value for the batch of coke to be stockpiled; This indicates the total demand for the use of the stockpiled batches of coke within a future preset period, expressed in tons. This represents the current total inventory of batches of coke awaiting storage, expressed in tons.

[0089] In this embodiment, the formula for calculating the comprehensive value assessment of the batch of coke to be stored, based on the quality assessment value, cost assessment value, and demand assessment value, is as follows:

[0090] ;

[0091] in, This represents the comprehensive value assessment of the batch of coke to be stored. , and These represent the weighting coefficients for the quality assessment value, the cost assessment value, and the demand assessment value, respectively. We can assume .

[0092] The beneficial effects of the above technical solution are as follows: The coke information acquisition submodule can comprehensively collect process source data, quality index data, cost data, and production demand data of the batches of coke to be stored, providing complete data support for subsequent classification; the primary classification submodule, based on the process source data, clearly allocates the batches of coke to be stored to the coke storage area silos of top-loading coke ovens or tamping coke ovens, avoiding the mixing of coke from different process sources and affecting the quality of subsequent ironmaking processes; the secondary classification submodule calculates the quality assessment value of the quality assessment unit, the cost assessment value of the cost assessment unit, and the demand assessment value of the demand assessment unit, combined with preset weights, to derive... The comprehensive value assessment enables dynamic value grading of coke, ensuring that high-value coke can be allocated to easily accessible upper or middle storage spaces. For example, coke with urgent production needs can have its comprehensive value increased through demand assessment, allowing it to be placed in upper or middle high-frequency storage for rapid retrieval and reducing production waiting time. At the same time, the coordinated assessment of quality and cost avoids excessive costs due to focusing solely on quality or substandard quality due to focusing solely on cost. This ensures that the storage decisions for batches of coke are aligned with the company's overall benefits. In actual production, it can accurately match the demand for coke of different values ​​for different ironmaking processes, thereby improving the utilization rate of coke resources.

[0093] Example 3: Based on Example 1, the inventory management module includes:

[0094] The map building and maintenance submodule is used to build a dynamic coke inventory map based on the 3D model of the warehouse, and to receive the stacking or reclaiming operation completion signal from the stacking or reclaiming execution module. Based on the stacking or reclaiming operation completion signal, it updates the inventory and location of the corresponding coke batch in the dynamic coke inventory map in real time, and updates the dynamic coke inventory map.

[0095] The stacking planning submodule is used to determine the target storage cell for the batch of coke to be stacked based on the current dynamic coke inventory map, the stacking area corresponding to the batch of coke to be stacked, and the target storage space through sequential traversal and priority calculation, and to generate the stacking planning instruction corresponding to the batch of coke to be stacked.

[0096] The material requisition planning submodule is used to generate a coke blending strategy based on the production plan input from external sources, and to generate material requisition planning instructions based on the coke blending strategy and the current dynamic coke inventory map.

[0097] In this embodiment, constructing a dynamic coke inventory map using a 3D model of the warehouse as the base map specifically includes: using 3D modeling software or 3D laser scanning point cloud data to establish a 3D geometric model of the warehouse corresponding to the physical warehouse at a 1:1 scale. This 3D geometric model includes the warehouse building structure such as columns and beams, fixed facilities such as tracks and fire pipes, as well as the geometric outline and dimensions of all storage spaces. Based on this, the 3D geometric model is imported into the inventory management module as an immutable static base map. Subsequently, a dynamically updatable data layer is constructed on top of the static base map. This data layer is bound to the geometry of each material cell in each storage space in the static base map, and is used to store and display the batch information of coke stored in each material cell of each storage space (such as batch ID, variety, quantity, and entry time). The superposition of the static base map and the dynamic data layer constitutes the dynamic coke inventory map.

[0098] In this embodiment, the inventory and location of the corresponding coke batch in the dynamic coke inventory map are updated in real time based on the completion signal of the stacking or retrieving operation. The update of the dynamic coke inventory map specifically includes:

[0099] After the stacking and reclaiming execution module completes a stacking or reclaiming operation, it sends a structured operation completion signal to the inventory management module. This signal includes at least: the operation type (stacking / reclaiming), the equipment number used, the coke batch ID involved, the location of each storage compartment in the operated warehouse space, and the quantity operated (tons). Upon receiving this signal, the map building and maintenance submodule performs the following update operations:

[0100] If it is a stacking operation: based on the batch ID and the location of each cell in the storage space, find the data record of each cell in the corresponding storage space in the dynamic data layer, and increase its inventory quantity by the operation quantity. If it is a new batch entering the warehouse, create a new coke batch data record under each cell in the storage space.

[0101] If it is a material retrieval operation: based on the batch ID and the location of each material cell in the storage space, find the corresponding data record, reduce its inventory quantity by the number of operations, and if the inventory quantity is reduced to zero, mark the coke batch data record as "empty" or move it to the history record;

[0102] After the update is complete, immediately re-render the 3D visualization interface to ensure that the dynamic coke inventory map display remains synchronized with the actual status of the physical warehouse.

[0103] In this embodiment, the planning of the stockpiling scheme includes the planning of the stockpiling strategy and the stockpiling path. The planning of the material handling scheme includes the coke blending strategy, the material handling path planning, and the material handling action planning.

[0104] The beneficial effects of the above technical solution are as follows: A dynamic coke inventory map constructed using a 3D model of the warehouse as the base map, combined with real-time updates of inventory and location based on stacking and retrieving operation completion signals, solves the problems of information lag and disconnection from the physical warehouse in traditional inventory management. Operators can monitor inventory at any time through a visual interface, such as updating inventory data immediately after retrieving materials, avoiding operational chaos caused by duplicate or missed retrieving; the stacking planning submodule, based on the dynamic inventory map and classification results, determines the target material cell through sequential traversal and priority calculation, ensuring efficient utilization of storage space; the retrieving planning submodule generates coke blending strategies based on the production plan, replacing manual experience. In the production of specific steel grades, the system can quickly screen batches of coke that meet quality requirements and calculate the optimal blending ratio, reducing errors from manual calculations and improving the stability of blending quality. At the same time, the real-time nature of the dynamic inventory map provides accurate data support for material retrieval planning, such as accurately locating the material grid positions that meet the demand, avoiding material retrieval delays caused by inaccurate inventory information. In real-world scenarios, it can help management intuitively understand the inventory status. The stacking planning submodule and the material retrieval planning submodule realize intelligent and scientific stacking, retrieval, and blending of coke through stacking strategy planning, stacking path planning, coke blending strategy, material retrieval path planning, and material retrieval action planning.

[0105] Example 4: Based on Example 3, the heap planning submodule includes:

[0106] The storage feasibility verification unit is used to traverse all the material cells in the target storage space in the current coke dynamic inventory map according to the total amount of the batch of coke to be stored in a preset order, query the current remaining capacity of each material cell, and verify whether there is an empty material cell that can accommodate the batch of coke to be stored or a material cell with a remaining capacity that can accommodate the batch of coke to be stored.

[0107] If the stacking feasibility verification unit finds a target cell that meets the conditions, the stacking path planning unit performs collision detection and path simulation based on the current dynamic coke inventory map, the current position of the transport vehicle corresponding to the batch of coke to be stacked, and the position of the target cell. It generates a collision-free optimal path with the current position of the transport vehicle corresponding to the batch of coke to be stacked as the starting point and the position of the target cell as the ending point, and uses it as the stacking planning instruction for the batch of coke to be stacked.

[0108] If the stacking feasibility verification unit fails to find a suitable target cell, the stacking strategy optimization unit calculates the similarity in quality between the coke stored in each cell that already contains coke but is not full and the batch of coke to be stacked. It then calculates a comprehensive priority evaluation coefficient based on the remaining capacity of each cell, identifying the cell with the highest comprehensive priority evaluation coefficient as the target cell for the batch of coke to be stacked. Subsequently, the stacking path planning unit generates a stacking planning instruction for the batch of coke to be stacked based on the location of the target cell, and marks the batch of coke to be stacked as mixed stacking on the current dynamic coke inventory map. If there are no cells in the target storage space that meet the requirements for mixed stacking, an expansion warning for the target storage space is issued.

[0109] In this embodiment, the stockpiling planning submodule determines the target storage cell for the batch of coke to be stockpiled through sequential traversal and priority calculation. The specific process includes:

[0110] This invention predefines the traversal order of storage compartments within each storage space. When a batch of coke needs to be stored, the system strictly follows this order to scan all compartments within the target storage space. The scanning process consists of two stages: the first stage prioritizes finding completely empty compartments; the second stage, if no empty compartments are found, it searches for compartments that already contain the same batch of coke (i.e., have the same batch ID) and still have remaining capacity. This sequential traversal method ensures the systematic and efficient nature of the storage operation, prioritizing the use of vacant resources and achieving centralized stacking of the same batch of coke. The target storage space refers to the storage space corresponding to the batch of coke to be stored.

[0111] In this embodiment, collision detection and path simulation specifically include:

[0112] Collision detection refers to automatically checking whether the planned transport vehicle movement path will geometrically intersect with fixed obstacles such as pillars and walls, or dynamic obstacles such as the planned paths of other operating equipment in the current dynamic coke inventory map. The specific process is as follows: the transport vehicle is abstracted into a three-dimensional envelope containing its physical dimensions, and this envelope is moved along the planned transport vehicle movement path. The spatial distance between the vehicle and the three-dimensional models of all obstacles in the environment is calculated in real time. Once the distance is detected to be lower than the preset safety threshold, it is judged as a potential collision.

[0113] Path simulation, on the other hand, simulates the entire process of a transport vehicle performing a stacking task in a digital twin environment after collision detection has passed. This includes acceleration, constant speed movement, deceleration, positioning, and execution of stacking actions, thereby estimating the execution time, energy consumption, and other indicators of the task to evaluate the quality of the path.

[0114] In this embodiment, the collision-free optimal path, starting from the current position of the transport vehicle corresponding to the batch of coke to be stored and ending at the position of the target material cell, refers to a continuous trajectory planned in the three-dimensional space of the current dynamic coke inventory map, connecting the current position of the transport vehicle corresponding to the batch of coke to be stored and the position of the target material cell. This trajectory satisfies two conditions: (1) Collision-free: at any point on the path, the transport vehicle maintains a safe distance from all static and dynamic objects in the environment; (2) Optimal: the path is planned based on the main optimization objective of achieving the shortest path length or the least expected travel time. The collision-free optimal path is represented in the form of a series of ordered spatial coordinate points.

[0115] In this embodiment, if the storage feasibility verification unit fails to find a target cell that meets the conditions, it calculates the similarity in quality between the coke stored in each cell that already contains coke but is not full within the target storage space and the batch of coke to be stored. Combined with the remaining capacity of each cell, a comprehensive priority evaluation coefficient is calculated, specifically including:

[0116] ;in, This represents the overall priority evaluation coefficient for the k-th storage cell within the target storage space that already contains coke but is not full. and These are the weighting coefficients corresponding to capacity matching degree and quality similarity degree, respectively; The remaining capacity of the k-th compartment in the target storage space that already contains coke but is not full, in tons; The total weight of the batch of coke to be stored is in tons; N represents the number of quality indicators included in the quality index data, specifically including crush strength (M40), abrasion resistance (M10), reactivity (CRI), post-reaction strength (CSR), ash content, and sulfur content. N=6. and These are the preset maximum and minimum benchmark values ​​for the j-th quality indicator within the evaluation system, respectively. This represents the actual measured value of the j-th quality index of the batch of coke to be stored. This represents the weighted average of the j-th quality index of the coke stored in the k-th cell that already contains coke but is not full within the target storage space. When the cell contains a single type of coke, the weighted average is the actual measured value of the j-th quality index of the coke. If the cell contains a mixture of several types of coke, the weighted average is the weighted average of the actual measured values ​​of the j-th quality index of each type of coke.

[0117] In this embodiment, the specific implementation method for the stacking strategy optimization unit to mark the batch of coke to be stacked as mixed stacking on the current coke dynamic inventory map can be as follows:

[0118] In the dynamic data layer of the current coke dynamic inventory map, a "mixed storage" status label is added to the data record of the batch of coke to be stored. At the same time, the operation log and management interface of the multi-variety coke blending storage and retrieval system in this embodiment of the invention highlight or display the target cell with a special icon and generate a prompt message, such as "Batch ID: XXX has been mixed with batch ID: YYY in [cell location] due to the lack of a dedicated cell", thereby reminding the operators to pay attention to this non-ideal storage status, which facilitates subsequent tracking and management.

[0119] The beneficial effects of the above technical solution are as follows: The storage feasibility verification unit traverses the storage cells in a preset order, prioritizing empty cells or cells from the same batch, thus achieving centralized storage of coke from the same batch. This eliminates the need to retrieve multiple cells during subsequent material handling, saving operation time and avoiding the potential impact of mixed storage on the quality of subsequent blending. The storage path planning unit performs collision detection and path simulation based on the current dynamic inventory map, generating a collision-free optimal path. In complex environments where multiple transport vehicles and storage / retrieval machines operate simultaneously in the storage area, this effectively avoids equipment collisions, reduces equipment maintenance costs and the risk of production interruption. For example, the path of a transport vehicle from its current location to the target cell can avoid other operating equipment, improving transportation efficiency and safety. This system avoids the risk of overturning of the transport vehicle due to the shift in the center of gravity of the coke pile inside the compartment caused by inertia during obstacle avoidance. It also prevents spontaneous combustion of coke due to increased friction between coke during obstacle avoidance. When there are no matching material cells, the stacking strategy optimization unit determines the mixed stacking cells by calculating the similarity of quality and the remaining capacity, ensuring full utilization of storage space. If storage space is tight, coke of similar quality can be mixed and stacked without affecting the subsequent blending quality. At the same time, the expansion warning can remind enterprises to expand storage space in advance, avoiding production delays caused by the inability to store coke. In practical applications, it can significantly improve the flexibility and space utilization of storage operations in the warehouse area, and meet the storage needs under different inventory conditions.

[0120] Example 5: Based on Example 3 or Example 4, the material requisition planning submodule includes:

[0121] The strategy generation and matching unit is used to generate coke blending strategies based on the production plan input from external sources, determine the required quantity of each type of coke, and locate and lock the coke batch that meets the requirements and the location of its specific cell based on the current dynamic coke inventory map.

[0122] The multi-objective material handling path planning unit takes at least one of the following optimization objectives: shortest total material handling path, least total operation time, and least impact on subsequent stacking operations. It performs global optimization calculations based on the location of all specific material compartments to generate a material handling path sequence.

[0123] The material taking verification unit simulates the material taking action in the current coke dynamic inventory map for each specific material cell in the material taking path sequence, and verifies whether the activity space of the material taking port of each specific material cell is sufficient and whether the material taking will cause the coke to collapse or overflow.

[0124] The final material picking instruction generation unit integrates the verified material picking path sequence, the material picking type and quantity information of each specific material cell, and generates a material picking planning instruction.

[0125] In this embodiment, a coke blending strategy is generated based on the externally input production plan to determine the required amount of each type of coke. Specifically, the following steps are included: (1) Demand analysis: First, analyze the production plan and determine the target quality index range (such as the lower limit of CSR and the upper limit of sulfur content) and the total demand tonnage of coke required for the production of the steel grade according to the steel grade formula database; (2) Candidate batch screening: Query the current dynamic coke inventory map, and quickly screen out all coke batches that meet the quality requirements according to the above target quality index range, and obtain information such as the real-time inventory, cost unit price, and location of these batches to form a candidate batch set; (3) Cost optimization blending calculation: Start the cost optimization algorithm based on linear programming. The core objective of the cost optimization algorithm is to: select from the candidate batches... In the batch set, a material combination scheme is calculated so that the total cost is minimized under the premise of satisfying all the following constraints: quality constraint: the weighted average of the key quality indicators of the coke after combination must fall within the target range; quantity constraint: the sum of the material quantities of all batches is equal to the total demand tons; inventory constraint: the material quantity taken from any batch must not exceed its current inventory; cost target: the total cost is the minimum of the sum of the products of the material quantity taken from each batch and its unit cost (which may include material cost and estimated operating cost); (4) Strategy output: after the cost optimization algorithm is solved, the final coke blending strategy is output. The coke blending strategy clearly lists which candidate batches should be taken from and the specific values ​​of the material quantity taken from each batch, so as to simultaneously meet the requirements of optimal quality, quantity and cost.

[0126] In this embodiment, the optimization objective of minimizing the total material retrieval path is to minimize the total geometric path length traversed by the storage and retrieval machine to access all specific material cells; the optimization objective of minimizing the total operation time is to further consider the material retrieval operation time and acceleration / deceleration time of the equipment at each material cell point based on the path length, and to minimize the total time to complete the entire material retrieval task; and the optimization objective of minimizing the impact on subsequent stacking operations is to prioritize emptying certain material cells to make room for coke that will arrive later. This objective can be achieved by setting different "emptying priority" weights for different material cells.

[0127] In this embodiment, for each specific material cell in the material picking path sequence, the material picking action is simulated in the current dynamic coke inventory map to verify whether the activity space of the material picking port of each specific material cell is sufficient and whether the material picking will cause coke to collapse or overflow. Specifically, this includes:

[0128] Activity Space Verification: In the current three-dimensional digital twin environment of the dynamic coke inventory map, a precise 3D model of the storage and retrieval machine is loaded, simulating its movement along the planned path to the retrieval port of the specific material cell, and simulating the extension, rotation, and pitch of the retrieval arm. The system will detect whether there is any interference between the retrieval arm model and the three-dimensional models of the material cell door frame, surrounding equipment and other obstacles. At the same time, it verifies whether the retrieval arm can effectively touch the surface of the coke pile in the specific material cell within its physical movement limits.

[0129] Collapse and Overflow Verification: Based on the physical model of the angle of repose of the coke pile in the specific material cell, the shape change of the remaining material pile after the planned amount of coke is removed is simulated, and the slope of the pile after material removal is calculated. If the slope exceeds the angle of repose of the coke, it is determined that there is a risk of collapse. If the simulation shows that the material removal operation may cause coke to scatter from the edge of the material cell, it is determined that there is a risk of overflow. If the verification fails, the system will automatically issue an alarm and suggest adjusting the material removal amount or the location of the material removal point.

[0130] The beneficial effects of the above technical solutions are as follows: The strategy generation and matching unit analyzes the production plan requirements, combines the current dynamic inventory map to screen candidate coke batches, and generates a coke blending strategy through a cost optimization algorithm. Under the premise of meeting the quality requirements of ironmaking process, it can accurately select coke batches with lower costs and reduce coke blending costs. For example, when producing specific steel grades, it can quickly determine the optimal amount of coke used for each grade. The multi-objective material picking path planning unit generates a material picking path sequence with the objectives of shortest path, least time, and least impact on subsequent storage. For example, when picking material from multiple specific material compartments, the optimal path can reduce the moving distance and time of the material picking machine, improve material picking efficiency, and at the same time, prioritizing the emptying of some material compartments can free up space for subsequent storage and avoid congestion in the storage area. The material picking verification unit judges the risk of collapse based on the angle of repose model by simulating material picking actions and detects motion interference through the equipment three-dimensional model. When the material picking volume is large or the material compartment structure is special, it can detect safety hazards in advance. For example, if the simulation finds that a collapse will occur after material picking, the material picking plan can be adjusted in time, which significantly improves the safety, efficiency, and quality stability of material picking operations and ensures the continuous operation of ironmaking production.

[0131] Example 6: Based on Example 5, the material inspection unit includes:

[0132] The physical simulation verification subunit is used to simulate the shape change of the remaining pile after the planned amount of coke is removed, based on the physical model of the angle of repose of the coke pile in the current dynamic coke inventory map, for each specific material cell in the material picking path sequence. If the calculated pile slope exceeds the angle of repose of the coke, it is determined that there is a risk of collapse. If the simulation shows that the material picking operation causes the coke to scatter from the edge of the material cell, it is determined that there is a risk of overflow.

[0133] The motion interference verification subunit is used to load the accurate 3D model of the storage and retrieval machine into the current coke dynamic inventory map, simulate the entire process of the storage and retrieval machine's retrieval arm moving to the retrieval port of a specific material cell and performing the retrieval action, detect in real time whether there is geometric interference between the retrieval arm model in the accurate 3D model of the storage and retrieval machine and the 3D models of the specific material cell door frame, surrounding facilities and other obstacles, and verify whether the retrieval arm can effectively touch the surface of the coke pile within its physical motion limit range;

[0134] The dynamic adjustment subunit is used to automatically adjust the parameters in the material picking plan instruction when the physical simulation verification subunit or the motion interference verification subunit fails the verification. The parameters include the material picking point position, the single material picking amount, the cutting angle and posture of the material picking arm, and then re-perform the simulation verification until it passes.

[0135] In this embodiment, the angle of repose is the maximum angle that the slope of a bulk material (such as coke) can form with the horizontal plane when it is naturally piled up. If the angle exceeds this, the material will collapse. The angle of repose value used in this embodiment is a preset empirical parameter, which can be obtained very directly: through industry standard tests or by consulting the coke material characteristic manual. For example, through the standard "funnel method" experiment: let coke fall freely from a certain height to form a pile, and measure its slope to obtain the angle of repose, which is usually between 35 and 45 degrees. A conservative angle of repose value (such as 38 degrees) is preset for different types of coke and stored in the database to form a physical model of the angle of repose. When performing collapse verification, this preset value is directly called for calculation.

[0136] In this embodiment, simulating the shape change of the remaining pile after the planned amount of coke is removed and calculating the pile slope specifically includes: first, generating a virtual three-dimensional model of the coke pile based on the shape of the specific material compartments and the inventory of stored coke. The initial state is usually a regular cone shape. Then, according to the instruction, the planned amount of coke to be removed is "dug out" from a designated location in this virtual coke pile. After the removal, the shape of the remaining coke pile will change and become irregular. Then, the newly generated irregular remaining coke pile model is automatically analyzed to find the steepest slope and calculate the inclination angle of this slope (i.e., the pile slope). Finally, the calculated pile slope is compared with the preset angle of repose. If the calculated pile slope is greater than the angle of repose, it is determined that there is a "risk of collapse".

[0137] In this embodiment, the parameter adjustment strategy of the dynamic adjustment subunit specifically includes: First step: fine-tuning the position and posture. If the material picking arm will hit the door frame or cause interference, the picking point will be automatically fine-tuned (moved by tens of centimeters near the entrance) or the extension angle of the material picking arm will be adjusted. This adjustment does not change the total amount of material picked up and is the optimal choice. Second step: reducing the amount of material picked up at one time. If the simulation shows that the coke pile will still collapse when picking up material after the adjustment in the first step, the amount of material picked up at one time will be reduced. The one-time material picking task will be split into multiple times. For example, if it was originally planned to pick up 10 tons at a time, now it is changed to pick up 6 tons first, transport them away, and then come back to pick up the remaining 4 tons. The amount of material picked up each time is smaller, and the impact on the stability of the material pile is smaller.

[0138] Each time parameters are adjusted, the system will automatically perform a simulation verification to ensure that the new solution is safe. This process will be repeated until a solution that can pass all simulation verifications is found. If all preset solutions have been tried and still fail, the system will issue an alarm to remind the administrator to intervene manually.

[0139] The beneficial effects of the above technical solution are as follows: The physical simulation verification subunit, based on the coke angle of repose physical model, simulates the change in the shape of the material pile after material removal, accurately judges the risk of collapse and spillage. Different types of coke can be preset with corresponding angle of repose values ​​to avoid safety accidents caused by misjudgment of the angle of repose. For example, high-ash coke and low-ash coke have different angles of repose, and different angles of repose can be set separately to ensure accurate judgment. The motion interference verification subunit loads a precise three-dimensional model of the material handling machine, simulates the entire process of the material handling arm movement, and detects interference with the specific material grid frame and surrounding facilities. When the extension angle of the material handling arm is large or the specific material grid position is special, it can avoid... Wear caused by equipment collisions, such as when the material handling arm of the storage and retrieval machine needs to penetrate deep into a specific material compartment, ensure that its range of motion is within physical limits and does not touch obstacles; when the dynamic adjustment subunit fails the verification, first fine-tune the material handling point and the posture of the material handling arm. If it still fails, reduce the amount of material handled in a single operation and split the task to solve the problem without affecting the total amount of material handled. If fine-tuning the position can pass the test, there is no need to split the task, saving operation time. When splitting the material handling, it can also ensure the safety of each operation and avoid accidents caused by forced material handling. In practical applications, it can provide multiple safety guarantees for material handling operations, reduce equipment failure and coke loss, and improve operational reliability.

[0140] Example 7: Based on Example 1, the stacking and reclaiming execution module includes:

[0141] The transportation submodule is used to receive and control the transportation vehicle to move between the storage area and the material grid based on the stacking planning instructions or the material picking planning instructions, so as to complete the stacking or material picking transportation of coke.

[0142] The material storage and retrieval auxiliary submodule is used to transfer the coke carried by the transport vehicle to the target material cell during the stacking operation, or to transfer the coke in the specific material cell to the transport vehicle during the retrieval operation.

[0143] The mixing and blending execution submodule is used to physically mix different types of coke that are conveyed from specific material compartments to transport vehicles by the storage and retrieval machine during material handling operations, forming a uniform mixed coke.

[0144] The beneficial effects of the above technical solution are as follows: The transportation submodule controls the movement of the transport vehicle based on the stacking planning command or the material retrieval planning command. The material storage and retrieval auxiliary submodule transfers the coke from the transport vehicle to the target material cell during stacking and transfers the coke from the specific material cell to the transport vehicle during retrieval, replacing manual handling, reducing labor costs and labor intensity, and improving conveying efficiency. For example, after the transport vehicle reaches the target material cell, the material storage and retrieval machine can automatically complete the loading and unloading of coke without manual intervention. The blending execution submodule physically mixes different types of coke through mechanical stirring, turning, etc., to ensure uniform mixing. When producing a specific ratio of blended coke, the uniform mixing quality can ensure the stability of the ironmaking process and avoid fluctuations in steel quality due to uneven mixing. In summary, the stacking and reclaiming execution module realizes the automated connection of stacking, transportation, and blending, improves the smoothness and efficiency of the overall operation process, and reduces the errors and risks caused by manual operation.

[0145] Example 8: Based on Example 1, the intelligent safety assurance module for the transport vehicle includes:

[0146] The multi-source information acquisition and processing submodule is used to collect information on the interior of the transport vehicle, vehicle operation information, and road conditions ahead in real time. It processes the collected information to obtain several parameter processing values ​​corresponding to each time point. The processing values ​​of each parameter are arranged in chronological order to obtain several sequences. These sequences include the coke pile unevenness coefficient sequence, the coke average temperature sequence, the coke maximum temperature difference sequence, the average dust concentration in the compartment sequence, the transport vehicle speed sequence, the transport vehicle three-dimensional acceleration sequence, the transport vehicle body tilt angle sequence, and the road condition ruggedness index sequence.

[0147] The risk prediction and matching submodule is used to construct a transportation safety risk prediction matrix based on the sequences obtained by the multi-source information acquisition and processing submodule through time series prediction, extract key risk feature vectors that characterize the future risk situation of the transport vehicle, and match the key risk feature vectors with the corresponding risk pattern vectors in the historical risk pattern library to identify the current risk pattern to be adjusted of the transport vehicle and its risk matching degree.

[0148] The adaptive control submodule determines the adaptive adjustment strategy of the transport vehicle based on the current risk mode to be adjusted and its risk matching degree.

[0149] In this embodiment, the collected information is processed to obtain several parameter processing values ​​corresponding to each time point, including:

[0150] Optical flow analysis is performed on images from adjacent time frames to calculate the variance of the motion vector field on the surface of the coke pile, and the variance value is used as the roughness coefficient.

[0151] Based on the coke temperature distribution information inside the carriage, the average coke temperature and the maximum coke temperature difference are obtained. Specifically, the arithmetic mean of all readings from the temperature sensor array is taken as the average coke temperature, and the difference between the maximum and minimum values ​​is taken as the maximum coke temperature difference.

[0152] The average dust concentration in the carriage is obtained based on the dust concentration information in the air inside the carriage. Specifically, the average dust concentration in the carriage is obtained by performing a moving average filter on the readings of the dust concentration sensor.

[0153] Based on vehicle operation information, the vehicle speed, three-dimensional acceleration, and tilt angle of the transport vehicle are obtained by directly parsing the CAN bus data packets to obtain the corresponding values.

[0154] The road condition ruggedness index is obtained based on the road condition information ahead. Specifically, after performing ground segmentation on the lidar point cloud data, the height variance of non-ground points is calculated as the road condition ruggedness index.

[0155] In this embodiment, the adaptive adjustment strategy includes the basic control strategy types and their specific execution parameters corresponding to different risk modes to be adjusted.

[0156] The beneficial effects of the above technical solution are as follows: The multi-source information acquisition and processing submodule collects information about the interior of the vehicle compartment, vehicle operation, and road conditions in real time. Even if the transport vehicle is traveling on temporary rugged roads outside the storage area or in an environment with excessive dust concentration, it can comprehensively grasp the status of the vehicle and the coke, providing complete data for risk assessment. The risk prediction and matching submodule constructs a transportation safety risk prediction matrix through time-series prediction, extracts key risk feature vectors, and matches them with a historical risk pattern library. This can predict the risks of overturning, overheating of the compartment, and dust explosion in advance, avoiding passive responses after risks occur. For example, it can detect abnormal tilt angles of the vehicle in advance. Normally, timely intervention can prevent overturning; the adaptive control submodule determines the adaptive adjustment strategy of the transport vehicle based on the risk mode to be adjusted and the risk matching degree, so as to adjust the working parameters of the transport vehicle, such as increasing the power of the cooling device when the temperature in the compartment is too high, increasing the spray intensity of the dust suppression device when the dust concentration is high, and reducing the vehicle speed when the speed is too high. It can accurately respond to different risk scenarios, avoiding excessive control that wastes energy or insufficient control that leads to accidents. In practical applications, it can significantly reduce the safety risks of transport vehicles, reduce personnel casualties, equipment damage and production interruptions caused by safety accidents, and ensure the safety and stability of the transportation process.

[0157] Example 9: Based on Example 8, the risk prediction and matching submodule includes:

[0158] The time series prediction unit is used to form M fixed-dimensional real-time joint feature vectors based on the sequence values ​​corresponding to the most recent M time points in each sequence obtained by the multi-source information acquisition and processing submodule. The M fixed-dimensional real-time joint feature vectors are then input into the trained risk prediction model according to the time series. The risk prediction model outputs K predicted joint feature vectors corresponding to the next K time points.

[0159] The risk matrix construction unit is used to combine K predictive joint feature vectors in a time sequence to construct a K-row, J-column transportation safety risk prediction matrix. , where J is the dimension of the predicted joint feature vector;

[0160] The key risk feature vector acquisition unit is used to calculate the transportation safety risk prediction matrix. covariance matrix And for the covariance matrix Eigenvalue decomposition is performed to obtain the covariance matrix. Several eigenvalues ​​and their corresponding eigenvectors are obtained. The eigenvalues ​​are arranged in descending order, the top Y largest eigenvalues ​​are selected, and the eigenvectors corresponding to the top Y largest eigenvalues ​​are extracted as key risk eigenvectors that characterize the future risk situation of the transport vehicle.

[0161] The unit for determining the risk pattern to be adjusted is used to calculate the cosine similarity between each key risk feature vector and each reference risk pattern vector pre-stored in the historical risk pattern library, and use it as the risk matching degree corresponding to each key risk feature vector. If any risk matching degree is greater than the preset threshold value of the risk matching degree, then the risk pattern corresponding to the reference risk pattern vector in the historical risk pattern library is used as the risk pattern to be adjusted for the corresponding key risk feature vector.

[0162] In this embodiment, the most recent M time points are referenced to the current time. t The real-time joint feature vector at each time point can be represented as: Where J=8, They represent the first t The data included the coke pile unevenness coefficient, average coke temperature, maximum coke temperature difference, average dust concentration in the truck compartment, truck speed, three-dimensional acceleration modulus of the truck, truck tilt angle, and road roughness index at various time points.

[0163] In this embodiment, the risk prediction model is a gated recurrent unit neural network model. This model is trained using a large amount of historical normal transportation data. It takes M historical joint feature vectors arranged in sequence at fixed time intervals (e.g., 60 seconds) as input and outputs K real joint feature vectors arranged in sequence at subsequent fixed time intervals. The trained risk prediction model has predictive capabilities. By inputting the M fixed-dimensional real-time joint feature vectors into the trained risk prediction model in time sequence, K predicted joint feature vectors corresponding to the next K time points can be obtained. Specifically:

[0164] Will M fixed-dimensional real-time joint feature vectors arranged in time sequence are input into the trained risk prediction model to obtain... These K predict joint feature vectors, where J=8, For the first t + K The predicted joint feature vector at each time point The risk prediction model predicts the first t + K The data included the coke pile unevenness coefficient, average coke temperature, maximum coke temperature difference, average dust concentration in the truck compartment, truck speed, three-dimensional acceleration modulus of the truck, truck tilt angle, and road roughness index at various time points.

[0165] In this embodiment, the transportation safety risk prediction matrix :

[0166] ;

[0167] In this embodiment, the transportation safety risk prediction matrix is ​​calculated. covariance matrix And for the covariance matrix Eigenvalue decomposition is performed, specifically: the transportation safety risk prediction matrix is ​​analyzed. Centering is performed on each column, i.e., subtracting the mean of the sum of the elements in each column from each element to obtain the centered matrix H, and the covariance matrix H. covariance matrix It is a J×J, or 8×8, matrix. for The transpose of the covariance matrix Eigenvalue decomposition satisfies the formula ,in b =1, 2, ..., 8 and They are respectively represented by the covariance matrix The eigenvalues ​​and corresponding eigenvectors.

[0168] In this embodiment, the covariance matrix The eigenvalues ​​are respectively The corresponding feature vectors are respectively The eigenvectors corresponding to the top Y largest eigenvalues ​​are extracted as key risk feature vectors, where the key risk feature vectors are assumed to include... .

[0169] In this embodiment, the historical risk pattern library stores the reference risk pattern vectors extracted from historical abnormal transportation events (including vehicle overturning and spontaneous combustion of coke in the carriage) and their corresponding risk patterns (such as overturning risk, carriage overheating risk, and dust explosion risk). The risk patterns are manually labeled based on the direct attribution of each historical abnormal transportation event. Each reference risk pattern vector can correspond to multiple risk patterns.

[0170] The risk pattern vector is constructed as follows: extract information about the interior of the carriage, vehicle operation, and road conditions ahead for a period of time before the occurrence of the abnormal transportation event; construct a transportation safety risk prediction matrix and calculate the covariance matrix using the same method; and after eigenvalue decomposition of the covariance matrix, take the eigenvector corresponding to the largest eigenvalue among all eigenvalues ​​as the risk pattern vector.

[0171] In this embodiment, the cosine similarity calculation formula between the key risk feature vector and each reference risk pattern vector pre-stored in the historical risk pattern library has the numerator being the inner product of the two vectors (key risk feature vector and reference risk pattern vector) and the denominator being the product of the magnitudes of the two vectors. The closer the cosine similarity is to 1, the more similar the directions of the two vectors are, and the higher the risk matching degree. If the risk matching degree is greater than the preset threshold value of the risk matching degree, the risk pattern corresponding to the corresponding reference risk pattern vector in the historical risk pattern library is taken as the risk pattern to be adjusted for the corresponding key risk feature vector. The risk pattern to be adjusted includes one or more of the following: overturning risk, carriage overheating risk, and dust explosion risk.

[0172] The risk matching degree preset threshold is a manually set value. When the risk matching degree is greater than the risk matching degree preset threshold, it proves that the transport vehicle will experience abnormal transport events such as overturning or spontaneous combustion of coke in the compartment within the next K time points.

[0173] In this embodiment, the risk matching degree includes the matching degree of overturning risk, the matching degree of carriage overheating risk, and the matching degree of dust explosion risk, which correspond to the risk modes to be adjusted as overturning risk, carriage overheating risk, and dust explosion risk, respectively.

[0174] The beneficial effects of the above technical solution are as follows: The time-series prediction unit adopts a trained gated recurrent unit neural network model, which accurately predicts the predicted joint feature vectors for the next K time points based on the real-time joint feature vectors of the most recent M time points. When the operating parameters of the transport vehicle suddenly change (such as a sudden increase in vehicle speed or a larger vehicle tilt angle), it can accurately capture the future state change trend, providing a reliable basis for risk prediction; The risk matrix construction unit integrates the K predicted joint feature vectors into a K-row J-column matrix, comprehensively reflecting future multi-dimensional risk information, avoiding the limitations of single time points or single-dimensional data, such as simultaneously reflecting information such as temperature, dust concentration, and vehicle speed for the next K time points, facilitating overall risk assessment; Key risk features are directed to... The quantity acquisition unit calculates the covariance matrix of the transportation safety risk matrix and performs eigenvalue decomposition, extracting the eigenvectors corresponding to the top Y largest eigenvalues ​​to focus on key risk factors and avoid judgment interference caused by information redundancy. The risk mode determination unit determines the risk mode to be adjusted by calculating cosine similarity and matching the key risk feature vectors with the corresponding risk mode vectors in the historical risk mode library. With abundant historical data, it can accurately identify the current risk type, such as matching the high dust concentration risk mode and the dust explosion risk mode, avoiding incorrect response strategies due to misjudgment of risk type. In real-world scenarios, it can improve the accuracy and pertinence of risk prediction, provide accurate risk basis for subsequent adaptive control, and reduce false or missed warnings.

[0175] This invention first constructs a transportation safety risk prediction matrix through time-series prediction, then extracts principal features (key risk feature vectors) through covariance analysis, and finally achieves a leap from passive response to proactive intelligent early warning through cosine matching of historical risk patterns. It uses a risk prediction model to predict the joint feature vectors of K future time points, constructing a complete spatiotemporal picture of future risks, i.e., the transportation safety risk prediction matrix. Furthermore, by calculating the covariance matrix of the transportation safety risk prediction matrix and performing eigenvalue decomposition, noise interference can be eliminated, and key risk feature vectors representing the essence of the risk can be extracted through dimensionality reduction. Finally, by calculating the cosine similarity between the key risk feature vectors and the corresponding risk pattern vectors in the historical risk pattern library, accurate qualitative and quantitative risk matching based on complex patterns rather than simple rules is achieved. This not only significantly improves the accuracy and foresight of risk identification but also provides precise strength basis for adaptive control, thereby enabling early, accurate, and intelligent early warning and prevention of various risks such as overturning, overheating, and dust explosions.

[0176] Example 10: Based on Example 8 or Example 9, the adaptive control submodule includes:

[0177] The strategy mapping unit pre-stores the correspondence between different risk modes to be adjusted and basic control strategy types, and is used to map the corresponding basic control strategy type based on the current risk mode to be adjusted of the transport vehicle.

[0178] The control parameter calculation unit is used to calculate the specific execution parameters of the basic control strategy type based on the basic control strategy type and risk matching degree of the transport vehicle.

[0179] The control instruction synthesis and issuance unit is used to generate an adaptive adjustment strategy for the transport vehicle based on the basic control strategy type determined by the strategy mapping unit and the specific execution parameters obtained by the control parameter calculation unit, and then issue the strategy to the transport vehicle.

[0180] In this embodiment, the correspondence between different risk modes to be adjusted and basic regulatory strategy types includes:

[0181] When the risk mode to be adjusted is overturning risk, the corresponding basic control strategy type is speed limit and route adjustment; when the risk mode to be adjusted is overheating risk of the carriage, the corresponding basic control strategy type is speed limit and activation of cooling device; when the risk mode to be adjusted is dust explosion risk, the corresponding basic control strategy type is speed limit and activation of dust suppression device.

[0182] In this embodiment, the specific execution parameters for each basic control strategy type are calculated based on the basic control strategy type and risk matching degree of the transport vehicle, including:

[0183] When the risk mode to be adjusted is capsizing risk, a path replanning instruction is generated, requiring that the maximum gradient of the new path must not exceed a preset threshold. in, The maximum value of the slope reference. Risk matching degree for the risk of capsizing;

[0184] When the risk mode to be adjusted is the risk of overheating of the vehicle compartment, the cooling device of the transport vehicle is activated. The formula for calculating the power of the cooling device is: ;in, The adjusted power of the cooling device. The rated power of the cooling device, Risk matching degree for the risk of overheating in the carriage. The larger the adjustment, the greater the power of the cooling device, thus achieving precise cooling and preventing the carriage from overheating;

[0185] When the risk mode to be adjusted is dust explosion risk, the dust suppression device is activated: The formula for calculating the spray intensity of the dust suppression device is: ;in, The adjusted spray intensity of the dust suppression device The baseline spray intensity for the dust suppression device. Risk matching degree for dust explosion risk The larger the size of the adjusted dust suppression device, the greater the spray intensity, thus achieving precise dust suppression;

[0186] The adjusted formula for calculating the target vehicle speed is as follows: ;in, The adjusted target speed, The current speed of the transport vehicle. , and These are the risk weights for overturning, overheating of the carriage, and dust explosion. , and The value is determined based on the risk pattern to be adjusted corresponding to the key risk feature vector: if the risk pattern vector corresponding to the key risk feature vector includes two types of risk patterns in the historical risk pattern library, such as overturning risk and carriage overheating risk, then... , , If the corresponding risk pattern vector to the key risk feature vector in the historical risk pattern library only includes capsizing risk, then , , ; The larger the value, the lower the target vehicle speed after adjustment, thus ensuring that the transport vehicle will not overturn. The larger the adjustment, the smaller the target speed, thus mitigating overheating of the carriage. The larger the adjustment, the smaller the target vehicle speed, thus avoiding exacerbating the risk of dust explosion.

[0187] The beneficial effects of the above technical solution are as follows: The strategy mapping unit pre-stores the correspondence between different risk modes to be adjusted and basic control strategy types, which can quickly determine the control direction when a risk occurs, without the need to formulate a strategy on the spot, saving response time and avoiding the expansion of risk due to decision delay; The control parameter calculation unit accurately calculates the specific execution parameters of the basic control strategy type based on the risk matching degree and preset formula. For example, when the risk matching degree of the overheating of the carriage is high, the power of the cooling device is increased according to the formula. When the risk matching degree of overturning is high, the target vehicle speed is reduced. This can ensure the control effect and avoid over-control (such as using high power cooling when the risk is low, resulting in energy waste) or under-control (such as only slightly adjusting the vehicle speed when the risk is high, which cannot avoid the risk), thus achieving refined control; The control instruction synthesis and issuance unit integrates the control type (basic control strategy type) and specific execution parameters into an executable instruction and issues it to the transport vehicle in a timely manner, ensuring that the transport vehicle can operate safely and stably under different risk scenarios, further improving the safety and reliability of the transportation process.

[0188] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A multi-species coke blending and reclaiming system, characterized in that, include: The classification and storage module is used to classify the batches of coke to be stored based on the acquired process source data, quality index data, cost data and production demand data, and to determine the storage area of ​​the batches of coke to be stored and the target storage space of the corresponding storage area. The inventory management module is used to determine the target storage cell for the batch of coke to be stored based on the current dynamic coke inventory map, the storage area corresponding to the batch of coke to be stored, and the target storage space, and to generate the storage planning instruction corresponding to the batch of coke to be stored. The stacking and reclaiming execution module is used to receive and execute stacking planning instructions or reclaiming planning instructions, and control the transport vehicle, storage and reclaiming machine and mixing unit to complete the transportation, mixing and coke conveying operations related to the stacking or reclaiming of coke. The intelligent safety assurance module for transport vehicles is used to predict the future risk situation of transport vehicles based on information about the interior of the transport vehicle, vehicle operation information, and road conditions ahead, and to determine the adaptive adjustment strategy of the transport vehicle based on the future risk situation. The storage area includes the coke storage area of ​​the top-loading coke oven and the coke storage area of ​​the tamping coke oven. The storage space includes the upper high-value storage area, the middle high-frequency storage area and the lower high-substitutability storage area. Each storage space is divided into several compartments by partitions. The inventory management module includes a material picking planning submodule, which is used to generate a coke blending strategy based on the externally input production plan, and to generate material picking planning instructions based on the coke blending strategy and the current dynamic coke inventory map. The material requisition planning submodule includes: The strategy generation and matching unit is used to generate coke blending strategies based on the production plan input from external sources, determine the required quantity of each type of coke, and locate and lock the coke batch that meets the requirements and the location of its specific cell based on the current dynamic coke inventory map. The multi-objective material handling path planning unit takes at least one of the following as optimization objectives: shortest total material handling path, least total operation time, and least impact on subsequent stacking operations. It performs global optimization calculations based on the location of all specific material compartments to generate a material handling path sequence. The material taking verification unit simulates the material taking action in the current coke dynamic inventory map for each specific material cell in the material taking path sequence, and verifies whether the activity space of the material taking port of each specific material cell is sufficient and whether the material taking will cause the coke to collapse or overflow. The final material picking instruction generation unit integrates the verified material picking path sequence, the material picking type and quantity information of each specific material cell, and generates a material picking planning instruction. The material handling verification unit includes: The physical simulation verification subunit is used to simulate the shape change of the remaining pile after the planned amount of coke is removed, based on the physical model of the angle of repose of the coke pile in the current dynamic coke inventory map, for each specific material cell in the material picking path sequence. If the calculated pile slope exceeds the angle of repose of the coke, it is determined that there is a risk of collapse. If the simulation shows that the material picking operation causes the coke to scatter from the edge of the material cell, it is determined that there is a risk of overflow. The motion interference verification subunit is used to load the accurate 3D model of the storage and retrieval machine into the current coke dynamic inventory map, simulate the entire process of the storage and retrieval machine's retrieval arm moving to the retrieval port of a specific material cell and performing the retrieval action, detect in real time whether there is geometric interference between the retrieval arm model in the accurate 3D model of the storage and retrieval machine and the 3D models of the specific material cell door frame, surrounding facilities and other obstacles, and verify whether the retrieval arm can effectively touch the surface of the coke pile within its physical motion limit range; The dynamic adjustment subunit is used to automatically adjust the parameters in the material picking plan instruction when the physical simulation verification subunit or the motion interference verification subunit fails the verification. The parameters include the material picking point position, the single material picking amount, the cutting angle and posture of the material picking arm, and re-perform the simulation verification until it passes. The intelligent safety protection module for transport vehicles includes: The multi-source information acquisition and processing submodule is used to collect information on the interior of the transport vehicle, vehicle operation information, and road conditions ahead in real time. It processes the collected information to obtain several parameter processing values ​​corresponding to each time point. The processing values ​​of each parameter are arranged in chronological order to obtain several sequences. These sequences include the coke pile unevenness coefficient sequence, the coke average temperature sequence, the coke maximum temperature difference sequence, the average dust concentration in the compartment sequence, the transport vehicle speed sequence, the transport vehicle three-dimensional acceleration sequence, the transport vehicle body tilt angle sequence, and the road condition ruggedness index sequence. The risk prediction and matching submodule is used to construct a transportation safety risk prediction matrix based on the sequences obtained by the multi-source information acquisition and processing submodule through time series prediction, extract key risk feature vectors that characterize the future risk situation of the transport vehicle, and match the key risk feature vectors with the corresponding risk pattern vectors in the historical risk pattern library to identify the current risk pattern to be adjusted of the transport vehicle and its risk matching degree. The adaptive control submodule determines the adaptive adjustment strategy of the transport vehicle based on the current risk mode to be adjusted and its risk matching degree. The risk prediction and matching submodule includes: The time series prediction unit is used to form M fixed-dimensional real-time joint feature vectors based on the sequence values ​​corresponding to the most recent M time points in each sequence obtained by the multi-source information acquisition and processing submodule. The M fixed-dimensional real-time joint feature vectors are then input into the trained risk prediction model according to the time series. The risk prediction model outputs K predicted joint feature vectors corresponding to the next K time points. The risk matrix construction unit is configured to combine the K prediction joint feature vectors in time sequence to construct a K-row and J-column transport safety risk prediction matrix. wherein J is the dimension of the prediction joint feature vector. The key risk feature vector acquisition unit is used to calculate the transportation safety risk prediction matrix. covariance matrix And for the covariance matrix Eigenvalue decomposition is performed to obtain the covariance matrix. Several eigenvalues ​​and their corresponding eigenvectors are obtained. The eigenvalues ​​are arranged in descending order, the top Y largest eigenvalues ​​are selected, and the eigenvectors corresponding to the top Y largest eigenvalues ​​are extracted as key risk eigenvectors that characterize the future risk situation of the transport vehicle. The unit for determining the risk pattern to be adjusted is used to calculate the cosine similarity between each key risk feature vector and each reference risk pattern vector pre-stored in the historical risk pattern library, and use it as the risk matching degree corresponding to each key risk feature vector. If any risk matching degree is greater than the preset threshold value of the risk matching degree, then the risk pattern corresponding to the reference risk pattern vector in the historical risk pattern library is used as the risk pattern to be adjusted for the corresponding key risk feature vector.

2. The multi-variety coke blending, storage, and reclaiming system according to claim 1, characterized in that, The categorized storage module includes: The coke information acquisition submodule is used to acquire process source data, quality index data, cost data, and production demand data of the batch of coke to be stockpiled. The primary classification submodule is used to perform primary classification of the coke batch to be stored based on the process source data of the coke batch to be stored, and to determine the storage area corresponding to the coke batch to be stored. The secondary classification submodule is used to perform dynamic value assessment of the batch of coke to be stored based on the quality index data, cost data, and production demand data of the batch of coke to be stored, and to determine the target storage space corresponding to the batch of coke to be stored based on the dynamic value assessment results.

3. The multi-variety coke blending, storage, and reclaiming system according to claim 1, characterized in that, The inventory management module includes: The map building and maintenance submodule is used to build a dynamic coke inventory map based on the 3D model of the warehouse, and to receive the stacking or reclaiming operation completion signal from the stacking or reclaiming execution module. Based on the stacking or reclaiming operation completion signal, it updates the inventory and location of the corresponding coke batch in the dynamic coke inventory map in real time, and updates the dynamic coke inventory map. The stacking planning submodule is used to determine the target storage cell for the batch of coke to be stacked based on the current dynamic coke inventory map, the stacking area corresponding to the batch of coke to be stacked, and the target storage space through sequential traversal and priority calculation, and to generate the stacking planning instruction corresponding to the batch of coke to be stacked.

4. The multi-variety coke blending, storage, and reclaiming system according to claim 3, characterized in that, The heap planning submodule includes: The storage feasibility verification unit is used to traverse all the material cells in the target storage space in the current coke dynamic inventory map according to the total amount of the batch of coke to be stored in a preset order, query the current remaining capacity of each material cell, and verify whether there is an empty material cell that can accommodate the batch of coke to be stored or a material cell with a remaining capacity that can accommodate the batch of coke to be stored. If the stacking feasibility verification unit finds a target cell that meets the conditions, the stacking path planning unit performs collision detection and path simulation based on the current dynamic coke inventory map, the current position of the transport vehicle corresponding to the batch of coke to be stacked, and the position of the target cell. It generates a collision-free optimal path with the current position of the transport vehicle corresponding to the batch of coke to be stacked as the starting point and the position of the target cell as the ending point, and uses it as the stacking planning instruction for the batch of coke to be stacked. If the stacking feasibility verification unit fails to find a suitable target cell, the stacking strategy optimization unit calculates the similarity in quality between the coke stored in each cell that already contains coke but is not full and the batch of coke to be stacked. It then calculates a comprehensive priority evaluation coefficient based on the remaining capacity of each cell, identifying the cell with the highest comprehensive priority evaluation coefficient as the target cell for the batch of coke to be stacked. Subsequently, the stacking path planning unit generates a stacking planning instruction for the batch of coke to be stacked based on the location of the target cell, and marks the batch of coke to be stacked as mixed stacking on the current dynamic coke inventory map. If there are no cells in the target storage space that meet the requirements for mixed stacking, an expansion warning for the target storage space is issued.

5. A multi-variety coke blending, storage, and reclaiming system according to claim 1, characterized in that, The stacking and reclaiming execution module includes: The transportation submodule is used to receive and control the transportation vehicle to move between the storage area and the material grid based on the stacking planning instructions or the material picking planning instructions, so as to complete the stacking or material picking transportation of coke. The material storage and retrieval auxiliary submodule is used to transfer the coke carried by the transport vehicle to the target material cell during the stacking operation, or to transfer the coke in the specific material cell to the transport vehicle during the retrieval operation. The mixing and blending execution submodule is used to physically mix different types of coke that are conveyed from specific material compartments to transport vehicles by the storage and retrieval machine during material handling operations, forming a uniform mixed coke.

6. The multi-variety coke blending, storage, and reclaiming system according to claim 1, characterized in that, The adaptive control submodule includes: The strategy mapping unit pre-stores the correspondence between different risk modes to be adjusted and basic control strategy types, and is used to map the corresponding basic control strategy type based on the current risk mode to be adjusted of the transport vehicle. The control parameter calculation unit is used to calculate the specific execution parameters of the basic control strategy type based on the basic control strategy type and risk matching degree of the transport vehicle. The control instruction synthesis and issuance unit is used to generate an adaptive adjustment strategy for the transport vehicle based on the basic control strategy type determined by the strategy mapping unit and the specific execution parameters obtained by the control parameter calculation unit, and then issue the strategy to the transport vehicle.

Images