Multi-source solid waste collaborative anaerobic digestion-methane residue utilization integrated system based on internet of things monitoring, control method thereof and digital twin modeling method

Abstract

The present application relates to solid waste resource processing and environmental engineering technical field, especially to multi-source solid waste collaborative anaerobic digestion-methane residue utilization integrated system based on internet of things monitoring and its control method, digital twin modeling method. The system includes intelligent sorting module, collaborative anaerobic digestion reactor group, internet of things monitoring platform, methane residue treatment line and central controller. The collaborative anaerobic digestion reactor group completes hydrolysis acidification and methanogenesis process through hierarchical reaction; the internet of things monitoring platform realizes operation state perception, data acquisition and remote control; the methane residue treatment line resources the residual solid into fertilizer, soil conditioner or fuel product; the central controller performs multi-objective optimization based on raw material parameters, dynamically generates feeding strategy and operation instruction, and has abnormal prediction and intelligent control ability. The system has high identification precision, high degradation efficiency and high resource degree, and has broad engineering application value.

Description

Technical Field

[0001] This invention relates to the field of solid waste resource utilization and environmental engineering technology, and in particular to an integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on Internet of Things monitoring, as well as its control method and digital twin modeling method. Background Technology

[0002] With urbanization and agricultural modernization, the discharge of organic solid waste such as kitchen waste, livestock and poultry manure, municipal sludge, and fruit and vegetable waste continues to increase. Improper disposal will cause a series of environmental problems, including water pollution, soil degradation, and greenhouse gas emissions. Currently, anaerobic digestion technology has become an important means of treating such waste. It can not only stabilize organic matter but also produce biogas for energy use and recycle biogas residue, offering the dual advantages of resource recovery and waste reduction.

[0003] However, the following technical problems still exist in existing anaerobic digestion systems: 1. Most existing anaerobic digestion systems are designed for single types of raw materials (such as kitchen waste or livestock manure). In actual operation, when faced with mixed materials from multiple sources, the gas production rate is often unstable, the system acidification is unstable, and the microbial community is unbalanced due to factors such as C / N ratio imbalance and large fluctuations in dry matter. They lack the ability to adapt to the co-treatment of multi-source organic solid waste.

[0004] 2. In terms of feed control, traditional operations generally rely on manual experience. The raw material mixing ratio and load setting lack real-time identification and modeling analysis of key parameters such as C / N, VS, and heavy metal concentration. The control process has obvious lag, making it difficult to cope with load shocks and easily causing a decrease in degradation efficiency.

[0005] 3. At the process monitoring level, most existing devices lack a high-frequency, multi-parameter continuous monitoring mechanism. In particular, there are blind spots in the real-time sensing of core indicators such as pH, VFA, and methane concentration, making it impossible to intervene in and optimize the operation of anaerobic reaction in a timely manner. At the same time, the lack of remote diagnosis and trend analysis functions restricts the implementation of unattended operation and centralized management.

[0006] 4. In the by-product treatment stage, traditional systems usually only perform simple dewatering treatment on biogas residue, without classifying and modifying or deeply processing it according to its nutrient composition and pollution risks. This results in low product added value, limited land use, and affects the overall resource recovery benefits.

[0007] Overall, existing systems are insufficient to meet the engineering requirements of "multi-point distribution, remote scheduling, and high stability" in solid waste co-processing scenarios. There is an urgent need to build a co-processing anaerobic treatment system that is oriented towards multi-source solid waste, has intelligent sensing and dynamic control capabilities, and can realize the closed-loop utilization of biogas residue resources. Summary of the Invention

[0008] Given that existing multi-source organic solid waste treatment systems generally suffer from poor adaptability, reliance on experience for proportioning control, insufficient monitoring of process parameters, and incomplete resource recovery pathways, this invention proposes an integrated system for multi-source solid waste co-processing anaerobic digestion and biogas residue utilization based on Internet of Things (IoT) monitoring, along with its control method and digital twin modeling method. Based on IoT and optimization modeling technology, a complete process chain is formed from front-end identification to back-end resource recovery. Specifically, it is applicable to the co-degradation, energy recovery, and comprehensive utilization of by-products of various organic wastes such as kitchen waste, livestock and poultry manure, municipal sludge, and fruit and vegetable waste.

[0009] To achieve the above objectives, the present invention specifically adopts the following solution: An integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring includes: The intelligent sorting module is used for the automated classification of multi-source organic solid waste, including a machine vision recognition unit and a robotic arm execution unit; The co-anaerobic digestion reactor group, connected to the intelligent sorting module, is used to receive the sorted solid waste and perform anaerobic digestion treatment. The co-anaerobic digestion reactor group includes at least two reactors arranged in series, each with a hydrolysis acidification zone and a methanogenic zone, and each reactor has a multi-parameter sensor array inside. The Internet of Things (IoT) monitoring platform is connected to the multi-parameter sensor array via a wireless communication network for real-time monitoring of the reactor's operating status. It includes an edge computing unit and a cloud database. The edge computing unit is used to optimize control parameters in real time, and the cloud database is used to store historical process data. The biogas residue treatment line, connected to the co-anaerobic digestion reactor group, is used to treat the biogas residue produced by anaerobic digestion for resource utilization, and includes a dewatering device, a modification device and a molding device in sequence. The central controller is connected to the intelligent sorting module, the collaborative anaerobic digestion reactor group, the Internet of Things monitoring platform, and the biogas residue treatment line to uniformly execute process control commands.

[0010] Furthermore, the central controller performs the following methods: Real-time acquisition of raw material characteristic data, including carbon-nitrogen ratio C / N (dimensionless), volatile solids content VS (mass percentage, on a dry basis), and heavy metal concentration HMi, where i represents the i-th heavy metal (mg / kg, on a dry basis). The optimal feed ratio is calculated based on a hybrid optimization model. The objective function of the model is:

[0011] in, , , These are weighting coefficients used to characterize the impact of C / N ratio deviation, VS deviation of volatile solids content, and heavy metal risk, respectively; HM i This represents the concentration of the i-th heavy metal; The reactor operating parameters are dynamically adjusted according to the optimal ratio, including pH control at 6.8~7.4, temperature control at 35±1℃, and stirring intensity control at 10~30 rpm. When the concentration of volatile fatty acids (VFA) monitored online exceeds 3000 mg / L, the alkali dosing system is automatically activated to neutralize acidic substances and inhibit acidification instability; the alkali solution is one or more of sodium hydroxide or calcium hydroxide; The central controller sends the results of the optimization model as control commands to the actuators, thereby achieving comprehensive optimization of gas production efficiency, system stability, and environmental risks.

[0012] The output results include the feed ratio, pH setting, stirring speed, temperature setting, and alkali dosage or frequency.

[0013] Furthermore, the reactors in the synergistic anaerobic digestion reactor group include: The shell is double-layered, with the inner and outer layers made of 304L stainless steel and the interlayer filled with polyurethane insulation material. The air deflector array has each deflector tilted at an angle of 45° ± 5° to the horizontal plane. A jet stirring device is installed at the bottom of the reactor, with a nozzle diameter of 10~15 mm; A microwave pretreatment unit is located at the top of the reactor and is equipped with a temperature feedback control system for dynamically adjusting the target temperature during the heating process. The operating parameters of the microwave pretreatment unit include: operating frequency of 2450±50 MHz; single radiation time of 90-120s; and energy density of 0.5-1.5W / g raw material.

[0014] Furthermore, the multi-parameter sensing array includes: Embedded pH sensor with a measurement accuracy of ±0.05; Infrared methane analyzer, with a measurement range of 0~100% and a resolution of 0.1%; Ultrasonic sludge concentration meter with a measurement error of less than 3%.

[0015] Furthermore, the hybrid optimization model is trained using a machine learning algorithm, and the training dataset includes: No fewer than 2000 sets of experimental data with different raw material ratios; Climate condition parameters, including temperature, humidity and atmospheric pressure; Historical operational failure records.

[0016] Furthermore, the modification device includes: The biochar addition unit is used to add biochar to the biogas residue at a mass ratio of 5-8%; the plasma treatment chamber has a power of 2-5kW; and the humic acid extraction module has an extraction rate of not less than 65%.

[0017] Correspondingly, this invention also provides a control method for an integrated system of multi-source solid waste co-anaerobic digestion and biogas residue utilization based on Internet of Things monitoring, including the following steps: S1: Real-time acquisition of raw material characteristic data, including carbon-to-nitrogen ratio (C / N), volatile solids content (VS), and heavy metal concentration (HMi); S2: Calculate the optimal feed ratio based on the hybrid optimization model. The objective function is:

[0018] S3: Dynamically adjust reactor operating parameters, including pH, temperature, and stirring intensity, according to the optimal ratio; S4: When the concentration of volatile fatty acids exceeds the preset threshold, the alkali dosing system will be automatically activated; S5: The optimization results are sent as control commands to the actuators to achieve comprehensive optimization of system operation.

[0019] Furthermore, it also includes: abnormal operating condition prediction steps: based on real-time data collected by the IoT platform, potential faults are identified and warnings are issued at least 30 minutes in advance through machine learning models; Furthermore, it also includes: a multi-objective optimization step: based on gas production, processing efficiency, and unit energy consumption, a Pareto front solution method is used to generate a multi-objective optimization strategy.

[0020] Correspondingly, this invention also provides a digital twin modeling method for an integrated system of multi-source solid waste co-anaerobic digestion and biogas residue utilization based on Internet of Things monitoring, including the following steps: Establish a three-dimensional fluid dynamics simulation model of the system, which should include at least the reactor structure, fluid domain and boundary conditions; Multiphysics simulation tools are used to perform joint simulation of the temperature field, pressure field and electromagnetic field of the system during operation, so as to obtain the virtual operating state of multiphysics coupling; Based on the real-time collected operating data from the Internet of Things platform, the key parameters of the three-dimensional fluid simulation model are dynamically updated through a data assimilation algorithm, so that the deviation between the model output and the actual operating state is controlled within a preset tolerance. The updated digital twin model is used to predict the future operating trend of the system and output early warning information or optimize control strategies.

[0021] Compared with the prior art, the present invention has the following beneficial technical effects: This invention constructs a complete solution for the co-processing anaerobic digestion of multi-component, multi-source solid waste through a full-process technology chain encompassing "source identification—intelligent sorting—reactor coordination—IoT sensing—intelligent algorithm control—resource closed-loop utilization." The system utilizes a control system to dynamically identify raw material characteristic parameters, including carbon-to-nitrogen ratio, volatile solids content, and heavy metal concentration, and constructs a multi-objective optimization model with a penalty function. This model can generate the optimal feeding strategy in real time, ensuring that the reaction process always operates within the optimal operating range. This model integrates parameter estimation, machine learning, and multi-objective control logic, effectively avoiding the problems of poor system stability and low output efficiency caused by the reliance on human experience in traditional methods.

[0022] The synergistic anaerobic reactor employs a staged reaction zone design, establishing separate hydrolysis-acidification and methanogenesis reaction environments in different areas to achieve targeted enrichment and metabolic optimization of functional microorganisms, thereby improving organic matter conversion efficiency. The microwave pretreatment unit integrated at the top of the reactor enables rapid cell disruption of the substrate, effectively improving the biodegradability of recalcitrant materials and enhancing the system's resistance to shock loads. The internal baffle array, double-layer insulation structure, and jet stirring device optimize the material flow field, temperature field, and microbial community spatial distribution, ensuring excellent physical and biological stability of the reaction system.

[0023] The system's IoT monitoring platform and edge computing unit work together to achieve integrated intelligent control from perception to decision-making to execution. The platform can connect to hundreds of sensor terminals to continuously monitor key operating parameters such as temperature, pH value, redox potential, volatile fatty acid concentration, methane concentration, and material height, and performs trend modeling and status prediction based on historical data. The platform features intelligent alarm and remote intervention capabilities, reducing the need for manual intervention and improving the system's automation level and safety.

[0024] In the end-of-pipe treatment stage, the biogas residue resource utilization module integrates dewatering, modification, and molding processes, which can transform biogas residue into various by-products such as bio-organic fertilizer, soil conditioner, seedling substrate, or solid fuel, depending on its composition and characteristics. Through modification methods such as biochar addition, plasma treatment, and humic acid extraction, this module achieves the harmless treatment and functional enhancement of biogas residue, truly realizing the recovery and reuse of the "carbon-nitrogen-energy" triple resources.

[0025] In summary, the system of this invention possesses broad adaptability to the co-processing of multi-source solid waste, supporting the mixed anaerobic treatment of various organic wastes such as kitchen waste, livestock and poultry manure, fruit and vegetable waste, and municipal sludge. Its control algorithm integrates optimization modeling and machine learning methods, achieving dynamic closed-loop control of feed ratio and reaction parameters. Compared with traditional systems, methanogenesis efficiency can be increased by 20% to 30%, and system load tolerance is enhanced. The biogas residue products are diversified and meet national standards for organic fertilizer, resulting in more thorough resource recovery. The IoT platform possesses high-frequency sampling, low-latency response, and historical data backtracking capabilities, achieving a high degree of integration of automation and informatization. The system's modular design and remote centralized operation and maintenance capabilities make it suitable for various application scenarios such as urban kitchen waste treatment centers, rural livestock and poultry waste disposal sites, and agricultural parks, offering advantages such as low operation and maintenance costs and good scalability. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall structure of a specific embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an intelligent sorting module according to a specific embodiment of the present invention; Figure 3 This is a flowchart of a central control method according to a specific embodiment of the present invention; Figure 4 This is a structural diagram of an IoT monitoring platform according to a specific embodiment of the present invention; Figure 5 This is a schematic diagram of the process flow of a biogas residue treatment line according to a specific embodiment of the present invention; Figure 6 This is a data interaction diagram of the digital twin modeling method and system of the present invention. Detailed Implementation

[0027] To provide a clearer understanding of the technical solution of this invention, the various functional modules and control methods of this invention will be described in detail below with reference to the accompanying drawings and claims. It should be understood that the following description is intended to illustrate multiple possible embodiments of this invention and is not intended to limit the scope of protection of this invention.

[0028] Example 1 The overall system structure of the present invention is as follows: Figure 1 As shown, it includes five core modules: intelligent sorting module, collaborative anaerobic digestion reactor group, Internet of Things monitoring platform, biogas residue treatment line and central controller, which together constitute an intelligent solid waste treatment system integrating information perception, decision control and resource utilization.

[0029] In this embodiment, the integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring includes: an intelligent sorting module for automatically classifying multi-source solid waste, the intelligent sorting module including a machine vision recognition unit and a robotic arm execution unit; a co-anaerobic digestion reactor group connected to the intelligent sorting module for receiving the sorted solid waste and performing anaerobic digestion treatment, the co-anaerobic digestion reactor group including at least two reactors arranged in series, each reactor having a hydrolysis acidification zone and a methanogenesis zone, and each reactor having a multi-parameter sensor array; and an IoT monitoring platform, using LoRaWAN. Alternatively, a 5G network can be used to communicate with the multi-parameter sensor array for real-time monitoring of the reactor's operating status. The IoT monitoring platform includes an edge computing unit and a cloud database, where the edge computing unit is used to optimize control parameters in real time, and the cloud database is used to store historical process data. A biogas residue treatment line, connected to the co-anaerobic digestion reactor group, is used to process the biogas residue produced after anaerobic digestion for resource utilization. The biogas residue treatment line includes a dewatering device, a modification device, and a molding device. A central controller is communicated with the intelligent sorting module, the co-anaerobic digestion reactor group, the IoT monitoring platform, and the biogas residue treatment line to uniformly execute process control commands.

[0030] like Figure 1 As shown, the system integrates five functional units: intelligent sorting module, collaborative anaerobic digestion reactor group, Internet of Things monitoring platform, biogas residue treatment line and central controller, forming a closed-loop collaborative treatment process that runs through "identification-degradation-monitoring-resource utilization-control feedback". Figure 2 The intelligent sorting module shown is located at the system entrance and includes a machine vision recognition unit and a robotic arm execution unit. Through image acquisition, depth recognition, and automatic grasping, it accurately classifies and directs raw materials such as kitchen waste, municipal sludge, and livestock manure. The sorted organic solid waste is then transported to... Figure 1 The synergistic anaerobic digestion reactor assembly shown includes two reactors arranged in series: the first stage has a hydrolysis acidification zone to break down macromolecules into fermentable organic acids; the second stage has a methanogenic zone to promote methanogenic biogas production by optimizing pH and temperature. Each reactor stage is equipped with a multi-parameter sensor array, including a MEMS pH sensor, an infrared methane analyzer, and an ultrasonic sludge concentration meter. The collected data is transmitted via LoRaWAN or 5G to... Figure 4 The IoT monitoring platform shown includes an edge computing unit that processes and adjusts control commands in real time, and a cloud database for historical data storage and trend prediction. The biogas residue produced after the reaction is processed... Figure 5 The biogas residue processing line shown in the figure sequentially undergoes dewatering, modification, and molding processes to produce bio-organic fertilizer, soil conditioner, or fuel products. Figure 3The control logic executed by the central controller is demonstrated: based on the characteristics of raw materials such as C / N, VS, and heavy metals, the controller generates the optimal ratio and operating parameters through an optimization model, coordinates and controls each unit module, and achieves efficient and stable operation of the system as a whole.

[0031] The system described in this invention comprises five integrated modules, systematically connecting the entire chain of "source sorting—coordinated digestion—information sensing—intelligent control—resource utilization." Compared to traditional single anaerobic treatment technologies, it possesses stronger adaptability to multi-source waste and automated operation capabilities. Through the collaborative operation of machine vision and robotic arms, it achieves accurate identification and targeted diversion of different types of waste, greatly reducing human error in sorting and interference from foreign matter in the system. The multi-stage anaerobic reactor series design, combined with intelligent parameter control, optimizes the zoning of hydrolysis and methanogenesis processes, improving organic matter conversion efficiency and methane yield. A multi-parameter sensor array and an IoT monitoring platform together constitute a high-frequency, low-latency data sensing system, ensuring that the entire reaction process is visible and controllable. The biogas residue treatment line realizes a closed-loop utilization path for solid by-products from dehydration to productization, significantly increasing the added value of by-products. The system as a whole is highly integrated, possessing remote scheduling and unattended operation capabilities, reducing operation and maintenance costs, and improving resource recovery rate and environmental benefits.

[0032] Based on the above implementation methods, the image recognition unit in the intelligent sorting module can use other types of image processing algorithms, such as YOLO and ResNet, to adapt to material scenarios with different complexities; the robotic arm gripping device can adopt electric grippers, negative pressure suction cups, or magnetic end effectors, dynamically switching according to the type of material; the number of reactors in the collaborative anaerobic reactor group can be increased to three or more stages to improve load tolerance; the multi-parameter sensor array can be expanded to include oxidation-reduction potential (ORP) sensors or temperature / level sensors to further enrich the sensing dimensions; in addition to LoRaWAN and 5G, network connection methods can also adopt NB-IoT or Wi-Fi Mesh networks to maintain data stability in complex communication environments; the modification device in the biogas residue treatment line can use other passivating materials such as bentonite and lime to replace biochar to achieve different heavy metal adsorption pathways; the central controller can be deployed on a local edge gateway or industrial PC platform, or integrated with a host computer system such as SCADA to meet the deployment needs of different scale scenarios.

[0033] like Figure 3 As shown, the central controller collects data on the characteristics of each raw material and executes a hybrid optimization model based on a weighted objective function. The raw material characteristics include three key indicators: C / N ratio (carbon to nitrogen ratio), VS (volatile solids content), and HM (heavy metal concentration). The controller performs a standardized evaluation of the current raw material state and substitutes it into the following objective function for minimization calculation:

[0034] in, , , Weighting coefficients, set empirically, are used to balance the impact of different factors on system performance. The controller obtains the optimal feed ratio through gradient optimization or evolutionary algorithms and adjusts the feed ratio and reactor operating parameters accordingly, including stabilizing the pH at 6.8-7.4, maintaining the temperature at 35±1°C, and controlling the stirring intensity at 10-30 rpm. When the VFA concentration exceeds 3000 mg / L, the controller automatically activates the alkali dosing system to neutralize the acidic substances with alkaline agents such as NaOH or Ca(OH)2, preventing system acidification and instability.

[0035] This optimized control method uses a mathematical model to drive the control logic, comprehensively balancing gas production efficiency, system stability, and environmental risk through a weighted penalty mechanism. The model is designed with three objectives in mind: C / N balance, VS effectiveness, and HM environmental safety, enhancing the system's adaptability to various complex organic feedstocks. Compared to traditional experience-based control methods, the central controller can automatically sense changes in the feed structure and output optimal operating parameters in real time, ensuring the system dynamically adjusts within its optimal operating range, effectively improving methane yield and processing stability. Simultaneously, the integrated VFA (Vacuum-Free Anomalous Response) mechanism significantly enhances the system's resilience, reduces the frequency of manual intervention, and is suitable for long-cycle, continuous operation scenarios.

[0036] In the above implementation, the weighting coefficient , , The system can be adjusted according to the system's operational objectives to suit different scenarios with varying weight biases (such as prioritizing heavy pollutant control or maximizing gas yield). The C / N and VS target values ​​can also be modified to other suitable ranges based on the reactor type, such as C / N=25 and VS=12%. The optimization algorithm can employ intelligent algorithms such as genetic algorithms, simulated annealing, and particle swarm optimization to replace linear programming, obtaining stable solutions under more complex objective functions. The VFA threshold can be set in the 2500-4000 mg / L range, and the response mechanism can be adjusted according to the acidification resistance of different substrates. The type of alkali solution can be selected based on the availability of materials in different regions, such as Na2CO3 or KOH, and an intelligent flow control pump can be configured to precisely control the dosage. The entire system can also be integrated with a digital twin platform to achieve closed-loop control driven by model and verified by simulation.

[0037] In the above embodiments, the reactor in the synergistic anaerobic digestion reactor group includes: a double-layer shell, both the inner and outer layers of which are made of 304L stainless steel, with polyurethane insulation material filling the interlayer; a baffle array, disposed inside the reactor, with each baffle having an inclination angle relative to the horizontal plane; a jet stirring device, disposed at the bottom of the reactor, with a nozzle diameter of 10-15mm; and a microwave pretreatment unit, disposed at the top of the reactor, with an operating frequency of 2450±50 MHz.

[0038] like Figure 1 As shown, the co-anaerobic digestion reactor unit adopts a double-layer structure design. Each reactor shell consists of an inner and outer layer made of 304L stainless steel, which has good corrosion resistance and mechanical strength. The interlayer between the two layers is filled with polyurethane material, forming a high-efficiency insulation layer that reduces heat loss and stabilizes the reaction temperature environment. An array of inclined guide vanes is installed inside the reactor, with the vanes preferably set at an angle of 45° relative to the horizontal plane, forming a spiral upward flow field. This helps optimize the substrate flow path, avoids short-circuiting and dead zones, and improves the uniformity of material mixing and the efficiency of microbial contact. A jet stirring device with a nozzle diameter of 10-15 mm is installed at the bottom of the reactor. High-pressure reflux liquid is ejected from the bottom to form a circulation, which enhances the mixing effect and reduces energy consumption, saving more than 20% energy compared to traditional mechanical stirring. Furthermore, a microwave pretreatment unit is integrated at the top of the reactor. The microwave operates at a frequency of 2450 MHz. Electromagnetic waves heat and break down the cell walls of the substrate, accelerating the dissolution of organic matter, significantly increasing the reaction rate, and enhancing the treatment effect on high-oil or recalcitrant materials.

[0039] The aforementioned structural design integrates multiple functions, including physical insulation, fluid guidance, efficient mixing, and intelligent pretreatment, significantly improving the overall performance of the reactor in the synergistic anaerobic process. The combination of a double-layer shell and polyurethane insulation ensures temperature stability and reduces energy consumption; the angled guide vanes optimize fluid flow within the reactor, reducing dead zones and sedimentation, and improving the uniformity of organic matter degradation; the jet stirring device provides a highly efficient mixing method without mechanical contact, reducing equipment wear and maintenance requirements; and the microwave pretreatment unit breaks through the limitations of traditional pretreatment pathways, achieving efficient synergy between physical and biochemical processes, ensuring rapid system startup and stable operation under high loads. The overall structure provides multi-source solid waste treatment systems with strong adaptability, high efficiency, and modularity.

[0040] In the above embodiments, the outer layer of the double-layer shell can be replaced with a more weather-resistant 316L stainless steel material according to environmental requirements; the polyurethane insulation layer can be replaced with vacuum insulation panels or mineral wool, etc., selected according to specific temperature zones and insulation level requirements; the inclination angle of the guide plate can be adjusted to 30°-60° to match different reactant viscosities and flow characteristics, and the guide plate material can be made of corrosion-resistant materials such as reinforced polypropylene and PTFE; the nozzle shape of the jet stirring device can be designed as conical, annular, or adjustable to meet different mixing intensity requirements; the frequency of the microwave pretreatment unit can also be adjusted to industrial bands such as 915 MHz according to the dielectric properties of the target substrate, and the equipment can also integrate dual-frequency control or frequency conversion control modules to achieve higher energy utilization efficiency. In addition, the pretreatment module can be equipped with a temperature sensor and feedback control loop to achieve intelligent control of automatic power adjustment and heating time.

[0041] As a structural complement, the multi-parameter sensing array includes: an embedded pH sensor manufactured using MEMS technology with a measurement accuracy of ±0.05; an infrared methane analyzer with a measurement range of 0–100% (volume fraction) and a resolution of 0.1%; and an ultrasonic sludge concentration meter with a measurement error of less than 3%.

[0042] like Figure 1 As shown, a multi-parameter sensor array is deployed in each stage of the synergistic anaerobic digestion reactor group for real-time monitoring of key operating parameters. Specifically, it includes three types of core sensors: first, an embedded pH sensor, manufactured using MEMS miniaturization technology, possessing high accuracy (±0.05) and rapid response characteristics, which can be directly inserted into the reactor for continuous pH monitoring, with data uploaded to the IoT platform in real time; second, an infrared methane analyzer, based on the non-dispersive infrared absorption (NDIR) principle, with a measurement range covering 0-100% volume fraction and a resolution of 0.1%, suitable for continuously assessing methane gas yield and reactivity during anaerobic fermentation; and third, an ultrasonic sludge concentration meter, which uses ultrasonic reflection echo time and intensity to invert sludge content within the reactor, with measurement errors controlled within 3%, suitable for monitoring substrate concentration fluctuations and judging deposition trends. These three types of sensors collect and preprocess data through a local edge controller, then transmit it to the monitoring platform via LoRaWAN or 5G networks to achieve status awareness, trend analysis, and anomaly early warning functions.

[0043] This multi-parameter sensor array constructs a multi-dimensional, continuous reactor operation data acquisition system, significantly improving the informatization level and operational visibility of the anaerobic system. A high-precision pH sensor enables early warning of acid-base fluctuations, ensuring microbial activity; an infrared methane analyzer can be used to assess gas production efficiency in real time, providing a basis for optimizing feeding strategies and load management; and an ultrasonic sludge concentration meter can effectively identify sedimentation and accumulation trends, guiding adjustments to the stirring strategy. These three types of data together constitute a key indicator system for system status. While ensuring measurement accuracy, it adapts to continuous operation scenarios, effectively supporting the edge computing module in achieving dynamic parameter optimization, fault identification, and control decision-making, thus constructing a "digital perception layer" for system operation.

[0044] In the above embodiments, the pH sensor can be replaced with a fouling-resistant gel electrode structure to adapt to environments with high sludge concentration and high solids content; the infrared methane analyzer can be replaced with a laser gas analyzer to obtain higher accuracy and anti-interference performance; the ultrasonic sludge concentration meter can be selected with a multi-frequency composite structure according to the density range of the treated object to improve measurement resolution. The number and placement of sensors can also be customized according to the reactor volume and mixing method, such as arranging multiple pH and concentration sensors in each stage of the reactor to achieve three-dimensional regional monitoring. In addition, the system can be expanded with ORP sensors, level gauges, temperature sensors, etc. to build a more comprehensive monitoring network. The relevant data can be used to build predictive models to improve the overall intelligence and operational stability of the system.

[0045] In the above implementation, the hybrid optimization model is trained by a machine learning algorithm, and the training dataset includes: no less than 2,000 sets of experimental data with different raw material ratios; climatic condition parameters, including temperature, humidity and atmospheric pressure; and historical operation failure records.

[0046] like Figure 3 As shown, the hybrid optimization model deployed inside the central controller is an adaptive decision-making system built on machine learning algorithms. Its core is a set of supervised learning-based regression algorithms (such as XGBoost, LightGBM, or Random Forest), using no fewer than 2000 sets of historical experimental data as training data. This data includes key response variables such as anaerobic gas production rate, stability, and VFA accumulation of multi-source organic waste (kitchen waste, livestock manure, etc.) under different proportions. In addition to raw material proportion data, the model input also incorporates climatic condition parameters (temperature, humidity, atmospheric pressure) to reflect the impact of the environment on fermentation efficiency; simultaneously, historical system operation failure records are included as negative samples to improve the model's ability to identify abnormal operating conditions. After training, the model can predict the gas production efficiency, risk level, and parameter deviation of different proportion schemes based on real-time collected raw material characteristics and operating environment, and finally output the optimal feed formula for the controller to execute.

[0047] This machine learning optimization model outperforms traditional empirical formulas and static weight models in terms of input dimensionality, sample size, and algorithm structure, improving raw material adaptability and prediction accuracy. Driven by large-scale experimental data, the model accurately reflects complex nonlinear relationships, achieving an effective mapping between feed ratios and reaction results. Introducing climate parameters enhances the model's adaptability to external environments and strengthens the system's stability under seasonal changes or at different deployment locations. Historical failure data serves as training labels, improving the model's ability to perceive potential system instability trends and the robustness of response strategies. Overall, the optimization model provides the central controller with powerful "feedforward prediction capabilities," effectively avoiding delays and errors caused by manual experience-based control, and driving the system's operation from "passive response" to "active optimization."

[0048] In the above implementation, the training algorithm can be replaced with a neural network (such as a multilayer perceptron MLP) or an ensemble learning model (such as a Bagging or Boosting ensemble structure), selecting the optimal structure based on sample distribution and task complexity. The training dataset size can be further expanded to over 10,000 sets to improve generalization ability. The model input parameters can be expanded to include more physical indicators such as C / N, VS, TS, pH, and EC, or time series information can be introduced for dynamic modeling. Climate data can be automatically updated by accessing local weather stations or third-party API platforms. Fault labels can be refined into specific anomaly types (such as VFA accumulation, temperature deviation, sudden drop in methane concentration, etc.), and the model can use a multi-output structure to simultaneously predict different fault risks. The final output format can also be expanded from the optimal single-value ratio to a Pareto front solution set to support the generation of multi-objective trade-off strategies.

[0049] As a further optimization of the treatment of biogas residue, the modification device includes: a biochar addition unit for adding biochar to biogas residue at a mass ratio of 5-8%; a plasma treatment chamber with a power of 2-5kW; and a humic acid extraction module with an extraction rate of not less than 65%.

[0050] like Figure 5As shown, the modification unit, a key component of the biogas residue treatment line, is responsible for improving the resource utilization performance of biogas residue. First, the biogas residue from the dewatering unit is conveyed to the biochar addition unit via a screw conveyor. This unit consists of a metering feeder and a mixer, which uniformly mixes biochar and biogas residue at a set mass ratio of 5–8%. Biochar, with its high specific surface area and strong adsorption capacity, effectively fixes heavy metal ions and organic pollutants, improving the physicochemical properties of the biogas residue. The mixture then enters the plasma treatment chamber, which uses a high-energy thermal plasma source (2-5 kW) to perform instantaneous high-temperature treatment on the biogas residue in a sealed cavity. This process breaks down harmful organic residues (such as antibiotics and pathogens), achieving harmless and stable treatment. After treatment, the biogas residue is then conveyed to the humic acid extraction module, where humic acid is extracted from the organic matter using an alkali-dissolution-acid precipitation method, achieving an extraction rate of over 65%. The obtained humic acid can be used as a bio-fertilizer additive or to prepare soil conditioners, while the remaining biogas residue is processed into end products using molding equipment.

[0051] This modification device integrates physical mixing, thermal treatment, and chemical extraction to comprehensively modify biogas residue, significantly enhancing its resource properties and environmental safety. The addition of biochar not only improves the residue's structure but also enhances its nutrient slow-release and heavy metal passivation capabilities. Plasma treatment, characterized by its rapid reaction speed, thorough processing, and low energy consumption, can degrade harmful components quickly, preventing secondary pollution. The efficient extraction of humic acid enables the recycling and reuse of organic functional components, increasing the added value and functional diversity of biogas residue products. The synergistic effect of these three processes transforms biogas residue from raw waste into commercially viable organic fertilizer or conditioner, truly achieving high-value conversion and closed-loop utilization of byproducts.

[0052] In the above embodiments, the biochar raw material can be replaced by various sources such as rice husk charcoal, bamboo charcoal, or sludge charcoal, and its addition ratio can be adjusted to 3-10% according to the target application. The plasma treatment chamber can adopt DC plasma, radio frequency plasma, or microwave plasma structures, with different types having different energy consumption and treatment depths. The treatment power can also be set from 1-10kW according to the treatment scale, and is equipped with automatic temperature control and overload protection mechanisms. In addition to alkali dissolution-acid precipitation, the humic acid extraction process can also adopt green processes such as membrane separation and bio-enzymatic hydrolysis to improve product purity and environmental friendliness. In terms of modular structure, the three units can be integrated into an integrated skid-mounted module, which facilitates system deployment and expansion; or the units can be combined in a skip-and-go manner according to different biogas residue properties to achieve customized treatment paths.

[0053] In the above embodiments, the operating parameters of the microwave pretreatment unit include: a single radiation time of 90-120s; an energy density of 0.5-1.5W / g raw material; and a temperature feedback control system for dynamically adjusting the target temperature during the heating process.

[0054] like Figure 1 As shown, the microwave pretreatment unit is located at the top of the co-anaerobic digestion reactor group, used to physically enhance the substrate before solid waste enters the reactor. This unit operates at a frequency of 2450 MHz, utilizing microwaves to polarize and excite water molecules in the organic waste, causing them to vibrate and generate heat rapidly under the influence of a high-frequency electromagnetic field, thereby achieving rapid heating and cell structure destruction. Its operating parameters are set as follows: processing time per batch is 90-120 seconds, energy density is controlled between 0.5-1.5 W / g of raw material, and the input energy is dynamically adjusted according to the type of raw material. To avoid overheating or localized coking, the equipment integrates a temperature feedback control system, which monitors the material temperature in real time and adjusts the microwave power output according to a set threshold (e.g., 65–85°C) to achieve closed-loop temperature control. This treatment process is completed before the material enters the anaerobic reactor, which not only improves the solubility of the substrate but also enhances its biodegradability, providing better reaction conditions for subsequent hydrolysis, acidification, and methanogenesis reactions.

[0055] Compared to traditional heat treatment or mechanical crushing processes, this microwave pretreatment unit offers advantages such as rapid heating, low energy consumption, and thorough structural destruction. Through precise control of radiation time and energy density settings, it achieves a pretreatment method highly adaptable to various materials (such as oily kitchen waste and cellulose-based fruit and vegetable residues). The temperature feedback control system enhances the safety and uniformity of the heating process, preventing localized carbonization or thermal decomposition and ensuring the stability of pretreatment quality. Microwave treatment enhances the hydrolysis efficiency of raw materials, releasing more fermentable components, thereby increasing biogas production by over 20%. Simultaneously, it effectively shortens the anaerobic reaction start-up time and the cycle to reach stable operation, improving the overall system processing efficiency and dynamic response capability.

[0056] In the above embodiments, the microwave operating frequency can be adjusted to 915 MHz according to the dielectric properties of the raw material to achieve deeper penetration heating; the single radiation time can be set to 60-180 seconds according to the material's thermosensitivity and initial moisture content; the energy density range can also be extended to 0.3-2.0 W / g to adapt to high-concentration substrate processing scenarios. The temperature feedback system can employ various temperature measurement methods such as infrared thermometers, thermocouple arrays, or fiber optic sensors to improve measurement accuracy and response speed; the control method can also integrate fuzzy control or PID control algorithms to further optimize the heating curve. In addition, the microwave cavity can be designed as a batch processing or continuous transmission structure according to the processing volume requirements, and the inner wall of the cavity can be coated with a reflective coating to improve heating uniformity; the equipment can also be linked with the raw material conveying system to achieve automatic matching of material flow rate with microwave power, constructing an efficient collaborative pretreatment module.

[0057] To further enhance the system's operational intelligence, this invention also provides a digital twin modeling method for the system, including: establishing a three-dimensional fluid dynamics simulation model of the system; using ANSYS and COMSOL to jointly simulate the temperature field, pressure field, and electromagnetic field; and dynamically updating the model parameters based on real-time data collected by the Internet of Things to ensure that the model deviation tolerance is less than 5%.

[0058] like Figure 6 As shown, this digital twin modeling method constructs a system simulation platform based on physical process modeling and real-time data fusion. First, a three-dimensional structural model of the system is established, encompassing the anaerobic reactor, jet stirring device, guide vane structure, microwave unit, and insulation layer. Computational fluid dynamics (CFD) is employed, using ANSYS software to simulate and analyze flow velocity distribution, temperature conduction, and pressure changes. Simultaneously, the COMSOL Multiphysics platform is used to model the electromagnetic field distribution, temperature rise curve, and energy density distribution of the microwave heating zone, achieving coupled simulation of multiple physical fields (fluid field, thermal field, and electric field). The model dynamically updates key boundary conditions and physical parameters by integrating real-time data such as pH, temperature, VFA, and CH4 concentration collected from an IoT monitoring platform, combined with regression fitting and Kalman filtering algorithms, achieving "real-virtual synchronization" of the model. The model output results can be fed back to the central controller in real time, serving as an auxiliary decision-making basis for optimizing operating parameters, predicting faults, and suggesting operational strategies, ensuring that the simulation result deviation is controlled within 5%, possessing high reliability and application guidance value.

[0059] The proposed modeling method integrates 3D structural modeling, coupled simulation, and real-time data fusion to construct a visualized, dynamically adjustable digital twin platform. This platform not only supports the analysis of the thermal and hydrodynamic field distribution and structural optimization within the reactor, but also has functions such as predicting future operating trends, identifying abnormal states, and simulating different feeding strategies. Through interaction with IoT data, it achieves "optimization while operating," improving the scientific and intelligent level of operation management. Compared to static modeling or experience-based control, this method can simulate system responses under different operating conditions, substrate ratios, and control strategies, guiding maintenance personnel to conduct pre-commissioning, pre-drills, and optimal strategy selection, reducing commissioning cycles and operational risks, and possessing broad engineering applicability and expansion potential.

[0060] In the aforementioned digital twin modeling methods, in addition to ANSYS and COMSOL, open-source or commercial software such as OpenFOAM and SolidWorks Flow Simulation can be used as alternatives for 3D structural modeling tools. Thermal and electromagnetic field simulation modules can be modeled separately or subjected to weak coupling as needed. Dynamic parameter update algorithms can be replaced with particle filtering, Bayesian estimation, or LSTM neural networks for more complex time-series behavior prediction. Data integration can be achieved through REST API, OPC UA, or MQTT protocols to enable high-frequency interaction with the monitoring platform. Model deviation tolerance can also be adjusted to 1-10% according to application requirements to balance computational accuracy and real-time performance. Furthermore, the twin platform can be deployed on local servers, edge computing nodes, or industrial cloud platforms, and supports multi-user remote access, hierarchical access control, and graphical user interface configuration, enhancing model usability and engineering integration capabilities.

[0061] In practical applications of this invention, the system is used to process combinations of two or more of the following wastes: kitchen waste with a total solids content of 15-25%; livestock and poultry manure with a carbon-to-nitrogen ratio of 6-10; fruit and vegetable waste with a volatile solids content of more than 80%; and municipal sludge with a heavy metal content of less than 50 mg / kg.

[0062] like Figure 1 As shown, the system of this invention uses an intelligent sorting module to perform image recognition and automated classification of mixed organic solid waste (such as kitchen waste, municipal sludge, livestock and poultry manure, and fruit and vegetable waste), achieving refined source management. Different types of solid waste, after pretreatment, enter a co-processing anaerobic digestion reactor. This reactor has a staged treatment function; the hydrolysis and acidification zone is suitable for the rapid decomposition of high-fat, highly hydrolyzable kitchen waste and fruit and vegetable waste, while the methanogenic zone provides a stable degradation environment for nitrogen-rich livestock and poultry manure and complex municipal sludge. The central controller dynamically generates feed ratios and operating parameters based on the physicochemical properties of various raw materials (such as TS, C / N, VS, heavy metal content, etc.) to ensure the system operates under optimal conditions. System operating data is collected and processed by an IoT monitoring platform, realizing the co-processing of multi-source mixed raw materials and intelligent management throughout the entire process. The final biogas residue is modified and molded to form commercially viable by-products, achieving a closed-loop "waste-energy-resource" path.

[0063] This invention's system possesses strong adaptability to raw materials and co-processing capabilities, enabling stable and efficient treatment of complex combinations of components such as kitchen waste (TS 15-25%), livestock and poultry manure (C / N 6-10), fruit and vegetable waste (VS>80%), and municipal sludge (heavy metals <50mg / kg) under actual operating conditions. Supported by a central control strategy, the system can dynamically adjust the feed ratio, pH, temperature, and stirring parameters according to the properties of various raw materials, avoiding adverse reactions such as system acidification and heavy metal accumulation, thus improving treatment efficiency and operational stability. The co-processing mode not only increases biogas production by 20-30% but also effectively reduces operating costs and resource waste. After further processing, the biogas residue can be converted into organic fertilizer, soil conditioner, or solid fuel, improving the resource utilization rate of solid waste and enhancing the system's environmental and economic benefits.

[0064] In the above embodiments, in addition to the four typical wastes listed, the system is also applicable to other types of organic waste, such as food processing residues, crop straw, slaughterhouse waste, and catering waste. The ranges of parameters such as TS, C / N, and VS can be expanded and adjusted according to the target raw material combination. The system supports stable operation within the ranges of TS 10-30%, C / N 5-15, and VS 60-90%. The mixing ratio can be dynamically updated through model optimization. The reactor can adopt a modular structure to support rapid switching and adjustment of different raw material quantities. When deployed in different regions, system parameters can be adapted to local climate conditions and raw material sources. Simultaneously, the system supports unified management of multi-site deployments via a remote platform, making it suitable for urban and rural distributed waste co-processing scenarios.

[0065] The above implementation also includes an abnormal operating condition prediction algorithm module, which is used to issue an early warning at least 30 minutes before the system malfunctions; and a multi-objective optimization module, which is used to solve the Pareto front for gas production, processing efficiency and unit energy consumption.

[0066] To achieve automated control using the above method, the present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is used to implement the system control method. like Figure 3 and Figure 6As shown, the abnormal operating condition prediction algorithm module and the multi-objective optimization module are deployed in the central controller or edge computing unit, serving as the core algorithm module carrying the intelligent control logic. Firstly, the abnormal operating condition prediction algorithm module accesses multi-dimensional operational data (such as pH, VFA, temperature, methane concentration, sludge concentration, etc.) collected by the IoT platform, combines historical fault records with model training results, and uses machine learning methods (such as Support Vector Machine (SVM), Random Forest, or LSTM time series models) to construct a state prediction model. This model can predict the system's operating trend within the next 30 minutes and trigger early warning logic before identifying characteristic patterns such as abnormal VFA accumulation, sudden drops in methane concentration, and temperature deviations, thereby achieving proactive response and risk avoidance.

[0067] Secondly, the multi-objective optimization module constructs a Pareto optimization model with the objective functions of maximizing gas production, optimizing processing efficiency, and minimizing unit energy consumption. It employs evolutionary algorithms (such as NSGA-II and MOEA / D) to perform multi-dimensional optimization of feed ratios and operating parameters (pH, temperature, stirring frequency, etc.). This module can output a set of non-dominated solutions (Pareto front), providing multiple optimal control strategies for system operation. Operators or automatic control systems can then make trade-offs and choices to achieve the best performance balance under different operating objectives.

[0068] The program module integrates predictive and multi-objective decision-making functions into traditional automated control, significantly improving the system's intelligence and operational robustness. The abnormal operating condition prediction function effectively reduces the probability of sudden system instability, improving sensitivity and safety redundancy under complex operating conditions. The multi-objective optimization module solves the problem of traditional single-objective optimization easily getting trapped in local optima, achieving a dynamic balance between resource utilization efficiency, gas production performance, and energy consumption. Combined with the central controller and IoT platform, the program supports adaptive switching of operating strategies and continuous optimization in long-term operating modes, driving the system's transformation from experience-based control to intelligent scheduling.

[0069] In the above implementation, the abnormal operating condition prediction algorithm can enhance its ability to identify nonlinear time series relationships based on more complex deep learning structures (such as GRU and Transformer), and the prediction time window can be adjusted to 15-60 minutes according to the actual operating conditions; the multi-objective optimization module can be expanded to introduce other objective dimensions, such as heavy metal inhibition effect, product quality or carbon emissions, and supports online learning capabilities to achieve adaptive optimization of the model; the program can be deployed on local industrial computers, edge servers or industrial cloud platforms, and supports unified cross-site calls; the data interface is compatible with industrial communication protocols such as OPC UA, Modbus TCP / IP, and MQTT, realizing integrated applications with various control systems and sensing platforms; the program can also integrate graphical interfaces and decision support tools to provide users with visual optimization suggestions and abnormal response paths.

[0070] Example 2 This embodiment focuses on system integration and engineering applications, combined with the appendix. Figures 1-5 This paper elaborates in detail on the structural composition, functional linkage and control logic of the IoT-based intelligent treatment system for multi-source solid waste synergistic anaerobic digestion and biogas residue resource utilization in actual deployment scenarios.

[0071] like Figure 1 As shown, the system integrates five core subsystems: intelligent sorting module ( Figure 2 ), co-anaerobic digester group, Internet of Things monitoring platform ( Figure 4 ), biogas residue treatment line ( Figure 5 The system includes a central controller. Each subsystem operates collaboratively through material conveying channels and a communication network, constructing an integrated processing path covering "source identification—collaborative degradation—state sensing—intelligent control—resource recycling".

[0072] like Figure 2 The intelligent sorting module, deployed at the front end of the system, is a key component for achieving refined source management and automated preprocessing. Its structure mainly includes a high-speed image acquisition system, a depth image recognition algorithm platform, a multi-degree-of-freedom industrial robotic arm, a sorting path control unit, and a multi-station buffer device.

[0073] For image acquisition, the module is equipped with a multi-angle high-definition industrial camera and an adaptive LED light source system, which can acquire images of solid waste entering the sorting area via conveyor belt in real time. To adapt to the differences in color, surface texture, and shape of different materials, the system uses a self-adjusting exposure and filtering strategy to ensure stable image data quality.

[0074] The image recognition processing unit integrates deep learning algorithms (such as YOLOv5 or ResNet) to quickly classify, detect, and locate acquired images. This algorithm model builds a classifier using a large amount of training data, capable of identifying fruit peels and bones, vegetable scraps, straw and hair in kitchen waste, and interfering materials such as plastics and metals in municipal sludge, with an accuracy exceeding 95%.

[0075] After identification, the system sends the coordinates and category information to the robotic arm control platform. The multi-degree-of-freedom robotic arm quickly performs grasping, sucking, or pushing actions based on the positioning information. The end effector can switch between suction cups, grippers, magnetic devices, and other tools according to the material type to complete the classified transfer. All classified materials are automatically delivered to the corresponding temporary storage bins or pre-processing inlets according to preset paths.

[0076] To further enhance adaptability, this module is equipped with an online model self-learning mechanism. By continuously collecting actual sorting images and feedback data, the system can automatically fine-tune the recognition model and supplement samples, thereby adapting to seasonal fluctuations in raw material structure, regional changes, or sudden pollution sources.

[0077] Furthermore, the system supports multi-material collaborative identification and dynamic sorting mechanisms. In complex raw material scenarios, priority can be given to identifying the main material category and removing impurities in batches. The accompanying sorting trajectory optimization module can adjust the robotic arm scheduling path and cycle time control in real time according to the material flow rate, ensuring sorting efficiency and stability.

[0078] In summary, the intelligent sorting module not only improves the automation level of front-end material handling, but also provides homogeneous and stable raw material input for the back-end anaerobic reactor, significantly enhancing the continuity, adaptability, and resource recovery efficiency of the overall system operation.

[0079] like Figure 1 The co-anaerobic digestion reactor unit: Following the intelligent sorting module, the co-anaerobic digestion reactor unit serves as the core biochemical reaction unit of this system, undertaking the crucial task of converting organic solid waste into biomass energy (mainly methane). This reactor unit adopts a two-stage series structure to achieve the staged co-degradation of complex substrates from multiple sources, thereby improving the organic matter conversion rate and gas production stability.

[0080] 1. Partition structure design The first-stage reactor (hydrolysis and acidification zone) primarily aims to hydrolyze and initially ferment large organic molecules such as proteins, polysaccharides, and fats. By controlling the pH at 5.5-6.5 and the reaction temperature at 35±1°C, a slightly acidic environment conducive to the synergistic metabolism of hydrolytic and acidifying bacteria is created. This zone effectively increases soluble chemical oxygen demand (sCOD), providing easily degradable intermediates for downstream methane generation.

[0081] Second-stage reactor (methanogenic zone): This zone mainly carries out the methanation reaction of intermediate products such as acetic acid, hydrogen, and formic acid. The operating temperature is maintained in the mesophilic range (35-37℃), and the pH is controlled between 6.8 and 7.4 to ensure the activity of methanogenic bacteria and the stability of the system.

[0082] By cascading the reactors, the hydraulic retention time (HRT) and load intensity can be independently adjusted between the upstream and downstream reactors, allowing each type of microorganism to function in its optimal metabolic environment. This avoids microbial competition and system imbalance, and significantly improves the efficiency of synergistic degradation.

[0083] 2. Internal structure optimization To further improve the internal physical and biological properties of the reactor, the following key units are incorporated into the system's structural design: Inclined baffle array: Several baffles with inclination angles of 30°-45° are arranged along the vertical direction of the reactor to optimize the substrate flow path and avoid dead zones and short-circuiting. This structure helps to form a spiral upward flow field, improving substrate mixing uniformity and microbial contact efficiency.

[0084] Jet mixing system: A jet pipe located at the bottom of the reactor forms an axial circulation with high-pressure reflux liquid, and the nozzle diameter is controlled at 10-15 mm. Compared with traditional paddle agitation, this system has advantages such as no mechanical contact, low maintenance cost, and low energy consumption, and is suitable for uniform mixing and preventing sedimentation of high-concentration, high-viscosity organic substrates.

[0085] Temperature-controlled insulation structure: The reactor adopts a double-layer 304L stainless steel shell structure, with polyurethane insulation material filling the interlayer, which can effectively prevent heat loss and maintain a stable reaction temperature. Combined with a temperature-controlled heating system, it ensures normal operation of the system even in winter or low-temperature regions.

[0086] To verify the impact of the "baffle angle" on the internal flow field and material residence behavior of the reactor, this embodiment compares three angles: 30°, 45°, and 60°, while maintaining consistent conditions such as nozzle diameter, stirring speed, and HRT. Considering that 45° is the preferred configuration given in the specification, we aim to evaluate its trade-off effect between methane yield, specific energy consumption, and operational stability through measured data, and calibrate the optimal range of structural parameters accordingly. Key operating indicators for the three angles are shown in Table 2-1.

[0087] Comparative test: Deflector angle optimization test (30° / 45° / 60°) Experimental design: Under the condition that the reactor structure and flow components (nozzle diameter, stirring speed, HRT, etc.) are consistent, three guide vane angles are set: 30°, 45° (preferred), and 60°.

[0088] Evaluation metrics: methane yield (mL CH4 / gVS), specific energy consumption (kWh / kg TS), and stability (peak VFA mg / L and pH RSD%).

[0089] Results and Analysis: The comparison results of the three guide vane angles are shown in Table 2-1.

[0090] Table 2-1 Effects of different guide vane angles on gas production, energy consumption and stability (mean ± SD, n=3)

[0091] Note: Room temperature (25±2)°C; substrate on dry basis; energy consumption calculated as total system power consumption per unit TS; data are mean ± SD, n=3.

[0092] As shown in Table 2-1, 45° has better overall performance than 30° and 60° (higher gas production, lower VFA peak, and energy consumption decreases / maintains a low level). Based on this, 45° is determined to be the preferred angle and will be used in subsequent operating condition combinations.

[0093] 3. Microwave Preprocessing Unit To enhance the processing capacity for recalcitrant substrates (such as oily kitchen waste and high-fiber fruit and vegetable residues), a microwave pretreatment module is integrated at the top of the reactor. This module operates at a frequency of 2450 MHz and has an energy density controlled between 0.5 and 1.5 W / g. This module possesses the following technical characteristics: Rapid heating and chain breaking: Microwave action causes water molecules in the substrate to vibrate at high frequency, achieving cell wall disruption and structural breakage, and increasing the proportion of soluble organic matter.

[0094] Processing time is adjustable: the duration of a single radiation session is 90–120 seconds, which can be dynamically adjusted according to the characteristics of the raw materials.

[0095] Intelligent feedback control: Integrating a temperature sensor and a feedback closed-loop controller ensures that the microwave processing temperature is within the optimal range, avoiding the generation of coking or pyrolysis byproducts.

[0096] The introduction of the microwave pretreatment unit significantly shortens the system start-up time, improves the substrate degradation rate, and enhances the resistance to load shocks, which is one of the important technological innovations of the system of this invention.

[0097] Given clearly defined structural parameters, the introduction of microwave pretreatment directly impacts substrate degradability and system start-up / steady-state performance. To avoid relying solely on empirical judgment, this embodiment sets up two control groups under the same feed load and control conditions: one with and one without microwave pretreatment. The microwave parameters were 2450±50 MHz, and the treatment time was 8 min. We focused on changes in gas production efficiency, VFA peak value, pH fluctuations, and other stability indicators, as well as the corresponding differences in unit energy consumption. The comparison results of the two groups are shown in Table 2-2.

[0098] Comparative Experiment: Feed Pretreatment Comparison (with / without Microwave) Experimental objective: To evaluate the impact of microwave pretreatment on the gas production efficiency, system stability, and energy consumption of anaerobic digestion, and to provide a basis for determining whether to configure a microwave unit.

[0099] Apparatus and conditions: The overall apparatus and control strategy of the embodiment were adopted; the microwave parameters were 2450±50 MHz, the processing time was 8 min (the control group was without microwave), and other operating conditions (substrate ratio, feed load, temperature 35±1 ℃, pH setting 6.8-7.4, stirring 10-30 rpm) were kept consistent.

[0100] Evaluation metrics: methane yield (mL CH4 / gVS), specific energy consumption (kWh / kg TS), and stability (measured by peak VFA mg / L and pH fluctuation RSD%).

[0101] Results and analysis: Key operating indicators for microwave pretreatment and the control group are shown in Table 2-2.

[0102] Table 2-2 Effect of microwave pretreatment on anaerobic digestion performance (mean ± SD, n=3)

[0103] Note: Room temperature (25±2)℃; substrate on dry basis; energy consumption calculated as total system power consumption / unit TS; data are mean ± SD, n=3.

[0104] As shown in Table 2-2, microwave pretreatment significantly improves methane yield, reduces VFA peak, and narrows pH fluctuations with an acceptable increase in energy consumption, verifying the necessity and rationality of configuring a microwave unit in this system.

[0105] 4. Integration with multi-parameter sensing systems Each stage of the reactor is equipped with multi-parameter sensors, including: pH sensor (MEMS structure, accuracy ±0.05); Infrared methane concentration analyzer (0-100% volume fraction, resolution 0.1%). Ultrasonic sludge concentration meter (error <3%).

[0106] The aforementioned sensor data will be uploaded to the IoT monitoring platform in real time for visualization of reactor operating status, parameter optimization, and formulation of anomaly intervention strategies.

[0107] In summary, the synergistic anaerobic digestion reactor unit, through biochemical staged design, fluid structure optimization, intelligent pretreatment, and multi-parameter sensing integration, effectively improves the treatment capacity, gas production efficiency, and system operational stability of multi-source organic solid waste, making it a core unit for achieving the dual goal of "resource-energy" conversion. Furthermore, combining it with a digital twin model platform can further enhance the system's simulation prediction and optimized scheduling capabilities.

[0108] like Figure 4 The IoT monitoring platform serves as the information nerve center of this system, running through various subsystems such as intelligent sorting, collaborative anaerobic digestion, biogas residue treatment, and central control. It constructs a closed-loop data link of "sensing-transmission-computation-feedback," which is the fundamental guarantee for realizing intelligent operation management and remote collaborative control.

[0109] 1. Communication Structure and Network Configuration like Figure 4As shown, the platform uses LoRaWAN (Low Power Wide Area Network) or 5G cellular network as the primary communication method, and can flexibly switch wireless protocols according to the application environment: LoRaWAN is suitable for scenarios with campus-level, multi-point deployment, long transmission distance, and low power consumption requirements; 5G communication is suitable for scenarios with large data volumes, high concurrency, and high real-time requirements, such as high-frequency sampling and high-resolution image transmission.

[0110] The platform reliably connects with various field sensors (pH, VFA, methane concentration, temperature, sludge concentration, etc.), actuators (pumps, valves, heaters, dosing devices), and edge controllers through the aforementioned communication technologies. It also supports mechanisms such as breakpoint resume, disconnection reconnection, and dual-channel fault tolerance to ensure system data stability and uninterrupted control chain.

[0111] 2. Platform Functional Module Composition The platform consists of two core modules: Edge Computing Unit: Deployed in an on-site control cabinet or on a nearby server; Real-time acquisition of sensor data, execution of local preliminary calculations, anomaly detection, and issuance of control commands; It supports independent operation, has local control policy caching, and can maintain core functions without interruption when the network is disconnected; Embedded lightweight machine learning models (such as decision trees and support vector machines) can be used for short-term trend prediction or anomaly identification.

[0112] Cloud Database and Remote Operations Center: Connect to the cloud platform via MQTT, HTTPS, or industrial protocols (such as OPC UA); Enables efficient storage of historical data, visualization of trend charts, parameter comparison, energy consumption statistics, and remote management of multiple sites; It comes with a web-based visual dashboard and a mobile app interface, supporting hierarchical user permission management and mobile alert push notifications.

[0113] 3. Data Flow and Control Logic The platform's data interaction and control process can be divided into the following three levels: Front-end sensing layer: Various sensors achieve a sampling frequency of seconds to continuously collect and preprocess key parameters (pH, temperature, VFA, gas yield) locally.

[0114] Local Response Layer: Edge computing units provide rapid feedback based on set thresholds, trend models, or control logic. For example, when the pH is below 6.6 and the VFA rise rate exceeds the limit, the system will automatically execute the alkali dosing command and issue a local warning.

[0115] Cloud-based optimization layer: Regularly synchronize all monitoring data to the cloud, combine AI models to perform trend modeling, energy efficiency analysis and push operation optimization suggestions, and can perform actions such as "predicting gas yield in the next 48 hours" or "suggesting optimization of feed structure".

[0116] 4. Multi-site deployment and remote management capabilities The IoT platform supports multi-site access and is suitable for urban centralized processing centers or county-level distributed processing networks, possessing the following capabilities: Supports concurrent access and control of dozens of terminal devices; A unified cloud-based interface manages the operational status of different sites; Automatically generates daily, weekly, and monthly reports, and supports PDF export; Provide rule-based automated task triggering (e.g., "When gas production rate drops by more than 20%, activate the data diagnostic model and notify management personnel").

[0117] 5. Intelligent early warning and operation and maintenance assistance The platform is equipped with an intelligent early warning system, which has the following functions: Anomaly detection based on multiple indicator combinations (e.g., pH decrease + temperature increase → acidification trend identification). Time series models (such as LSTM) predict potential risks; The alert push supports multiple channels, including SMS, app notifications, and email. Fault-assisted diagnosis: By combining historical similar data, it generates a ranking of possible causes of faults and corresponding solutions.

[0118] In summary, the IoT monitoring platform of this invention provides a highly reliable, highly responsive, and sustainably optimized information infrastructure for the entire collaborative processing system through its flexible communication architecture, edge-cloud collaborative computing, full-process perception and control, and remote intelligent operation and maintenance functions. It is a key support platform for promoting the system towards intelligent solid waste treatment.

[0119] The following is the expanded section: "IV. Biogas Sludge Treatment Line ( Figure 5 The content enhances the explanation of its technological process, structural composition, control logic, and resource utilization effects. like Figure 5 The biogas residue treatment line, located at the end of the system, receives the effluent from the co-processing anaerobic reactor and is a key component in achieving waste reduction, harmlessness, and resource recovery. This treatment line integrates three major process stages: dewatering, modification, and molding, enabling in-depth treatment and functional material conversion of anaerobic residue, and allows for flexible deployment through a modular structure.

[0120] 1. First stage: Dehydration treatment Primary dewatering uses a spiral filter press to compress the biogas residue, which has a moisture content of over 80% after the reaction, to a moisture content of less than 40%. The specific configuration is as follows: Equipment structure: It consists of a feeding screw, a filter drum, a hydraulic pressure roller, and a liquid collection tray; Operating parameters: Processing capacity 10-20 m³ / h, adjustable inlet and outlet pressure; Control method: Linked with the discharge pump of the front-end reactor, the dewatering load is dynamically adjusted according to the discharge volume; Wastewater recycling: The extruded liquid is directly returned to the reaction system or sent to the wastewater treatment module to achieve internal recycling.

[0121] This step significantly reduces the volume of the material, creating favorable physical conditions for subsequent modification and heat treatment.

[0122] 2. Second stage: Modification treatment The dehydrated, semi-dried biogas residue enters the modification system, primarily aimed at improving its environmental safety and agricultural value. This module consists of three main units: (1) Biochar blending unit: Dosage range: 5-8% (by mass); Raw material sources: rice husk charcoal, straw charcoal, sludge charcoal, etc.; Mechanism of action: It adsorbs heavy metal ions and some organic pollutants through its porous structure, thereby enhancing physical stability and nutrient slow-release capacity.

[0123] (2) Plasma processing chamber: Principle: A high-energy thermal plasma source (2-5kW) is used to instantaneously decompose the surface and interior of the material; Treatment effect: High temperature environment (1000°C) can effectively lyse pathogenic microorganisms, antibiotic residues and residual harmful organic matter; Control mechanism: Real-time temperature control + closed-loop energy regulation, with supporting cooling protection and exhaust gas purification modules to ensure safe and stable operation.

[0124] (3) Humic acid extraction module: Extraction method: Alkali dissolution-acid precipitation process; Extraction rate: ≥65%, reaction temperature and pH can be adjusted to suit different raw materials; Product Uses: The extracted humic acid solution is used to formulate foliar fertilizers, seedling substrates, or soil conditioners to increase the added value of the final product.

[0125] This modified system, while ensuring environmental friendliness, "functionalizes" biogas residue, laying the foundation for its resource utilization.

[0126] 3. Third stage: Molding process The modified biogas residue enters the molding section to achieve standardization of its physical form and facilitate transportation and storage. The specific process is as follows: Equipment configuration: Roll forming machine, compression molding equipment and low temperature drying system are used; Product forms: granular organic fertilizer (3-5mm in diameter, moisture content <15%); soil conditioner (powder / granular); solid fuel rods (20-50mm in length, calorific value approximately 12-15 MJ / kg). Process control: Material moisture content, particle size distribution, and extrusion pressure are all intelligently regulated by a central controller; Packaging and unloading: Equipped with an online weighing and automatic packaging system to achieve standardized production and batch traceability.

[0127] 4. System integration and intelligent control mechanism The entire processing line is jointly scheduled by a central controller and an IoT platform, and has the following operational characteristics: Load Adaptive: The operating status of each processing section can be dynamically matched according to the discharge rhythm of the anaerobic reactor to avoid congestion or stagnation; Data Monitoring: Real-time collection of key parameters such as dewatering pressure, power consumption, humic acid concentration, and moisture content of the formed product; Energy Consumption Assessment and Optimization: Based on data backtracking, the unit processing energy consumption, carbon footprint, and other indicators are assessed to provide a basis for operation optimization; Abnormal Response Mechanism: Equipment failure and processing abnormalities (such as material blockage and high temperature warning) trigger alarms and automatic shutdown protection.

[0128] 5. Resource utilization and economic benefits The biogas residue treatment line in this invention not only meets the national standard NY525-2021 for organic fertilizer, but also has the following advantages: High resource utilization rate: The utilization rate of residue is over 90%; diversified products: The output path can be flexibly switched according to the properties of raw materials, such as fertilizer, fuel, and conditioner; increased added value: The extraction of humic acid and the addition of functional biochar make the end products competitive in the market; significant carbon emission reduction benefits: Compared with landfill / incineration, each ton of residue can reduce CO2 equivalent emissions by about 0.5 tons.

[0129] In summary, the biogas residue treatment line of this invention achieves efficient reduction and high-value utilization of by-products through a closed-loop process chain of physical concentration, biological function enhancement, and structural shaping, thus constructing a true "resource closed loop" for anaerobic digestion systems. This treatment line is highly modular and suitable for end-of-pipe resource recovery solutions in distributed deployments or centralized treatment centers, making it a key component in promoting the green, efficient, and intelligent development of solid waste treatment systems.

[0130] like Figure 3The central controller is the intelligent decision-making core of this system, undertaking key functions such as end-to-end data fusion, work status scheduling, fault early warning, and adaptive optimization. Its hardware platform adopts industrial-grade embedded edge computing devices, coupled with a highly available redundant architecture, to ensure high system reliability and real-time response.

[0131] 1. Control Logic like Figure 3 As shown, the working logic of the central controller is divided into five main stages: (1) Data collection: It receives multi-source data from various subsystems in real time, including raw material characteristics (C / N, VS, HM), operating status (temperature, pH, VFA, methane concentration), equipment load, etc.; and establishes a complete time-series operating condition information flow by combining it with historical databases.

[0132] (2) Feed optimization and proportion control: A hybrid optimization algorithm (linear programming + genetic algorithm) was used to calculate the optimal feeding strategy. The objective functions included: maximizing biogas yield; reducing pH fluctuation; and reducing the risk of ammonia inhibition. The output includes control commands such as "proportion of feed categories, total amount added, and order of batching", which automatically link the intelligent sorting module and the feeding system.

[0133] (3) Linkage control of operating parameters: Based on the optimization results, the parameters of the co-reactor were adjusted synchronously: the temperature of the reaction zone was controlled (accuracy ±0.5°C); the jet stirring frequency was adjusted; the pH control range was 6.8-7.2, and the alkali addition frequency and flow rate were automatically adjusted.

[0134] (4) Anomaly identification and intelligent intervention: When a key parameter (such as VFA) is detected to exceed a set threshold (e.g., 3000 mg / L) or its rate of change is abnormal: the system initiates intelligent intervention; triggers multiple measures in tandem, such as alkali compensation, intermittent feeding, and stirring frequency adjustment; if these measures are ineffective in the short term, the system automatically enters "protection mode" to limit the load and prevent system acidification instability. After the structure and pretreatment are determined, the key to operational stability lies in timely response to abnormal fluctuations. Based on the control logic given in the manual, when the online monitored VFA concentration is ≥3000 mg / L, a "flow restriction + alkali addition" intervention is triggered to suppress acidification instability and stabilize pH. To evaluate the effectiveness of this threshold setting and linkage strategy, this embodiment compares two operating modes: "threshold strategy not enabled" and "VFA ≥3000 mg / L trigger strategy enabled." Relevant indicators are shown in Tables 2-3.

[0135] Comparative experiment: Validation of VFA threshold-triggered control strategy (threshold ≥3000 mg / L).

[0136] Control logic: When the online monitored VFA concentration is ≥3000 mg / L, the central controller triggers the flow restriction + alkali addition strategy (alkali is NaOH or Ca(OH)2), and maintains the pH at 6.8-7.4.

[0137] Comparison settings: Threshold triggering policy is not enabled.

[0138] Evaluation metrics: methane yield (mL CH4 / gVS), specific energy consumption (kWh / kg TS), and stability (peak VFA mg / L and pH RSD%).

[0139] Results and Analysis: The verification results of the threshold triggering strategy are shown in Table 2-3.

[0140] Table 2-3 Comparison of system operation with and without VFA threshold triggering strategy (mean ± SD, n=3)

[0141] Note: Room temperature (25±2)℃; substrate on dry basis; energy consumption calculated as total system power consumption / unit TS; data are mean ± SD, n=3.

[0142] As shown in Table 2-3, after enabling threshold triggering, the VFA peak value decreased significantly, the pH fluctuation (RSD) converged, and the methane yield increased, proving that setting the threshold = 3000 mg / L and the linkage intervention have a positive effect on improving system stability and efficiency.

[0143] (5) Predictive analysis and remote scheduling: It embeds an anomaly trend prediction module based on Long Short-Term Memory (LSTM) networks; it has a minimum warning window of 30 minutes and can respond to future fluctuations in key parameters; all scheduling instructions can be issued remotely and are retained through operation logs to support traceability and auditing.

[0144] 2. Control Algorithm Model Integration The central controller integrates the following algorithm models to achieve intelligent optimization and stable operation:

[0145] 3. Platform integration and operational visualization The controller is highly integrated with the IoT monitoring platform, supporting: control strategy distribution and recording; historical control-response relationship comparison and analysis; one-click switching of operating modes (automatic / semi-automatic / manual); cloud-based control strategy updates and hot deployment (control logic replacement can be completed without stopping the system). Through the web-based management interface, users can view the current feeding structure, parameter settings, and execution status at any time, and manually confirm, adjust strategies, or intervene in operations based on prediction results and warnings.

[0146] 4. Engineering Applications and Deployment Examples The central control system of this invention has been successfully deployed in the following typical scenarios: Urban centralized processing center (City A project): Daily processing capacity reaches 100 tons, employing a dual-reactor parallel + dual-controller hot backup architecture; achieving 24-hour unattended operation, 8-hour self-diagnosis of operational data and report generation; daily biogas production increased by 15%, and system stability improved by 30%. County-level distributed collaborative processing station (County B project): Configured with a simplified central controller, supporting local / cloud hybrid control; automatically adjusting the daily processing load of each station and uniformly uploading operational data to the district-level platform; cost savings of 20%, and remote control efficiency improved by over 50%. Agricultural park crop-livestock cycle station (Town C project): Emphasizing the co-processing of manure and straw; adjusting fertilizer production and parameter preferences according to the planting cycle, with a system fertilizer blending accuracy of up to 95%.

[0147] This invention's central controller not only achieves adaptive regulation and intelligent scheduling of the system's entire lifecycle operating conditions, but also significantly improves the system's stability, resource conversion efficiency, and operational intelligence level through AI-driven optimization and prediction mechanisms. Its flexible deployment and strong adaptability make it particularly suitable for high-quality operation requirements in various scenarios such as urban-rural integration, agricultural recycling, and regional solid waste collaborative treatment.

[0148] In summary, the intelligent treatment system for multi-source solid waste co-processing anaerobic digestion and biogas residue resource utilization proposed in this invention integrates IoT sensing technology, edge computing, multi-objective optimization algorithms, digital twin modeling, and a closed-loop resource utilization path. It not only effectively solves the problems of poor adaptability, low control precision, and low resource utilization in existing technologies for multi-source solid waste co-processing, but also establishes a complete technology chain covering "raw material identification—reaction regulation—byproduct utilization—intelligent early warning." The system has a reasonable structural design, a high degree of intelligent operation, and good environmental, economic, and promotional adaptability, making it suitable for various organic waste treatment scenarios in urban and rural areas.

[0149] It should be noted that although this invention has been described in detail with reference to specific structures, parameters and implementation methods, those skilled in the art should understand that various equivalent substitutions and local adjustments can be made to the structural form, control method or system deployment method without departing from the basic concept of this invention, and all such variations and modifications should also be considered to fall within the protection scope of this invention.

Claims

1. An integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on Internet of Things monitoring, characterized in that, include: The intelligent sorting module is used for the automated classification of multi-source organic solid waste, including a machine vision recognition unit and a robotic arm execution unit; The co-anaerobic digestion reactor group, connected to the intelligent sorting module, is used to receive the sorted solid waste and perform anaerobic digestion treatment. The co-anaerobic digestion reactor group includes at least two reactors arranged in series, each with a hydrolysis acidification zone and a methanogenic zone, and each reactor has a multi-parameter sensor array inside. The Internet of Things (IoT) monitoring platform is connected to the multi-parameter sensor array via a wireless communication network for real-time monitoring of the reactor's operating status. It includes an edge computing unit and a cloud database. The edge computing unit is used to optimize control parameters in real time, and the cloud database is used to store historical process data. The biogas residue treatment line, connected to the co-anaerobic digestion reactor group, is used to treat the biogas residue produced by anaerobic digestion for resource utilization, and includes a dewatering device, a modification device and a molding device in sequence. The central controller is connected to the intelligent sorting module, the collaborative anaerobic digestion reactor group, the Internet of Things monitoring platform, and the biogas residue treatment line to uniformly execute process control commands.

2. The integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring as described in claim 1, characterized in that, include: The central controller performs the following method: Real-time acquisition of raw material characteristic data, including carbon-to-nitrogen ratio (C / N), volatile solids content (VS), and heavy metal concentration (HM). i , where i represents the i-th heavy metal; The optimal feed ratio is calculated based on a hybrid optimization model. The objective function of the model is: in, , , These are weighting coefficients used to characterize the impact of C / N ratio deviation, VS deviation of volatile solids content, and heavy metal risk, respectively; HM i This represents the concentration of the i-th heavy metal; The reactor operating parameters are dynamically adjusted according to the optimal ratio, including pH control at 6.8~7.4, temperature control at 35±1℃, and stirring intensity control at 10~30 rpm. When the concentration of volatile fatty acids monitored online exceeds the threshold, the alkali dosing system is automatically activated to neutralize acidic substances and inhibit acidification instability. The central controller sends the results of the optimization model as control commands to the actuators, thereby achieving comprehensive optimization of gas production efficiency, system stability, and environmental risks.

3. The integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring as described in claim 1, characterized in that, The reactors in the synergistic anaerobic digestion reactor group include: The shell is double-layered, with the inner and outer layers made of stainless steel and the sandwich layer filled with polyurethane insulation material. The air deflector array has each deflector at a certain angle to the horizontal plane. A jet agitator is installed at the bottom of the reactor; The microwave pretreatment unit, located at the top of the reactor, is equipped with a temperature feedback control system for dynamically adjusting the target temperature during the heating process.

4. The integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring as described in claim 1, characterized in that, The multi-parameter sensing array includes: Embedded pH sensor with a measurement accuracy of ±0.05; Infrared methane analyzer, with a measurement range of 0~100% and a resolution of 0.1%; Ultrasonic sludge concentration meter with a measurement error of less than 3%.

5. The integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring according to claim 2, characterized in that, The hybrid optimization model was trained using a machine learning algorithm, and the training dataset includes: No fewer than 2000 sets of experimental data with different raw material ratios; Climate condition parameters, including temperature, humidity and atmospheric pressure; Historical operational failure records.

6. The integrated system for multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring as described in claim 1, characterized in that, The modification device includes: A biochar addition unit is used to add biochar to biogas residue at a certain mass ratio; a plasma treatment chamber; and a humic acid extraction module.

7. A control method for an integrated system of multi-source solid waste co-anaerobic digestion and biogas residue utilization based on Internet of Things monitoring, characterized in that, Includes the following steps: S1: Real-time acquisition of raw material characteristic data, including carbon-to-nitrogen ratio (C / N), volatile solids content (VS), and heavy metal concentration (HMi); S2: Calculate the optimal feed ratio based on the hybrid optimization model. The objective function is: S3: Dynamically adjust reactor operating parameters, including pH, temperature, and stirring intensity, according to the optimal ratio; S4: When the concentration of volatile fatty acids exceeds the preset threshold, the alkali dosing system will be automatically activated; S5: The optimization results are sent as control commands to the actuators to achieve comprehensive optimization of system operation.

8. The control method for the integrated system of multi-source solid waste synergistic anaerobic digestion and biogas residue utilization based on Internet of Things monitoring according to claim 7, characterized in that, Also includes: Abnormal operating condition prediction steps: Based on real-time data collected by the IoT platform, potential faults are identified and warnings are issued at least 30 minutes in advance through machine learning models.

9. The control method for the integrated system of multi-source solid waste synergistic anaerobic digestion and biogas residue utilization based on Internet of Things monitoring according to claim 7, characterized in that, Also includes: Multi-objective optimization steps: Based on gas production, processing efficiency, and unit energy consumption, a multi-objective optimization strategy is generated using the Pareto front solution method.

10. A digital twin modeling method for an integrated system of multi-source solid waste co-anaerobic digestion and biogas residue utilization based on IoT monitoring, characterized in that, Includes the following steps: Establish a three-dimensional fluid dynamics simulation model of the system, which should include at least the reactor structure, fluid domain and boundary conditions; Multiphysics simulation tools are used to perform joint simulation of the temperature field, pressure field and electromagnetic field of the system during operation, so as to obtain the virtual operating state of multiphysics coupling; Based on the real-time collected operating data from the Internet of Things platform, the key parameters of the three-dimensional fluid simulation model are dynamically updated through a data assimilation algorithm, so that the deviation between the model output and the actual operating state is controlled within a preset tolerance. The updated digital twin model is used to predict the future operating trend of the system and output early warning information or optimize control strategies.

Images