Waste incineration treatment front-end pollution source blocking and resource recycling method and system

Abstract

The invention relates to a waste incineration treatment front-end pollution source blocking and resource recycling method, which is used for carrying out front-end pollution source blocking treatment and resource recycling treatment before waste incineration treatment, and is characterized in that a control center with an intelligent learning function is used for carrying out whole-process automatic control on the front-end pollution source blocking treatment and the resource recycling treatment; the front-end pollution source blocking treatment comprises the following steps: S1, pollution source component identification and separation; s2, humidity regulation and control; s3, performing multiple physical sorting; s4, pollution monitoring and feedback; and S5, comprehensive reactant pretreatment. The resource recycling treatment refers to plastic cleaning granulation, metal magnetic separation and electronic part thermal separation of the screened pollution source components. The front-end pollution source blocking treatment before waste incineration treatment can be accurate, thorough and controllable, stable treatment can be achieved in the front-end treatment process and the incineration process after the front-end pollution source blocking treatment, and therefore respective adverse effects caused by residual pollutants in the waste incineration process are completely eradicated.

Description

Technical Field

[0001] This invention relates to the field of waste treatment technology, and in particular to methods and systems for blocking pollution sources and recycling resources at the front end of waste incineration. Background Technology

[0002] Currently, waste incineration, especially municipal solid waste, is the mainstream method for harmless waste treatment, including post-incineration power generation. However, waste materials contain large amounts of plastics, discarded electronic products, and metallic impurities. Chlorine and bromine in plastics and PVC, along with metallic copper, chromium, and cadmium, catalyze the formation of dioxins. These substances react with chlorine during incineration. Small household appliances and microelectronic products containing heavy metals such as lead and mercury are prone to evaporation at high temperatures, forming secondary pollution; these are the main sources of dioxin formation and heavy metal evaporation pollution. Simultaneously, there are also issues with emissions of sulfur oxides, nitrogen oxides, and heavy metals such as lead.

[0003] Current methods for pollution removal during waste incineration include pre-incineration treatment of pollution sources, in-process treatment of pollution, and post-incineration treatment of pollutants in fly ash and flue gas. Existing technologies primarily focus on end-of-pipe flue gas treatment, making it difficult to effectively suppress pollution at the source. Traditional front-end sorting equipment is mostly based on manual labor or single air separation, unable to handle complex mixed components. Especially in waste environments with high humidity and high impurity ratios, sorting accuracy and stability are significantly reduced, resulting in large amounts of waste plastics, waste electronics, and small appliances in the incinerated waste. Specifically, chlorine / bromine precursors exist in PVC, chlorine-containing films, and bromine-containing flame-retardant plastics (WEEE, home textiles), providing halogen sources upon entering the furnace. HCl / HBr and metal ash co-promote dioxin regeneration. Metal catalytic centers, such as Cu, Fe, Zn, and Pb oxides, act as catalytic centers in the ash, accelerating the chlorination / rearrangement of polycyclic aromatic hydrocarbons. Mercury, including mercury (Hg), cadmium (Cd), and phosphorus (Pb), is directly volatilized from batteries, fluorescent lamps, and button cells. These substances can only be intercepted using activated carbon, dry, or wet methods. Failure to remove them at the source significantly increases the load and cost at the downstream stages.

[0004] Existing patented technologies for waste incineration include front-end processing such as sorting, like separating dry and wet waste, to allow different types of waste to be processed using different incineration methods, thus addressing issues of processing efficiency and quality. For example, patent number 202322990428.2, entitled "Waste Sorting and Screening Device for Waste Incineration," filters out plastic bags before incineration to prevent the generation of harmful gases during combustion. However, household waste also contains other plastic waste, as well as batteries, small appliances, and microelectronic products containing heavy metals. The chlorine in these other plastic products, along with the heavy metal-containing waste, also produces dioxins, a toxic substance, during combustion. Furthermore, current resource utilization in waste incineration focuses on the resource utilization of post-incineration products. Since dioxins, even when decomposed at high temperatures, can reform during cooling, and rapid cooling processes cannot guarantee against their reformation, the resource utilization of dioxin-containing products such as fly ash faces either very limited utilization or high costs.

[0005] In Europe, the MVA (Municipal Waste-to-Energy) system for waste incineration incorporates PVC and metal sorting at the front end to block pollution sources. Dioxin emissions can be reduced from 0.3 ng TEQ / Nm³ to 0.015 ng TEQ / Nm³, a 95% reduction. However, due to the random composition of mixed municipal solid waste (plastics, PVC, chlorinated / bromine-containing fuel inhibitors, lithium batteries, WEEE heavy metal parts), the sorting process remains coarse, incomplete, and unstable. Lightweight plastics, in particular, are difficult to remove. Consequently, the input of chlorine / bromine, mercury / cadmium / lead to the incinerator fluctuates constantly, leading to a series of problems. For example, the "memory effect" of filter materials and filter cake accumulation can cause the instantaneous release of harmful substances such as PCDD / F and Hg when operating conditions are unstable. Unstable operating conditions can also create a dioxin regeneration stage within the incinerator. Some heavy metals volatilize within the furnace and accumulate in fly ash / APC residue. SNCR / SCR systems also experience ammonia escape and catalyst poisoning, among other issues. Therefore, even though the EU has carried out in-depth treatment according to IED / BAT, there are still inherent defects / weak links in the incineration front-end and furnace-waste heat-flue gas treatment chain, and the control of dioxins and heavy metals at the source is still insufficient.

[0006] Taking a typical European municipal solid waste incineration power plant as an example, even with PVC and some metal sorting equipment installed at the front end, the plastic content in the waste entering the furnace remains at around 15% on a dry basis. A significant proportion of this consists of film, packaging bag fragments, and composite materials, which are difficult to completely remove through air separation or manual sorting alone. These residual plastics, along with copper, iron, and cadmium-containing metal components, enter the grate together, causing not only continuous fluctuations in the dioxin formation rate but also making it difficult to reduce SO2, HCl, and heavy metal emissions to lower levels. This has become a common bottleneck in front-end treatment faced by waste incineration plants in Europe, America, Japan, and South Korea.

[0007] Therefore, even with front-end pollution source blocking and sorting processes in waste incineration in Europe, America, and Japan, there is still a problem of about 15% plastic residue after front-end sorting. Therefore, it is necessary to develop systems and methods for municipal solid waste incineration, especially waste-to-energy plants, to thoroughly, accurately, and stably identify, block, and recycle pollution sources at the front end of the incineration process, so as to completely eliminate pollution sources and meet the dual goals of pollution reduction and resource recovery. Summary of the Invention

[0008] This invention addresses the technical problems of inaccurate, incomplete, and unstable front-end pollution source blocking in existing waste incineration processes, which fail to eliminate various adverse effects caused by residual pollutants during incineration. It provides a method and system for blocking front-end pollution sources and recycling resources in waste incineration, enabling precise, thorough, and stable blocking of front-end pollution sources in waste incineration, thereby reducing and eliminating various adverse effects caused by residual pollutants in subsequent waste incineration processes.

[0009] The technical solution of the present invention is as follows:

[0010] A method for blocking pollution sources and recycling resources at the front end of waste incineration, comprising front-end pollution source blocking treatment and resource recycling treatment before waste incineration, characterized by fully automatic control of the front-end pollution source blocking treatment and resource recycling treatment through a control center with intelligent learning function; the front-end pollution source blocking treatment includes the following steps:

[0011] S1. Pollution Source Component Identification and Separation: The control center uses intelligent identification and AI robotic arms to break open bags, identify, and sort waste materials, separating pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances; the remaining waste materials proceed to the next step; the intelligent identification also includes obtaining identification data of the waste materials, and the control center also obtains the operating parameters of this step;

[0012] S2. Humidity control: After the control center comprehensively calculates the pollution potential index based on the identification data of the remaining waste material in S1, it automatically triggers the adjustment of the moisture content of the remaining waste material to 50%–150%. The moisture content is on a dry basis, i.e., water / dry solids = 0.5–1.5. The operating parameters of the remaining waste material are continuously monitored, including the moisture content.

[0013] S3. Multiple physical sorting: This includes the following steps (S31 and S32) and the detection of operating parameters, either individually or in combination:

[0014] S31, Gravity differential air separation: The residual waste material after humidity control in S2 is separated from the residual waste material by low temperature hot air using density difference. The pollution source components also include lightweight plastics.

[0015] S32, Ionization Response Electrostatic Separation: refers to the further separation of residual lightweight plastics from the remaining waste material in S2 or S31 by using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics.

[0016] S4. Pollution monitoring and feedback: At the sorting outlet of S32, the content of chlorine, sulfur oxides, heavy metals and NOx are detected online. Based on the detection data, the dioxin formation trend is formed. The control center combines the identification data of S1 and the operating parameters of each step to monitor pollution and adjust the operating parameters of the above four steps in order to suppress and control the pollution trend in real time.

[0017] S5. Pretreatment with Integrated Reactant: Before the waste material after multiple physical sorting in S3 enters the waste incinerator, an integrated reactant is added by spraying or mixing. The integrated reactant includes one or more of alkaline adsorbents, porous silica-alumina framework materials, and heavy metal trapping agents, used to adsorb and fix hydrogen chloride precursors, sulfur oxide precursors, and free heavy metal ions in the material before incineration, thereby reducing the available chlorine, available sulfur, and active heavy metal content of the material entering the furnace. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained in the pollution monitoring and feedback step in S4.

[0018] The control center calculates the pollution potential index PI based on the identified data, and uses this index as the control target to dynamically correct the operating parameters of at least one step from S2 to S4, forming a closed-loop control of identification-separation-feedback. The pollution source components are removed at the front end through physical means in S1 to S3, and combined with the comprehensive reactant pretreatment based on pollution trends in S4 to S5, the graded blocking and furnace-front reduction of dioxin precursors and heavy metals are achieved.

[0019] Preferably, the resource recycling process refers to the plastic washing and granulation, metal magnetic separation, and electronic component thermal separation of the screened pollution source components; the method also includes the steps of sending the remaining waste material after S5 treatment into a waste incinerator for incineration and using the high heat energy of the incineration process to generate electricity; S1 also includes the step of AI robotic arm pre-spreading the waste; the control center adopts an AI control strategy based on a deep learning model.

[0020] Preferably, the intelligent identification in S1 includes using AI visual recognition, near-infrared spectroscopy detection, and metal response detection with a metal response detector to obtain identification data and transmit it to the control center, which then controls the AI ​​robotic arm to perform bag breaking and sorting. The humidity control in S2 automatically calculates the optimal humidity range based on the moisture content. Specifically, S1 includes:

[0021] S11. Identification and Separation of Large Solid Items: AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Items with dimensions greater than 150 mm and featuring shells or cables are identified as large discarded household appliances or electronic waste. Metal response detectors identify items with regular metal reflection characteristics as metal parts. The identification data includes the size, shape, and reflection characteristics of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robotic arm, which prioritizes grabbing these pollution source components (large household appliances, electronic waste, and metal parts) and transports them along with their identification data to the resource recycling process, thus achieving priority removal of large-sized objects, electronic parts, and metal pollution source components.

[0022] S12. Separation of rigid plastics and film plastics: For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter the resource recycling process for plastic washing and granulation. For film plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is retained by the control center in step S3 for further rejection of film plastics with this marking data in the humidity control and multiple physical sorting steps.

[0023] S13. The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the downstream magnetic separator and eddy current separator to separate the ferromagnetic metal and non-ferrous metal to the corresponding resource recycling channels, thereby achieving complete separation from plastic and combustible waste materials.

[0024] Preferably, the operating parameters of gravity differential air separation in S31 include an air velocity of 2–8 m / s, and the operating parameters of ionization response electrostatic separation in S32 include a high-voltage electrostatic field voltage of 20–50 kV. The high-voltage electrostatic field elucidates the difference in charge response, causing lightweight plastics to deflect and separate from other waste materials. The operating parameters of gravity differential air separation and ionization response electrostatic separation also include a voltage adjustment frequency, which achieves dynamic response within the range of 0.5–2 Hz. 5. The method for blocking pollution sources and recycling resources at the front end of waste incineration treatment according to claim 4, characterized in that the humidity control in S2 and the air separation in S31 are linked; the operating parameters include the material layer pressure difference signal, stratification rules, material density, and air velocity in S31 of the waste materials in each step; the control center uses AI to optimize online with unit separation power consumption / separation purity as the objective function, and dynamically fine-tunes the moisture content setpoint under the pollution monitoring and feedback triggering in S4.

[0025] Preferably, the ionization response electrostatic separation of S32 includes traditional electrostatic separation, deflection plate type and drum electrode type separation; the heavy metal content mentioned in S4 includes the content of mercury, lead and cadmium.

[0026] This invention also provides a front-end pollution source blocking and resource recovery system for waste incineration, including an AI-enabled control center. The system further includes a front-end pollution source blocking treatment device and a resource recovery device automatically controlled by the control center, both equipped with intelligent learning capabilities. The front-end pollution source blocking treatment device performs front-end pollution source blocking treatment on the waste before incineration, comprising a pollution source component identification and separation module, a humidity control module, a multiple physical sorting module, a pollution monitoring and feedback module, and a comprehensive reactant pretreatment module connected in sequence. The control center, through signal interaction and parameter linkage with each device and module, achieves the identification, separation, extraction, and feedback control of pollution source components, thereby blocking the pollution source from entering the waste incineration process.

[0027] The pollution source component identification and separation module includes an intelligent identification device and an AI robotic arm to perform bag breaking, identification, and sorting of waste materials, separating out pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances. The remaining waste materials after separating out the pollution source components enter the humidity control module. The intelligent identification also includes obtaining identification data and operating parameters of the waste materials.

[0028] The humidity control module automatically adjusts the moisture content of the remaining waste material after the control center calculates the pollution potential index based on the identification data of the remaining waste material. The operating parameters include moisture content.

[0029] The multi-physical sorting module implements physical sorting, including a gravity differential air separation submodule and an ionization response electrostatic sorting submodule. The gravity differential air separation submodule uses density difference to separate lightweight plastics from the remaining waste material using low-temperature hot air. The pollution source components also include lightweight plastics. The ionization response electrostatic sorting submodule further separates the remaining lightweight plastics from the waste material from the gravity differential air separation module using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics.

[0030] The pollution monitoring and feedback module detects chlorine, sulfur oxides, heavy metals and NOx online. Based on the detection data, it forms a dioxin formation trend. The control center combines the identified data and the operating parameters of each module to monitor pollution and adjust the operating parameters of the four modules to suppress and regulate the pollution trend in real time.

[0031] The integrated reactant pretreatment module adds an integrated reactant to the remaining waste material after it has been sorted by the multiple physical sorting module, either by spraying or mixing, before it enters the waste incinerator. The integrated reactant includes one or more of the following: alkaline adsorbent, porous silica-alumina framework material, and heavy metal trapping agent. It is used to adsorb and fix hydrogen chloride precursors, sulfur oxide precursors, and free heavy metal ions in the remaining waste material before incineration, thereby reducing the available chlorine, available sulfur, and active heavy metal content of the material entering the furnace. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained from the pollution monitoring and feedback module.

[0032] The control center calculates the pollution potential index PI based on the identified data, and uses this index as the control target to dynamically correct the operating parameters of each module, forming a closed-loop control of identification-separation-feedback. The pollution source components are physically removed at the front end through the pollution source component identification and separation module, humidity control module and multiple physical sorting module. Combined with the pollution monitoring and feedback module and the comprehensive reactant pretreatment module based on pollution trends, the system achieves graded blocking and furnace-front reduction of dioxin precursors and heavy metals.

[0033] Preferably, the resource recycling device includes a plastic washing and granulating machine, a metal magnetic separator and eddy current separator, and an electronic component thermal separation device, which respectively wash and granulate the plastic, perform metal magnetic separation, and thermal separation of electronic components screened by the front-end pollution source blocking treatment device; the system also includes a waste incineration module, which sends the remaining waste material from the integrated reaction agent pretreatment module into the waste incinerator for incineration and power generation; the pollution source component identification and separation module also uses an AI robotic arm to pre-spread the waste.

[0034] Preferably, the intelligent identification device of the pollution source component identification and separation module includes an AI visual recognition component, a three-dimensional contour scanning device, a near-infrared spectroscopy detection device, and a metal response detector. These are used to perform AI image recognition, spectral feature analysis, and metal detection on the incoming waste material, respectively, to obtain identification data of the waste material and pollution source components, including plastics, electronic components, and metal impurities. This data is then transmitted to the control center, which controls an AI robotic arm to perform bag breaking and sorting. For the remaining waste material after separating the pollution source components, the humidity control module triggers humidity control, and the multiple physical sorting module performs physical sorting based on the identification data. The identification data includes the quantity, size, and status data of the waste material and its pollution source components, including plastics, batteries, metals, and electronic components. The division of labor and cooperation among the components of the pollution source component identification and separation module include:

[0035] AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Individuals with a size greater than 150 mm and with shell or cable features are identified as large discarded household appliances or electronic waste. Metal response detectors identify individuals with regular metal reflection features as metal parts. The identification data includes the identification results of the size, shape, and reflection features of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robot arm, which prioritizes picking up the above-mentioned large household appliances, electronic waste, and metal parts, and transports these pollution source components and their identification data to the resource recycling and processing device to achieve priority removal of large-sized objects, electronic parts, and metal pollution source components.

[0036] For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance, color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter a resource recycling device for plastic cleaning and granulation. For film-type plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is transmitted to a humidity control module and a multi-physical sorting module for further rejection of film-type plastics with the marked data.

[0037] The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the metal magnetic separator and eddy current separator in the resource recycling and processing device, thereby leading the ferromagnetic metal and non-ferrous metal to the processing channel of the resource recycling device, achieving complete separation from plastic and combustible waste materials.

[0038] Preferably, the humidity control module operates in conjunction with the gravity differential air separation submodule in the multi-physical sorting module. The humidity control module automatically adjusts the airflow humidity based on an AI control algorithm by the control center, with a humidity control range of 50%–150%, preferably 60–120%, to meet the gravity differential air separation conditions without requiring a fixed drying ratio. The pollution source component identification and separation module is connected to both the humidity control module and the gravity differential air separation submodule via an industrial bus to form a feedback closed loop, automatically correcting the humidity and wind speed control parameters when changes in material characteristics are detected.

[0039] Preferably, the gravity differential air separation submodule has a wind speed range of 2–8 m / s and a differential pressure range of 0.1–0.3 MPa; the ionization response electrostatic separation submodule operates at a voltage of 20–50 kV and utilizes the difference in surface conductivity of different materials to generate a shift, thereby achieving the separation of plastics and electronic components.

[0040] Preferably, the pollution monitoring and feedback module includes a dioxin formation trend prediction algorithm. This module monitors the chlorine and heavy metal content in the remaining waste material after sorting by the multiple physical sorting module (i.e., the material before incineration) in real time and feeds the results back to the control center to dynamically adjust the parameters of each module. Based on the detected changes in dioxin or heavy metal concentration, the pollution monitoring and feedback module automatically adjusts the wind speed / volume (or fan frequency) of the gravity differential air separation submodule and the electric field strength of the ionization response electrostatic separation submodule through the control center. The control center generates pollution reduction strategies based on pollution monitoring data using AI algorithms and records operating parameters for traceability.

[0041] Technical effects of the invention:

[0042] Dioxin formation requires three core conditions: chlorinated organic matter, a metal catalyst (mainly copper, iron, cadmium, etc.), and a temperature range of 250–450°C. In municipal solid waste, chlorinated plastics such as PVC and PVDC account for approximately 15–25%, while metal components containing Cu, Fe, and Ni, such as electronic waste, circuit boards, and wires, account for approximately 2–3%. The front-end pollution source blocking treatment of this invention separates all potentially dioxin-generating pollutant components—plastics, batteries, metal parts, small appliances, and electronic products—after obtaining accurate identification data through intelligent recognition. This includes an AI robotic arm that breaks open bags, identifies, and sorts the waste materials, separating the vast majority of pollutant components, including plastics, batteries, metals, electronic parts, and large and small appliances. Solids in the plastics can be separated by the AI ​​robotic arm, but soft or lightweight plastics may remain, such as the plastic film remaining after the garbage bag is broken. The method and apparatus of this invention enhance the gas-solid coupling and buoyancy stability of lightweight particles (especially films and sheet plastics) by controlling the humidity of residual waste materials. This allows for further separation of these lightweight plastic pollutant components through multiple physical sorting methods. Simultaneously, pollution detection and feedback are implemented throughout the process. Dioxin and heavy metal content are detected at the separation outlet, and dioxin formation trends are identified and controlled online in real time to suppress these pollution trends. This ensures precise, thorough, and controllable front-end pollution source blocking before waste incineration, achieving stable treatment both in the front-end process and the subsequent incineration process. Therefore, it eliminates the adverse effects caused by residual pollutants during waste incineration.

[0043] After removing the main plastics, metals, and WEEE at the front end, a comprehensive reactant is sprayed onto the remaining combustible matrix (paper fiber, organic matter, a small amount of low-chlorinated plastics, etc.) to further pre-fix / neutralize chlorine, sulfur, and free heavy metals before the furnace. Specifically, after implementing the S5 comprehensive reactant pretreatment, the mass fraction of soluble chlorine and sulfur in the furnace feed material is further reduced by 20-30% on top of the front-end pollution source blocking. The leaching concentration of heavy metals in fly ash decreases by approximately 50% compared to the condition without the comprehensive reactant. Simultaneously, the peak concentrations of SO2 and HCl become smoother, further reducing the load on the desulfurization, dry / semi-dry absorption towers, and activated carbon injection system by 10-15%.

[0044] After processing with the method and system of this invention, compared with existing front-end pollution source blocking treatments, dioxin formation is reduced by ≥95%; heavy metal emissions are reduced by ≥80%; SO2 emissions are reduced by approximately 80%-85%; NOx emissions are reduced by approximately 40%; flue gas purification agent dosage is reduced by 60%; operating costs are reduced by approximately 50%; heavy metal content in ash is reduced by ≥80%; plastic extraction rate is ≥90%; energy recovery efficiency is improved by approximately 12%; furnace corrosion rate is reduced by 50%; and equipment lifespan is extended by approximately 3 years. Moreover, due to the automated operation of the system, human intervention is reduced, ensuring both safety and continuity. The method and system of this invention not only improve the environmental indicators of the waste incineration process but also achieve high-value resource recovery and carbon reduction.

[0045] Because this invention separates the components of each pollution source before waste incineration, different pollution sources can be recycled and treated separately, including plastic washing and granulation, metal magnetic separation, and electronic component thermal separation, thereby effectively improving the efficiency of energy recovery and utilization.

[0046] Furthermore, the pollution source component identification and separation steps further adopt a three-level identification and sorting process: (1) S11, identification and separation of large solid components: large household appliances / waste microelectronics (electronic waste) / metal parts are first identified and separated; (2) then rigid plastics are separated from thin film plastics; and (3) finally, metal parts are further separated. Moreover, these identification and separation processes also provide identification data and control means for subsequent multiple physical sorting and resource recycling. This enables various data in the method and system of this invention to support each other and form a coordinated overall processing method and system.

[0047] Physical sorting methods include gravity differential air separation and ionization response electrostatic separation. Gravity differential air separation uses density difference to separate light plastics from heavy plastics with low-temperature hot air, while ionization response electrostatic separation uses a high-voltage electrostatic field to separate plastics from metals and electronic components. These physical sorting methods are simple, low-cost, and highly efficient.

[0048] Humidity control and air separation are linked, which can better separate lightweight plastics and heavy impurities according to density.

[0049] The control center uses AI to perform online optimization with unit separation power consumption / separation purity as the objective function, and dynamically fine-tunes the moisture content setpoint under the pollution detection and feedback of S4, which enables more thorough separation of plastics and other components.

[0050] Ionization-responsive electrostatic separation can separate plastic and metal electronic components using a high-voltage electrostatic field.

[0051] The pollution monitoring and feedback described in S4 uses the detection of dioxin content to obtain the result by detecting chlorine content. The heavy metal content includes the content of mercury, lead, and cadmium, thereby accurately reflecting the degree to which the components of the pollution source are blocked. Attached Figure Description

[0052] Figure 1 This is a flowchart of the method of the present invention;

[0053] Figure 2 This is a schematic diagram of the system structure of the present invention; Detailed Implementation

[0054] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings:

[0055] See Figure 1 The present invention relates to a method for blocking pollution sources and recycling resources at the front end of waste incineration. This method involves blocking pollution sources and recycling resources before waste incineration. The method is characterized by fully automated control of both processes by a control center with intelligent learning capabilities. The blocking of pollution sources includes the following steps:

[0056] S1. Pollution Source Component Identification and Separation: The control center uses intelligent identification and AI robotic arms to break open bags, identify, and sort waste materials, separating pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances; the remaining waste materials proceed to the next step; the intelligent identification also includes obtaining identification data of the waste materials, and the control center also obtains the operating parameters of this step;

[0057] S1's intelligent identification includes AI visual recognition, near-infrared spectroscopy detection, and metal response detection using a metal response detector. AI visual recognition, based on a deep learning model, identifies the appearance characteristics of plastics, metals, and electronic waste; near-infrared spectroscopy detects the type of plastic and chlorine-containing characteristic peaks; and the metal response detector identifies conductive materials such as discarded small household appliances, circuit boards, and metal casings. After obtaining the identification data of the above-mentioned pollution source components, the identification data is transmitted to the control center, which then controls the AI ​​robotic arm to perform bag breaking and sorting. S1, pollution source component identification and separation, also includes the AI ​​robotic arm spreading the waste material. The intelligent identification and the AI ​​robotic arm's identification, bag breaking, spreading, and sorting of waste material can be implemented using another patent of the applicant of this invention: patent number 202510582822.3, invention titled "Waste Feeding, Bag Breaking, and Spreading System and Method Based on Artificial Intelligence Bionic Robotic Arm". The AI ​​robotic arm can be the first, second, or third robotic arm in that patent. For example, foreign objects and large items removed by the third robotic arm can be set as pollution source components through the control center. These include front-end pollution source components such as plastics, batteries, metals, electronic components, and large and small household appliances, which may adversely affect the combustion of the remaining waste materials after subsequent sorting and separation. This allows for the identification and separation of most pollution source components before waste incineration.

[0058] However, while AI robotic arms can sort out hard plastics, block plastics, and fragments of some bagged plastics, they cannot effectively sort out soft plastics such as candy wrappers and small films embedded in the remaining waste materials. This is the fundamental reason why, although other waste incineration systems achieve good results in blocking pollution sources at the front end, the process is not thorough, and the precision of front-end pollution source blocking is not high, thus affecting the instability of the subsequent waste incineration process and leading to uncontrollable pollution. Therefore, the method of this invention further performs the following more thorough and precise treatment of soft plastics: S2, Humidity Control: After the control center comprehensively calculates the pollution potential index based on the identification data of the remaining waste materials in S1, it automatically triggers the adjustment of the moisture content of the remaining waste materials to 50%–150%, where the moisture content is on a dry basis, i.e., water / dry solids = 0.5–1.5. The operating parameters of the remaining waste materials, including the moisture content, are continuously monitored. Humidity control can enhance the gas-solid coupling and floating stability of these lightweight particles (especially films and sheet plastics), allowing for the separation of these lightweight plastic pollution source components through further multiple physical sorting methods. The humidity control described in S2 automatically calculates the optimal humidity range based on the moisture content to improve sorting accuracy.

[0059] The pollution potential index can be calculated using the following formula:

[0060] PI = Ti × Li × Qi

[0061] PI: Pollution potential index. Ti: Toxicity classification of pollutant components (dimensionless); Li: Probability classification of pollutant component release (depending on the adequacy of containment measures); Qi: Release quantity classification of pollutant components.

[0062] After the identification data is comprehensively calculated by the control center to obtain the pollution potential index, the humidity control and the operation parameters of each sorting step are automatically triggered to achieve intelligent closed-loop control of identification-separation-feedback.

[0063] S3. Multiple physical sorting: This includes the following steps (S31 and S32) and the detection of operating parameters, either individually or in combination:

[0064] S31. Gravity differential air separation: The density difference of the remaining waste material after humidity control in S2 is used to separate lightweight plastics from the remaining waste material through low-temperature hot air. Therefore, the separated pollution source components also include lightweight plastics. This step can use the air separator and its separation method in the applicant's application number 202410257794.3, entitled "A Separation Device and Separation Method for a Mixture of Plastic, Paper and Textile Waste", to separate lightweight plastics, such as candy wrappers and plastic bag fragments, from the remaining waste material.

[0065] S32, Ionization Response Electrostatic Separation: refers to the further separation of residual lightweight plastics from the remaining waste material in S2 or S31 by using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics.

[0066] The sorting in this step can be implemented using the ionization-responsive electrostatic sorting method and apparatus described in the invention patent of the applicant, No. 202510632269.X, entitled "Electrostatic Sorting System, Method and Apparatus Based on Internet of Things and Artificial Intelligence". This further separates the residual lightweight plastics from the remaining waste material after step S31, making the front-end pollution source blocking treatment more thorough and precise than existing front-end pollution source blocking methods. The ionization-responsive electrostatic sorting in step S32 includes traditional electrostatic sorting, deflection plate type, and drum electrode type sorting.

[0067] The operating parameters of S31 gravity differential air separation include an air velocity of 2–8 m / s. The operating parameters of S32 ionization response electrostatic separation include a high-voltage electrostatic field voltage of 20–50 kV. The high-voltage electrostatic field explains the difference in charge response, causing lightweight plastics to be separated from other waste materials. The operating parameters of gravity differential air separation and ionization response electrostatic separation also include a voltage adjustment frequency, which achieves dynamic response in the range of 0.5–2 Hz.

[0068] The humidity control described in S2 and the air separation described in S31 are linked; the operating parameters include the material layer pressure difference signal, stratification rules, material density, and wind speed in S31 of the waste material in each step; the control center uses AI to optimize online with unit separation power consumption / separation purity as the objective function, and dynamically fine-tunes the moisture content setpoint under the pollution monitoring and feedback trigger in S4.

[0069] This invention introduces S2 humidity control and S3 gravity differential air separation + ionization response electrostatic separation linkage after S1. By adjusting the material moisture content to 50-150% (dry basis water / dry solid ratio 0.5-1.5), the gas-solid coupling of lightweight sheet plastics is enhanced. Then, low-temperature hot air at 2-8 m / s and an electrostatic field of 20-50 kV are used to classify, deflect and completely remove such residual lightweight plastics, fundamentally solving the problem of difficult removal of 'film residue' in mixed plastics.

[0070] Preferably, in step S1, hierarchical identification and channel-specific processing are also adopted, specifically:

[0071] S11. Identification and Separation of Large Solid Items: AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Items with dimensions greater than 150 mm and featuring shells or cables are identified as large discarded household appliances or electronic waste. Metal response detectors identify items with regular metal reflection characteristics as metal parts. The identification data includes the size, shape, and reflection characteristics of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robotic arm, which prioritizes grabbing these pollution source components (large household appliances, electronic waste, and metal parts) and transports them along with their identification data to the resource recycling process, thus achieving priority removal of large-sized objects, electronic parts, and metal pollution source components.

[0072] S12. Separation of rigid plastics and film plastics: For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter the resource recycling process for plastic washing and granulation. For film plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is retained by the control center in step S3 for further rejection of film plastics with this marking data in the humidity control and multiple physical sorting steps.

[0073] S13. The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the downstream magnetic separator and eddy current separator to separate the ferromagnetic metal and non-ferrous metal to the corresponding resource recycling channels, thereby achieving complete separation from plastic and combustible waste materials.

[0074] This involves three-level identification and sorting. Existing AI-based robotic arm-based front-end sorting systems primarily target hard plastics, bulk combustibles, and large WEEE (Waste Effective Waste). When processing lightweight sheet materials such as broken soft plastic films and candy wrappers, the sorting efficiency and accuracy significantly decrease due to their tendency to entangle and difficulty in stable gripping. As a result, even with multiple robotic arms deployed at the front end, approximately 10-15% of mixed plastics remain in the incinerator waste. Therefore, during the three-level identification process in step S1, for film-type plastics with a large area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area. This facilitates the clear monitoring of the processing status of these residual lightweight plastics during the S2 humidity control and S3 gravity differential air separation + ionization response electrostatic sorting linkage processes introduced after S1, ensuring their complete removal and fundamentally solving the problem of difficult-to-remove 'film residue' in mixed plastics.

[0075] S4. Pollution Monitoring and Feedback: At the sorting outlet of S32, online detection of dioxin content, chlorine content, sulfur oxide content, heavy metal content, and NOx index is performed. Based on the detection data, a dioxin formation trend is formed. This formation trend can also be generated by combining a preset model. The control center combines the identification data of S1 and the operating parameters of each step to perform pollution monitoring and adjust the operating parameters of the above four steps to perform real-time pollution trend suppression and control. The detection of dioxin content in the pollution monitoring and feedback is obtained by online detection of chlorine content at the outlet. The heavy metal content includes the content of mercury, lead, and cadmium, and also includes the detection of SOx and NOx indexes.

[0076] S5. Pretreatment with Integrated Reactant: Before the waste material after multiple physical sorting in S3 enters the waste incinerator, an integrated reactant is added by spraying or mixing. The integrated reactant includes one or more of alkaline adsorbents, porous silica-alumina framework materials, and heavy metal trapping agents, which are used to adsorb and fix hydrogen chloride precursors, sulfur oxide precursors, and free heavy metal ions in the material before incineration, thereby reducing the available chlorine, available sulfur, and active heavy metal content of the material entering the furnace. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained in the pollution monitoring and feedback step in S4.

[0077] In a preferred embodiment, the combined reactant comprises, by mass fraction: 50–80% of an alkaline adsorbent (such as one or more of CaO, Ca(OH)2, and NaHCO3), 15–40% of a porous silica-alumina framework powder (such as fly ash, zeolite powder, or bentonite), and 2–10% of a heavy metal trapping agent (such as iron- or aluminum-based inorganic salts or organic chelating agents), while avoiding the use of metal salts containing copper, nickel, or other substances that easily promote dioxin formation. The combined reactant is added at a ratio of 1–5 wt.% of the dry basis mass of the remaining waste material.

[0078] The resource recycling process refers to the process of washing and granulating plastics, magnetically separating metals, and thermally separating electronic components from the screened pollutant components. The method also includes the steps of sending the remaining waste material after the S5 process into a waste incinerator for incineration and using the high heat energy from the incineration process to generate electricity.

[0079] See Figure 2 The present invention also relates to a front-end pollution blocking and resource recovery system for waste incineration, comprising an AI-enabled control center, characterized in that it further comprises a front-end pollution source blocking treatment device and a resource recovery device automatically controlled by the control center with intelligent learning capabilities; the front-end pollution source blocking treatment device performs front-end pollution source blocking treatment on the waste before incineration, comprising a pollution source component identification and separation module, a humidity control module, a multiple physical sorting module, a pollution monitoring and feedback module, and a comprehensive reactant pretreatment module connected in sequence; the control center, through signal interaction and parameter linkage with each device and its modules, realizes the identification, separation, extraction, and feedback control of pollution source components, thereby blocking the pollution source from entering the waste incineration;

[0080] The pollution source component identification and separation module uses intelligent identification and AI robotic arms to break open bags, identify, and sort waste materials, separating out pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances. The remaining waste materials after separating the pollution source components enter the humidity control module. The intelligent identification also includes obtaining identification data of the waste materials, and the control center also obtains the operating parameters of each module.

[0081] The pollution source component identification and separation module includes an intelligent identification device and an AI robotic arm. The intelligent identification device includes an AI visual recognition component, a 3D contour scanning device, a near-infrared spectroscopy detection device, and a metal response detector. These are used to perform AI image recognition (AI visual recognition), spectral feature analysis, and metal detection on the incoming waste material, respectively. The AI ​​visual recognition component uses a deep learning model to identify the appearance characteristics of plastics, metals, and electronic waste. The near-infrared spectroscopy detection device detects the type of plastic and chlorine-containing characteristic peaks. The metal response detector identifies conductive materials such as discarded small household appliances, circuit boards, and metal casings. This module obtains identification data of the waste material and its pollution source components, including plastics, electronic components, and metal impurities, and transmits this data to the control center. The control center then controls the AI ​​robotic arm to perform bag breaking and sorting. For the remaining waste material after separating the pollution source components, the humidity control module triggers the humidity control module and the multiple physical sorting module to perform physical sorting based on the identification data. The identification data includes the quantity, size, and state data of the waste material and its pollution source components, such as plastics, batteries, metals, and electronic components. The pollution source component identification and separation module also uses the AI ​​robotic arm to pre-spread the waste.

[0082] The intelligent identification device and AI robotic arm for identifying, breaking bags, spreading, and sorting waste materials can utilize corresponding components from another patent of the applicant: Patent No. 202510582822.3, entitled "Waste Feeding, Bag Breaking, and Spreading System and Method Based on Artificial Intelligence Bionic Robotic Arm". The AI ​​robotic arm can be the first, second, and third robotic arms in that patent. For example, foreign objects and large items removed by the third robotic arm can be set as pollution source components through the control center, including front-end pollution source components such as plastics, batteries, metals, electronic components, and large and small household appliances that will adversely affect the combustion of the remaining waste materials after subsequent sorting and separation, thereby achieving the identification and separation of most pollution source components before waste incineration.

[0083] However, while AI robotic arms can sort out hard plastics, block-shaped plastics, and fragments of some bagged plastics, they cannot effectively sort out soft plastics such as candy wrappers and small films embedded in the remaining waste materials. This is the most important reason why, although other waste incineration systems achieve good results in blocking pollution sources at the front end, the process is not thorough, and the precision of front-end pollution source blocking is not high, thus affecting the instability of the subsequent waste incineration process and leading to uncontrollable pollution. Therefore, the method of this invention further improves the thoroughness and precision of processing soft plastics as follows:

[0084] The division of labor and cooperation among the components of the pollution source component identification and separation module includes:

[0085] AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Individuals with dimensions greater than 150 mm and featuring shells or cables are identified as large discarded household appliances or electronic waste. A metal response detector identifies individuals with regular metal reflection characteristics as metal parts. The identification data includes the identification results of the size, shape, and reflection characteristics of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robotic arm, which prioritizes picking up the aforementioned large household appliances, electronic waste, and metal parts, and transports these pollution source components and their identification data to the resource recycling and processing device, thereby achieving the priority removal of large-sized objects, electronic parts, and metal pollution source components.

[0086] For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance, color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter a resource recycling device for plastic cleaning and granulation. For film-type plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is transmitted to a humidity control module and a multi-physical sorting module for further rejection of film-type plastics with the marked data.

[0087] The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the metal magnetic separator and eddy current separator in the resource recycling and processing device, thereby leading the ferromagnetic metal and non-ferrous metal to the processing channel of the resource recycling device, achieving complete separation from plastic and combustible waste materials.

[0088] The humidity control module automatically adjusts the moisture content of the remaining waste material after the control center calculates the pollution potential index based on the identification data of the remaining waste material; for the specific functions of this module, please refer to the content of the humidity control step S2.

[0089] The multi-physical sorting module implements physical sorting, including a gravity differential air separation submodule and an ionization response electrostatic sorting submodule. The gravity differential air separation submodule uses density difference to separate lightweight plastics from the remaining waste material through low-temperature hot air. The pollution source components also include lightweight plastics. The gravity differential air separation submodule can use the air separator and its separation method in the applicant's application number 202410257794.3, entitled "A Separation Device and Separation Method for Waste Mixtures of Plastics, Paper, and Textiles," to separate lightweight plastics, such as candy wrappers and plastic bag fragments, from the remaining waste material. The ionization-responsive electrostatic separation submodule further separates residual lightweight plastics from the remaining waste material from the gravity differential air separation submodule using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics. The ionization-responsive electrostatic separation submodule can be implemented using the ionization-responsive electrostatic separation method and device described in the invention patent with patent number 202510632269.X, entitled "Electrostatic Separation System, Method and Device Based on Internet of Things and Artificial Intelligence".

[0090] The humidity control module operates in conjunction with the gravity differential air separation submodule in the multi-physical sorting module. The humidity control module automatically adjusts the airflow humidity based on an AI control algorithm by the control center. Its humidity control range is 50% to 150%, preferably 60-120%, to meet the gravity differential air separation conditions without the need for a fixed drying ratio. The pollution source component identification and separation module is connected to the humidity control module and the gravity differential air separation submodule through an industrial bus to form a feedback closed loop. When changes in material characteristics are detected, the humidity and wind speed control parameters are automatically corrected.

[0091] The gravity differential air separation submodule has a wind speed range of 2–8 m / s and a differential pressure range of 0.1–0.3 MPa; the ionization response electrostatic separation submodule operates at a voltage of 20–50 kV and utilizes the difference in surface conductivity of different materials to generate a shift, thereby achieving the separation of plastics and electronic components.

[0092] The pollution monitoring and feedback module detects chlorine levels online to determine dioxin content, and also detects sulfur oxides, heavy metals, and NOx levels. Based on the detection data, it generates a dioxin formation trend, which can also be generated based on a preset model. The control center combines the identified data with the operating parameters of each module to monitor pollution and adjust the operating parameters of the four modules for real-time pollution trend suppression and control. The pollution monitoring and feedback module includes a dioxin formation trend prediction algorithm. This module detects the chlorine and heavy metal content in the remaining waste material after sorting by the multiple physical sorting module (i.e., the material before incineration) in real time and feeds the results back to the control center for dynamic adjustment of module parameters. Based on the detected changes in dioxin or heavy metal concentrations, the pollution monitoring and feedback module automatically adjusts the wind speed / volume (or fan frequency) of the gravity differential air separation submodule and the electric field strength of the ionization response electrostatic separation submodule through the control center. The control center uses an AI algorithm to generate pollution reduction strategies based on pollution monitoring data and records operating parameters for traceability.

[0093] The resource recycling device is automatically controlled by the control center and includes a plastic washing and granulation device, a metal magnetic separation and separation device, and an electronic component thermal separation device, which respectively wash and granulate the plastic, perform metal magnetic separation, and thermal separation of electronic components screened by the front-end pollution source blocking treatment device.

[0094] The integrated reactant pretreatment module adds an integrated reactant to the remaining waste material after it has been sorted by the multiple physical sorting module, either by spraying or mixing, before it enters the waste incinerator. The integrated reactant includes one or more of the following: alkaline adsorbent, porous silica-alumina framework material, and heavy metal trapping agent. It is used to adsorb and fix hydrogen chloride precursors, sulfur oxide precursors, and free heavy metal ions in the remaining waste material before incineration, thereby reducing the available chlorine, available sulfur, and active heavy metal content of the material entering the furnace. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained from the pollution monitoring and feedback module.

[0095] The control center calculates the pollution potential index PI based on the identified data, and uses this index as the control target to dynamically correct the operating parameters of each module, forming a closed-loop control of identification-separation-feedback. The pollution source components are physically removed at the front end through the pollution source component identification and separation module, humidity control module and multiple physical sorting module. Combined with the pollution monitoring and feedback module and the comprehensive reactant pretreatment module based on pollution trends, the system achieves graded blocking and furnace-front reduction of dioxin precursors and heavy metals.

[0096] The system also includes a waste incineration module, which feeds the remaining waste material from the integrated reactant pretreatment module into the waste incinerator for incineration and power generation.

[0097] Explanation of the basis for the technical effects of the method and system of this invention:

[0098] (i) Scientific inference that the dioxin (PCDD / Fs) reduction rate is ≥95%

[0099] 1. Testing revealed that this invention reduces the metal pollution components in electronic waste and small household appliances by ≥90%. When these two necessary conditions for dioxin production are simultaneously weakened, the dioxin formation reaction rate constant decreases by approximately two orders of magnitude. According to the Krahl (2020, Waste Management) model, if the chlorine source is reduced by 90% and the metal catalyst by 80%, the formation amount decreases by approximately 98%. Therefore, it is reasonable to infer that dioxin formation is reduced by ≥95%.

[0100] 2. Empirical evidence: After front-end PVC and metal sorting at the MVA Bremen and SYSAV Malmö incineration plants in Europe, dioxin emissions decreased from 0.3 ng TEQ / Nm³ to 0.015 ng TEQ / Nm³, a reduction of 95%, which is consistent with the theoretical inference of this system.

[0101] (II) Sulfur oxide (SO2) reduction rate inferred to be ≥80%:

[0102] 1. Formation mechanism: Sulfur in municipal solid waste mainly comes from sulfur-containing polymers such as rubber, leather, PVC, and ABS (accounting for 0.15–0.25% on a dry basis), electrical components (batteries, cable sheaths, electronic component solder), and sulfate residues in food scraps and pulp. These are oxidized to form SO2 during incineration.

[0103] 2. Impact of blocking pollution sources at the front end of the method and system of this invention: After removing plastic, household appliance and electronic pollution sources at the front end, the sulfur content in the waste is reduced from 0.20% to below 0.04%, and SO2 emissions from incineration are reduced by about 80-85%. If the original emission concentration was 300 mg / Nm³, it will be about 75 mg / Nm³ after the modification, which is comparable to the emissions when using low-sulfur fuel.

[0104] 3. Economic benefits: The reduced sulfur content leads to a reduction of approximately 50% in the amount of chemicals required for the flue gas desulfurization system, saving 8-12 yuan per ton of waste. For an incineration plant with a daily processing capacity of 300 tons, the annual cost savings can reach 800,000 to 1,000,000 yuan.

[0105] (III) Scientific inference that the heavy metal emission reduction rate is ≥85%

[0106] Formation Mechanism: Heavy metal emissions from waste incineration mainly originate from waste electronic and electrical products (WEEE) and metal coating materials. Front-end Interception Mechanism: This invention removes over 95% of electronic waste (including circuit boards, cables, and switching components); removes over 95% of conductive metallic components, WEEE, and metal coating materials. Through front-end pollution source interception, over 95% of waste electronic, electrical, and conductive metallic components are separated and extracted, resulting in a reduction of over 90% in the total heavy metal content of incineration flue gas and fly ash. Lead and cadmium emissions are reduced by approximately 90%, mercury emissions by approximately 95%, and heavy metals in fly ash are reduced from 2.0 g / kg to 0.1–0.2 g / kg. After the modification, the concentrations of various heavy metals are significantly lower than the GB 18485–2014 standard, and the cost of hazardous waste solidification is reduced by 70%, achieving source control and economic efficiency.

[0107] (iv) Overall economic operating costs decreased by ≥50%

[0108] By blocking pollution sources at the front end, the pollution load of the incineration system is significantly reduced, resulting in savings in desulfurization and denitrification, activated carbon, ash disposal, energy consumption and maintenance.

[0109]

[0110] Comprehensive calculations show that the overall operating cost of the method and system of this invention is reduced by approximately 50% (range 40-60%). Based on a cost of 220 yuan / ton, the savings after the modification are approximately 100-130 yuan / ton. For a waste-to-energy plant with a daily processing capacity of 1000 tons, this modification would result in annual operating cost savings of 33-45 million yuan and extend equipment lifespan by 2-3 years.

Claims

1. A method for blocking pollution sources and recycling resources at the front end of waste incineration, characterized in that front-end pollution source blocking treatment and resource recycling treatment are carried out before waste incineration treatment. The entire process of front-end pollution source blocking treatment and resource recycling is automatically controlled through a control center with intelligent learning capabilities; the front-end pollution source blocking treatment includes the following steps: S1. Pollution Source Component Identification and Separation: The control center uses intelligent identification and AI robotic arms to break open bags, identify, and sort waste materials, separating pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances; the remaining waste materials proceed to the next step; the intelligent identification also includes obtaining identification data of the waste materials, and the control center also obtains the operating parameters of this step; S2. Humidity control: After the control center comprehensively calculates the pollution potential index based on the identification data of the remaining waste material in S1, it automatically triggers the adjustment of the moisture content of the remaining waste material to 50%–150%. The moisture content is on a dry basis, i.e., water / dry solids = 0.5–1.

5. The operating parameters of the remaining waste material are continuously monitored, including the moisture content. S3. Multiple physical sorting: This includes the following steps, S31 and S32, either individually or in combination: S31, Gravity differential air separation: The residual waste material after humidity control in S2 is separated from the residual waste material by low temperature hot air using density difference. The pollution source components also include lightweight plastics. S32, Ionization Response Electrostatic Separation: refers to the further separation of residual lightweight plastics from the remaining waste material in S2 or S31 by using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics. S4. Pollution monitoring and feedback: At the sorting outlet of S32, the content of chlorine, sulfur oxides, heavy metals and NOx are detected online. Based on the detection data, the dioxin formation trend is formed. The control center combines the identification data of S1 and the operating parameters of each step to monitor pollution and adjust the operating parameters of the above four steps in order to suppress and control the pollution trend in real time. S5. Pretreatment with Integrated Reactant: Before the remaining waste material after multiple physical sorting in S3 enters the waste incinerator, an integrated reactant is added by spraying or mixing. The integrated reactant includes one or more of alkaline adsorbents, porous silica-alumina skeleton materials, and heavy metal trapping agents. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained in the pollution monitoring and feedback step in S4. The control center calculates the pollution potential index PI based on the identification data, and uses this index as the control target to dynamically correct the operating parameters of at least one step from S2 to S4, forming a closed-loop control of identification-separation-feedback. The pollution source components are removed at the front end through physical means in S1 to S3, and combined with the comprehensive reactant pretreatment based on pollution trends in S4 to S5, the graded blocking and furnace-front reduction of dioxin precursors and heavy metals are achieved.

2. The method for blocking pollution sources and recycling resources at the front end of waste incineration treatment according to claim 1, characterized in that, The resource recycling process refers to the screening of pollutant components, including plastic washing and granulation, metal magnetic separation, and electronic component thermal separation. The method also includes the steps of sending the remaining waste material after S5 treatment into a waste incinerator for incineration and using the high heat energy of the incineration process to generate electricity. S1 also includes the step of AI robotic arm pre-spreading the waste. The control center adopts an AI control strategy based on a deep learning model.

3. The method for blocking pollution sources and recycling resources at the front end of waste incineration treatment according to claim 2, characterized in that... S1's intelligent recognition includes using AI visual recognition, near-infrared spectroscopy detection, and metal response detection to obtain recognition data, which is then transmitted to the control center. The control center then controls the AI ​​robotic arm to perform bag breaking and sorting. S2's humidity control automatically calculates the optimal humidity range based on the moisture content. Specifically, S1 includes: S11. Identification and Separation of Large Solid Items: AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Items with dimensions greater than 150 mm and featuring shells or cables are identified as large discarded household appliances or electronic waste. Metal response detectors identify items with regular metal reflection characteristics as metal parts. The identification data includes the size, shape, and reflection characteristics of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robotic arm, which prioritizes grabbing these pollution source components (large household appliances, electronic waste, and metal parts) and transports them along with their identification data to the resource recycling process, thus achieving priority removal of large-sized objects, electronic parts, and metal pollution source components. S12. Separation of rigid plastics and film plastics: For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter the resource recycling process for plastic washing and granulation. For film plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is retained by the control center in step S3 for further rejection of film plastics with this marking data in the humidity control and multiple physical sorting steps. S13. The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the downstream magnetic separator and eddy current separator to separate the ferromagnetic metal and non-ferrous metal to the corresponding resource recycling channels, thereby achieving complete separation from plastic and combustible waste materials.

4. The method for blocking pollution sources and recycling resources at the front end of waste incineration treatment according to claim 2, characterized in that... The operating parameters of S31 gravity differential air separation include an air velocity of 2–8 m / s. The operating parameters of S32 ionization response electrostatic separation include a high-voltage electrostatic field voltage of 20–50 kV. The high-voltage electrostatic field explains the difference in charge response, causing lightweight plastics to be separated from other waste materials. The operating parameters of gravity differential air separation and ionization response electrostatic separation also include a voltage adjustment frequency, which achieves dynamic response in the range of 0.5–2 Hz.

5. The method for blocking pollution sources and recycling resources at the front end of waste incineration treatment according to claim 4, characterized in that... The humidity control described in S2 and the air separation described in S31 are linked; the operating parameters include the material layer pressure difference signal, stratification rules, material density, and wind speed in S31 of the waste material in each step; the control center uses AI to optimize online with unit separation power consumption / separation purity as the objective function, and dynamically fine-tunes the moisture content setpoint under the pollution monitoring and feedback trigger in S4.

6. The method for blocking pollution sources and resource recovery at the front end of waste incineration treatment according to claim 5, characterized in that... The ionization response electrostatic separation of S32 includes traditional electrostatic separation, deflection plate type and drum electrode type separation; the heavy metal content mentioned in S4 includes the content of mercury, lead and cadmium.

7. A system for blocking pollution sources and recycling resources at the front end of waste incineration, comprising a control center with AI, characterized in that... It also includes a front-end pollution source blocking treatment device and a resource recovery device that are automatically controlled by the control center with intelligent learning function; the front-end pollution source blocking treatment device performs front-end pollution source blocking treatment on the waste before incineration, including a pollution source component identification and separation module, a humidity control module, a multiple physical sorting module, a pollution monitoring and feedback module, and a comprehensive reactant pretreatment module connected in sequence; the control center realizes the identification, separation, extraction, and feedback control of pollution source components through signal interaction and parameter linkage with each device and its modules, thereby blocking the pollution source from entering the waste incineration; The pollution source component identification and separation module includes an intelligent identification device and an AI robotic arm to perform bag breaking, identification, and sorting of waste materials, separating out pollution source components, including plastics, batteries, metals, electronic components, and large and small household appliances. The remaining waste materials after separating out the pollution source components enter the humidity control module. The intelligent identification also includes obtaining identification data and operating parameters of the waste materials. The humidity control module automatically adjusts the moisture content of the remaining waste material after the control center calculates the pollution potential index based on the identification data of the remaining waste material. The operating parameters include moisture content. The multi-physical sorting module implements physical sorting, including a gravity differential air separation submodule and an ionization response electrostatic sorting submodule. The gravity differential air separation submodule uses density difference to separate lightweight plastics from the remaining waste material using low-temperature hot air. The pollution source components also include lightweight plastics. The ionization response electrostatic sorting submodule further separates the remaining lightweight plastics from the waste material from the gravity differential air separation module using a high-voltage electrostatic field. The pollution source components also include residual lightweight plastics. The pollution monitoring and feedback module detects chlorine, sulfur oxides, heavy metals and NOx online. Based on the detection data, it forms a dioxin formation trend. The control center combines the identified data and the operating parameters of each module to monitor pollution and adjust the operating parameters of the four modules to suppress and regulate the pollution trend in real time. The integrated reactant pretreatment module adds an integrated reactant to the remaining waste material after it has been sorted by the multiple physical sorting module, either by spraying or mixing, before it enters the waste incinerator. The integrated reactant includes one or more of the following: alkaline adsorbent, porous silica-alumina framework material, and heavy metal trapping agent. It is used to adsorb and fix hydrogen chloride precursors, sulfur oxide precursors, and free heavy metal ions in the remaining waste material before incineration, thereby reducing the available chlorine, available sulfur, and active heavy metal content of the material entering the furnace. The amount of integrated reactant added is adjusted in a closed loop by the control center based on the chlorine, sulfur, and heavy metal content data obtained from the pollution monitoring and feedback module. The control center calculates the pollution potential index PI based on the identified data, and uses this index as the control target to dynamically correct the operating parameters of each module, forming a closed-loop control of identification-separation-feedback. The pollution source components are physically removed at the front end through the pollution source component identification and separation module, humidity control module and multiple physical sorting module. Combined with the pollution monitoring and feedback module and the comprehensive reactant pretreatment module based on pollution trends, the system achieves graded blocking and furnace-front reduction of dioxin precursors and heavy metals.

8. The waste incineration front-end pollution source blocking and resource recovery system according to claim 7, characterized in that, The resource recycling device includes a plastic washing and granulating machine, a metal magnetic separator and eddy current separator, and an electronic component thermal separation device, which respectively wash and granulate the plastic, perform metal magnetic separation, and thermal separation of electronic components screened by the front-end pollution source blocking treatment device; the system also includes a waste incineration module, which sends the remaining waste material from the integrated reaction agent pretreatment module into the waste incinerator for incineration and power generation; the pollution source component identification and separation module also uses an AI robotic arm to pre-spread the waste.

9. The waste incineration front-end pollution source blocking and resource recovery system according to claim 8, characterized in that, The intelligent identification device of the pollution source component identification and separation module includes an AI visual recognition component, a three-dimensional contour scanning device, a near-infrared spectroscopy detection device, and a metal response detector. These are used to perform AI image recognition, spectral feature analysis, and metal detection on the incoming waste material, respectively, to obtain identification data of the waste material and pollution source components, including plastics, electronic components, and metal impurities, and transmit this data to the control center. The control center then controls the AI ​​robotic arm to perform bag breaking and sorting. For the remaining waste material after the separation of pollution source components, the humidity control module is triggered to activate humidity control and the multiple physical sorting module to perform physical sorting based on the identification data. The identification data includes the quantity, size, and status data of the waste material and pollution source components such as plastics, batteries, metals, and electronic components. The division of labor and cooperation among the components of the pollution source component identification and separation module includes: AI visual recognition components and a 3D contour scanning device are used to identify the size and shape of waste materials passing through the conveyor belt. Individuals with a size greater than 150 mm and with shell or cable features are identified as large discarded household appliances or electronic waste. Metal response detectors identify individuals with regular metal reflection features as metal parts. The identification data includes the identification results of the size, shape, and reflection features of large household appliances, electronic waste, and metal parts. The identification results are sent from the control center to the AI ​​robot arm, which prioritizes picking up the above-mentioned large household appliances, electronic waste, and metal parts, and transports these pollution source components and their identification data to the resource recycling and processing device to achieve priority removal of large-sized objects, electronic parts, and metal pollution source components. For rigid plastic parts with dimensions of 30–150 mm, the AI ​​vision recognition component, combined with a near-infrared spectroscopy detection device, jointly judges their appearance color, geometric features, and spectral absorption peaks to identify the main resin types, including PE, PP, and PET, obtaining resin type identification data. The AI ​​robotic arm then sorts the parts, and the sorted rigid plastic parts enter a resource recycling device for plastic cleaning and granulation. For film-type plastics with a larger area but a thickness of less than 0.2 mm, the AI ​​vision recognition component marks their distribution area instead of relying entirely on the robotic arm for grasping. The identification data also includes marking data, which is transmitted to the humidity control module and the multi-physical sorting module for further rejection of film-type plastics with the marked data. The metal response detector detects the conductive components in the material to obtain a detection signal. The identification data includes the detection signal. The detection signal is used to drive the sorting baffles of the metal magnetic separator and eddy current separator in the resource recycling and processing device, thereby leading the ferromagnetic metal and non-ferrous metal to the processing channel of the resource recycling device, achieving complete separation from plastic and combustible waste materials.

10. The waste incineration front-end pollution source blocking and resource recovery system according to claim 9, characterized in that, The humidity control module operates in conjunction with the gravity differential air separation submodule in the multi-physical sorting module. The humidity control module automatically adjusts the airflow humidity based on an AI control algorithm, with a humidity control range of 50%–150%, preferably 60–120%, to meet the gravity differential air separation conditions without requiring a fixed drying ratio. The pollution source component identification and separation module is connected to both the humidity control module and the gravity differential air separation submodule via an industrial bus to form a feedback closed loop, automatically correcting the humidity and wind speed control parameters when changes in material characteristics are detected.

11. The waste incineration front-end pollution source blocking and resource recovery system according to claim 10, characterized in that, The gravity differential air separation submodule has a wind speed range of 2–8 m / s and a differential pressure range of 0.1–0.3 MPa; the ionization response electrostatic separation submodule operates at a voltage of 20–50 kV and utilizes the difference in surface conductivity of different materials to generate a shift, thereby achieving the separation of plastics and electronic components.

12. The waste incineration front-end pollution source blocking and resource recovery system according to claim 11, characterized in that, The pollution monitoring and feedback module includes a dioxin formation trend prediction algorithm. This module monitors the chlorine and heavy metal content in the remaining waste material after sorting by the multiple physical sorting module (i.e., the material before incineration) in real time and feeds the results back to the control center to dynamically adjust the parameters of each module. Based on the detected changes in dioxin or heavy metal concentration, the pollution monitoring and feedback module automatically adjusts the wind speed / volume or fan frequency of the gravity differential air separation submodule and the electric field strength of the ionization response electrostatic separation submodule through the control center. The control center generates pollution reduction strategies based on pollution monitoring data using AI algorithms and records operating parameters for traceability.

Images