A scrap metal self-adaptive multi-stage compression capacity reduction control method

Abstract

The application discloses a kind of waste metal self-adapting multistage compression capacity reduction control methods, it is related to waste metal recovery processing technical field technical field, including obtaining the initial profile image of the waste metal stack to be compressed and the displacement response data in pre-compression stroke, based on exposed protruding part and local displacement lag section partition and generate compression unit set Stack;Utilize compression unit adjacency relationship, exposed protruding part cross-zone lap joint relationship and local displacement lag section distribution identify bridging compression unit, closed candidate unit and limited compression unit, construct heterogeneous force transmission diagram;According to the connection direction of heterogeneous force transmission diagram, execute hierarchical heuristic compression, extract delay transmission chain;Delay transmission chain and closed candidate unit, limited compression unit are mapped to generate segmented pressure release window set;Generation adaptive compression sequence.The application realizes the structural mechanics self-adapting regulation and control of waste metal stack compression process, improves the capacity reduction efficiency and block density uniformity.

Description

Technical Field

[0001] This invention relates to the field of waste metal recycling and treatment technology, and in particular to an adaptive multi-stage compression and volume reduction control method for waste metal. Background Technology

[0002] Baling and volume reduction of scrap metal is a crucial pre-processing step in the scrap steel recycling industry chain. Existing scrap metal compression control methods generally employ an integrated compression strategy with a preset fixed stroke or constant pressure threshold, neglecting the non-uniform mechanical transmission structure formed within the compressible pile due to irregular geometry, overlapping sections, and localized voids. In actual compression, the distribution of exposed protrusions and internal closed cavities on the pile surface is highly random, leading to lag in displacement response and localized stress concentration when the indenter load is transmitted along the compression direction. This causes some areas to reach their compaction limit prematurely, forming closed barriers that hinder the transmission of subsequent loads to uncompacted areas within the pile. This disordered load transmission further leads to problems such as redundant compression energy consumption, uneven bale density distribution, increased risk of lateral bulging, and frequent overloads in the baler's hydraulic system, making it difficult to achieve balanced volume reduction across the entire pile within a single compression cycle.

[0003] Furthermore, traditional control methods lack pre-identification and adaptive adjustment mechanisms for the internal mechanical heterogeneity of scrap metal piles. Bridging force transmission paths, the shielding effect of closed elements on loads, and inefficient compression deformation in edge regions due to lack of constraints during compression cannot be effectively distinguished and differentiated within the pre-defined control logic. Existing technologies have not yet proposed a compression element partitioning method based on the collaborative analysis of pile contour features and displacement response data, nor have they established a topological model of the force transmission relationship between bridging compression elements, closed candidate elements, and constrained compression elements. This results in a lack of structural mechanics basis for arranging the multi-stage compression sequence, ultimately affecting the stability of the volume reduction ratio and the quality of the bale forming. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an adaptive multi-stage compression and volume reduction control method for scrap metal. It aims to solve the technical problems of insufficient ability to identify non-uniform mechanical structures within the scrap metal pile, disordered compression load transmission paths, and a lack of structural mechanical basis for the multi-stage compression sequence in existing scrap metal compression and volume reduction control methods. The invention provides a volume reduction control method that can adaptively generate a multi-stage compression sequence based on the initial profile and pre-compression displacement response of the scrap metal pile.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an adaptive multi-stage compression and volume reduction control method for waste metal, comprising the following steps: S1. Obtain the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, partition the scrap metal pile to be compressed to generate a set of compression units. S2. Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, identify bridging compression units, closed candidate units, and confined compression units in the compression unit set, and generate a heterogeneous force transmission map. S3. According to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, perform hierarchical trial compression on the set of compression units, record the order of displacement transmission of each compression unit in adjacent compression stages, and extract the delayed transmission chain based on the order of displacement transmission. S4. Map the delayed transmission chain to the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented release window set with the chain segment boundary as a constraint. S5. Adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. S6. Perform multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, check the closure state of the corresponding chain segment boundary according to the decompression window set, and output the capacity reduction control result.

[0006] This invention achieves dual-modal perception of the structural and mechanical response characteristics of a scrap metal pile by acquiring its initial contour image and indenter displacement response data during the pre-compression stroke. Based on exposed protrusions and local displacement lag segments, it partitions the pile and generates a set of compression units, transforming non-uniform and irregular scrap metal piles into quantifiable discrete unit models. By identifying bridging compression units, closed candidate units, and constrained compression units and generating heterogeneous force transmission maps, it reveals the force transmission paths and constraint relationships between different regions within the pile. Furthermore, by using graded trial compression to extract delayed transmission chains, it further quantifies the transmission sequence of displacement between different compression units. A segmented pressure release window set is generated using chain segment boundaries as constraints, and the multi-level compression sequence is adjusted accordingly. This achieves decoupling of bridging units, subsequent destabilization of closed candidate units, and follow-up compensating compression of constrained units. This method solves problems such as force transmission hindrance, local overload, and rebound dispersion caused by the heterogeneity of the scrap metal pile in traditional scrap metal compression, improving the uniformity and controllability of compression efficiency and volume reduction effect.

[0007] Furthermore, the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke are obtained. Based on the exposed protrusions in the initial contour image and the local displacement lag segments in the pressure head displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units, including: The initial contour image of the scrap metal pile to be compressed is obtained by an image acquisition device, and the boundary contour line of the pile in the initial contour image is extracted by the Canny edge detection operator. Perform convex hull detection on the stack boundary contour line to obtain the stack convex hull boundary. Calculate the non-overlapping area between the stack boundary contour line and the stack convex hull boundary. Mark the stack surface geometry corresponding to the non-overlapping area as the exposed protrusion. According to the effective coverage of the pressure head of the scrap metal baler, the area enclosed by the boundary outline of the pile is divided into multiple adjacent areas, and the boundary of each adjacent area is the boundary of the action area of ​​the pressure head in the pre-compression stroke. During the pre-compression stroke, displacement sensors are used to collect pressure head displacement response data for each adjacent region, and pressure head displacement-time history curves for each adjacent region are constructed. Piecewise linear fitting is performed on the head displacement-time history curves corresponding to each adjacent region. The time intervals in which the head displacement value increases continuously but the ratio of displacement increment to time increment is lower than the average of the ratio in adjacent time periods are preset proportions are extracted. These time intervals are marked as local displacement lag segments in adjacent regions. The exposed protrusion and the local displacement lag segment are bound according to their spatial correspondence in adjacent regions. The adjacent regions are used as basic units, and the location information of the bound exposed protrusion and the time period information of the local displacement lag segment are used as unit attributes to generate a set of compressed units.

[0008] The above scheme accurately extracts exposed protrusions from the scrap metal pile by combining Canny edge detection and convex hull detection, effectively identifying the impact of protruding structures on the compression process. It divides the area according to the effective coverage of the pressure head, matching the spatial scale of the compression unit with the pressure head's action area, facilitating subsequent refined control. By collecting the pressure head displacement-time history curves for each region and performing piecewise linear fitting, it accurately locates local displacement lag segments, identifying time intervals where displacement response is significantly delayed under continuous loading. Exposed protrusions are spatially and temporally bound to local displacement lag segments, assigning structural and response attributes to each compression unit. This provides a reliable data foundation for subsequent heterogeneous force transmission map construction and delayed transmission chain extraction, improving the perception accuracy and adaptability of the entire control method.

[0009] Furthermore, based on the adjacency relationships of each compression unit in the compression unit set, the cross-regional overlapping relationships of exposed protrusions, and the corresponding distribution of local displacement lag segments, the bridging compression units, closed candidate units, and confined compression units in the compression unit set are identified, and a heterogeneous force transmission map is generated, including: Extract the spatial adjacency relationship of each compression unit in the set of compression units within the outline of the stack body. Mark two compression units that share a boundary region as adjacent compression unit pairs. Construct a compression unit adjacency topology graph with compression units as nodes and adjacent compression unit pairs as edges. Based on the location information of the exposed protrusions bound in the set of compression units, identify the geometric structure of the exposed protrusions that cross the boundary of adjacent regions and whose spatial projection simultaneously covers at least two adjacent regions, and mark the compression unit pairs with such geometric structures as cross-regional overlapping compression unit pairs. Based on the local displacement lag segment time information bound in the compression unit set, the start time and end time of the local displacement lag segment corresponding to each compression unit are extracted, and the average value of the ratio of displacement increment to time increment within the local displacement lag segment is calculated. A compressed unit in the adjacent topology graph that meets the following conditions is identified as a bridged compressed unit: there is a cross-regional overlapping compressed unit pair relationship between the compressed unit and at least two non-directly adjacent compressed units, and the start time of the local displacement lag segment corresponding to the compressed unit is earlier than the start time of the local displacement lag segment corresponding to the overlapping compressed unit. Compressed elements whose spatial location is not adjacent to the boundary contour of the stack are identified as internal compressed elements. Compressed elements whose local displacement lag segment start time is later than the local displacement lag segment start time of all their adjacent compressed elements are identified as closed candidate elements. The compression units that are spatially adjacent to the boundary contour of the stack are identified as edge compression units. The compression units in the edge compression units whose average ratio of displacement increment to time increment in the local displacement lag segment is lower than the average ratio of the previous period by more than a preset attenuation threshold are identified as restricted compression units. Using the adjacent topology of the compression units as the base map, the identification results of bridging compression units, cross-regional overlapping compression unit pairs, closed candidate units, and confined compression units are mapped and labeled according to the spatial coordinates of each compression unit within the contour line of the stack body, generating a heterogeneous force transmission map.

[0010] This study constructs an adjacency topology graph of compression units, transforming the spatial structural relationships of the scrap metal pile into a computable graphical model. This lays the topological foundation for subsequent force transmission path analysis. By identifying overlapping compression unit pairs across regions, the bridging effect of force transmission formed by exposed protrusions crossing the boundaries of different regions is accurately captured. Based on this, and combined with the temporal characteristics of local displacement lag segments, discrimination criteria for bridging compression units, closed candidate units, and confined compression units are proposed. Among them, bridging compression units represent the main force transmission nodes in the pile, closed candidate units represent the enclosed and delayed response internal regions, and confined compression units represent the edge regions that are difficult to fully compress due to boundary constraints. Mapping these identification results onto the heterogeneous force transmission diagram generated on the topology graph intuitively and quantitatively reflects the mechanical heterogeneity distribution inside the pile, providing clear guidance for subsequent graded trial compression and compression sequence optimization.

[0011] Furthermore, the step of performing graded trial compression on the set of compression units according to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, recording the displacement transmission order of each compression unit in adjacent compression stages, and extracting the delayed transmission chain based on the displacement transmission order includes: Starting from the centroid coordinates of the adjacent region corresponding to the bridging compression unit in the heterogeneous force transmission diagram, and ending from the centroid coordinates of the adjacent region corresponding to the closed candidate unit that has a cross-regional overlapping compression unit relationship with the bridging compression unit, a direction vector is generated as the direction for the trial suppression connection. Following the order of the bridging compression unit pointing to the closed candidate unit in the direction of the trial pressing connection, the compression load of the pressure head is increased step by step and the displacement response data of the pressure head is collected simultaneously to determine the starting stage of the displacement response; The compression units are sorted from smallest to largest according to the values ​​of the initial displacement response levels. Adjacent compression units in the sort are extracted to construct a directed displacement transfer pair, where the compression unit with the smaller initial displacement value is the transfer start point and the compression unit with the larger initial displacement value is the transfer end point. All displacement transfer directed pairs are arranged in ascending order according to the starting level of the displacement response corresponding to the starting point of the transfer. Displacement transfer directed pairs with a difference of one between the starting level of the displacement response corresponding to the starting point and the ending point of the transfer are retained to obtain the delayed transfer chain.

[0012] This scheme uses the direction from the bridging compression unit to the closed candidate unit in the heterogeneous force transmission diagram as the direction of the trial compression connection. It fully considers the physical law of force propagation from the transmission node to the delayed response region. By progressively increasing the compression load and recording the displacement response initial level of each compression unit, it achieves quantitative detection of the force transmission path inside the reactor body. The displacement response initial level characterizes the load level at which a certain compression unit begins to produce significant compressive deformation. Its magnitude directly reflects the force sequence and constraint degree of the unit in the reactor body. By extracting the displacement transmission directional relationship pairs with an initial level difference of one, a delayed transmission chain is constructed, which accurately describes the order of force transmission between different compression units. This delayed transmission chain provides a direct temporal basis for subsequently determining the chain segment boundary of the first compression and then release, making the division of the release window more scientific and reasonable.

[0013] Furthermore, the stepwise increasing of the indenter compression load and simultaneous acquisition of indenter displacement response data to determine the initial stage of the displacement response includes: The first-level compression load is set as a preset ratio of the maximum compression load value in the pre-compression stroke. The first-level compression load is applied to the compression unit set. After loading, the head displacement data of the adjacent areas of each compression unit are collected. The compression unit whose head displacement increment before and after loading exceeds the displacement response threshold is marked as the first-level response unit. The compression load value is increased by a preset step size as the second-level compression load. The second-level compression load is applied to the remaining compression units after removing the first-level response units in the compression unit set. The head displacement data of the adjacent areas of the remaining compression units after loading are collected. The compression units whose head displacement increment before and after loading exceeds the displacement response threshold are marked as second-level response units. The compression load is increased repeatedly and the remaining compression units are loaded until all compression units in the set of compression units have obtained response unit markings. The compression load level when each compression unit is marked is recorded as the displacement response starting level of that compression unit.

[0014] This scheme achieves precise measurement of the initial displacement response level of each compression unit by designing a trial compression process with progressively increasing loads and tiered loading of remaining unresponsive units. The first-level compression load is set as a preset proportion of the maximum load of the pre-compression stroke, avoiding structural damage due to excessive initial loads or failure to elicit an effective response due to insufficient initial loads. After each loading, only responding units are removed, ensuring that subsequent loads only act on areas that have not yet deformed significantly, avoiding repeated over-compression of responding units. By recording the load level corresponding to the first time each compression unit reaches the displacement response threshold, a quantitative sequence of the response capability of each unit is constructed. This sequence directly reflects the order of stress and activation difficulty of different regions within the reactor body during the compression process, providing high-precision time-series data support for the construction of delayed transmission chains and the division of segmented decompression windows.

[0015] Furthermore, the step of mapping the delayed transmission chain to closed candidate units and confined compression units in the heterogeneous force transfer diagram, determining the chain segment boundaries in the delayed transmission chain that are first compressed and then released, and generating a set of segmented release windows using the chain segment boundaries as constraints, includes: Each compression unit in the delayed transmission chain is associated with the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram. The compression units belonging to the delayed transmission chain and the closed candidate unit are marked as closed transmission nodes, and the compression units belonging to the delayed transmission chain and the confined compression unit are marked as confined transmission nodes. Extract the order of closed-loop transmission nodes and restricted-loop transmission nodes in the delayed transmission chain, and identify adjacent node pairs with closed-loop transmission nodes first and restricted-loop transmission nodes later as chain segment boundaries that are first compressed and then released. A segmented pressure relief window is constructed by taking the displacement response start level corresponding to the closed transmission node in the chain segment boundary as the start level of the pressure relief window and the displacement response start level corresponding to the restricted transmission node as the end level of the pressure relief window. Arrange the segmented pressure relief windows corresponding to all chain segment boundaries in ascending order of their starting level to generate a set of segmented pressure relief windows.

[0016] The above scheme accurately identifies key nodes that simultaneously possess delayed response characteristics and closed or confined characteristics by associating and mapping the compression units in the delayed transmission chain with closed candidate units and confined compression units in the heterogeneous force transmission diagram. Closed transmission nodes represent regions within the stack that are enclosed and have a delayed response, while confined transmission nodes represent regions that are constrained by boundaries and have a sluggish response. By extracting adjacent node pairs where the closed transmission node precedes the confined transmission node, the chain segment boundary of the first compression followed by the second compression is accurately identified. This boundary characterizes the physical turning point where force accumulates and then releases within the stack. Using the displacement response start levels of the two nodes within this boundary as the start and end levels of the pressure release window, the constructed segmented pressure release window set precisely defines the load ranges requiring pressure release verification in different compression stages, providing clear spatiotemporal constraints for subsequent multi-level compression sequence adjustments.

[0017] Furthermore, the construction of a segmented pressure relief window, using the displacement response start level corresponding to the closed transmission node in the chain segment boundary as the start level of the pressure relief window and the displacement response start level corresponding to the restricted transmission node as the end level of the pressure relief window, includes: The starting stage of the displacement response of the compression unit corresponding to the closed transfer node during the graded trial compression is obtained as the starting stage of the pressure relief window. The starting stage of the displacement response of the compression unit corresponding to the restricted transmission node during the graded trial compression is obtained as the ending stage of the pressure relief window. Calculate the numerical difference between the end stage and the beginning stage of the pressure relief window, and use it as the window stage difference of the segmented pressure relief window; By binding the start level, end level, and window level difference of the pressure release window, a single segmented pressure release window is generated. This scheme accurately locates the start and end positions of the pressure release window by obtaining the displacement response start levels of both closed-transfer nodes and confined-transfer nodes. The start level corresponds to the load level at which the closed-transfer node begins to deform, representing the starting point of effective compression in that region. The end level corresponds to the load level at which the confined-transfer node begins to exhibit significant deformation, representing the end point of the response at the end of the force transmission path. The window level difference is obtained by calculating the numerical difference between the two, quantifying the length of the transition interval from compressive force accumulation to release. Binding the start level, end level, and window level difference to generate a single segmented pressure release window ensures that each window not only includes the load range requiring pressure release verification but also implicitly contains information about the force transmission span within the window. This provides a complete parameterized basis for setting load increments and determining the timing of pressure release in subsequent compression sequences.

[0018] Furthermore, the step of adjusting the multi-level suppression order of the compression unit set based on the segmented pressure release window set, performing decoupling suppression on bridging compression units, subsequent destabilization suppression on closed candidate units, and follow-up compensation suppression on constrained compression units to generate an adaptive suppression sequence includes: Based on the starting and ending levels of each segmented pressure release window, the compression level of the bridging compression unit in the heterogeneous force transmission diagram is set to no more than the compression load level corresponding to the starting level of the pressure release window minus one, thus generating a decoupled compression sequence segment. The suppression level of the closed candidate unit in the heterogeneous force transfer diagram is set to the compressive load level between the start level and the end level of the pressure relief window, and a post-stability suppression sequence segment is generated. In the heterogeneous force transmission diagram, the compression level of the confined compression unit is set to be no lower than the compression load level corresponding to the end level of the pressure release window, and a follow-up compensation compression sequence segment is generated. The adaptive suppression sequence is generated by merging the first decoupling suppression sequence segment, the second destabilizing suppression sequence segment, and the third follow-up compensating suppression sequence segment.

[0019] This scheme, based on the constraints of a segmented decompression window set, performs differentiated compression sequence adjustments on three types of key compression units. The bridging compression unit, as the pivot node for force transmission, has its compression level set to no more than one level below the initial stage of the decompression window. This ensures decoupling compression is completed before the decompression window begins, preventing unnecessary force locking in subsequent compression. The closed candidate unit, as the enclosed delayed response region, has its compression level set between the initial and final stages of the decompression window, achieving post-stability failure compression within the window interval and ensuring that this region is activated for compression at the appropriate load stage. The confined compression unit, as the boundary confined region, has its compression level set to no less than the final stage of the decompression window, achieving follow-up compensation compression after the force transmission path is fully established, ensuring sufficient compaction of the edge region. The adaptive compression sequence generated by merging in the order of decoupling, then stability failure, and follow-up compensation optimizes the force transmission sequence within the reactor body, effectively avoiding local overload, stress concentration, and rebound dispersion, significantly improving the overall volume reduction effect.

[0020] Furthermore, the process involves performing multi-stage compression based on an adaptive compression sequence, and after each stage of compression, verifying the closure state of the corresponding chain segment boundaries according to the decompression window set, and outputting the capacity reduction control result, including: Following the order of first decoupling the compression sequence segment, then breaking the stability of the compression sequence segment, and finally compensating for the compression sequence segment in the adaptive compression sequence, the compression load is applied to the compression unit set step by step to complete the multi-stage compression process. After each level of compression load is applied, the compression load level of that level is obtained. The compression level that is equal to the termination level of any segmented pressure release window in the segmented pressure release window set is marked as the pressure release check level. After the pressure relief check stage compression, the closed transfer node in the chain segment boundary to which the window belongs is determined according to the segmented pressure relief window corresponding to the pressure relief window termination stage. The pressure head displacement data of the adjacent area corresponding to the closed transfer node is collected, and the difference between the pressure head displacement data and the pressure head displacement data before loading of the pressure relief window start stage of the same segmented pressure relief window is calculated as the pressure head displacement difference. When the head displacement difference is less than or equal to the closing displacement threshold, the chain segment boundary is determined to be closed, and the next level of compression load loading continues. When the displacement difference of the pressure head is greater than the closed displacement threshold, supplementary compression is performed on the adjacent area corresponding to the closed transfer node before the next level of compression load is applied. The incremental compression load of the supplementary compression is a preset step size. Record the head displacement data and chain segment boundary closure judgment results after each level of compression load is applied, and generate the volume reduction control results.

[0021] After completing multi-stage compression according to the adaptive compression sequence, the above scheme introduces a verification mechanism based on a segmented pressure release window set after each compression stage. By marking the level where the compression load level equals the termination level of any segmented pressure release window as the pressure release verification level, the key load nodes that need to be checked for closure are accurately located. After compression at the pressure release verification level, the indenter displacement data of the closed-end transfer node in the corresponding chain segment boundary is collected, and the difference is calculated with the displacement data before loading at the beginning level of the window to quantitatively evaluate the degree of closure of the chain segment boundary. When the indenter displacement difference is less than or equal to the closure displacement threshold, closure is determined and subsequent compression continues; when it is greater than the threshold, supplementary compression is performed on the corresponding area before the next compression stage, ensuring sufficient closure of the chain segment boundary. This closed-loop verification and compensation mechanism effectively prevents rebound and loosening caused by insufficient compression, ensuring the consistency and stability of the volume reduction results. The final output volume reduction control results provide reliable data for subsequent processes.

[0022] This invention also introduces an adaptive multi-stage compression and volume reduction control system for waste metals, comprising: Compression partitioning module: Used to acquire the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units. Force transmission diagram module: Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, it identifies bridging compression units, closed candidate units, and confined compression units in the compression unit set and generates heterogeneous force transmission diagrams. The transmission chain module is used to perform graded trial compression on the set of compression units according to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, record the displacement transmission order of each compression unit in the adjacent compression stage, and extract the delayed transmission chain based on the displacement transmission order. Segmented pressure relief module: used to map the delayed transmission chain to the closed candidate unit and confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented pressure relief window set with the chain segment boundary as constraint; Suppression sequence module: used to adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. The capacity reduction control module is used to complete multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, it checks the closure state of the corresponding chain segment boundary according to the decompression window set and outputs the capacity reduction control result.

[0023] The aforementioned control system, through the coordinated operation of six functional modules, fully automates the entire process of the control method described above. The compression partitioning module handles initial sensing and unit division; the force transmission diagram module analyzes and visualizes mechanical heterogeneity; the transmission chain module quantifies and detects the force transmission timing; the segmented decompression module divides the decompression window and generates constraints; the compression sequence module adaptively optimizes the compression sequence; and the volume reduction control module performs execution and closed-loop verification. These modules communicate via standardized data interfaces, forming a complete control loop from sensing, analysis, and decision-making to execution and verification. This system can automatically generate optimal compression strategies for scrap metal piles of different shapes and materials without human intervention, improving the intelligence level, compression efficiency, and volume reduction quality of scrap metal baling and processing, while reducing the reliance on operator experience and labor intensity.

[0024] The present invention provides an adaptive multi-stage compression and volume reduction control method and system for waste metal, the beneficial effects of which are mainly reflected in the following aspects: 1. This invention introduces a collaborative analysis mechanism between the initial contour image of the stack and the pre-compression displacement response data to divide the scrap metal stack to be compressed into a set of compression units with clear geometric and mechanical properties. This decouples the macroscopic compression process into a differentiated compression control problem of local structural units. Based on the corresponding binding of exposed protrusions and local displacement lag segments, this method can accurately identify bridging compression units, closed candidate units, and confined compression units in the stack, and generate heterogeneous force transmission diagrams to explicitly characterize the force transmission direction and topological dependence between each compression unit. This preliminary identification step provides a quantitative basis at the structural mechanics level for the subsequent formulation of graded compression strategies, avoiding the blind assumptions about the load transmission path inside the stack by traditional methods.

[0025] 2. Based on the heterogeneous force transfer map and the delayed transmission chain extracted by graded trial suppression, this method further constructs a segmented pressure release window set constrained by the chain segment boundaries, and generates an adaptive suppression sequence consisting of decoupling suppression, subsequent destabilization suppression, and follow-up compensation suppression. This suppression sequence prioritizes disrupting the bridging force transfer path to release the cross-regional overlap constraint, then applies concentrated loads to closed candidate units to achieve internal cavity destabilization and closure, and finally performs follow-up compensation suppression on edge-restricted areas to eliminate density gradients. The above differentiated suppression logic effectively suppresses the negative effects of prematurely shielding loads in closed units, ensures the orderly transmission of compressive loads along the internal structure of the reactor, and improves the volume reduction ratio, bale density uniformity, and energy efficiency of the packing process. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating an adaptive multi-stage compression and volume reduction control method for waste metal according to the present invention. Figure 2This is a functional block diagram of an adaptive multi-stage compression and volume reduction control system for waste metals according to the present invention. Figure 3 This is a schematic diagram of the delayed transmission chain and segmented decompression window generation process of the present invention; Figure 4 This is a flowchart of the stability check process for the pressure relief boundary of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] like Figures 1-4 As shown, this invention provides an adaptive multi-stage compression and volume reduction control method for waste metals, comprising the following steps: S1. Obtain the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, partition the scrap metal pile to be compressed to generate a set of compression units. In this embodiment, an initial contour image of the scrap metal pile to be compressed and pressure head displacement response data during the pre-compression stroke are acquired. Based on the exposed protrusions in the initial contour image and the local displacement lag segments in the pressure head displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units, including: The initial contour image of the scrap metal pile to be compressed is acquired by an image acquisition device, and the boundary contour line of the pile in the initial contour image is extracted by the Canny edge detection operator.

[0029] Specifically, the image acquisition device refers to the industrial camera installed above the feed inlet of the scrap metal baler. The initial contour image refers to the orthographic digital image of the baler body taken without any load applied. The Canny edge detection operator is a conventional algorithm for extracting pixels with abrupt grayscale changes in image processing. The baler body boundary contour refers to the set of closed pixel coordinates of the outer edge of the baler body extracted by this operator.

[0030] Perform convex hull detection on the stack boundary contour line to obtain the stack convex hull boundary. Calculate the non-overlapping area between the stack boundary contour line and the stack convex hull boundary. Mark the stack surface geometry corresponding to the non-overlapping area as the exposed protrusion.

[0031] Specifically, convex hull detection refers to finding the smallest convex polygon containing a set of planar points. The boundary of the convex hull of the stack is the boundary of this smallest convex polygon. The non-overlapping region refers to the area difference between the inside of the convex hull boundary and the outside of the actual boundary of the stack. Exposed protrusions refer to the local protruding structures on the surface of the stack corresponding to this region; these structures bear pressure first in the initial stage of compression.

[0032] Based on the effective coverage of the scrap metal baler's pressure head, the area enclosed by the outline of the stack is divided into multiple adjacent areas, and the boundary of each adjacent area is the boundary of the pressure head's domain during the pre-compression stroke.

[0033] Specifically, the effective coverage area of ​​the pressure head refers to the orthographic projection area of ​​the pressure head end face onto the surface of the reactor body. Adjacent areas refer to sub-regions obtained by non-overlapping subdivision of the area enclosed by the reactor body boundary outline according to this coverage area. The domain boundary refers to the dividing line between adjacent areas that coincides with the projection edge of the pressure head end face.

[0034] During the pre-compression stroke, displacement sensor data is used to collect pressure head displacement response data for each adjacent region, and pressure head displacement-time history curves for each adjacent region are constructed.

[0035] Specifically, the pre-compression stroke refers to the initial compression of the reactor body by the indenter under a preset low load. The displacement sensor is a linear displacement measuring element on the indenter drive mechanism. Indenter displacement response data refers to the sequence of axial displacements of the indenter recorded at a fixed frequency. The indenter displacement-time history curve is the continuous trajectory of displacement changing over time.

[0036] Piecewise linear fitting is performed on the head displacement-time history curves corresponding to each adjacent region. The time intervals in which the head displacement value increases continuously but the ratio of displacement increment to time increment is lower than the average of the ratio in adjacent time periods are preset. These time intervals are marked as local displacement lag segments in adjacent regions.

[0037] Specifically, piecewise linear fitting refers to dividing the curve into approximately straight line segments and determining the slope of each segment using the least squares method. The ratio of displacement increment to time increment is the slope of the fitted straight line segment, which physically represents the instantaneous average propulsion rate. The local displacement lag segment refers to the time interval during which the pressure head displacement remains positive but the propulsion rate decreases compared to adjacent time periods, indicating that the scrap metal in this area is experiencing an increase in resistance caused by the closure of internal cavities or local buckling.

[0038] The exposed protrusion and the local displacement lag segment are bound according to their spatial correspondence in adjacent regions. The adjacent regions are used as basic units, and the location information of the bound exposed protrusion and the time period information of the local displacement lag segment are used as unit attributes to generate a set of compressed units.

[0039] Specifically, spatial correspondence refers to the association between exposed protrusions and local displacement lag segments within the same adjacent region. Binding refers to establishing a data association between the location coordinates of exposed protrusions and the start and end times of local displacement lag segments. The location information of exposed protrusions refers to their coordinates and area description in the stack coordinate system. The time period information of local displacement lag segments refers to their start and end times. A compressed unit refers to a single adjacent region analysis unit carrying geometric and temporal attributes. The set of compressed units is the sum of all compressed units, providing a partitioning basis for subsequent heterogeneous force transfer map construction and hierarchical suppression.

[0040] S2. Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, identify bridging compression units, closed candidate units, and confined compression units in the compression unit set, and generate a heterogeneous force transmission map. In this embodiment, based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlap relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, bridging compression units, closed candidate units, and confined compression units in the compression unit set are identified, and a heterogeneous force transmission map is generated, including: Extract the spatial adjacency relationships of each compressed unit in the compressed unit set within the stack boundary outline, mark two compressed units that share a boundary as adjacent compressed unit pairs, and construct a compressed unit adjacency topology graph with compressed units as nodes and adjacent compressed unit pairs as edges.

[0041] Specifically, spatial adjacency refers to whether two compressed units share the same adjacent region for boundary delineation. An adjacent compressed unit pair is two compressed units that satisfy the shared boundary condition. The compressed unit adjacency topology graph is an undirected graph structure with compressed units as vertices and adjacency relationships as edges, used to describe the mechanical adjacency relationships between compressed units.

[0042] Based on the location information of the exposed protrusions bound in the set of compression units, the geometric structure of the exposed protrusions that cross the boundary of adjacent regions and whose spatial projection simultaneously covers at least two adjacent regions is identified, and the compression unit pairs with such geometric structures are marked as cross-regional overlapping compression unit pairs.

[0043] Specifically, "crossing adjacent area boundaries" refers to the continuous distribution of the geometric projection of the same exposed protrusion across two or more adjacent areas. "Cross-area overlapping compression unit pair" refers to a pair of compression units that are associated with lateral load transfer due to the cross-area distribution of the same exposed protrusion.

[0044] Based on the local displacement lag segment time information bound in the compression unit set, the start and end times of the local displacement lag segment corresponding to each compression unit are extracted, and the average value of the ratio of displacement increment to time increment within the local displacement lag segment is calculated.

[0045] Specifically, the local displacement lag period information refers to the start and end time data bound in the aforementioned steps. The mean value refers to the arithmetic mean of the ratio of displacement increment to time increment in each sampling interval within the lag period, reflecting the overall propulsion rate of the compression unit in the lag phase.

[0046] A compressed unit in the adjacent topology graph that meets the following conditions is identified as a bridged compressed unit: there is a cross-regional overlapping compressed unit pair relationship between the compressed unit and at least two non-directly adjacent compressed units, and the start time of the local displacement lag segment corresponding to the compressed unit is earlier than the start time of the local displacement lag segment corresponding to the overlapping compressed unit.

[0047] Specifically, non-directly adjacent compressed elements refer to two compressed elements that do not have a direct connection in the compressed element adjacency topology graph. Bridged compressed elements refer to compressed elements that mechanically connect multiple non-adjacent elements through cross-regional overlapping relationships, and they exhibit displacement lag earlier in the early stages of compression.

[0048] Compressed elements whose spatial location is not adjacent to the boundary contour of the stack are identified as internal compressed elements. Compressed elements whose local displacement lag segment start time is later than the start time of the local displacement lag segment of all their adjacent compressed elements are identified as closed candidate elements.

[0049] Specifically, internal compression elements refer to compression elements whose geometric extent does not intersect with the boundary contour line of the stack. Closed candidate elements refer to internal compression elements that exhibit displacement hysteresis only after all adjacent elements have entered the displacement hysteresis stage; such elements are prone to forming closed cavities.

[0050] Compression units adjacent to the spatial boundary contour line are identified as edge compression units. Compression units in which the average ratio of displacement increment to time increment in the local displacement lag segment is lower than the average ratio of the previous period by more than a preset attenuation threshold are identified as restricted compression units.

[0051] Specifically, an edge compression element refers to a compression element whose geometric range intersects with the boundary contour line of the stack. The preceding time period refers to the adjacent time period before the occurrence of the local displacement lag segment of this compression element. The preset attenuation threshold is a pre-set rate attenuation judgment threshold. A confined compression element refers to an edge element with insufficient lateral constraint and reduced compression efficiency due to proximity to the free boundary.

[0052] Using the adjacent topology of the compression units as the base map, the identification results of bridging compression units, cross-regional overlapping compression unit pairs, closed candidate units, and confined compression units are mapped and labeled according to the spatial coordinates of each compression unit within the contour line of the stack body, generating a heterogeneous force transmission map.

[0053] Specifically, mapping annotation refers to overlaying the above four types of identification results onto the adjacency topology graph of the compression unit as graphical node markers or edges. The heterogeneous force transmission graph refers to a composite graph model that simultaneously includes the adjacency topology relationship of the compression unit, the cross-regional overlapping relationship, and the unit mechanical type label, providing a structured basis for the connection direction and load transmission path for graded trial compression.

[0054] S3. According to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, perform hierarchical trial compression on the set of compression units, record the order of displacement transmission of each compression unit in adjacent compression stages, and extract the delayed transmission chain based on the order of displacement transmission. In this embodiment, according to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, a graded trial compression is performed on the compression unit set, and the displacement transmission order of each compression unit in adjacent compression stages is recorded. Based on the displacement transmission order, a delayed transmission chain is extracted, including: Starting from the centroid coordinates of the adjacent regions corresponding to the bridging compression unit in the heterogeneous force transmission diagram, and ending from the centroid coordinates of the adjacent regions corresponding to the closed candidate unit that has a cross-regional overlapping compression unit relationship with the bridging compression unit, a direction vector is generated as the direction for probing the compression connection.

[0055] Specifically, centroid coordinates refer to the planar coordinates of the geometric center of the adjacent polygon region. The direction vector refers to the directed line segment from the starting point to the ending point, representing the preferred axial direction of load transfer from the bridging compression element to the closed candidate element.

[0056] Following the order of the bridging compression unit pointing to the closed candidate unit in the direction of the trial pressing connection, the compression load of the pressure head is increased step by step, and the displacement response data of the pressure head is collected simultaneously to determine the starting stage of the displacement response.

[0057] Specifically, the incremental indentation head compression load refers to gradually increasing the indentation head force in preset steps. The displacement response initial level refers to the compression load level corresponding to the first time the indentation head displacement increment exceeds the displacement response threshold during the incremental loading process.

[0058] The compression units are sorted from smallest to largest according to the values ​​of the initial displacement response levels. Adjacent compression units in the sort are extracted to construct directed displacement transfer pairs, where the compression unit with the smaller initial displacement value is the transfer start point and the compression unit with the larger initial displacement value is the transfer end point.

[0059] Specifically, sorting refers to arranging the compression elements in ascending order of their displacement response initiation level. The directional relationship of displacement transfer characterizes the transfer dependence of compressive deformation along the sorting direction.

[0060] All displacement transfer directed relation pairs are arranged in ascending order according to the starting level of the displacement response corresponding to the transfer starting point. Displacement transfer directed relation pairs that satisfy the level difference condition are retained to obtain the delayed transfer chain.

[0061] The grade difference condition is determined by the following formula: In the formula, This represents the initial stage difference in displacement response; The starting stage is used to transmit the displacement response corresponding to the endpoint. To transmit the initial stage of the displacement response corresponding to the starting point. (Reserved) A displacement transfer pair equal to one indicates that the two compression elements respond continuously at adjacent load levels, forming a transfer relationship without jumps. A delayed transfer chain refers to a chain that satisfies... An ordered sequence of directed displacement transfer pairs, which are equal to a given condition, reflects the temporal path of load transfer within the stack.

[0062] In this embodiment, the compression load of the indenter head is increased step by step while the displacement response data of the indenter head is collected simultaneously to determine the initial stage of the displacement response, including: The first-level compression load is set as a preset ratio of the maximum compression load value in the pre-compression stroke. The first-level compression load is applied to the set of compression units. The displacement data of the indenter in the adjacent area of ​​each compression unit after loading is collected. The compression units whose indenter displacement increment before and after loading exceeds the displacement response threshold are marked as first-level response units.

[0063] Specifically, the preset ratio is a pre-defined empirical coefficient. The indentation displacement increment before and after loading refers to the difference between the indentation displacement value after the current load is applied and the indentation displacement value before loading. The first-level response unit refers to the compression unit whose displacement increment under the first-level compressive load exceeds the displacement response threshold.

[0064] The compression load value is increased by a preset step size as the second-level compression load. The second-level compression load is applied to the remaining compression units after removing the first-level response units in the compression unit set. The head displacement data of the adjacent areas of the remaining compression units after loading are collected. The compression units whose head displacement increment before and after loading exceeds the displacement response threshold are marked as second-level response units.

[0065] Specifically, the preset step size refers to the fixed increment value between two adjacent levels of compressive load. The remaining compressive elements refer to the subset of compressive elements that have not yet been marked as response elements. The second-level response element refers to the compressive element that first reaches the displacement response threshold under the second-level compressive load.

[0066] The compression load is increased repeatedly and the remaining compression units are loaded until all compression units in the set of compression units have obtained response unit markings. The compression load level when each compression unit is marked is recorded as the displacement response starting level of that compression unit.

[0067] Specifically, the compressive load level refers to the sequence number of the currently applied load in the increasing sequence. The displacement response initiation level is the compressive load level corresponding to the first time a compressive element is marked as a response element. This value reflects the sensitivity of different regions within the reactor body to compressive loads and the differences in deformation timing.

[0068] S4. Map the delayed transmission chain to the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented release window set with the chain segment boundary as a constraint. In this embodiment, the delayed transmission chain is mapped to the closed candidate units and confined compression units in the heterogeneous force transmission diagram to determine the chain segment boundaries in the delayed transmission chain that are first compressed and then released. A set of segmented release windows is generated using these chain segment boundaries as constraints, including: Each compression unit in the delayed transmission chain is associated with the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram. The compression unit that belongs to both the delayed transmission chain and the closed candidate unit is marked as a closed transmission node, and the compression unit that belongs to both the delayed transmission chain and the confined compression unit is marked as a confined transmission node.

[0069] Specifically, the association refers to comparing and matching the compression element numbers in the delayed transmission chain with the closed candidate element numbers and constrained compression element numbers already marked in the heterogeneous force transmission diagram. A closed transmission node refers to a compression element that simultaneously satisfies the closed candidate element attribute and is located on the delayed transmission chain. A constrained transmission node refers to a compression element that simultaneously satisfies the constrained compression element attribute and is located on the delayed transmission chain.

[0070] Extract the order of closed-loop transmission nodes and restricted-loop transmission nodes in the delayed transmission chain, and identify adjacent node pairs with closed-loop transmission nodes first and restricted-loop transmission nodes later as chain segment boundaries where pressure is applied first and then released.

[0071] Specifically, the order of arrangement refers to the sequential arrangement of closed-loop and restricted-loop nodes in the delayed-loop chain. Adjacent node pairs refer to combinations of closed-loop and restricted-loop nodes that are adjacent and appear sequentially in the delayed-loop chain. The chain segment boundary where pressure is applied first and then released refers to the endpoint of the path segment transitioning from a closed-loop node to a restricted-loop node.

[0072] A segmented pressure relief window is constructed by taking the starting level of the displacement response corresponding to the closed transmission node in the chain segment boundary as the starting level of the pressure relief window and the starting level of the displacement response corresponding to the restricted transmission node as the ending level of the pressure relief window.

[0073] The window increments for segmented decompression windows are determined by the following formula: In the formula, The window level difference for segmented decompression windows; The final displacement response level of the confined transfer node is the initial level of the compression load recorded during the first deformation of the confined transfer node in the graded trial compression. This represents the initial displacement response level corresponding to the closed-loop transfer node, i.e., the compressive load level recorded during the first deformation of the closed-loop transfer node in the graded trial compression. The segmented pressure release window refers to the compressive load level from... to Within a certain range, the corresponding chain segment region is in a transitional state where it is first compressed and closed, and then needs to be released and adjusted.

[0074] Arrange the segmented pressure relief windows corresponding to all chain segment boundaries in ascending order of their starting level to generate a set of segmented pressure relief windows.

[0075] Specifically, ascending order means arranging each segmental pressure release window in ascending order of its initial pressure release level value. The segmental pressure release window set is an ordered set containing the segmental pressure release windows corresponding to all chain segment boundaries, providing a constraint basis for adjusting the pressure release timing for multi-level suppression sequence adjustments.

[0076] In this embodiment, the construction of a segmented pressure relief window, using the starting level of the displacement response corresponding to the closed transmission node in the chain segment boundary as the starting level of the pressure relief window and the starting level of the displacement response corresponding to the restricted transmission node as the ending level of the pressure relief window, includes: The starting stage of the displacement response of the compression unit corresponding to the closed transfer node during the graded trial compression is obtained as the starting stage of the pressure relief window.

[0077] Specifically, "acquisition" refers to extracting the initial displacement response level value corresponding to a specified compression unit from the stored compression unit response record table during the graded trial compression phase. The initial level of the pressure relief window is the number of the initial compression load level of the segmented pressure relief window.

[0078] The starting stage of the displacement response of the compression unit corresponding to the restricted transmission node during the graded trial compression is obtained as the ending stage of the pressure relief window.

[0079] Specifically, the pressure relief window termination level is the number of the end compression load level of the segmented pressure relief window.

[0080] Calculate the numerical difference between the termination stage and the beginning stage of the pressure relief window, and use it as the window stage difference for the segmented pressure relief window.

[0081] Specifically, the numerical difference refers to the integer difference between the end stage number and the beginning stage number of the pressure relief window. The window stage difference represents the number of compressive load levels spanned by the segmented pressure relief window.

[0082] Bind the start level, end level, and window level difference of the decompression window to generate a single segmented decompression window.

[0083] Specifically, binding refers to combining the above three values ​​into a single data structure. A single segmented decompression window is a complete decompression interval definition generated for a certain chain segment boundary, including the start and end levels and the number of levels crossed.

[0084] S5. Adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. In this embodiment, the multi-level suppression order of the compression unit set is adjusted according to the segmented pressure release window set. Decoupling suppression is performed first on bridging compression units, followed by destabilization suppression on closed candidate units, and follow-up compensation suppression is performed on constrained compression units, generating an adaptive suppression sequence, including: Based on the starting and ending levels of each segmented pressure release window, the compression level of the bridging compression unit in the heterogeneous force transmission diagram is set to no more than the compression load level corresponding to the starting level of the pressure release window minus one, thus generating a decoupled compression sequence segment.

[0085] Specifically, the initial stage of the pressure relief window is the displacement response initial stage number corresponding to the closed transmission node. "First decoupling suppression" refers to applying load to the bridging compression element before the closed element enters the response, in order to release the cross-regional overlapping force transmission constraint. The "first decoupling suppression sequence" refers to the suppression order of the bridging compression elements at the corresponding load level.

[0086] The suppression level of the closed candidate element in the heterogeneous force transmission diagram is set to a compressive load level between the start and end stages of the pressure relief window, generating a post-stability suppression sequence segment.

[0087] Specifically, post-stability suppression refers to applying a concentrated load to a closed candidate element after the bridging element has been decoupled, in order to destroy its internal closed cavity. The post-stability suppression sequence segment is the order in which the closed candidate elements are suppressed within this load range.

[0088] The compression level of the confined compression unit in the heterogeneous force transmission diagram is set to be no lower than the compression load level corresponding to the end level of the pressure relief window, and a follow-up compensation compression sequence segment is generated.

[0089] Specifically, follow-up compensation suppression refers to applying supplementary loads to edge-confined elements after the closed elements have become unstable, in order to eliminate the density gradient. The follow-up compensation suppression sequence segment is the order in which confined compression elements are suppressed at the corresponding load levels.

[0090] The adaptive suppression sequence is generated by merging the first decoupling suppression sequence segment, the second destabilizing suppression sequence segment, and the third follow-up compensating suppression sequence segment.

[0091] Specifically, merging refers to splicing three sequence segments into a complete compression instruction set in chronological order. The adaptive compression sequence is a multi-level compression control instruction sequence generated based on the segmented release window set, which executes differentiated compression timings for the three types of compression units.

[0092] S6. Perform multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, check the closure state of the corresponding chain segment boundary according to the decompression window set, and output the capacity reduction control result.

[0093] In this embodiment, multi-level compression is performed based on an adaptive compression sequence. After each level of compression, the closure state of the corresponding chain segment boundary is checked according to the segmented decompression window set, and the capacity reduction control result is output, including: Following the order of first decoupling the compression sequence segment, then breaking the stability of the compression sequence segment, and finally compensating for the compression sequence segment in the adaptive compression sequence, the compression load is applied to the compression unit set step by step to complete the multi-stage compression process.

[0094] Specifically, stepwise loading refers to applying pressure sequentially according to the load levels specified in the adaptive compression sequence. The multi-stage compression process is the complete compression process from the first stage of compression load to the completion of all stages.

[0095] After each level of compressive load is applied, the level of compressive load for that level is obtained. The level of compressive load that is equal to the termination level of any segmented pressure release window in the segmented pressure release window set is marked as the pressure release check level.

[0096] Specifically, the compression load level refers to the number of the currently applied load in the incrementing sequence. The decompression check level refers to the load level for which the closure status of the chain segment boundaries needs to be checked.

[0097] After the compression of the pressure relief check stage is completed, the closed transfer node in the chain segment boundary to which the window belongs is determined according to the segmented pressure relief window corresponding to the pressure relief window termination stage. The pressure head displacement data of the adjacent area corresponding to the closed transfer node is collected, and the pressure head displacement difference is calculated.

[0098] The displacement difference of the pressure head is calculated using the following formula: ΔD = D check - D before start In the formula, ΔD is the head displacement difference, representing the cumulative compression deformation in the closed-end transfer node region within the pressure relief window interval. Dcheck is the head displacement value collected after the pressure relief check stage compression is completed. Dbefore is the last head displacement value recorded before loading at the beginning stage of the pressure relief window in the same segmented pressure relief window.

[0099] When ΔD is less than or equal to the closing displacement threshold, the chain segment boundary is determined to be closed, and the next level of compressive load loading continues.

[0100] Specifically, the closure displacement threshold is a pre-set displacement difference judgment threshold used to determine whether the boundary area of ​​the chain segment has reached a dense closure state.

[0101] When ΔD is greater than the closed displacement threshold, supplementary suppression is performed on the adjacent area corresponding to the closed transfer node before the next level of compressive load is applied. The incremental compressive load of the supplementary suppression is a preset step size.

[0102] Specifically, supplementary compression refers to adding an extra local compression operation to the region outside of the original compression sequence. The preset step size is the fixed increment value of the compression load between two adjacent stages in the graded trial compression.

[0103] Record the head displacement data and chain segment boundary closure judgment results after each level of compression load is applied, and generate the volume reduction control results.

[0104] Specifically, the volume reduction control results include the displacement history of the indenter under each level of compression load, the boundary closure status of each chain segment, and supplementary compression execution records, which are used to characterize the complete state and compaction quality of this compression and volume reduction process.

[0105] Please see Figure 2 The present invention also provides an adaptive multi-stage compression and volume reduction control system for waste metals, comprising: Compression partitioning module: Used to acquire the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units. Force transmission diagram module: Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, it identifies bridging compression units, closed candidate units, and confined compression units in the compression unit set and generates heterogeneous force transmission diagrams. The transmission chain module is used to perform graded trial compression on the set of compression units according to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, record the displacement transmission order of each compression unit in the adjacent compression stage, and extract the delayed transmission chain based on the displacement transmission order. Segmented pressure relief module: used to map the delayed transmission chain to the closed candidate unit and confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented pressure relief window set with the chain segment boundary as constraint; Suppression sequence module: used to adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. The capacity reduction control module is used to complete multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, it checks the closure state of the corresponding chain segment boundary according to the decompression window set and outputs the capacity reduction control result.

[0106] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0107] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for adaptive multi-stage compression and volume reduction control of scrap metal, characterized in that, Includes the following steps: S1. Obtain the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, partition the scrap metal pile to be compressed to generate a set of compression units. S2. Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, identify bridging compression units, closed candidate units, and confined compression units in the compression unit set, and generate a heterogeneous force transmission map. S3. According to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, perform hierarchical trial compression on the set of compression units, record the order of displacement transmission of each compression unit in adjacent compression stages, and extract the delayed transmission chain based on the order of displacement transmission. S4. Map the delayed transmission chain to the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented release window set with the chain segment boundary as a constraint. S5. Adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. S6. Perform multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, check the closure state of the corresponding chain segment boundary according to the decompression window set, and output the capacity reduction control result.

2. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 1, characterized in that, The process involves acquiring the initial contour image of the scrap metal pile to be compressed and the indenter displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segments in the indenter displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units, including: The initial contour image of the scrap metal pile to be compressed is obtained by an image acquisition device, and the boundary contour line of the pile in the initial contour image is extracted by the Canny edge detection operator. Perform convex hull detection on the stack boundary contour line to obtain the stack convex hull boundary. Calculate the non-overlapping area between the stack boundary contour line and the stack convex hull boundary. Mark the stack surface geometry corresponding to the non-overlapping area as the exposed protrusion. According to the effective coverage of the pressure head of the scrap metal baler, the area enclosed by the boundary outline of the pile is divided into multiple adjacent areas, and the boundary of each adjacent area is the boundary of the action area of ​​the pressure head in the pre-compression stroke. During the pre-compression stroke, displacement sensors are used to collect pressure head displacement response data for each adjacent region, and pressure head displacement-time history curves for each adjacent region are constructed. Piecewise linear fitting is performed on the head displacement-time history curves corresponding to each adjacent region. The time intervals in which the head displacement value increases continuously but the ratio of displacement increment to time increment is lower than the average of the ratio in adjacent time periods are preset proportions are extracted. These time intervals are marked as local displacement lag segments in adjacent regions. The exposed protrusion and the local displacement lag segment are bound according to their spatial correspondence in adjacent regions. The adjacent regions are used as basic units, and the location information of the bound exposed protrusion and the time period information of the local displacement lag segment are used as unit attributes to generate a set of compressed units.

3. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 1, characterized in that, The method, based on the adjacency relationships of each compression unit in the compression unit set, the cross-regional overlapping relationships of exposed protrusions, and the corresponding distribution of local displacement lag segments, identifies bridging compression units, closed candidate units, and confined compression units in the compression unit set, and generates a heterogeneous force transmission map, including: Extract the spatial adjacency relationship of each compression unit in the set of compression units within the outline of the stack body. Mark two compression units that share a boundary region as adjacent compression unit pairs. Construct a compression unit adjacency topology graph with compression units as nodes and adjacent compression unit pairs as edges. Based on the location information of the exposed protrusions bound in the set of compression units, identify the geometric structure of the exposed protrusions that cross the boundary of adjacent regions and whose spatial projection simultaneously covers at least two adjacent regions, and mark the compression unit pairs with such geometric structures as cross-regional overlapping compression unit pairs. Based on the local displacement lag segment time information bound in the compression unit set, the start time and end time of the local displacement lag segment corresponding to each compression unit are extracted, and the average value of the ratio of displacement increment to time increment within the local displacement lag segment is calculated. A compressed unit in the adjacent topology graph that meets the following conditions is identified as a bridged compressed unit: there is a cross-regional overlapping compressed unit pair relationship between the compressed unit and at least two non-directly adjacent compressed units, and the start time of the local displacement lag segment corresponding to the compressed unit is earlier than the start time of the local displacement lag segment corresponding to the overlapping compressed unit. Compressed elements whose spatial location is not adjacent to the boundary contour of the stack are identified as internal compressed elements. Compressed elements whose local displacement lag segment start time is later than the local displacement lag segment start time of all their adjacent compressed elements are identified as closed candidate elements. The compression units that are spatially adjacent to the boundary contour of the stack are identified as edge compression units. The compression units in the edge compression units whose average ratio of displacement increment to time increment in the local displacement lag segment is lower than the average ratio of the previous period by more than a preset attenuation threshold are identified as restricted compression units. Using the adjacent topology of the compression units as the base map, the identification results of bridging compression units, cross-regional overlapping compression unit pairs, closed candidate units, and confined compression units are mapped and labeled according to the spatial coordinates of each compression unit within the contour line of the stack body, generating a heterogeneous force transmission map.

4. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 1, characterized in that, The step involves performing graded trial compression on the set of compression units according to the connection direction between the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, recording the displacement transmission order of each compression unit in adjacent compression stages, and extracting the delayed transmission chain based on the displacement transmission order, including: Starting from the centroid coordinates of the adjacent region corresponding to the bridging compression unit in the heterogeneous force transmission diagram, and ending from the centroid coordinates of the adjacent region corresponding to the closed candidate unit that has a cross-regional overlapping compression unit relationship with the bridging compression unit, a direction vector is generated as the direction for the trial suppression connection. Following the order of the bridging compression unit pointing to the closed candidate unit in the direction of the trial pressing connection, the compression load of the pressure head is increased step by step and the displacement response data of the pressure head is collected simultaneously to determine the starting stage of the displacement response; The compression units are sorted from smallest to largest according to the values ​​of the initial displacement response levels. Adjacent compression units in the sort are extracted to construct a directed displacement transfer pair, where the compression unit with the smaller initial displacement value is the transfer start point and the compression unit with the larger initial displacement value is the transfer end point. All displacement transfer directed pairs are arranged in ascending order according to the starting level of the displacement response corresponding to the starting point of the transfer. Displacement transfer directed pairs with a difference of one between the starting level of the displacement response corresponding to the starting point and the ending point of the transfer are retained to obtain the delayed transfer chain.

5. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 4, characterized in that, The stepwise increasing of the indenter head compression load and simultaneous acquisition of indenter head displacement response data to determine the initial stage of the displacement response include: The first-level compression load is set as a preset ratio of the maximum compression load value in the pre-compression stroke. The first-level compression load is applied to the compression unit set. After loading, the head displacement data of the adjacent areas of each compression unit are collected. The compression unit whose head displacement increment before and after loading exceeds the displacement response threshold is marked as the first-level response unit. The compression load value is increased by a preset step size as the second-level compression load. The second-level compression load is applied to the remaining compression units after removing the first-level response units in the compression unit set. The head displacement data of the adjacent areas of the remaining compression units after loading are collected. The compression units whose head displacement increment before and after loading exceeds the displacement response threshold are marked as second-level response units. The compression load is increased repeatedly and the remaining compression units are loaded until all compression units in the set of compression units have obtained response unit markings. The compression load level when each compression unit is marked is recorded as the displacement response starting level of that compression unit.

6. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 1, characterized in that, The step of mapping the delayed transmission chain to closed candidate units and confined compression units in the heterogeneous force transfer diagram, determining the chain segment boundaries in the delayed transmission chain that are first compressed and then released, and generating a set of segmented release windows using the chain segment boundaries as constraints includes: Each compression unit in the delayed transmission chain is associated with the closed candidate unit and the confined compression unit in the heterogeneous force transmission diagram. The compression units belonging to the delayed transmission chain and the closed candidate unit are marked as closed transmission nodes, and the compression units belonging to the delayed transmission chain and the confined compression unit are marked as confined transmission nodes. Extract the order of closed-loop transmission nodes and restricted-loop transmission nodes in the delayed transmission chain, and identify adjacent node pairs with closed-loop transmission nodes first and restricted-loop transmission nodes later as chain segment boundaries that are first compressed and then released. A segmented pressure relief window is constructed by taking the displacement response start level corresponding to the closed transmission node in the chain segment boundary as the start level of the pressure relief window and the displacement response start level corresponding to the restricted transmission node as the end level of the pressure relief window. Arrange the segmented pressure relief windows corresponding to all chain segment boundaries in ascending order of their starting level to generate a set of segmented pressure relief windows.

7. The adaptive multi-stage compression and volume reduction control method for scrap metal according to claim 6, characterized in that, The segmented pressure relief window is constructed by using the displacement response start level corresponding to the closed transmission node in the chain segment boundary as the start level of the pressure relief window and the displacement response start level corresponding to the restricted transmission node as the end level of the pressure relief window, including: The starting stage of the displacement response of the compression unit corresponding to the closed transfer node during the graded trial compression is obtained as the starting stage of the pressure relief window. The starting stage of the displacement response of the compression unit corresponding to the restricted transmission node during the graded trial compression is obtained as the ending stage of the pressure relief window. Calculate the numerical difference between the end stage and the beginning stage of the pressure relief window, and use it as the window stage difference of the segmented pressure relief window; Bind the start level, end level, and window level difference of the decompression window to generate a single segmented decompression window.

8. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 1, characterized in that, The process of adjusting the multi-level suppression order of the compression unit set based on the segmented pressure release window set, performing decoupling suppression on bridging compression units, subsequent destabilization suppression on closed candidate units, and follow-up compensation suppression on constrained compression units, generates an adaptive suppression sequence, including: Based on the starting and ending levels of each segmented pressure release window, the compression level of the bridging compression unit in the heterogeneous force transmission diagram is set to no more than the compression load level corresponding to the starting level of the pressure release window minus one, thus generating a decoupled compression sequence segment. The suppression level of the closed candidate unit in the heterogeneous force transfer diagram is set to the compressive load level between the start level and the end level of the pressure relief window, and a post-stability suppression sequence segment is generated. In the heterogeneous force transmission diagram, the compression level of the confined compression unit is set to be no lower than the compression load level corresponding to the end level of the pressure release window, and a follow-up compensation compression sequence segment is generated. The adaptive suppression sequence is generated by merging the first decoupling suppression sequence segment, the second destabilizing suppression sequence segment, and the third follow-up compensating suppression sequence segment.

9. The adaptive multi-stage compression and volume reduction control method for waste metal according to claim 8, characterized in that, The process involves multi-stage compression based on an adaptive compression sequence, and after each compression stage, the closure state of the corresponding chain segment boundaries is checked according to the decompression window set. The resulting volume reduction control output includes: Following the order of first decoupling the compression sequence segment, then breaking the stability of the compression sequence segment, and finally compensating for the compression sequence segment in the adaptive compression sequence, the compression load is applied to the compression unit set step by step to complete the multi-stage compression process. After each level of compression load is applied, the compression load level of that level is obtained. The compression level that is equal to the termination level of any segmented pressure release window in the segmented pressure release window set is marked as the pressure release check level. After the pressure relief check stage compression, the closed transfer node in the chain segment boundary to which the window belongs is determined according to the segmented pressure relief window corresponding to the pressure relief window termination stage. The pressure head displacement data of the adjacent area corresponding to the closed transfer node is collected, and the difference between the pressure head displacement data and the pressure head displacement data before loading of the pressure relief window start stage of the same segmented pressure relief window is calculated as the pressure head displacement difference. When the displacement difference of the pressure head is less than or equal to the closing displacement threshold, it is determined that the boundary of the chain segment is closed, and the next level of compression load loading continues; When the displacement difference of the pressure head is greater than the closed displacement threshold, supplementary compression is performed on the adjacent area corresponding to the closed transfer node before the next level of compression load is applied. The incremental compression load of the supplementary compression is a preset step size. Record the head displacement data and chain segment boundary closure judgment results after each level of compression load is applied, and generate the volume reduction control results.

10. A waste metal adaptive multi-stage compression and volume reduction control system, used in conjunction with the waste metal adaptive multi-stage compression and volume reduction control method according to any one of claims 1 to 9, characterized in that, include: Compression partitioning module: Used to acquire the initial contour image of the scrap metal pile to be compressed and the pressure head displacement response data during the pre-compression stroke. Based on the exposed protrusions in the initial contour image and the local displacement lag segment in the pressure head displacement response data, the scrap metal pile to be compressed is partitioned to generate a set of compression units. Force transmission diagram module: Based on the adjacency relationship of each compression unit in the compression unit set, the cross-regional overlapping relationship of exposed protrusions, and the corresponding distribution of local displacement lag segments, it identifies bridging compression units, closed candidate units, and confined compression units in the compression unit set and generates heterogeneous force transmission diagrams. The transmission chain module is used to perform graded trial compression on the set of compression units according to the connection direction of the bridging compression unit and the closed candidate unit in the heterogeneous force transmission diagram, record the displacement transmission order of each compression unit in the adjacent compression stage, and extract the delayed transmission chain based on the displacement transmission order. Segmented pressure relief module: used to map the delayed transmission chain to the closed candidate unit and confined compression unit in the heterogeneous force transmission diagram, determine the chain segment boundary in the delayed transmission chain that is first compressed and then released, and generate a segmented pressure relief window set with the chain segment boundary as constraint; Suppression sequence module: used to adjust the multi-level suppression order of the compression unit set according to the segmented pressure release window set, perform decoupling suppression on the bridging compression unit, perform destabilization suppression on the closed candidate unit, and perform follow-up compensation suppression on the restricted compression unit to generate an adaptive suppression sequence. The capacity reduction control module is used to complete multi-level compression based on the adaptive compression sequence, and after each level of compression is completed, it checks the closure state of the corresponding chain segment boundary according to the decompression window set and outputs the capacity reduction control result.

Images