Cloud-based collaborative management and sorting methods and systems for waste recycling

By using a fuzzy inference model optimized by real-time data acquisition and genetic algorithm, combined with dynamic cost graphs and an improved Dijkstra algorithm, adaptive optimization of waste collection paths and dynamic adjustment of sorting lines were achieved. This solved the problems of static decision-making models and fragmented collaborative levels in existing waste recycling systems, and improved the intelligence level and efficiency of waste recycling and sorting.

CN121766713BActive Publication Date: 2026-11-10浙江虎哥废物管理有限公司
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202610001432.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-01-04
Publication Date
2026-11-10
Estimated Expiration
2046-01-04

AI Technical Summary

Technical Problem

Existing cloud-based collaborative management and sorting methods for waste recycling suffer from static decision-making models, fragmented collaboration levels, and insufficient ability to handle uncertainties. This leads to scheduling failures or a decline in overall system efficiency, making it difficult to support the intelligent waste recycling and sorting needs in large-scale, complex urban environments.

Method used

By collecting real-time status data from multiple waste disposal sites, a fuzzy inference model optimized by a genetic algorithm is used to predict the risk of waste overflow. A dynamic cost map is constructed by combining geographical location and real-time traffic information. An improved Dijkstra algorithm is used to optimize the waste collection path, and the waste flow on the sorting line is monitored by Euler equations, so as to achieve dynamic coordination and overall optimization of waste collection and sorting.

Benefits of technology

It has improved the timeliness of garbage collection and the efficiency of resource utilization, reduced garbage overflow and ineffective collection, and enhanced the intelligence level and operational reliability of the garbage recycling and sorting process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121766713B_ABST
    Figure CN121766713B_ABST
Patent Text Reader

Abstract

This invention provides a cloud-based collaborative management and sorting method and system for waste recycling, relating to the field of waste management technology. The method includes: calculating the waste filling rate and waste filling speed of each waste container; uploading the waste filling rate and waste filling speed to a cloud management platform; generating a waste collection priority set using a fuzzy inference model; constructing a dynamic cost graph and path cost function based on the waste collection priority set, combined with the geographical location information of waste disposal sites and real-time traffic information; solving the dynamic cost graph using Dijkstra's algorithm to determine the waste collection and transportation path; transporting waste from each waste disposal site to the sorting line by controlling waste collection vehicles according to the waste collection and transportation path; monitoring the waste flow rate and waste density on the sorting line; and optimizing the fuzzy inference model, path cost function, and flow balance model based on waste collection process data and waste sorting results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of waste management technology, and in particular to a cloud-based collaborative management and sorting method and system for waste recycling. Background Technology

[0002] With the acceleration of urbanization and the continuous increase in the amount of household waste generated, the traditional waste recycling model that relies on manual inspection and fixed-cycle collection is no longer able to meet the needs of refined urban management. On the one hand, there are many waste collection sites, which are scattered and waste generation has obvious spatial and temporal unevenness. On the other hand, the implementation of waste sorting policies has placed higher demands on front-end sensing, collection and dispatching, and back-end sorting coordination.

[0003] By using multi-site status sensing, cloud-based intelligent analysis, and vehicle route optimization, dynamic coordination and overall optimization of the waste collection and sorting process can be achieved, which helps to improve the timeliness of waste collection and resource utilization efficiency, and reduce problems such as waste overflow, empty runs, and ineffective collection.

[0004] Existing cloud-based collaborative management and sorting methods for waste recycling generally suffer from static decision-making models, fragmented collaboration hierarchies, and insufficient ability to handle uncertainties. On the one hand, waste collection route planning is often based on fixed thresholds or single time costs, making it difficult to comprehensively consider changes in waste filling levels, overflow risks, and traffic dynamics. On the other hand, collection scheduling and sorting control are often independent of each other, lacking cross-stage status feedback and joint optimization mechanisms. Furthermore, existing methods lack robustness to sensor errors, traffic fluctuations, and the randomness of waste generation, which can easily lead to scheduling failures or a decrease in overall system efficiency, making it difficult to support the intelligent waste recycling and sorting needs in large-scale, complex urban environments. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a cloud-based collaborative management and sorting method for waste recycling, which can solve the technical problems that existing cloud-based collaborative management and sorting methods for waste recycling generally suffer from static decision-making models, fragmented collaborative levels, and insufficient ability to cope with uncertainties, resulting in scheduling results failure or overall system efficiency decline, making it difficult to support the intelligent waste recycling and sorting needs in large-scale, complex urban environments.

[0006] A first aspect of this invention proposes a cloud-based collaborative management and sorting method for waste recycling, comprising: S1: Collect real-time status data from multiple waste disposal sites and calculate the waste filling rate and waste filling speed of each waste container based on the real-time status data; S2: Upload the waste filling rate and waste filling speed to the cloud management platform; S3: Using a fuzzy inference model optimized by a genetic algorithm, predict the risk of garbage overflow in each garbage container within a preset time window, and generate a garbage collection priority set based on the risk of garbage overflow. S4: Based on the garbage collection priority set, combined with the geographical location information of garbage collection sites and real-time traffic information, construct a dynamic cost graph and path cost function; S5: With the goal of minimizing the path cost function, the dynamic cost graph is solved using the improved Dijkstra algorithm to determine the waste collection and transportation path; S6: Based on the waste collection and transportation route, control the waste collection vehicles to transport the waste from each waste disposal site to the sorting line, and simultaneously update the status information of the waste disposal sites on the cloud management platform; S7: By using a flow balance model based on the Euler equation, the waste flow rate and density on the sorting line are monitored, and the operating parameters of the sorting line are dynamically adjusted according to the monitoring results to automatically separate different waste components according to the preset sorting process sequence and generate waste sorting results; S8: A cloud-based management platform that optimizes fuzzy inference models, path cost functions, and flow balance models based on Euler equations, using waste collection process data and waste sorting results.

[0007] A second aspect of this invention provides a cloud-based collaborative management and sorting system for waste recycling, comprising: a processor and a memory; The memory stores programs or instructions that can run on the processor, and when the programs or instructions are executed by the processor, they implement the steps of the cloud-based collaborative management and sorting method for waste recycling, as described in the first aspect.

[0008] A third aspect of the present invention provides a readable storage medium on which a program or instructions are stored, and when the program or instructions are executed by a processor, the steps of the cloud-based collaborative management and sorting method for garbage collection as described in the first aspect are implemented.

[0009] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, the risk of garbage overflow is predicted by real-time perception and centralized cloud analysis of the operational status of garbage collection sites, combined with a fuzzy inference model optimized by a genetic algorithm. Based on this, a path cost function that integrates traffic conditions and dynamic characteristics of garbage is constructed to achieve adaptive optimization of garbage collection paths. At the same time, by collaboratively modeling and implementing closed-loop feedback control of the collection process and the sorting line operation status, sorting parameters are dynamically adjusted to match actual garbage flow changes. This reduces garbage overflow and ineffective collection, while improving vehicle utilization and the stability and processing efficiency of the sorting system, thereby comprehensively improving the intelligence level, operational reliability, and resource utilization efficiency of the garbage recycling and sorting process. Attached Figure Description

[0010] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0011] Figure 1 This is a flowchart illustrating the cloud-based collaborative management and sorting method for waste recycling provided in this embodiment of the invention.

[0012] Figure 2 This is a schematic diagram of the cloud-based collaborative management and sorting system for waste recycling provided in an embodiment of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions 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, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] The cloud-based collaborative management and sorting method for waste recycling provided by the present invention will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0015] Reference manual attached Figure 1 The diagram illustrates a flowchart of the cloud-based collaborative management and sorting method for waste recycling provided in an embodiment of the present invention.

[0016] This invention provides a cloud-based collaborative management and sorting method for waste recycling, which may include the following steps: S1: Collect real-time status data from multiple waste disposal sites and calculate the waste filling rate and waste filling speed of each waste container based on the real-time status data.

[0017] Among them, a waste disposal site refers to a physical location set up in urban public areas, residential communities, or commercial areas for the centralized disposal of household waste, equipped with waste containers and status sensing devices. Real-time status data refers to operational data continuously collected and periodically updated by sensors deployed in waste containers or sites, including at least the height of the waste container and timestamps, used to characterize the current loading status of the waste. Waste filling rate refers to the ratio between the space occupied by waste in the waste container and the effective volume of the waste container, used to reflect the current degree of fullness of the waste container. Waste filling rate is the ratio between the change in waste filling rate over a continuous time interval and the corresponding time length, used to characterize the speed characteristics of waste generation and accumulation.

[0018] It should be noted that by synchronously collecting real-time status data from multiple waste disposal sites and calculating the waste filling rate and rate based on the real-time status data, this step can transform the traditional waste status perception method that relies on manual inspection or fixed-period statistics into a continuous, quantitative, and time-resolution dynamic monitoring method. This can not only accurately depict the actual load level and changing trend of waste generated at different sites, but also provide a reliable data foundation for subsequent waste overflow risk prediction, collection priority determination, and route optimization, thereby significantly improving the timeliness, accuracy, and overall intelligence level of waste management decision-making.

[0019] In one possible implementation, real-time status data includes the height of the waste container, timestamps, ultrasonic echo time difference, and sound velocity.

[0020] Specifically, an ultrasonic sensor installed inside the waste container emits ultrasonic pulse signals into the waste container and receives the echo signals formed by the reflection of the ultrasonic pulse signals on the surface of the waste. The time difference between the emission time of the ultrasonic pulse signal and the reception time of the echo is collected as the ultrasonic echo time difference.

[0021] S1 specifically includes: S101: Calculate the distance from the ultrasonic sensor to the surface of the waste based on the ultrasonic echo time difference and sound velocity.

[0022] The surface distance refers to the vertical distance between the ultrasonic sensor and the surface of the waste inside the waste container.

[0023] Specifically, the distance from the ultrasonic sensor to the surface of the waste is: ,in, Indicates the first iThe surface distance of each garbage container Indicates the speed of sound. Indicates the first i The ultrasonic echo time difference of each garbage container.

[0024] S102: Calculate the waste filling rate based on the surface distance: in, Indicates the first i The waste filling rate of each waste container This indicates the geometric height of the trash can.

[0025] S103: Determine the waste filling rate based on the waste filling rate and the waste filling increment.

[0026] It should be noted that by using ultrasonic sensors to perform non-contact ranging of the internal state of the waste container, and combining the ultrasonic echo time difference with the speed of sound to calculate the distance to the waste surface, and then converting the distance information into waste filling rate and waste filling speed based on the geometric height of the waste bin, this solution can achieve continuous, objective and high-precision perception of the loading status of the waste container without interfering with waste disposal and collection operations. This not only effectively avoids the lag and subjective errors caused by manual inspection, but also provides a dynamic data foundation that reflects the waste growth trend for subsequent waste overflow risk prediction and collection decisions by introducing the filling rate as a time dimension indicator. This significantly improves the response capability of the waste management system to sudden high load situations and the overall reliability of operation.

[0027] S2: Upload the garbage filling rate and garbage filling speed to the cloud management platform.

[0028] Among them, the cloud management platform refers to a centralized management and analysis system deployed in a cloud computing environment. It is used to uniformly receive, store, process and schedule operational data from multiple waste disposal sites, and provide computing and management support for waste collection route planning, risk prediction and sorting collaborative control.

[0029] It should be noted that by uploading the waste filling rate and waste filling speed to the cloud management platform, this step can break the information isolation between individual waste disposal sites, realize the centralized aggregation and global visualization management of waste status data from multiple sites, and transform waste collection decisions from local experience-based judgments to a collaborative decision-making mode based on global data analysis. At the same time, it provides a continuous and traceable data source for subsequent trend analysis based on historical data, risk prediction model training, and dynamic scheduling algorithms, thereby significantly improving the scheduling efficiency, decision accuracy, and overall intelligence level of the waste recycling system in large-scale application scenarios.

[0030] S3: Using a fuzzy inference model optimized by a genetic algorithm, predict the risk of garbage overflow in each garbage container within a preset time window, and generate a garbage collection priority set based on the risk of garbage overflow.

[0031] It should be noted that the input data of the fuzzy inference model includes at least one of the following: the current garbage filling rate of the garbage container, the garbage filling rate, and the time interval since the last collection. The input data is fuzzified by the membership function and then participates in fuzzy inference. The output data of the fuzzy inference model is the risk level of garbage overflow in each garbage container within a preset time window. The risk level is used to generate a garbage collection priority set.

[0032] It should be noted that those skilled in the art can set the size of the preset time window according to actual needs, and this invention does not limit this.

[0033] Among them, the fuzzy inference model refers to a decision-making model based on fuzzy set theory and fuzzy rule base, which performs inference calculations on continuous, uncertain, or nonlinear input variables to map the waste filling status to risk assessment results. The genetic algorithm is a global optimization algorithm that simulates biological evolution, iteratively optimizing model parameters or rules through operations such as selection, crossover, and mutation. The preset time window refers to the time interval used for short-term prediction, considering the changing trends of waste filling rate and waste filling speed within this interval. The risk level of waste overflow refers to the probability of waste containers overflowing or becoming overloaded within the preset time window. The waste collection priority set refers to a site priority processing sequence formed by sorting or classifying different waste container overflow risks, used to guide subsequent waste collection scheduling decisions.

[0034] It should be noted that by introducing a fuzzy inference model optimized by a genetic algorithm to predict the risk of waste overflow, this step can comprehensively characterize the continuous changes and uncertainties in the waste filling state without relying on a single hard threshold. Furthermore, by adaptively optimizing the fuzzy rules and parameters through a genetic algorithm, the subjectivity and lack of adaptability caused by manually set rules are avoided, thereby achieving a more accurate and stable prediction of waste overflow risk. At the same time, a set of waste collection priorities is generated based on the predicted risk level, enabling waste collection scheduling to change from a passive response to a proactive, risk-driven scheduling, effectively reducing the probability of waste overflow and improving the overall utilization efficiency of collection resources.

[0035] In one possible implementation, S3 specifically includes: S301: Construct a container-time window input vector based on the corresponding surface distance of each waste container.

[0036] The container-time window input vector is a vector composed of multiple waste surface distance values ​​arranged in chronological order within the time window, used to describe the temporal evolution characteristics of the waste filling state.

[0037] Specifically, within a preset time window, the surface distance values ​​of the same waste container are collected multiple times, and the surface distance values ​​are arranged in chronological order to form an input vector to characterize the changing trend of the waste filling state in the time dimension.

[0038] S302: Map the container-time window input vector to membership degrees using a trapezoidal membership function: in, Represents the trapezoidal membership function. y The input real-valued variable represents the distance from the sensor to the surface of the waste in the waste container. a This represents the lower limit threshold at which the membership degree begins to rise, i.e., the maximum boundary value at which the input variable no longer belongs to the current fuzzy set. b This indicates the starting point where the membership degree reaches 1, that is, the lower limit where the input variable begins to clearly conform to fuzzy semantics (such as approaching full load). c This represents the endpoint where the membership degree remains at 1, i.e., the stable interval where the input variable completely belongs to the fuzzy set. d This represents the upper limit threshold at which the membership degree drops to 0, i.e., the upper limit at which the input variable begins to deviate from the current fuzzy semantic set.

[0039] Specifically, the trapezoidal membership function is used to map the continuous real-valued input variable of the distance to the trash can measured by the sensor in the trash can to... The membership degree within the interval is used to characterize the degree to which the current waste filling state conforms to a certain fuzzy semantic (such as "nearly full" or "high risk"). Specifically, by setting parameters a, b, c, and d corresponding to different distance threshold intervals, when the distance to the waste bin is less than or greater than a preset range, its membership degree is 0, indicating that the state does not belong to the fuzzy semantic. When the distance is in the middle interval, the membership degree increases or decreases linearly with distance or remains at 1, thus achieving a smooth transition from "not full" to "nearly full" waste filling state. This membership function avoids the abrupt state change problem caused by using a single hard threshold, providing a continuous, stable, and robust input foundation for subsequent full-load risk prediction and waste collection priority determination based on fuzzy reasoning.

[0040] S303: Based on membership degree and combined with preset risk semantic classification rules, construct a risk semantic set for garbage containers.

[0041] Specifically, the membership values ​​are mapped to the corresponding risk semantic categories, which include at least "low risk", "medium risk" and "high risk", thus forming a risk semantic set for subsequent fuzzy reasoning.

[0042] S304: Taking the risk semantic set as input, the corresponding fuzzy rules are triggered through the Mamdani-type fuzzy inference structure, and the fuzzy rules are optimized through a genetic algorithm to output the optimal rule subset.

[0043] Among them, the Mamdani-type fuzzy inference structure refers to a classic fuzzy inference framework based on a fuzzy rule base and fuzzy set operations. A fuzzy rule is a rule expression describing the logical relationship between the input risk semantics and the output risk result. The optimal rule subset refers to a set of fuzzy rules that contributes significantly to risk prediction, selected through a genetic algorithm.

[0044] Specifically, a genetic algorithm is used to iteratively search the selection, combination, and weight of fuzzy rules. Under the premise of ensuring the accuracy of risk prediction, rules with higher contribution are selected to form the optimal rule subset.

[0045] S305: Aggregate and defuzzify the optimal rule subset to generate a single-valued risk: in, Indicates single-value risk. This represents the defuzzing operator. B 0 indicates that the fuzzy output set is the optimal rule subset. Indicates the first r The fuzzy output set corresponding to each fuzzy rule B r In output variables y Membership degree of location n This represents the number of fuzzy rules involved in the reasoning. y n Indicates output variable y The first in its domain of definition n A discrete value point, This represents the integration operation over the universe of discourse of the output variable. d Represents the differential symbol.

[0046] S306: Individual risk is used as the degree of risk of waste overflow in each waste container within a preset time window.

[0047] S307: Generate a set of waste collection priorities based on the level of risk.

[0048] It should be noted that by constructing a container-time window input vector and introducing a trapezoidal membership function to fuzzify the waste filling state, this scheme can fully characterize the changing trend of waste generation in the time dimension and effectively handle the uncertainty and volatility of sensor data, avoiding the problem of sudden state changes caused by single threshold judgment. Simultaneously, by combining a risk semantic set with a Mamdani-type fuzzy inference structure and using a genetic algorithm to adaptively optimize the fuzzy rules, the risk prediction process possesses both good interpretability and strong global optimization capabilities. Finally, through defuzzification, a single-valued risk that can be directly used for scheduling decisions is obtained, thereby achieving a continuous, stable, and refined assessment of waste overflow risk and providing a reliable basis for generating a scientifically reasonable set of waste collection priorities.

[0049] S4: Based on the set of garbage collection priorities, combined with the geographical location information of garbage collection sites and real-time traffic information, construct a dynamic cost graph and path cost function.

[0050] Geographic location information refers to the coordinates of waste disposal sites, waste collection vehicles, and treatment plants within the urban space, used to describe the spatial relationships between these nodes. Real-time traffic information refers to dynamic data reflecting road traffic conditions, including road congestion levels, travel speed, and travel time. The path cost function is a mathematical function used to comprehensively characterize the cost of vehicles traveling on different paths; its cost factors include at least travel time, waste status, and relevant weight parameters. The dynamic cost graph is a graph structure that dynamically updates with time and environmental changes, using waste disposal sites, vehicle locations, and treatment plants as nodes and the path cost function as edge weights, used to support subsequent path optimization calculations.

[0051] It should be noted that by introducing geographical location information and real-time traffic information on the basis of the garbage collection priority set, and mapping multi-dimensional factors into a dynamic cost graph and path cost function, this step can transform garbage collection decision-making from a static planning based solely on distance or fixed time into a dynamic optimization problem that comprehensively considers risk urgency and road traffic conditions. This allows subsequent path planning results to prioritize coverage of high-risk sites while effectively avoiding congested road sections and high-cost routes, thereby significantly improving the adaptability, execution efficiency, and overall operational reliability of garbage collection scheduling in complex urban environments.

[0052] In one possible implementation, S4 specifically includes: S401: Obtain the node set and the node locations of the nodes in the node set, where the node locations include the locations of garbage collection vehicles, garbage disposal sites, and treatment plants.

[0053] Specifically, the location of garbage collection vehicles is obtained through vehicle positioning information, the location of garbage disposal sites is obtained through site registration information, and the location of treatment plants is obtained through treatment plant registration information, thus forming a node set.

[0054] S402: Calculate the travel time to the reachable node based on the node location and real-time traffic information of the waste disposal site.

[0055] The travel time to reachable nodes refers to the actual time required to travel from one node to another, taking into account real-time traffic conditions.

[0056] S403: Set a balancing factor based on different urban traffic conditions and waste collection cycles.

[0057] The balancing factor refers to the weighting parameter used to adjust the relative importance of different cost factors in the path cost function. Here, A represents the weight of traffic time, and B represents the weight of waste status. Cities with severe congestion should increase the weight of A, while cities with better road conditions should increase the weight of B.

[0058] S404: Combining the balancing factor, construct a path cost function with travel time, garbage filling rate, and garbage filling rate as weighting factors: in, Indicates from node x To the node y The path cost function, A This represents the traffic weighting balance factor. B This represents the garbage state weighting balance factor. Indicates from node x To the node y Travel time, Indicates the change in waste filling volume. Indicates the time since the last collection. This indicates the garbage filling rate.

[0059] S405: Use the path cost function as the edge cost of the dynamic cost graph, and combine it with the node set to construct the dynamic cost graph.

[0060] Among them, the dynamic cost graph refers to a graph model constructed with a set of nodes as the nodes of the graph and a path cost function as the edge weights. Its edge costs are dynamically updated as traffic and waste status change.

[0061] It should be noted that by abstracting garbage collection vehicles, garbage disposal sites, and treatment plants into a unified set of nodes, and combining real-time traffic information to calculate the actual travel time between nodes, and by introducing an adjustable balance factor to incorporate traffic factors and garbage filling status factors into the path cost function, this solution can flexibly adjust the focus of path optimization according to different urban traffic characteristics and collection needs. This allows the path planning results to effectively respond to changes in the external environment such as traffic congestion, while also taking into account the urgency and efficiency of garbage collection. Thus, in complex dynamic scenarios, a more reasonable, executable, and adaptive dynamic cost graph is constructed, providing a reliable foundation for subsequent efficient and stable garbage collection path solutions.

[0062] S5: With the goal of minimizing the path cost function, the dynamic cost graph is solved using an improved Dijkstra algorithm to determine the waste collection and transportation path.

[0063] It should be noted that the input data of the improved Dijkstra's algorithm includes a dynamic cost graph consisting of the locations of waste collection vehicles, waste disposal sites, and waste treatment plants; the path cost function corresponding to each edge in the dynamic cost graph; and a set of waste collection priorities. The output data of the improved Dijkstra's algorithm is the waste collection path that minimizes the path cost function under the constraints of vehicle load and time, along with its corresponding cumulative path cost. The traditional Dijkstra's algorithm only searches for the shortest path based on static edge weights, without considering the dynamic changes in path cost and node risk priorities, making it difficult to adapt to the real-time changes in waste filling status and traffic conditions during waste collection.

[0064] Specifically, the improved Dijkstra algorithm refers to a path search algorithm that, based on the traditional Dijkstra shortest path algorithm, incorporates risk priority, capacity constraints, or dynamic edge weight update mechanisms to find the minimum cost path while satisfying actual operational constraints. The waste collection and transportation route refers to the route taken by waste collection vehicles, which sequentially visit waste disposal sites according to the optimization results and ultimately drive to the treatment plant.

[0065] It should be noted that by minimizing the path cost function and using an improved Dijkstra algorithm to solve the dynamic cost graph, this step can organically integrate traffic state changes, garbage collection urgency, and operational constraints into the path search process while ensuring the computational efficiency and stability of the algorithm. This makes the generated garbage collection and transportation routes no longer limited to the static shortest distance or shortest time, but rather a comprehensive optimal route that better meets actual operational needs. This effectively reduces ineffective detours and resource waste, and improves the overall efficiency and feasibility of garbage collection and transportation operations.

[0066] In one possible implementation, S5 specifically includes: S501: Initialize the cumulative cost label, risk label, and predecessor node from the source point to each node in the dynamic cost graph.

[0067] In this context, the source point refers to the starting node in the waste collection and transportation route planning, typically corresponding to the current location of the waste collection vehicle. The cumulative cost label refers to the total path cost accumulated from the source point to a given node, used to measure the overall cost of reaching that node. The risk label refers to the waste spill risk assessment result corresponding to the node, used to reflect the urgency of collection during the route selection process. The predecessor node refers to the previous node recorded during the optimal route search process, used to backtrack and generate the complete route.

[0068] Specifically, the cumulative cost label of the source node is set to zero, and the cumulative cost label of the other nodes is set to infinity.

[0069] S502: Set the size of the scrolling optimization window and create a candidate set based on the garbage collection priority set.

[0070] The rolling optimization window refers to the time or number of steps used to periodically check and update edge weights and path results during the path planning process. The candidate set refers to the set of waste disposal sites that participate in the path optimization calculation after being selected based on waste collection priority.

[0071] S503: Based on the remaining capacity of waste collection vehicles and the demand for waste disposal sites, the candidate set is filtered to determine the feasible set: in, Represents a node v The feasibility assessment results I ( ) indicates an indicator function. Cap This indicates the maximum available capacity of the garbage collection vehicle. Path This represents the site access sequence (path) obtained from the current plan. j This indicates the garbage disposal sites that have been identified and visited in the path. Indicates garbage disposal site j Current waste filling volume.

[0072] The feasible set refers to the set of nodes that further satisfy the constraints of garbage collection vehicle capacity and collection time window on the basis of the candidate set.

[0073] The candidate set limits the range of waste disposal sites participating in the route optimization calculation, while the feasible set further filters site nodes that meet the vehicle capacity constraint and waste overflow time window constraint based on the candidate set. By processing only nodes in the feasible set during subsequent node expansion and route relaxation processes, the generated waste collection and transportation route is ensured to be executable and invalid route calculations are reduced.

[0074] S504: Based on the feasibility set and a joint selection function combining risk priorities, select path extension nodes from nodes whose minimum cumulative cost labels have not yet been determined. in, w Indicates an extended node. argmin ( ) denotes the minimization of the selection operator. Represents a node v The cumulative path cost This represents the risk weighting coefficient. This represents the risk penalty function. Represents a node v Spillover risk.

[0075] Among them, the path expansion node refers to the node selected during the path search process for further expansion and relaxation operations.

[0076] S505: Combining dynamic edge weights, perform relaxation updates and predecessor updates on nodes adjacent to the path extension node that have not been visited: in, Indicates an extended node w The cumulative path cost, min means minimizing. Represents a node u The cumulative path cost Indicates from node u To the node w Dynamic edge weights Indicates from node u To the extended node w The basic path cost, This represents the traffic uncertainty weighting coefficient. Indicates from node u To the extended node w The degree of fluctuation in travel time within historical or real-time data. The weighting coefficient represents the uncertainty of the garbage state. Indicates an extended node w The corresponding amount of waste filling is uncertain.

[0077] Relaxation update refers to the process in the shortest path algorithm of updating the cumulative cost and predecessor relationship of a node by comparing the costs of different paths. Dynamic edge weight refers to the edge cost that changes over time after introducing traffic fluctuations and garbage state uncertainties into the basic path cost.

[0078] S506: Determine if the relaxation update is successful. If yes, update the cumulative cost labels of adjacent nodes and set the path extension node as the predecessor node of the adjacent nodes, then proceed to step S507. Otherwise, keep the cumulative cost labels and predecessor nodes of the adjacent nodes unchanged.

[0079] S507: In the rolling optimization window, determine whether the change in travel time exceeds the travel time change threshold or whether the garbage filling rate exceeds the garbage filling rate threshold. If so, recalculate the dynamic edge weights and perform an incremental update on the current shortest path result based on the updated dynamic edge weights, then return to step S504. Otherwise, proceed to step S508.

[0080] It should be noted that those skilled in the art can set the values ​​of the driving time change threshold and the garbage filling rate threshold according to actual needs, and the present invention does not limit them.

[0081] Incremental update refers to an optimization method that, when some edge weights change, only locally updates the affected paths instead of recalculating the entire system.

[0082] S508: Based on the predecessor update results, perform backtracking processing to generate waste collection and transportation routes.

[0083] It should be noted that by introducing a hierarchical screening mechanism of candidate and feasible sets into the improved Dijkstra algorithm, and by jointly controlling the path expansion and relaxation process in conjunction with garbage collection vehicle capacity constraints, garbage spill risk, and traffic uncertainty, this scheme can significantly reduce invalid node expansion and redundant computation while ensuring the executability of the path planning results. Furthermore, by leveraging a rolling optimization window and a dynamic edge weight incremental update mechanism, the algorithm can quickly respond to changes in traffic conditions and garbage generation, ensuring that the generated garbage collection and transportation paths remain adaptable and stable to the real-world operating environment. This enables efficient, reliable, and risk-aware garbage collection and transportation path planning in complex dynamic scenarios.

[0084] S6: Based on the waste collection and transportation route, control the waste collection vehicles to transport the waste from each waste disposal site to the sorting line, and simultaneously update the status information of the waste disposal sites on the cloud management platform.

[0085] Among them, waste collection vehicles refer to transport vehicles used to perform waste collection tasks, possessing a certain load capacity and controllable by a dispatch system. Sorting lines refer to automated processing equipment installed within processing plants, used to classify collected mixed waste according to a predetermined process sequence. The status information of waste disposal sites refers to data used to characterize the current operating status of the site, including waste filling status, whether collection has been completed, and availability indicators.

[0086] It should be noted that by uniformly scheduling garbage collection vehicles based on the optimized garbage collection and transportation routes, and synchronously updating the status information of garbage disposal sites during garbage collection and transportation, this step enables real-time linkage between the garbage collection and transportation execution process and the cloud-based decision-making system. This avoids duplicate or missed collection due to information lag, while ensuring that the quantity and time of garbage received by the sorting line are more controllable. It provides accurate status feedback for subsequent sorting control and overall system optimization, thereby significantly improving the synergy, execution reliability, and operational efficiency of the garbage recycling and sorting process.

[0087] In one possible implementation, S6 specifically includes: S601: Calculate the number of garbage bins that can be allocated to a single garbage collection vehicle based on the load capacity of the garbage collection vehicle and the real-time garbage volume at each garbage disposal site.

[0088] Among these, the load capacity of a waste collection vehicle refers to the maximum weight or volume of waste that a waste collection vehicle can carry under safe and normal operating conditions. Real-time waste volume refers to the current waste filling status collected by sensors at each waste disposal site before collection. The number of waste bins refers to the number of waste containers that a single waste collection vehicle can be allocated and collected in a single collection task, under the constraint of vehicle load capacity.

[0089] S602: Based on the number of trash cans, select from the trash disposal sites covered by the trash collection and transportation route to determine the set of collection sites that the trash collection vehicles can use.

[0090] The collection of eligible collection sites refers to the set of waste disposal sites selected from those covered by the waste collection and transportation routes that meet the constraints of the vehicle's remaining load capacity.

[0091] S603: Distribute the garbage collection routes and collection sites to the garbage collection vehicles, and predict the arrival time of the garbage collection vehicles to each garbage disposal site based on the travel time along the garbage collection routes.

[0092] The arrival time refers to the time when the garbage collection vehicle arrives at each garbage disposal site, based on the predicted route travel time.

[0093] The travel time along the garbage collection route refers to the travel time required for garbage collection vehicles to travel along one side of the garbage collection route (i.e., between two adjacent stations / nodes).

[0094] S604: Based on the arrival time, the waste filling rate before collection at the corresponding waste disposal site is calculated by triggering the ultrasonic module on the top of the site.

[0095] S605: Performs garbage loading operations based on the garbage filling rate before collection, and updates the remaining load of the garbage collection vehicle in real time during the loading process.

[0096] S606: After loading is completed, calculate the waste filling rate of the waste disposal site after collection and compare the waste filling rate after collection with the emptying threshold.

[0097] Among them, the pre-collection waste filling rate and post-collection waste filling rate refer to the waste loading status indicators of the waste disposal site before and after the waste loading operation begins and ends, respectively. The emptying threshold is a preset threshold used to determine whether a waste disposal site is considered to have completed collection. Those skilled in the art can set the size of the emptying threshold according to actual needs, and this invention does not limit it.

[0098] Specifically, if the conditions for clearing are met, the status of the waste disposal site is updated to idle in the cloud, and the process proceeds to step S607.

[0099] Otherwise, mark the site as "partially cleared" and return to step S602 to continue clearing the remaining sites.

[0100] S607: Based on the comparison results, the path edge cost corresponding to the garbage disposal sites that have been cleared is set to infinity in the cloud, forming a collaborative shielding mechanism for occupied / completed garbage disposal sites.

[0101] Among them, the collaborative shielding mechanism refers to logically shielding the corresponding path edges of sites that have been completed or are temporarily no longer eligible for clearing in the cloud, so as to prevent them from being included in subsequent scheduling calculations again.

[0102] S608: When the collaborative shielding mechanism takes effect, the collection and transportation of waste at the collection and transportation sites is completed, or the remaining load of the vehicle is insufficient to continue transportation, control the waste collection vehicle to drive into the treatment plant and unload the waste into the front end of the sorting line.

[0103] S609: At the front end of the sorting line, the unloading quality is obtained by a weighing device.

[0104] Among them, unloading quality refers to the weight of garbage measured by a weighing device when garbage collection vehicles unload garbage into the front end of the sorting line at the treatment plant.

[0105] S610: Calculate the loading mass of this collection task based on the difference between the garbage filling rate before and after collection at each garbage collection site.

[0106] Among them, the loading mass refers to the actual loading mass of garbage calculated based on the difference in garbage filling rate before and after the removal.

[0107] S611: Perform a mass conservation consistency check based on the unloading mass and loading mass. If the check passes, update the status information of the waste disposal site to idle and end the current collection task. Otherwise, mark the collection batch as an abnormal batch and send the corresponding data back to the cloud management platform.

[0108] Among them, "quality conservation consistency verification" refers to the verification process of comparing the unloading quality with the loading quality to verify the consistency of the cleaning data and the reliability of the operation.

[0109] It should be noted that by introducing site selection under vehicle load constraints, arrival time prediction, and dynamic sensing of waste filling rate before and after collection during the waste collection execution phase, this solution ensures that the waste loading process remains controllable and verifiable. Simultaneously, a collaborative shielding mechanism prevents completed or partially completed collection sites from being repeatedly scheduled, further improving scheduling efficiency and system consistency. Furthermore, by verifying the consistency of unloading and loading quality according to mass conservation, abnormal collection behaviors or sensor data deviations can be detected promptly, and the results can be transmitted back to the cloud for subsequent analysis and model optimization. This significantly enhances the reliability, transparency, and closed-loop management capability of the overall system operation during the waste collection execution process.

[0110] S7: By using a flow balance model based on the Euler equation, the waste flow rate and density on the sorting line are monitored, and the operating parameters of the sorting line are dynamically adjusted according to the monitoring results to automatically separate different waste components according to the preset sorting process sequence and generate waste sorting results.

[0111] It should be noted that the input data of the flow balance model based on the Euler equation includes the waste flow rate, waste density, and corresponding spatial and temporal information on the sorting line. The flow balance model solves the input data and outputs the waste flow rate control quantity and waste density control quantity used to maintain the load balance of the sorting line. Based on the control quantity, the operating parameters of the sorting line are dynamically adjusted to achieve automatic separation of different waste components and generate waste sorting results.

[0112] The flow balance model based on Euler's equations is a mathematical model established using the Euler description method in continuum mechanics to characterize the flow state and conservation relationships of waste during the transport process in a sorting assembly line. Waste flow rate refers to the quantity or mass of waste passing through a cross-section of the sorting assembly line per unit time. Waste density refers to the distribution of waste per unit length or volume, reflecting the load on the assembly line. Operating parameters are adjustable control variables for the sorting assembly line, including conveyor belt speed, feeding intensity, and the working rhythm of the sorting units. The sorting process sequence refers to the order in which waste undergoes processing on the sorting assembly line. Waste components refer to the different physical or material categories that make up the mixed waste. Waste sorting result refers to the waste output separated by category after processing.

[0113] It should be noted that by introducing a flow balance model based on the Euler equation to continuously monitor and model the waste flow rate and density on the sorting line, this step can transform the traditional sorting control method that relies on experience or fixed parameters into an adaptive control method based on real-time status feedback. This allows the sorting line to dynamically adjust its operating parameters according to changes in waste load and flow state, thereby avoiding problems such as decreased sorting efficiency or equipment blockage caused by local accumulation or uneven feeding. It also ensures that different waste components can be automatically separated stably and continuously according to the preset process sequence, thus improving the overall stability, processing efficiency, and classification accuracy of the sorting system.

[0114] In one possible implementation, S7 specifically includes: S701: Establish waste status function for sorting production line.

[0115] Specifically, the waste on the sorting line is considered as a continuous medium moving along the x-direction of the conveyor belt, and the location of the waste is defined as follows: x The moment of the for (This represents the state quantity of waste within a unit space segment, such as "quantity / volume element"), serving as the basis for subsequent flow and density calculations.

[0116] S702: Calculate waste flow rate and waste density based on the waste state function of the sorting line.

[0117] Specifically, the limits are taken at fixed locations and at fixed times to obtain the waste flow rate and waste density.

[0118] S703: Based on waste flow rate and waste density, and combined with the characteristic line form of the Euler equations, construct the flow balance solution relationship: in, d Represents the differential symbol. rThe characteristic line parameters, introduced by the Euler equations in the characteristic line method, describe the evolution trajectory of the waste flow state along a specific direction. x This represents the spatial location variable of waste on the sorting line. y This represents a second spatial or state dimension related to the state of the waste flow, such as lateral distribution, hierarchical state, or sorting segment index. q Indicates garbage flow. v This represents the equivalent velocity of waste along the main conveying direction. b This represents a parameter indicating sensitivity to changes in flow rate. Represents the equivalent rate of change of garbage state. k 1 represents the coupling coefficient between waste flow rate and waste density. k 2 represents the compensation coefficient related to the basic conveyor or external drive.

[0119] Specifically, parameters k 1. , b、k 2. Data can be obtained through offline calibration or online adaptive updates based on historical operating data.

[0120] S704: Solve the flow balance relationship to determine the flow control quantity and density control quantity.

[0121] Specifically, the flow control quantity is The density control amount is: ,in, q This indicates the waste flow control result at a specific location and time on the sorting line, that is, the amount of waste passing through that cross-section per unit time. Indicates an exponentially decaying or stable term. z This represents the amount of external compensation related to the density of the waste flow.

[0122] Specifically, by modeling the waste flow control quantity using a combination of integral and exponential terms, joint compensation for long-term and instantaneous disturbances is achieved. Furthermore, by introducing a flow-density coupling coefficient, the flow control result is mapped to the waste density control quantity, thereby achieving stable control of the sorting line load state.

[0123] S705: Based on the flow rate control and density control, the conveyor belt speed and feeding intensity are adjusted to drive the waste flow in the sorting line to automatically separate different waste components according to the preset sorting process sequence, and generate waste sorting results.

[0124] It should be noted that by treating the waste on the sorting line as a continuous medium and constructing a waste state function, and then combining the characteristic line method of the Euler equation to uniformly model and solve the waste flow and density, this scheme can characterize the flow and distribution of waste in the sorting process from a mechanistic perspective. Based on this, flow and density adjustment quantities that can be directly used for control are formed, enabling the sorting line operating parameters to be dynamically and adaptively adjusted according to the actual waste load and disturbance conditions. This effectively suppresses waste accumulation, fluctuations, and load imbalance, ensures the stable execution of the sorting process sequence, and improves the overall stability, continuity, and sorting efficiency of the sorting system under complex working conditions.

[0125] In one possible implementation, the sorting process sequence includes mechanical pretreatment, magnetic separation, gas separation, and optical sorting.

[0126] The relevant metal components are determined by magnetic separation.

[0127] Lightweight / recyclable components are identified through gas separation and optical separation.

[0128] Waste components include organic matter, metals, plastics, paper, glass, and residues.

[0129] S8: Based on the waste collection process data and waste sorting results, the cloud management platform optimizes the fuzzy inference model, path cost function, and flow balance model.

[0130] Among them, garbage collection process data refers to the operational data generated during the garbage collection process, including collection time, vehicle travel trajectory, loading weight, and changes in site status.

[0131] It should be noted that by introducing a unified analysis and feedback mechanism for waste collection process data and waste sorting results into the cloud management platform, and by continuously optimizing the fuzzy inference model, path cost function, and flow balance model based on Euler equations, this solution enables the waste recycling and sorting system to achieve self-learning and adaptive evolution. This allows the risk prediction, path planning, and sorting control models to continuously adapt to the actual operating environment and changes in urban characteristics, thereby avoiding performance degradation caused by long-term model fixation and improving the overall decision-making accuracy, operational stability, and long-term operating effect of the system under complex and variable operating conditions.

[0132] Specifically, the cloud management platform uniformly aggregates and analyzes data from the waste collection process and waste sorting results. By comparing predicted results with actual collection effects, sorting load changes, and sorting output, it identifies risk assessment biases in fuzzy inference models, unreasonable weight configurations in path cost functions, and parameter response lags in flow balance models. Based on this, it dynamically adjusts or iteratively updates fuzzy membership parameters, fuzzy rule weights, traffic weights and waste state weights in path cost functions, and coupling coefficients and sensitivity parameters in Euler equation models. This enables each model to more accurately reflect waste generation patterns, traffic state changes, and sorting load characteristics in subsequent operations, achieving continuous optimization and closed-loop adaptive improvement of collection decisions, path planning, and sorting control.

[0133] In this embodiment of the invention, the risk of garbage overflow is predicted by real-time perception and centralized cloud analysis of the operational status of garbage collection sites, combined with a fuzzy inference model optimized by a genetic algorithm. Based on this, a path cost function that integrates traffic conditions and dynamic characteristics of garbage is constructed to achieve adaptive optimization of garbage collection paths. At the same time, by collaboratively modeling and implementing closed-loop feedback control of the collection process and the sorting line operation status, sorting parameters are dynamically adjusted to match actual garbage flow changes. This reduces garbage overflow and ineffective collection, while improving vehicle utilization and the stability and processing efficiency of the sorting system, thereby comprehensively improving the intelligence level, operational reliability, and resource utilization efficiency of the garbage recycling and sorting process.

[0134] Reference manual attached Figure 2 The diagram shows a schematic representation of the cloud-based collaborative management and sorting system for waste recycling provided in an embodiment of the present invention.

[0135] This invention provides a cloud-based collaborative management and sorting system 20 for waste recycling, comprising: a processor 201 and a memory 202; The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the cloud-based collaborative management and sorting method for waste recycling described above and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0136] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0137] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM).

[0138] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0139] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0140] 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. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0141] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0142] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0143] 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.

[0144] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0145] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0146] This invention provides a readable storage medium that stores a program or instructions on the readable storage medium. When the program or instructions are executed by a processor, they implement the steps of the cloud-based collaborative management and sorting method for garbage collection described above, and can achieve the same technical effect. To avoid repetition, this invention will not elaborate further.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A cloud-based collaborative management and sorting method for waste recycling, characterized in that, include: S1: Collect real-time status data from multiple waste disposal sites, and calculate the waste filling rate and waste filling speed of each waste container based on the real-time status data; S2: Upload the waste filling rate and the waste filling speed to the cloud management platform; S3: Using a fuzzy inference model optimized by a genetic algorithm, predict the risk level of garbage overflow in each garbage container within a preset time window, and generate a garbage collection priority set based on the risk level of garbage overflow. S4: Based on the set of garbage collection priorities, combined with the geographical location information and real-time traffic information of the garbage disposal sites, construct a dynamic cost graph and path cost function; S5: With the goal of minimizing the path cost function, the dynamic cost graph is solved using the improved Dijkstra algorithm to determine the waste collection and transportation path; S6: According to the waste collection and transportation route, the waste collection vehicles are controlled to transport the waste from each waste disposal site to the sorting line, and the status information of the waste disposal site is updated simultaneously on the cloud management platform; S7: By using a flow balance model based on the Euler equation, the waste flow rate and density on the sorting line are monitored, and the operating parameters of the sorting line are dynamically adjusted according to the monitoring results to automatically separate different waste components according to the preset sorting process sequence and generate waste sorting results; S8: Based on the waste collection process data and the waste sorting results, the cloud management platform optimizes the fuzzy inference model, the path cost function, and the flow balance model; Specifically, S7 includes: S701: Establish waste status function for sorting production line; S702: Calculate the waste flow rate and waste density based on the waste state function of the sorting line; S703: Based on the waste flow rate and waste density, and combined with the characteristic line form of the Euler equations, construct the flow balance solution relationship; S704: Solve the flow balance relationship to determine the flow control quantity and density control quantity; S705: Adjust the conveyor belt speed and feeding intensity according to the flow control amount and the density control amount to drive the waste flow in the sorting line to automatically separate different waste components according to the preset sorting process sequence, and generate the waste sorting result.

2. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, The real-time status data includes the height of the waste container, timestamp, ultrasonic echo time difference, and speed of sound. S1 specifically includes: S101: Calculate the distance from the ultrasonic sensor to the surface of the waste based on the ultrasonic echo time difference and sound velocity; S102: Calculate the waste filling rate based on the surface distance; S103: Determine the waste filling rate based on the waste filling rate and the waste filling increment.

3. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, S3 specifically includes: S301: Construct a container-time window input vector based on the surface distances of each of the aforementioned waste containers; S302: Map the container-time window input vector to membership degrees using a trapezoidal membership function; S303: Based on the membership degree and combined with the preset risk semantic classification rules, construct the risk semantic set of the waste container; S304: Taking the risk semantic set as input, trigger the corresponding fuzzy rules through the Mamdani-type fuzzy reasoning structure, and optimize the fuzzy rules through the genetic algorithm to output the optimal rule subset; S305: Perform aggregation and defuzzification processing on the optimal rule subset to generate a single-value risk; S306: The single-value risk is used as the degree of risk of garbage overflow in each of the garbage containers within the preset time window; S307: Generate the waste collection priority set based on the risk level.

4. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, S4 specifically includes: S401: Obtain a set of nodes and the node positions of the nodes in the set of nodes, wherein the node positions include the positions of garbage collection vehicles, garbage disposal sites, and treatment plants; S402: Based on the node location and the real-time traffic information of the waste disposal site, calculate the travel time to the reachable node; S403: Set a balancing factor based on different urban traffic conditions and waste collection cycles; S404: Combine the balance factor to construct a path cost function with the travel time, the garbage filling rate and the garbage filling rate as weight factors; S405: Use the path cost function as the edge cost of the dynamic cost graph, and combine it with the node set to construct the dynamic cost graph.

5. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, S5 specifically includes: S501: Initialize the cumulative cost label, risk label, and predecessor node from the source point to each node in the dynamic cost graph; S502: Set the size of the scrolling optimization window and establish a candidate set based on the garbage collection priority set; S503: Based on the remaining capacity of garbage collection vehicles and the demand for garbage disposal sites, the candidate set is screened to determine the feasible set; S504: Based on the aforementioned feasibility set, and combined with the joint selection function of risk priority, select path extension nodes from the nodes whose minimum cumulative cost index has not yet been determined; S505: Combining dynamic edge weights, perform relaxation update and predecessor update on nodes adjacent to the path extension node that have not been visited; S506: Determine whether the relaxation update is valid; if yes, update the cumulative cost label of the adjacent node and set the path extension node as the predecessor node of the adjacent node, and proceed to step S507; otherwise, keep the cumulative cost label and predecessor node of the adjacent node unchanged. S507: In the rolling optimization window, determine whether the change in travel time is greater than the travel time change threshold or whether the garbage filling rate is greater than the garbage filling rate threshold; if so, recalculate the dynamic edge weights and perform incremental updates on the current shortest path result based on the updated dynamic edge weights, and return to step S504; otherwise, proceed to step S508. S508: Based on the predecessor update result, perform backtracking processing to generate the waste collection and transportation path.

6. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, S6 specifically includes: S601: Calculate the number of garbage bins that can be allocated to a single garbage collection vehicle based on the load capacity of the garbage collection vehicle and the real-time garbage volume at each of the garbage disposal sites. S602: Based on the number of trash cans, filter from the trash disposal sites covered by the trash collection and transportation route to determine the set of collection sites that the trash collection vehicle can collect; S603: The garbage collection and transportation route and the set of garbage collection stations are sent to the garbage collection vehicle, and the arrival time of the garbage collection vehicle to each garbage disposal station is predicted based on the travel time along the garbage collection and transportation route. S604: Based on the arrival time, the pre-collection waste filling rate of the corresponding waste disposal site is calculated by triggering the ultrasonic module on the top of the site. S605: Perform garbage loading operation according to the garbage filling rate before collection, and update the remaining load of the garbage collection vehicle in real time during the loading process; S606: After loading is completed, calculate the waste filling rate of the waste disposal site after collection, and compare the waste filling rate after collection with the emptying threshold. S607: Based on the comparison results, the path edge cost corresponding to the garbage disposal sites that have been cleared is set to infinity in the cloud, forming a collaborative shielding mechanism for occupied / completed garbage disposal sites; S608: When the collaborative shielding mechanism takes effect, the collection of waste collection sites is completed, or the remaining load of the vehicle is insufficient to continue collection, control the waste collection vehicle to drive into the treatment plant and unload the waste into the front end of the sorting line. S609: At the front end of the sorting line, the unloading mass is obtained by a weighing device; S610: Calculate the loading mass of this collection task based on the difference between the garbage filling rate before collection and the garbage filling rate after collection at each of the garbage collection sites. S611: Perform a mass conservation consistency check based on the unloading mass and the loading mass; if the check passes, update the status information of the waste disposal site to idle and end the current collection task; otherwise, mark the collection batch as an abnormal batch and send the corresponding data back to the cloud management platform.

7. The cloud-based collaborative management and sorting method for waste recycling according to claim 1, characterized in that, The sorting process sequence includes mechanical pretreatment, magnetic separation, air separation, and optical sorting; The relevant metal components were determined by magnetic separation; Lightweight / recyclable components were identified through gas separation and optical separation. The waste components include organic matter, metals, plastics, paper, glass, and residues.

8. A cloud-based collaborative management and sorting system for waste recycling, characterized in that, include: Processor and memory; The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the cloud-based collaborative management and sorting method for waste recycling as described in any one of claims 1 to 7.

9. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the cloud-based collaborative management and sorting method for waste recycling as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Garbage classification resource optimization management method based on cloud platform

    CN118585852A

  • Intelligent garbage collection and transportation system meeting low-carbon requirement

    CN120725252A

  • Background garbage classification resource optimization decision-making system based on big data

    CN120851271A