Intelligent solid waste sorting management system

Abstract

The invention discloses an intelligent solid waste sorting management system, and relates to the technical field of solid waste resource utilization, and the system comprises a full-link data acquisition module for integrating multi-dimensional data; a carbon footprint accurate accounting module calculates material carbon emission and emission reduction coefficients, and a parameter library is dynamically updated; the AI dual-objective optimization decision module is combined with a carbon-economic objective to generate a sorting instruction; the intelligent sorting execution module is used for accurately sorting and collecting; the carbon emission reduction evidence storage transaction and state feedback module is used for accounting the emission reduction amount, performing evidence storage transaction and feeding back the equipment state to form closed-loop management; according to the invention, through full-link data acquisition and carbon footprint accurate accounting, collaborative optimization of carbon emission reduction and economic benefits is realized, and the goal of dynamically matching enterprise carbon reduction is achieved; and meanwhile, a credible sorting management closed loop is constructed by utilizing distributed account book evidence storage and equipment state monitoring, so that the sorting precision and the system stability are improved, and reliable data support is provided for an enterprise to participate in carbon transaction.

Description

Technical Field

[0001] This invention relates to the field of solid waste resource utilization technology, specifically to an intelligent sorting and management system for solid waste. Background Technology

[0002] As a crucial node in resource recycling and carbon emission reduction, the sorting process's level of intelligence and precision directly impacts resource recycling efficiency and carbon reduction effectiveness. Currently, with the gradual improvement of the carbon trading market, enterprises conducting solid waste sorting operations not only need to achieve efficient material separation but also need to consider the quantitative accounting and trading conversion of carbon emission reduction effects. This places higher demands on the sorting system's data integration capabilities, carbon footprint accounting accuracy, scientific decision-making, and the credibility of emission reduction documentation.

[0003] Traditional solid waste sorting systems have significant limitations: data collection focuses primarily on the basic physical properties of materials, lacking full-chain integration of transfer information, energy consumption data, and multi-platform economic data, resulting in a single data dimension; carbon footprint accounting relies on a fixed parameter library, failing to consider regional specific differences, and parameter updates are lagging, leading to a disconnect between accounting results and actual scenarios; sorting decisions only focus on economic benefits or a single environmental goal, without establishing a collaborative mechanism for carbon-economic dual objectives; emission reduction certificates mostly adopt a centralized management model, making data prone to tampering and lacking traceability, failing to meet the credibility requirements of carbon trading, and exhibiting insufficient communication efficiency between modules and untimely equipment status feedback, thus hindering the improvement of sorting efficiency and system adaptability. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent sorting and management system for solid waste. This system integrates material characteristics, transfer information, and economic data from multiple platforms by constructing a full-link data acquisition and carbon footprint accurate accounting system. Combined with a carbon-economic dual-objective optimization algorithm, it dynamically balances carbon reduction benefits and economic benefits to generate intelligent sorting strategies. At the same time, it uses a consortium blockchain distributed ledger to achieve tamper-proof evidence of carbon emission reductions, connects to a carbon trading platform, and monitors equipment status in real time through the GRU algorithm, forming a data-driven, reliable traceability closed-loop management mechanism that effectively improves the resource utilization rate of solid waste and the effectiveness of carbon emission reduction.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an intelligent sorting and management system for solid waste, the system comprising:

[0006] The end-to-end data acquisition module is equipped with multiple sensors and data preprocessing units to collect material characteristics, transfer information, economic data, and equipment operation energy consumption data of solid waste. After standardized processing, the data is transmitted through a preset communication protocol.

[0007] Carbon footprint accurate accounting module: It has a built-in carbon footprint parameter library that conforms to the corresponding standards, receives the data transmitted by the above acquisition module, calculates the carbon emission coefficient and the carbon emission reduction coefficient of the material throughout its entire life cycle, and the parameter library is updated with regional specific coefficients through a federated learning mechanism;

[0008] AI Dual-Objective Optimization Decision Module: Receives carbon emission coefficients from the carbon footprint accounting module and economic and energy consumption data from the acquisition module. It adopts the core formula of carbon-economic dual-objective optimization, combines carbon weight-oriented dual-objective decision logic to calculate the comprehensive score of materials, and generates sorting priority and process parameter instructions.

[0009] Intelligent sorting execution module: includes an adaptive variable frequency conveyor belt, a six-axis robotic arm, a vision-tactile fusion recognition component and a graded collection bin. It receives instructions from the above decision module, executes sorting and graded collection operations, and records sorting process data synchronously.

[0010] Carbon emission reduction certificate storage and trading and status feedback module: Receives carbon emission data from carbon footprint accounting module and material processing volume data from sorting execution module, calculates emission reduction through the full-link carbon emission reduction calculation formula, stores the data in a distributed ledger and connects to the carbon trading platform, and monitors equipment operating status and feeds it back to the decision module.

[0011] Each module uses the MQTT protocol and 5G communication technology to achieve real-time data interaction.

[0012] Furthermore, the sensors in the end-to-end data acquisition module include an infrared spectral sensor, a weight sensor, a temperature and humidity sensor, and a GPS positioning unit.

[0013] Furthermore, the economic data in the full-link data acquisition module includes the unit price of recycled materials. This unit price is calculated using a multi-platform weighted average, with the weights allocated as follows: 60% for the primary renewable resource trading platform, 30% for the local trading platform, and 10% for the long-term cooperative enterprise quotation. The update cycle is 30 minutes.

[0014] Furthermore, the core formula for the carbon-economy dual-objective optimization in the AI ​​dual-objective optimization decision module is: ,in, For comprehensive evaluation of materials; The carbon emission reduction weight is set at a value ranging from 0.3 to 0.7. The initial value is preset according to the enterprise's carbon reduction target and policy requirements, and is dynamically adjusted every 7 days according to the proportion of carbon trading revenue. As the weight of economic returns, by =1−α is calculated to obtain; , is the carbon emission reduction coefficient for the i-th type of material regeneration, which is derived by the carbon footprint accurate accounting module based on the difference in carbon emission coefficients between virgin and recycled materials, combined with the material regeneration loss rate; The carbon emission coefficient for the entire life cycle of material i is calculated by the carbon footprint accurate accounting module based on material composition data, energy consumption data of each stage, and basic coefficients of the carbon footprint parameter library. The unit price for recycling the i-th type of material is obtained by the end-to-end data acquisition module from the renewable resource trading platform and through multi-platform weighted processing. The carbon trading price is captured from the carbon trading platform by the end-to-end data acquisition module and updated every 4 hours. The unit processing energy consumption for the i-th type of material is calculated by the end-to-end data acquisition module based on the total energy consumption of the equipment and the material processing volume. It is a unit energy consumption carbon emission coefficient, which is updated quarterly based on data released by the regional ecological and environmental departments.

[0015] Furthermore, the carbon-based dual-objective decision-making logic in the AI ​​dual-objective optimization decision-making module constructs carbon-based association rules based on carbon trading prices, carbon emission reductions, and material processing volumes, dynamically adjusts sorting priority judgment criteria, and prioritizes the allocation of sorting resources to materials with high carbon-based value.

[0016] Furthermore, the formula for calculating the total carbon emission reduction in the carbon emission reduction certificate trading and status feedback module is as follows: ,in, This represents the total carbon emission reduction; The processing volume of the i-th type of material is collected by the weight sensor of the collection bin of the intelligent sorting execution module; The industry benchmark full life cycle carbon emission coefficient is derived from the corresponding standard value built into the carbon footprint parameter library; This represents the actual life-cycle carbon emission coefficient. Let be the carbon emission reduction coefficient for the i-th type of material regeneration; The conversion efficiency for carbon emission reduction in regeneration is set to a value ranging from 0.85 to 0.95. The initial value is preset according to the material type and fine-tuned every six months based on actual feedback data from recycling companies.

[0017] Furthermore, the six-axis robotic arm in the intelligent sorting execution module is equipped with a force control feedback unit; the adaptive variable frequency conveyor belt uses a PID algorithm to dynamically correct its speed; the graded collection bin is equipped with a material level warning device, which has a first-level warning and a second-level warning. The trigger threshold for the first-level warning is 75% of the bin capacity, and the trigger threshold for the second-level warning is 85% of the bin capacity.

[0018] Furthermore, the carbon footprint parameter library in the carbon footprint accurate accounting module is updated through a two-level architecture of regional nodes and central nodes. When the processing volume of regional nodes accounts for ≥30%, the data contribution weight is set to 0.4; the material regeneration loss rate ranges from 0.03 to 0.08, preset according to the material type.

[0019] Furthermore, the distributed ledger of the carbon emission reduction certificate trading and status feedback module adopts a consortium blockchain architecture and uses SHA-256 encryption for certificate storage.

[0020] Furthermore, the equipment status monitoring in the carbon emission reduction certificate trading and status feedback module adopts the GRU algorithm. The input features include the correlation between carbon emissions and energy consumption. The monitoring data is fed back to the AI ​​dual-objective optimization decision module to adjust the sorting parameters.

[0021] Compared with existing technologies, this intelligent solid waste sorting and management system has the following advantages:

[0022] I. This invention integrates material characteristics, transfer information, economic data, and energy consumption data through a full-chain data acquisition module. Combined with the federated learning update mechanism of the carbon footprint accurate accounting module and the core formula for carbon-economic dual-objective optimization, it achieves synergistic optimization of carbon emission reduction and economic benefits. The system does not rely on a single decision dimension. By dynamically adjusting the weights of carbon emission reduction and economic benefits, it accurately matches enterprise carbon reduction goals with market demands, generating sorting priorities and process parameter instructions. Simultaneously, based on the difference in carbon emission coefficients between virgin and recycled materials and the adjustment of loss rates, the emission reduction coefficient calculation ensures the regional adaptability and accuracy of carbon footprint accounting, providing reliable data support for enterprises participating in carbon trading.

[0023] II. This invention constructs a reliable and efficient closed-loop sorting management system by combining distributed ledger storage and SHA-256 encryption technology with GRU algorithm-based equipment status monitoring and real-time inter-module interaction. The distributed ledger enables full-chain traceability of carbon emission reductions, eliminating the risk of data tampering and ensuring compliance with carbon trading platforms. Equipment status monitoring data is fed back to the decision-making module in real time, dynamically optimizing sorting parameters and improving system stability and sorting accuracy. The multi-platform weighted calculation of recycling unit price and the timed update mechanism ensure the accuracy of economic data. Combined with visual-tactile fusion recognition and adaptive sorting execution components, this further improves material sorting efficiency and grading quality.

[0024] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0026] Figure 1 A flowchart for an intelligent sorting and management system for solid waste;

[0027] Figure 2 Flowchart for parameter updates and coefficient calculations in the carbon footprint precision accounting module;

[0028] Figure 3 A flowchart for the comprehensive scoring calculation and sorting instruction generation of the AI ​​dual-objective optimization decision module. Detailed Implementation

[0029] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0030] Example 1:

[0031] This embodiment provides an intelligent sorting and management system for solid waste, applied to an urban recycling center. This center processes approximately 50 tons of recyclable solid waste daily from household and commercial waste, encompassing eight common materials including plastics, metals, paper, and glass. In current urban recycling scenarios, material composition is mixed, transportation links are numerous, and the center must simultaneously comply with carbon reduction policies and corporate economic benefit goals. Traditional sorting models struggle to meet these multi-dimensional needs. This system specifically addresses the challenges of intelligent and precise management in this scenario.

[0032] like Figure 1 As shown, the specific execution steps are as follows:

[0033] During the system hardware deployment phase, the infrared spectral sensors of the end-to-end data acquisition module are industrial-grade, high-stability models, installed 1.5 meters above the sorting conveyor belt, with one sensor every 5 meters along the conveyor belt length to form a full-coverage scanning area. The sensor lenses are equipped with dustproof protective covers and undergo optical cleaning and precision calibration weekly to ensure the accuracy of material composition identification meets actual operational requirements. Weight sensors are embedded in the conveyor belt feed rollers and the bottom of the grading collection bin. The feed roller sensors collect the total feed volume in real time, while the collection bin sensors accurately count the processing volume of individual materials, ensuring the accuracy of data measurement. Temperature and humidity sensors are deployed in the material storage area and sorting area, monitoring common environmental conditions with a data acquisition interval of 1 minute, providing a basis for optimizing material storage conditions. The GPS positioning unit is integrated into the onboard terminal of the transfer vehicle, interfacing with the traffic management platform to upload the real-time location, speed, departure and arrival times of the transfer vehicle, forming a complete transfer trajectory chain. The trajectory data is retained for future reference according to regulations.

[0034] The data preprocessing unit utilizes an industrial-grade embedded computer, equipped with a Linux operating system and dedicated data processing software. It employs a sliding window algorithm to denoise the raw sensor data, removing outliers and missing values. The standardized data is packaged in a common format and transmitted via an internal LAN using encryption algorithms, controlling transmission latency to meet real-time requirements. Economic data acquisition connects to the pricing systems of three primary-level renewable resource trading platforms, two local trading platforms, and five long-term cooperative recycling companies through API interfaces. The unit price of recycled materials is calculated according to preset weights and automatically updated every 30 minutes. When an interface on any platform fails, the system automatically triggers a redundancy switchover mechanism, activating a backup interface and logging the fault to ensure uninterrupted data acquisition.

[0035] The carbon footprint accurate accounting module is deployed on the core server cluster of the recycling center, employing a primary-backup dual-machine hot standby architecture to ensure the continuity of accounting services. The module's built-in carbon footprint parameter library initially draws data from greenhouse gas inventory compilation standards, general standards for carbon emission accounting in renewable resource recycling, and annual carbon emission data published by the ecological and environmental departments. It covers basic parameters such as carbon emission coefficients for virgin material production and recycled material production for over 200 materials. The parameter library is updated through a two-tier architecture of regional nodes and a central node. Three surrounding renewable resource processing enterprises are selected as regional nodes, and updated data is transmitted between the central node and regional nodes via an encrypted VPN tunnel. When a regional node's monthly material processing volume reaches a set percentage, that regional node is allocated a data contribution weight according to a preset standard, and the remaining nodes are allocated the remaining weight according to their processing volume percentage. Figure 2 As shown, during the calculation process, after receiving material composition data and equipment operating energy consumption data from the end-to-end data acquisition module, the module calculates the carbon emission coefficient of the material's entire life cycle using a weighted calculation method. It then performs a comprehensive calculation by combining multiple dimensions such as material composition, processing energy consumption, and transfer links. ,in, , , These are the weighting coefficients for material composition, processing energy consumption, and transfer links, respectively. This represents the basic carbon emission factor corresponding to the material composition. The carbon emission coefficient corresponding to a unit of energy consumption. This is the carbon emission coefficient for the transportation process. The recycled carbon emission reduction coefficient is derived by adjusting the difference between the carbon emission coefficients of virgin materials and recycled materials, combined with the material recycling loss rate. The loss rate is preset within a reasonable range according to the material type, and different materials such as plastics and metals are selected with corresponding values ​​according to their actual characteristics.

[0036] The AI ​​dual-objective optimization decision-making module adopts a collaborative architecture of edge computing nodes and cloud servers. Edge computing nodes are deployed in the control cabinet of the sorting workshop, responsible for real-time data reception and rapid decision-making, while the cloud server provides large-scale data storage and complex computational support. This module receives correlation coefficients output from the carbon footprint accounting module and various basic data from the end-to-end data acquisition module, and then uses the carbon-economic dual-objective optimization core formula to calculate the comprehensive material score. The carbon-economic dual-objective optimization core formula is: ,in, For comprehensive evaluation of materials; The carbon emission reduction weight is set at a value ranging from 0.3 to 0.7. The initial value is preset according to the enterprise's carbon reduction target and policy requirements, and is dynamically adjusted every 7 days according to the proportion of carbon trading revenue. As the weight of economic returns, by =1−α is calculated to obtain; , is the carbon emission reduction coefficient for the i-th type of material regeneration, which is derived by the carbon footprint accurate accounting module based on the difference in carbon emission coefficients between virgin and recycled materials, combined with the material regeneration loss rate; The carbon emission coefficient for the entire life cycle of material i is calculated by the carbon footprint accurate accounting module based on material composition data, energy consumption data of each stage, and basic coefficients of the carbon footprint parameter library. The unit price for recycling the i-th type of material is obtained by the end-to-end data acquisition module from the renewable resource trading platform and through multi-platform weighted processing. The carbon trading price is captured from the carbon trading platform by the end-to-end data acquisition module and updated every 4 hours. The unit processing energy consumption for the i-th type of material is calculated by the end-to-end data acquisition module based on the total energy consumption of the equipment and the material processing volume. The system uses a unit energy consumption carbon emission coefficient, linked to data released by regional environmental protection departments, and is updated quarterly. Simultaneously, based on a carbon-weighted dual-objective decision-making logic, a carbon weight-related rule base is constructed. This rule base contains multiple core rules covering the correlation logic between key influencing factors such as carbon trading prices and processing energy consumption and sorting priorities. Figure 3 As shown, during the decision-making process, the system dynamically adjusts sorting priorities by matching the rule base using a fuzzy reasoning algorithm. For example, when the carbon trading price reaches a set level, the system automatically prioritizes materials with high renewable carbon emission reduction coefficients to the highest level, giving priority to allocating robotic arm resources. The generated sorting instructions include parameters such as the material target collection bin number, the robotic arm gripping angle, and the conveyor belt speed. These instructions are transmitted to the intelligent sorting execution module via a 5G communication link, ensuring timely and reliable transmission.

[0037] The intelligent sorting execution module's adaptive variable frequency conveyor belt is made of wear-resistant rubber. Its length and width are configured according to the requirements of the work site. The drive motor power matches the conveyor belt's load capacity, and the speed adjustment range covers the required operational area. The conveyor belt speed control uses a PID algorithm, the algorithm expression of which is... , Output quantity for conveyor belt speed control The proportional coefficient is the core response parameter of the PID algorithm, used to quickly offset the current deviation. This is the deviation signal, which is the difference between the actual value of the material flow rate and the preset reference value. The integral time constant is used to eliminate the steady-state error of the system. The differential time constant is used to predict deviation trends and make adjustments in advance. The system dynamically adjusts parameters based on the material flow rate collected by the weight sensor at the feed end, matching the conveyor belt speed with the feed rate to avoid material accumulation or insufficient sorting. Four six-axis robotic arms are configured and arranged at sorting stations on both sides of the conveyor belt. Each robotic arm has the load capacity and repeatability accuracy to meet the sorting operation requirements and is equipped with a force control feedback unit. The gripping force threshold is preset within a reasonable range according to the material type, with corresponding threshold ranges selected for different materials such as plastics and metals. The vision-tactile fusion recognition component consists of a high-definition industrial camera and a pressure sensor. The industrial camera's shooting frequency meets the material recognition requirements, and the image feature extraction algorithm identifies information such as the material's shape and color. The pressure sensor collects material hardness data at the moment of gripping. The fusion of the two data enables accurate material identification. Eight graded collection bins are set up according to material type, with each bin's capacity adapted to the daily processing volume requirement. Ultrasonic level sensors are configured. When the bin capacity reaches the first-level warning threshold, a notification is displayed on the workshop screen; when the second-level warning threshold is reached, a cleaning notification is pushed to the mobile terminal of the management personnel.

[0038] The distributed ledger of the carbon emission reduction certificate storage, trading, and status feedback module adopts a consortium blockchain architecture. The nodes include a recycling center, a carbon trading platform, an environmental regulatory agency, a third-party carbon verification agency, recycling companies, and partner banks, totaling six core nodes. Each node is authenticated using digital certificates, and data is encrypted before being uploaded to the blockchain. Block generation time is controlled, and each block contains the hash value of the previous block, certificate storage data, and a timestamp, ensuring data immutability. Carbon emission reduction calculation uses a full-chain carbon emission reduction calculation formula, which is as follows: ,in, This represents the total carbon emission reduction; The processing volume of the i-th type of material is collected by the weight sensor of the collection bin of the intelligent sorting execution module; The industry benchmark full life cycle carbon emission coefficient is derived from the corresponding standard value built into the carbon footprint parameter library; The actual life-cycle carbon emission coefficient, and the core formula for optimizing the carbon-economy dual objectives. For the same variable; Let be the carbon emission reduction coefficient for the i-th type of material regeneration; To optimize carbon emission reduction conversion efficiency, a value ranging from 0.85 to 0.95 is used. Initial values ​​are preset according to material type, and minor adjustments are made every six months based on actual feedback data from recycling companies. After calculation, the system automatically generates a carbon emission reduction calculation result file, including material processing volume, carbon emission coefficient, and emission reduction details. The file is stored on the blockchain and then connected to the national carbon trading platform through a standardized interface. Equipment status monitoring uses the GRU algorithm. The algorithm's input features include equipment runtime, energy consumption fluctuation coefficient, correlation between carbon emissions and energy consumption, and component temperature trends. The algorithm is trained using historical fault data to form a predictive model, which can provide early warnings of potential equipment failures. Monitoring data is collected at fixed intervals. When the joint temperature or energy consumption fluctuation coefficient of a robotic arm exceeds the set range, the system immediately triggers an audible and visual alarm and automatically adjusts the sorting task at that workstation to other robotic arms to ensure operational continuity.

[0039] Each module achieves real-time data interaction via the MQTT protocol and 5G communication technology. The 5G network provides a transmission rate that meets operational requirements, while the MQTT protocol employs a reliable message transmission mechanism to ensure reliable data delivery. After the system was operational, daily material sorting efficiency was significantly improved compared to the traditional model, sorting accuracy reached a high level, carbon emission reduction calculation errors were controlled within a reasonable range, and stable emission reduction revenue was achieved through the carbon trading platform, significantly improving the resource utilization efficiency and economic benefits of the recycling center.

[0040] Example 2:

[0041] This embodiment provides an intelligent solid waste sorting and management system applied to a solid waste treatment center in an industrial park. This center primarily handles industrial solid waste generated by industries such as electronics manufacturing and machining, processing approximately 80 tons daily, encompassing 12 categories of materials including metal scrap, plastic offcuts, rubber waste, and industrial cardboard. Industrial materials are characterized by complex compositions, significant differences in hardness, and marked fluctuations in carbon emission coefficients, and must also meet the real-time data reporting requirements of environmental regulatory authorities. This system achieves precise and compliant industrial-grade sorting management through the integration of multiple technologies.

[0042] The hardware deployment of the end-to-end data acquisition module is optimized for industrial scenarios. The infrared spectral sensor is a high-temperature resistant model with a protection level meeting industrial environmental standards. It is installed 2 meters above the sorting conveyor belt, and the distance between the sensor and the material can be adjusted via an electric lifting frame to accommodate industrial materials of different heights. The weight sensor adopts an explosion-proof design and is installed at the feed end of the conveyor belt and the bottom of the collection bin. It can operate stably in industrial environments with high dust concentrations, and its measurement range and accuracy meet industrial data acquisition requirements. Temperature and humidity sensors are deployed in the material storage warehouse, sorting workshop, and transfer channels. The monitoring data is linked to the workshop's ventilation and dehumidification equipment. When the humidity exceeds a set value, the dehumidification equipment is automatically activated to ensure the quality of material storage. In addition to being integrated into the transfer vehicles, the GPS positioning unit is also deployed on the material turnover boxes within the workshop, tracking the material flow path within the workshop in real time, forming complete trajectory data of "external transfer + internal flow".

[0043] The data preprocessing unit adopts a server cluster architecture, consisting of multiple industrial servers. Processing tasks are distributed through a load balancing algorithm, and the hardware configuration of each server meets the rapid processing needs of massive amounts of industrial data. After cleaning and standardization, the raw data is stored in a time-series database, supporting rapid queries by time range, material type, and other dimensions. Economic data collection is integrated with multiple primary-level recycled resource trading platforms, local industrial waste trading platforms, and long-term cooperating industrial recycling companies. The unit price of recycled materials is calculated using a preset weighted average and updated every 30 minutes. To address the significant price fluctuations in industrial materials, the system includes a price anomaly monitoring mechanism. When the price fluctuation in a single update exceeds a set percentage, a manual review process is automatically triggered, and the update only takes effect after confirmation.

[0044] The carbon footprint accurate accounting module is deployed on a private cloud platform in the industrial park. Its built-in carbon footprint parameter library specifically collects data related to industrial materials, including carbon emission coefficients from the primary and recycled production of over 30 types of industrial materials such as steel, aluminum alloys, and engineering plastics. Initial data is sourced from general carbon emission accounting standards, life cycle assessment technical requirements, and publicly available data from leading companies in the recycling industry. The parameter library updates employ a two-tier architecture: regional nodes and a central node. Regional nodes are selected from the processing centers of two surrounding industrial parks, while the central node is located on the core server of its respective processing center. When the monthly material processing volume of a regional node reaches a set percentage, data contribution weights are allocated according to preset standards. Data transmission between nodes uses encryption algorithms to ensure industrial data security. During the accounting process, the full life cycle carbon emission coefficient is calculated through weighted summation, combining multi-dimensional data such as energy consumption in industrial material production processes, transportation distance, and recycling energy consumption. Energy consumption factors in corresponding production, transfer, and recycling stages are incorporated, taking into account the characteristics of different industrial wastes. The recycled carbon emission reduction coefficient is derived based on the difference in production carbon emission coefficients between primary and recycled industrial materials, adjusted for industrial waste recycling loss rates, ensuring that the accounting results closely reflect actual industrial production.

[0045] The AI ​​dual-objective optimization decision-making module adopts a collaborative model of local server and cloud computing power. The local server is responsible for handling real-time sorting decisions, while the cloud server performs big data analysis and decision model optimization. After receiving various data from the carbon footprint accounting module and the full-link data acquisition module, the module calls the core formula of carbon-economic dual-objective optimization to calculate the comprehensive score of materials. The carbon-weighted dual-objective decision-making logic is optimized for industrial scenarios, constructing a three-dimensional decision model of "carbon emission reduction benefits - economic benefits - processing costs". The model uses the analytic hierarchy process to determine the weight of each dimension. When industrial enterprises face emission reduction assessment pressure, they can manually increase the weight of carbon emission reduction benefits. During the decision-making process, the system optimizes the sorting scheme through a genetic algorithm. For example, when the carbon trading-related benefits of a certain type of industrial material meet the set conditions, its sorting priority is automatically set to the highest, giving it priority in occupying sorting resources. The generated process parameter instructions include robotic arm gripping force, conveyor belt speed, and sorting frequency. For metal scrap with high hardness, the instructions will automatically increase the robotic arm gripping force and conveyor belt speed to improve sorting efficiency.

[0046] The intelligent sorting execution module features an adaptive frequency conversion conveyor belt made of industrial-grade high-strength material with a thickness adapted to the impact requirements of industrial materials. It can withstand the impact and wear of industrial materials, and its speed adjustment range covers the needs of industrial sorting operations. It uses a frequency converter for speed control, combined with a PID algorithm to achieve precise speed adjustment. The six-axis robotic arm is an industrial collaborative model with a load capacity adapted to the heavier industrial materials. Equipped with a force control feedback unit, it can detect the gripping force in real time and automatically stop gripping when the gripping force exceeds a preset threshold multiple, preventing damage to the robotic arm or material breakage. The vision-tactile fusion recognition component is optimized for the irregular shape and rough surface of industrial materials. The industrial camera is equipped with a polarized lens to effectively eliminate the influence of surface reflection. The pressure sensor uses a high-sensitivity model to identify the hardness differences of different industrial materials. Twelve graded collection bins are set up according to the type of industrial material. Each bin has a capacity that meets industrial-grade storage requirements and is equipped with a hydraulic door. When the bin capacity reaches a warning threshold, the door automatically opens to transfer the material to the storage area, enabling continuous operation.

[0047] The distributed ledger consortium blockchain nodes for the carbon emission reduction certificate storage, trading, and status feedback module have been expanded to eight, with the addition of nodes from the industrial park management committee and industry associations to enhance data credibility. Block data adopts a "local storage + off-site backup" model. Local storage uses a disk array, while off-site backup is deployed in a backup data center at a safe distance to ensure no data loss. After carbon emission reduction calculation is completed, the system automatically organizes carbon emission reduction-related data files that meet the requirements of industrial carbon trading, including details of industrial material processing, the carbon emission calculation process, and emission reduction calculation results. These files are then submitted to the carbon trading platform after being stored on the consortium blockchain. Equipment status monitoring uses the GRU algorithm, with added industrial-specific parameters such as vibration frequency and lubricating oil temperature as input features. The algorithm model is trained on a large amount of industrial equipment fault data, achieving a high level of fault prediction accuracy. Monitoring data is fed back to the AI ​​dual-objective optimization decision module in real time. For example, when the vibration frequency of a robotic arm exceeds a threshold, the system automatically reduces its workload and adjusts the sorting tasks of other robotic arms to ensure overall system stability.

[0048] Each module interacts with the 5G industrial module via the MQTT protocol. The 5G industrial module supports edge computing, enabling data preprocessing on the device side and reducing bandwidth consumption. After the system is operational, the accuracy of industrial material sorting reaches a high standard, sorting efficiency is significantly improved, carbon emission reduction data is verified in real time by environmental regulatory authorities, and stable monthly revenue is obtained through the carbon trading platform. At the same time, the amount of industrial solid waste going to landfills is reduced, realizing the resource utilization and low-carbon management of industrial solid waste.

[0049] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A solid waste intelligent sorting and management system, characterized in that, The system includes: The end-to-end data acquisition module is equipped with multiple sensors and data preprocessing units to collect material characteristics, transfer information, economic data, and equipment operation energy consumption data of solid waste. After standardized processing, the data is transmitted through a preset communication protocol. Carbon footprint accurate accounting module: It has a built-in carbon footprint parameter library that conforms to the corresponding standards, receives the data transmitted by the above acquisition module, calculates the carbon emission coefficient and the carbon emission reduction coefficient of the material throughout its entire life cycle, and the parameter library is updated with regional specific coefficients through a federated learning mechanism; AI Dual-Objective Optimization Decision Module: Receives carbon emission coefficients from the carbon footprint accounting module and economic and energy consumption data from the acquisition module. It adopts the core formula of carbon-economic dual-objective optimization, combines carbon weight-oriented dual-objective decision logic to calculate the comprehensive score of materials, and generates sorting priority and process parameter instructions. Intelligent sorting execution module: includes an adaptive variable frequency conveyor belt, a six-axis robotic arm, a vision-tactile fusion recognition component and a graded collection bin. It receives instructions from the above decision module, executes sorting and graded collection operations, and records sorting process data synchronously. Carbon emission reduction certificate storage and trading and status feedback module: Receives carbon emission data from the carbon footprint accounting module and material processing volume data from the sorting execution module, calculates emission reductions using the full-chain carbon emission reduction calculation formula, stores the data in a distributed ledger and connects to the carbon trading platform, and simultaneously monitors equipment operating status and feeds it back to the decision-making module.

2. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The sensors in the end-to-end data acquisition module include an infrared spectral sensor, a weight sensor, a temperature and humidity sensor, and a GPS positioning unit.

3. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The economic data in the full-link data acquisition module includes the unit price of recycled materials. This unit price is calculated using a multi-platform weighted average, with the weights allocated as follows: 60% for the primary renewable resource trading platform, 30% for the local trading platform, and 10% for the long-term cooperation quotation from enterprises. The update cycle is 30 minutes.

4. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The core formula for carbon-economy dual-objective optimization in the AI ​​dual-objective optimization decision module is: ,in, For comprehensive evaluation of materials; As a weight for carbon emission reduction; Weighted by economic returns; Let be the carbon emission reduction coefficient for the i-th type of material regeneration; Let i be the carbon emission coefficient for the entire life cycle of material i. The unit price for recycling material of type i; For carbon trading prices; The unit processing energy consumption for the i-th type of material; The carbon emission coefficient per unit of energy consumption.

5. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The carbon-based dual-objective decision-making logic in the AI ​​dual-objective optimization decision-making module constructs carbon-based association rules based on carbon trading prices, carbon emission reductions, and material processing volumes, dynamically adjusts sorting priority judgment criteria, and prioritizes the allocation of sorting resources to materials with high carbon-based value.

6. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The formula for calculating the end-to-end carbon emission reduction in the carbon emission reduction certificate trading and status feedback module is as follows: ,in, This represents the total carbon emission reduction; This represents the processing volume of the i-th type of material; The industry benchmark for carbon emission coefficients throughout the entire life cycle; This represents the actual life-cycle carbon emission coefficient. Let be the carbon emission reduction coefficient for the i-th type of material regeneration; This refers to the efficiency of carbon emission reduction conversion in renewable energy.

7. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The six-axis robotic arm in the intelligent sorting execution module is equipped with a force control feedback unit; the adaptive variable frequency conveyor belt uses a PID algorithm to dynamically correct the speed; the graded collection bin is equipped with a material level warning device, which has a first-level warning and a second-level warning. The trigger threshold for the first-level warning is 75% of the bin capacity, and the trigger threshold for the second-level warning is 85% of the bin capacity.

8. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The carbon footprint parameter library in the carbon footprint accurate accounting module is updated through a two-level architecture of regional nodes and central nodes. When the processing volume of regional nodes accounts for ≥30%, the data contribution weight is set to 0.

4. The material regeneration loss rate ranges from 0.03 to 0.08 and is preset according to the material type.

9. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The distributed ledger of the carbon emission reduction certificate storage and trading and status feedback module adopts a consortium blockchain architecture and uses SHA-256 encryption for certificate storage.

10. The intelligent solid waste sorting and management system according to claim 1, characterized in that, The equipment status monitoring in the carbon emission reduction certificate trading and status feedback module adopts the GRU algorithm. The input features include the correlation between carbon emissions and energy consumption. The monitoring data is fed back to the AI ​​dual-objective optimization decision module to adjust the sorting parameters.

Images