Intelligent linkage control method and system of Internet of Things terminal and zeolite purification system

Abstract

The invention is suitable for the technical field of environmental governance and air purification, and provides an intelligent linkage control method and system of an Internet of Things terminal and a zeolite purification system, environmental parameters and zeolite module states are collected through a distributed sensor array, and a three-dimensional pollution situation map is constructed through space-time alignment and noise suppression; analyzing a dynamic coupling relationship between pollutant diffusion and zeolite adsorption, calculating a pollution treatment priority and generating a control instruction queue; dynamically adjusting a fan matrix airflow distribution and molecular sieve module activation strategy; when pollution suddenly occurs, a distributed consensus network is constructed, multiple nodes are coordinated to establish a directional adsorption barrier, and gradient regeneration is triggered; and the regeneration period and the energy consumption ratio are dynamically adjusted by combining the feedback of the dielectric sensor. According to the scheme, full-process intelligentization of pollution monitoring, priority scheduling, multi-node cooperative control and energy efficiency optimization is achieved, the method is suitable for industrial pollution prevention and control, urban air treatment and other scenes, the purification efficiency and the resource utilization rate are remarkably improved, and the system energy consumption and the operation and maintenance cost are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental management and air purification, and in particular relates to an intelligent linkage control method and system for an Internet of Things terminal and a zeolite purification system. Background Art

[0002] With the development of the Internet of Things, sensors, and intelligent control technologies, traditional passive purification models are gradually evolving towards an integrated, intelligent, and coordinated "monitoring-analysis-control" model. The industry is currently placing higher demands on the dynamic tracking of pollutants in complex environments, the coordinated control of multi-node purification units, and the optimization of system energy efficiency. This is particularly true in scenarios such as industrial pollution prevention and control, urban air quality improvement, and the maintenance of healthy indoor environments. Intelligent purification systems with real-time data-driven, adaptive adjustment capabilities are urgently needed to improve pollution treatment efficiency and resource utilization.

[0003] Existing technologies primarily employ single-point control strategies triggered by fixed thresholds. These strategies monitor local environmental parameters through a single sensor and adjust fan power or regeneration cycles based on preset rules. These strategies lack dynamic coupling analysis of pollutant diffusion characteristics and purification unit performance. In multi-node collaborative scenarios, existing solutions rely on centralized control architectures, resulting in high inter-node state synchronization latency and difficulty in rapidly responding to sudden pollution incidents. Regeneration strategies often employ single regeneration methods based on timing or fixed saturation, failing to incorporate the real-time characteristics of zeolite module performance degradation, leading to an imbalance between regeneration energy consumption and module losses. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent linkage control method for an Internet of Things terminal and a zeolite purification system, aiming to solve the technical problems existing in the prior art identified in the background technology.

[0005] The present invention is implemented as follows: an intelligent linkage control method for an Internet of Things terminal and a zeolite purification system, the method comprising:

[0006] Through the distributed sensor array carried by the IoT terminal, environmental parameters and zeolite module status indicators are collected synchronously, and multi-source data are aligned in time and space and noise is suppressed to construct a three-dimensional pollution situation map;

[0007] Analyze the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, calculate the pollutant treatment priority, and generate a purification control instruction queue containing the priority;

[0008] Based on the purification control instructions, it sends instructions to the distributed zeolite purification unit to dynamically adjust the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module;

[0009] When a sudden pollution event is detected, a distributed consensus network is constructed to coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy;

[0010] Combined with the zeolite performance attenuation curve fed back by the dielectric sensor, the regeneration cycle and energy consumption ratio are dynamically adjusted to establish a closed-loop control mechanism.

[0011] As a further solution of the present invention, the spatiotemporal alignment and noise suppression of multi-source data to construct a three-dimensional pollution situation map specifically includes:

[0012] Real-time collection of environmental parameters, including: volatile organic compound concentration, laser scattering particle counter to monitor particle distribution, digital temperature and humidity sensor to obtain temperature and humidity data, hot wire anemometer to measure airflow velocity, and monitoring the adsorption saturation and structural stress of the zeolite module;

[0013] Perform spatiotemporal alignment and noise suppression on multi-source heterogeneous data to limit time synchronization accuracy;

[0014] A three-dimensional spatial interpolation algorithm is used to construct a three-dimensional situation map of pollution concentration field, temperature and humidity field, and airflow velocity field.

[0015] As a further solution of the present invention, the analysis of the dynamic coupling relationship between the pollutant diffusion characteristics and the zeolite adsorption capacity calculates the pollutant treatment priority and generates a purification control instruction queue containing the priority, specifically including:

[0016] Establish a pollutant diffusion model to calculate the rate of change of any pollutant concentration in space over time

[0017]

[0018] Where C is the pollutant concentration, D is the diffusion coefficient, v is the air velocity vector, and S is the source term;

[0019] Construct a zeolite adsorption kinetic model to calculate the rate of change of zeolite adsorption over time

[0020]

[0021] Where q is the adsorption amount, k f is the adsorption rate constant, k d is the desorption rate constant, C * is the equilibrium concentration;

[0022] Perform dynamic priority calculation based on the edge computing engine to obtain the pollutant treatment priority:

[0023]

[0024] Among them, P is the treatment priority, T is the toxicity coefficient, is the normalized value of the pollutant concentration gradient, q is the current adsorption amount, q max is the maximum adsorption capacity, is the current adsorption saturation of zeolite, ɑ, β, γ, and δ are weight coefficients;

[0025] Based on the processing priority, a purge control instruction queue including a level priority is generated.

[0026] As a further solution of the present invention, the instructions are sent to the distributed zeolite purification unit to dynamically adjust the air flow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module, specifically including:

[0027] Read all purification control instructions and corresponding time series, send control instructions to distributed zeolite purification units, and dynamically adjust the airflow distribution ratio of the fan matrix;

[0028] A molecular sieve switching decision tree model was established, and the molecular sieve module was intelligently switched based on the pollutant type and concentration.

[0029] As a further solution of the present invention, the construction of a distributed consensus network, coordinating multi-node zeolite units to establish a directional adsorption barrier and triggering a gradient regeneration strategy, specifically includes:

[0030] Set pollutant concentration warning thresholds. When a single sensor detects that the pollutant concentration exceeds the warning threshold, a three-level warning mechanism is triggered;

[0031] Build a distributed consensus network, synchronize node states, and coordinate multi-node zeolite units to establish a directional adsorption barrier, where peripheral nodes act as annular airflow barriers.

[0032] A gradient regeneration strategy was triggered to restore the adsorption capacity of the zeolite purification system through thermal regeneration and solution chemical regeneration, respectively.

[0033] As a further solution of the present invention, the zeolite performance attenuation curve fed back by the dielectric sensor is combined to dynamically adjust the regeneration cycle and energy consumption ratio to establish a closed-loop control mechanism, specifically including:

[0034] A zeolite performance attenuation model was established to calculate the remaining adsorption capacity of zeolite:

[0035] q(t)=q0·e -λt ;

[0036] Where q(t) is the adsorption capacity at time t, q0 is the initial adsorption capacity, and λ is the attenuation coefficient;

[0037] Dynamically adjust the regeneration cycle to obtain the optimal regeneration cycle at each moment:

[0038]

[0039] Among them, T reg is the regeneration period, T max is the maximum regeneration period, T0 is the minimum regeneration period, λ max is the maximum attenuation coefficient, k is the adjustment coefficient;

[0040] Quantify the energy consumption during system operation and optimize the ratio of fan power to regeneration temperature:

[0041] E total =E fan +E reg =ρ·Q·H+σ·m·C p ΔT;

[0042] Among them, E total is the total energy consumption, E fan is the fan energy consumption, E reg is the regeneration energy consumption, Q is the air volume, H is the wind pressure, m is the regeneration mass, C p is the specific heat capacity, ΔT is the temperature rise, ρ and σ are the energy consumption coefficients

[0043] A deep reinforcement learning algorithm is used to continuously optimize control parameters.

[0044] Another object of the present invention is to provide an intelligent linkage control system for an Internet of Things terminal and a zeolite purification system, the system comprising:

[0045] The parameter data acquisition module is used to synchronously collect environmental parameters and zeolite module status indicators through the distributed sensor array carried by the IoT terminal, and to perform spatiotemporal alignment and noise suppression on multi-source data to construct a three-dimensional pollution situation map;

[0046] The treatment priority analysis module is used to analyze the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, calculate the pollutant treatment priority, and generate a purification control instruction queue containing the priority;

[0047] A control instruction issuing module is used to issue instructions to the distributed zeolite purification unit based on the purification control instructions, dynamically adjusting the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module;

[0048] The pollution event response module is used to build a distributed consensus network, coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy when a sudden pollution event is detected;

[0049] The energy consumption optimization closed-loop control module is used to dynamically adjust the regeneration cycle and energy consumption ratio based on the zeolite performance attenuation curve fed back by the dielectric sensor, thus establishing a closed-loop control mechanism.

[0050] The beneficial effects of the present invention are:

[0051] Through a distributed sensor array of IoT terminals, we enable global, simultaneous collection of environmental parameters and zeolite module status. Combined with spatiotemporal alignment and three-dimensional situation mapping techniques, we accurately capture pollution distribution and diffusion trends, providing a real-time data foundation for dynamic control. Based on coupled analysis of pollutant diffusion models and zeolite adsorption kinetics models, we construct a multi-dimensional priority calculation framework, enabling intelligent matching of pollution treatment sequences with purification resources, improving response efficiency in high-risk areas.

[0052] Dynamically adjusting fan matrix airflow distribution and molecular sieve module activation strategies allows for real-time reconfiguration of purification paths based on pollution type and concentration, avoiding energy waste and processing blind spots under fixed strategies. In the event of a sudden pollution event, a distributed consensus network and a targeted adsorption barrier mechanism achieve millisecond-level node coordination. Combined with a gradient regeneration strategy, this rapidly restores system processing capacity and significantly reduces the scope of the pollution impact.

[0053] The closed-loop control mechanism driven by dielectric sensor feedback and deep reinforcement learning dynamically optimizes the regeneration cycle and energy consumption ratio, thereby increasing the service life of the zeolite module while reducing overall energy consumption, enabling the system to maintain efficient and economical operation in complex environments.

[0054] The overall solution breaks through the response delay and resource mismatch bottlenecks of traditional purification systems through the intelligentization of the entire "data-model-control" process, significantly improving the accuracy, timeliness and sustainability of environmental governance. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of the intelligent linkage control method of the Internet of Things terminal and the zeolite purification system provided in an embodiment of the present invention;

[0056] Figure 2 A flowchart of the embodiment of the present invention for synchronously collecting environmental parameters and zeolite module status indicators, performing spatiotemporal alignment and noise suppression on multi-source data, and constructing a three-dimensional pollution situation map;

[0057] Figure 3 Provided for the embodiment of the present invention is a flowchart for calculating pollutant treatment priorities and generating a purification control instruction queue including the priorities;

[0058] Figure 4 A flowchart of an embodiment of the present invention providing instructions to a distributed zeolite purification unit to dynamically adjust the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module;

[0059] Figure 5 A flow chart of coordinating multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy provided by an embodiment of the present invention;

[0060] Figure 6A flowchart of a closed-loop control mechanism for dynamically adjusting the regeneration cycle and energy consumption ratio provided by an embodiment of the present invention;

[0061] Figure 7 This is a structural block diagram of the intelligent linkage control system of the Internet of Things terminal and the zeolite purification system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0063] Figure 1 Flowchart of the intelligent linkage control method of the Internet of Things terminal and the zeolite purification system provided by the embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0064] S100 uses a distributed sensor array on IoT terminals to simultaneously collect environmental parameters and zeolite module status indicators, and performs spatiotemporal alignment and noise suppression on multi-source data to construct a three-dimensional pollution situation map;

[0065] Distributed sensor arrays on IoT terminals enable high-density, multi-dimensional monitoring deployments in complex indoor and outdoor environments. For example, sensor nodes can be deployed at different heights in industrial plants, near ventilation outlets, and pollution sources, or grid-like monitoring networks can be formed at street corners and rooftops in urban areas. These networks simultaneously collect environmental parameters such as volatile organic compound concentration, particulate matter distribution, temperature and humidity, and airflow velocity, as well as state indicators such as the adsorption saturation and structural stress of zeolite modules. In data processing, high-precision time synchronization technology is employed to ensure microsecond-level alignment of multi-source data on the timeline. Adaptive filtering algorithms (such as Kalman filtering or wavelet denoising) are used to remove sensor noise and environmental interference signals. Three-dimensional spatial interpolation algorithms, such as Kriging interpolation or inverse distance weighted interpolation, are then used to transform discrete monitoring data into continuous three-dimensional situation maps of pollution concentration, temperature and humidity, and airflow velocity. These maps can be dynamically overlaid with geographic information data to present pollution diffusion trends and environmental parameter distributions in a three-dimensional visualization.

[0066] The distributed sensor array configuration overcomes the limitations of single-point monitoring, achieving full coverage of the monitoring area. This allows the capture of localized contamination hotspots that are often missed by traditional single-point detection. For example, even small concentration changes near the volatilization point of a reagent in a laboratory can be accurately captured.

[0067] Spatiotemporal alignment and noise suppression technologies ensure the spatiotemporal consistency and reliability of multi-source heterogeneous data, providing high-quality input data for subsequent pollutant diffusion models and zeolite adsorption kinetic models, thereby avoiding control strategy deviations caused by data errors. Furthermore, the three-dimensional pollution situation map transforms abstract environmental data into intuitive visual information, allowing operators to quickly grasp the overall distribution of pollution and analyze pollution evolution patterns through historical data. For example, in the event of a sudden leak in a chemical park, the map can display the pollutant diffusion path in real time, providing a basis for decision-making for emergency command.

[0068] The dynamic data foundation constructed in this step lays the foundation for the intelligent linkage control of the entire system. Subsequent priority calculations, airflow adjustment, regeneration strategies and other links all rely on the precise data support provided by this step, thereby achieving closed-loop optimization from data collection to control execution, and improving the system's responsiveness to complex environmental changes and purification efficiency.

[0069] like Figure 2 As shown, the spatiotemporal alignment and noise suppression of multi-source data to construct a three-dimensional pollution situation map specifically includes:

[0070] S110 collects environmental parameters in real time, including: collecting volatile organic compound concentrations, monitoring particle distribution with a laser scattering particle counter, acquiring temperature and humidity data with a digital temperature and humidity sensor, measuring airflow velocity with a hot-wire anemometer, and monitoring the adsorption saturation and structural stress of the zeolite module;

[0071] S120, performs spatiotemporal alignment and noise suppression on multi-source heterogeneous data to limit time synchronization accuracy;

[0072] S130, constructing a three-dimensional situation map of the pollution concentration field, the temperature and humidity field, and the airflow velocity field using a three-dimensional spatial interpolation algorithm.

[0073] S200, analyzing the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, calculating the pollutant treatment priority, and generating a purification control instruction queue including the priority;

[0074] When analyzing the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, the actual mapping of the model can be deepened by combining it with specific application scenarios. For example, in a chemical park scenario, for leaked benzene pollutants, the pollutant diffusion model can be used to calculate the concentration change rate in three-dimensional space in real time. Combined with real-time wind speed and direction data in the park (such as northwest wind 3m / s), the pollutant diffusion path to the southeast and the time when the concentration peak will arrive can be predicted;

[0075] At the same time, based on the zeolite adsorption kinetics model, the real-time adsorption changes of benzene series by the zeolite module in the area are monitored. When the zeolite adsorption saturation in a certain area is found to be close to 80%, the model can warn that the remaining adsorption capacity is insufficient and regeneration or switching is required as a priority. In the dynamic priority calculation, the edge computing engine can adaptively adjust the weight coefficient according to the pollution characteristics of different time periods. For example, during the morning rush hour traffic pollution period, the weight of the pollutant concentration gradient and toxicity coefficient is increased to quickly locate pollution hotspots such as road intersections.

[0076] During industrial nighttime production, the weighting of zeolite adsorption saturation is increased to prevent a sudden drop in purification efficiency due to module overload. The generated purification control command queue can be deeply integrated with time series. For example, during high summer temperatures, pre-set priority rules for processing highly volatile pollutants are used. When temperature and humidity field data trigger thresholds, the execution priority of related commands is automatically increased, forming a real-time "data-model-decision" linkage.

[0077] This step establishes a real-time mapping mechanism between pollutant migration and purification resources through dynamic coupling analysis. For example, during the spread of urban smog, the system can simultaneously track changes in PM2.5 concentration fields and the adsorption status of zeolite modules within the area, avoiding the resource mismatch problem of "overloaded modules not being treated while underloaded modules are idling" under traditional fixed strategies.

[0078] The priority calculation framework integrates multiple parameters, including toxicity, diffusion rate, and module status, to achieve intelligent classification of purification strategies. For example, for formaldehyde pollution in sensitive areas such as hospital operating rooms, due to its extremely high toxicity coefficient (T), the system will forcibly raise its treatment priority to the highest level to ensure instant response.

[0079] Time series management of the command queue gives the system forward-looking control capabilities. For example, before a rainstorm is predicted, the system can adjust the airflow distribution of the outdoor zeolite unit in advance to prioritize the treatment of accumulated industrial waste gas and avoid secondary diffusion of pollutants before precipitation.

[0080] The localized processing characteristics of the edge computing engine significantly reduce data transmission delays. In the event of a sudden chemical leak, the entire process from data collection to priority calculation can be completed in milliseconds, which is more than 90% faster than the response speed of cloud-based processing solutions, thus creating a critical time window for emergency response.

[0081] In addition, this mechanism can continuously optimize control strategies through model iteration. For example, in an industrial park application, after three months of data training, the system's efficiency in handling sudden pollution increased by 45%, and the average regeneration cycle of the zeolite module was extended by 20%, significantly improving the system's intelligence level and operational efficiency.

[0082] like Figure 3As shown, the analysis of the dynamic coupling relationship between the pollutant diffusion characteristics and the zeolite adsorption capacity calculates the pollutant treatment priority and generates a purification control instruction queue containing the priority, specifically including:

[0083] S210, establish a pollutant diffusion model and calculate the rate of change of any pollutant concentration in the space over time

[0084]

[0085] Where C is the pollutant concentration, D is the diffusion coefficient, v is the air velocity vector, and S is the source term;

[0086] S220, construct a zeolite adsorption kinetics model and calculate the rate of change of zeolite adsorption over time

[0087]

[0088] Where q is the adsorption amount, k f is the adsorption rate constant, k d is the desorption rate constant, C * is the equilibrium concentration;

[0089] S230, performs dynamic priority calculation based on the edge computing engine to obtain the pollutant treatment priority:

[0090]

[0091] Among them, P is the treatment priority, T is the toxicity coefficient, is the normalized value of the pollutant concentration gradient, q is the current adsorption amount, q max is the maximum adsorption capacity, is the current adsorption saturation of zeolite, ɑ, β, γ, and δ are weight coefficients;

[0092] S240 , generating a purge control instruction queue including a level priority based on the processing priority.

[0093] S300: Based on the purification control instructions, the system sends instructions to the distributed zeolite purification unit to dynamically adjust the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module;

[0094] When dynamically adjusting the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module, refined control can be achieved by combining the real-time pollution situation with the status of the zeolite unit. For example, in a smart building scenario, when the three-dimensional pollution situation map shows that the concentration of volatile organic compounds (VOCs) in the corridor of a certain floor exceeds the standard, the system sends a command to the fan matrix in that area through the control command distribution module, increasing the fan power in the corresponding area by 30%. At the same time, it adjusts the airflow guidance angle of the fans in adjacent areas to form a local negative pressure environment to prevent pollutants from spreading to other floors.

[0095] At the same time, based on the molecular sieve switching decision tree model, when the main component of the pollutant is detected to be formaldehyde, the amino-loaded molecular sieve module in the zeolite unit in this area is automatically activated, and its specific adsorption efficiency for formaldehyde is increased by more than 50% compared with the general module.

[0096] In the industrial waste gas treatment scenario, when multi-source data show that the concentration of nitrogen oxides emitted by a production line suddenly increases, the system can complete the command analysis within 0.5 seconds, synchronously adjust the fan matrix of the zeolite unit corresponding to the production line, and adjust the airflow distribution ratio from the default 1:1:1 to 3:1:0 (main processing unit: backup unit: regeneration unit), and trigger the thermal activation program of the high-efficiency denitrification molecular sieve module, so that the nitrogen oxide removal rate is restored to more than 95% within 10 minutes.

[0097] The time series control mechanism can further optimize the system's operating rhythm. For example, during the city's morning rush hour (7:00-9:00), the fan matrix of the zeolite units around the main traffic arteries is pre-set to be upgraded to high-speed operation mode, and the molecular sieve module combination for motor vehicle exhaust is activated in advance to achieve targeted and efficient treatment of nitrogen oxides and hydrocarbons.

[0098] This step uses a dynamic airflow adjustment mechanism to reconstruct the airflow path in real time based on pollution distribution, avoiding the dual drawbacks of "undertreatment" in high-pollution areas and "excessive energy consumption" in low-pollution areas under the traditional fixed wind speed mode. For example, in a certain warehousing and logistics park application, dynamic airflow allocation based on the cargo storage type of different warehouses (such as chemical warehouses and general cargo warehouses) reduced overall fan energy consumption by 28% while improving purification efficiency in high-risk areas by 40%.

[0099] The intelligent molecular sieve switching strategy achieves a precise match between pollutants and purification resources. For example, in a hospital operating room scenario, when an anesthetic gas leak is detected, the system can complete the switch from the general module to the halogenated hydrocarbon-specific molecular sieve module within 2 seconds, increasing the adsorption rate of anesthetic gas by 3 times and ensuring that the indoor air quality always meets medical standards. The deep integration of time series and instruction queues gives the system proactive predictive capabilities. For example, before periodic production fluctuations in chemical companies (such as equipment start-up and shutdown periods), the system automatically adjusts the module activation sequence and fan power curve of the corresponding zeolite unit, shortening the response time for sudden pollution from the traditional manual intervention mode of 15 minutes to within 3 minutes through real-time dynamic adjustment.

[0100] The distributed command issuance mechanism ensures the consistency of multi-node collaboration. In a large data center scenario covering 5,000 square meters, the system can simultaneously control the fan matrix and module status of 2,000+ zeolite units, with the execution delay error of each node command less than 50 milliseconds, ensuring the uniformity and stability of the purification effect across the entire area. In addition, this step enables the zeolite purification system to maintain efficient operation under different loads and different pollutant types through the three-dimensional coordinated control of "airflow-module-time". Actual measured data shows that compared with the fixed control strategy, the dynamic adjustment mechanism can increase the average adsorption capacity utilization of the zeolite module by 35%, reduce the regeneration frequency by 22%, and optimize the overall energy consumption by 18%-25%.

[0101] like Figure 4 As shown, the instructions are sent to the distributed zeolite purification unit to dynamically adjust the air flow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module, specifically including:

[0102] S310, reading all purification control instructions and corresponding time series, issuing control instructions to the distributed zeolite purification units, and dynamically adjusting the airflow distribution ratio of the fan matrix;

[0103] S320, establishing a molecular sieve switching decision tree model, and intelligently switching the molecular sieve module based on the pollutant type and concentration.

[0104] S400, when a sudden pollution event is detected, builds a distributed consensus network to coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy;

[0105] When responding to sudden pollution incidents, the system achieves intelligent upgrades in emergency response through multi-level early warning and distributed coordination mechanisms. For example, when a chemical leak occurs on a city's main road, if sensors on a certain road section detect benzene concentrations exceeding a preset warning threshold (e.g., 50 ppm), a three-level early warning mechanism is activated: a level one warning (warning threshold - 120%) triggers a local data review; a level two warning (120%-150%) activates the pre-start state of zeolite units within 500 meters; and a level three warning (>150%) immediately establishes a distributed consensus network based on the Byzantine Fault Tolerance (BFT) algorithm, ensuring state synchronization among 50+ nodes within 200 milliseconds.

[0106] At this point, the system dynamically dispatches peripheral nodes to form a circular airflow barrier based on the direction of pollution diffusion. For example, three layers of zeolite units are deployed upwind of the leak point. Through fan matrix adjustment, a reverse airflow wall with a speed of 8m / s is formed. In conjunction with the directional adsorption module, the diffusion rate of benzene is reduced by more than 70%.

[0107] At the same time, the gradient regeneration strategy is triggered - for core area modules with adsorption saturation greater than 70%, solution chemical regeneration (such as sodium hydroxide solution elution) is preferentially used to restore 85% of the adsorption capacity within 45 minutes; low-temperature thermal regeneration (60°C hot air circulation) is used for edge area modules, and the regeneration cycle is completed within 2 hours to avoid the system processing capacity gap caused by a single regeneration method.

[0108] In unmanned scenarios at night in industrial parks, this mechanism can automatically identify sudden VOCs explosion incidents and complete the entire process from early warning to barrier establishment within 10 seconds, shortening the handling time by 90% compared to the traditional manual response mode.

[0109] The significant advantages of this step are: the distributed consensus network breaks the single-point bottleneck of traditional centralized control, achieving state consistency among thousands of nodes with a network latency of less than 50ms. Even if communication with some nodes is interrupted, the remaining nodes can still maintain the integrity of the barrier through the consensus algorithm. For example, in a test at a chemical park, when 30% of the nodes were deliberately disconnected, the system was still able to reconstruct the network within 1 second and maintain the effectiveness of the barrier.

[0110] The dynamic construction capability of the directional adsorption barrier enables the system to function as a "pollution fence," controlling sudden contamination within a 50-meter diameter range, reducing the affected area by 80% compared to traditional diffusion control solutions. The gradient regeneration strategy intelligently matches the regeneration method with the degree of contamination, avoiding the waste of resources caused by "one-size-fits-all" regeneration. For example, in the event of a radon gas leak in a hospital's radiology department, chemical regeneration was prioritized for modules in highly radioactively contaminated areas. This reduced regeneration energy consumption by 60% compared to full thermal regeneration, while also shortening module recovery time from 4 hours to 1.5 hours.

[0111] The combination of the three-level warning mechanism and the time series gives the system a flexible response capability, and can dynamically adjust the disposal efforts according to the pollution evolution trend. In a container leakage incident at a certain port, the system gradually increased the warning level and invested purification resources of different scales in stages, reducing overall energy consumption by 35% compared to one-time full-load operation, while ensuring zero pollution spread.

[0112] In addition, through continuous learning of emergency response data (such as pollution type-barrier structure mapping model), this mechanism can improve disposal efficiency by more than 50% in repeated pollution scenarios, significantly enhancing the system's environmental adaptability and emergency reliability.

[0113] like Figure 5 As shown, the construction of a distributed consensus network coordinates multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy, specifically including:

[0114] S410, setting a pollutant concentration warning threshold. When a single sensor detects that the pollutant concentration exceeds the warning threshold, a three-level warning mechanism is triggered;

[0115] S420, constructing a distributed consensus network, while controlling node state synchronization, coordinating multi-node zeolite units to establish a directional adsorption barrier, where the peripheral nodes act as annular airflow barriers;

[0116] S430, triggering a gradient regeneration strategy to restore the adsorption capacity of the zeolite purification system through thermal regeneration and solution chemical regeneration, respectively.

[0117] S500, combined with the zeolite performance attenuation curve fed back by the dielectric sensor, dynamically adjusts the regeneration cycle and energy consumption ratio to establish a closed-loop control mechanism.

[0118] The closed-loop control mechanism enables intelligent management of the zeolite purification system throughout its entire life cycle through real-time feedback from dielectric sensors and dynamic model iteration. Dielectric sensors can monitor subtle changes in the zeolite's dielectric constant at high frequencies (e.g., 10 times per second). This parameter is strongly correlated with the degree of damage to the zeolite's internal microporous structure and the amount of pollutant attachment. For example, when a zeolite module reaches saturation with toluene adsorption, its dielectric constant increases by 25%-30% compared to its initial value. The system calculates the remaining adsorption capacity in real time using an attenuation model. When q(t) is detected to be less than 30% of the maximum capacity, the regeneration pre-assessment process is automatically triggered.

[0119] The dynamic regeneration cycle adjustment mechanism can shorten the regeneration cycle from the default 12 hours to 8 hours through a formula based on the real-time attenuation coefficient λ (for example, in a high-humidity environment of industrial waste gas, λ may increase from 0.01 / h to 0.03 / h), while optimizing the regeneration energy consumption ratio: during the low-load period at night, valley electricity is used to increase the regeneration temperature by 10°C, shortening the regeneration time by 30%; and during the peak period during the day, the "low-power preheating + pulsed regeneration" mode is adopted to reduce energy consumption by 15%.

[0120] The deep reinforcement learning algorithm continuously analyzes historical data (such as the regeneration cycle-λ value-energy consumption data pairs of the past 30 days) and automatically optimizes the adjustment coefficient and energy consumption coefficient. In a semiconductor clean room application, after training, the system controls the regeneration cycle prediction error within ±15 minutes, and the combined energy consumption of fans and regeneration is reduced by 18%.

[0121] The dielectric sensor establishes a real-time closed loop of "performance monitoring-attenuation prediction-regeneration decision-making," reducing ineffective regeneration operations by over 40% compared to traditional timed regeneration modes. For example, in laboratory scenarios, for intermittently used formaldehyde purification modules, the system can dynamically adjust the regeneration cycle based on the actual adsorption load, extending the module life by 35%.

[0122] The dynamic energy consumption ratio mechanism achieves precise control of "supply on demand." In the odor treatment system of a food processing plant, by real-time matching of fan power and regeneration temperature, the energy consumption per ton of waste gas treated was reduced from 0.8kWh to 0.5kWh, saving over 100,000 yuan in electricity bills annually.

[0123] The introduction of deep reinforcement learning gives the system the ability to evolve autonomously. In a cross-seasonal environment (such as switching from high temperature and high humidity in summer to low temperature and dryness in winter), the algorithm can automatically identify the correlation between environmental parameters and λ, adjust the regeneration strategy 7 days in advance, and reduce the module performance fluctuation by 50%; the full process digitization of closed-loop control (from sensor signals to control instructions are transmitted via digital quantities) avoids the lag and error of traditional analog control, and 2 In commercial complex applications, the system controls the standard deviation of the adsorption saturation of zeolite modules in each area within 5%, ensuring a balanced and stable purification effect across the entire area.

[0124] In addition, this mechanism achieves Pareto optimality in purification performance and energy efficiency through quantitative analysis. According to third-party testing, systems using this solution reduce energy consumption by 15%-22% and regeneration costs by 25% compared to traditional solutions at the same purification efficiency, significantly improving the economy and sustainability of the system.

[0125] like Figure 6 As shown, the zeolite performance attenuation curve fed back by the dielectric sensor is combined to dynamically adjust the regeneration cycle and energy consumption ratio to establish a closed-loop control mechanism, specifically including:

[0126] S510, establish a zeolite performance attenuation model and calculate the remaining adsorption capacity of the zeolite:

[0127] q(t)=q0·e -λt ;

[0128] Where q(t) is the adsorption capacity at time t, q0 is the initial adsorption capacity, and λ is the attenuation coefficient;

[0129] Dynamically adjust the regeneration cycle to obtain the optimal regeneration cycle at each moment:

[0130]

[0131] Among them, T reg is the regeneration period, T max is the maximum regeneration period, T0 is the minimum regeneration period, λ max is the maximum attenuation coefficient, k is the adjustment coefficient;

[0132] Quantify the energy consumption during system operation and optimize the ratio of fan power to regeneration temperature:

[0133] E total =E fan +E reg =ρ·Q·H+σ·m·C p ΔT;

[0134] Among them, E total is the total energy consumption, E fan is the fan energy consumption, E reg is the regeneration energy consumption, Q is the air volume, H is the wind pressure, m is the regeneration mass, C p is the specific heat capacity, ΔT is the temperature rise, ρ and σ are the energy consumption coefficients

[0135] A deep reinforcement learning algorithm is used to continuously optimize control parameters.

[0136] Figure 7 The block diagram of the intelligent linkage control system of the Internet of Things terminal and the zeolite purification system provided in the embodiment of the present invention is as follows: Figure 7 As shown, the system includes:

[0137] The parameter data acquisition module 100 is used to synchronously collect environmental parameters and zeolite module status indicators through the distributed sensor array carried by the IoT terminal, and perform spatiotemporal alignment and noise suppression on multi-source data to construct a three-dimensional pollution situation map;

[0138] The processing priority analysis module 200 is used to analyze the dynamic coupling relationship between the pollutant diffusion characteristics and the zeolite adsorption capacity, calculate the pollutant processing priority, and generate a purification control instruction queue containing the priority;

[0139] The control instruction issuing module 300 is used to issue instructions to the distributed zeolite purification unit based on the purification control instruction, dynamically adjusting the air flow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module;

[0140] The pollution event response module 400 is used to build a distributed consensus network, coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy when a sudden pollution event is detected;

[0141] The energy consumption optimization closed-loop control module 500 is used to dynamically adjust the regeneration cycle and energy consumption ratio based on the zeolite performance attenuation curve fed back by the dielectric sensor, thereby establishing a closed-loop control mechanism.

[0142] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent linkage control method for an Internet of Things terminal and a zeolite purification system, characterized in that: The method comprises: Through the distributed sensor array carried by the IoT terminal, environmental parameters and zeolite module status indicators are collected synchronously, and multi-source data are aligned in time and space and noise is suppressed to construct a three-dimensional pollution situation map; Analyze the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, calculate the pollutant treatment priority, and generate a purification control instruction queue containing the priority; Based on the purification control instructions, it sends instructions to the distributed zeolite purification unit to dynamically adjust the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module; When a sudden pollution event is detected, a distributed consensus network is constructed to coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy; Combined with the zeolite performance attenuation curve fed back by the dielectric sensor, the regeneration cycle and energy consumption ratio are dynamically adjusted to establish a closed-loop control mechanism.

2. The method according to claim 1, characterized in that The spatiotemporal alignment and noise suppression of multi-source data to construct a three-dimensional pollution situation map specifically includes: Real-time collection of environmental parameters, including: volatile organic compound concentration, laser scattering particle counter to monitor particle distribution, digital temperature and humidity sensor to obtain temperature and humidity data, hot wire anemometer to measure airflow velocity, and monitoring the adsorption saturation and structural stress of the zeolite module; Perform spatiotemporal alignment and noise suppression on multi-source heterogeneous data to limit time synchronization accuracy; A three-dimensional spatial interpolation algorithm is used to construct a three-dimensional situation map of pollution concentration field, temperature and humidity field, and airflow velocity field.

3. The method according to claim 2, characterized in that The analysis of the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity calculates the pollutant treatment priority and generates a purification control instruction queue containing the priority, specifically including: Establish a pollutant diffusion model to calculate the rate of change of any pollutant concentration in space over time Where C is the pollutant concentration, D is the diffusion coefficient, v is the air velocity vector, and S is the source term; Construct a zeolite adsorption kinetic model to calculate the rate of change of zeolite adsorption over time Where q is the adsorption amount, k f is the adsorption rate constant, k d is the desorption rate constant, C * is the equilibrium concentration; Perform dynamic priority calculation based on the edge computing engine to obtain the pollutant treatment priority: Among them, P is the treatment priority, T is the toxicity coefficient, is the normalized value of the pollutant concentration gradient, q is the current adsorption amount, q max is the maximum adsorption capacity, is the current adsorption saturation of zeolite, ɑ, β, γ, and δ are weight coefficients; Based on the processing priority, a purge control instruction queue including a level priority is generated.

4. The method according to claim 3, characterized in that The issuing of instructions to the distributed zeolite purification unit to dynamically adjust the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module specifically includes: Read all purification control instructions and corresponding time series, send control instructions to distributed zeolite purification units, and dynamically adjust the airflow distribution ratio of the fan matrix; A molecular sieve switching decision tree model was established, and the molecular sieve module was intelligently switched based on the pollutant type and concentration.

5. The method according to claim 4, characterized in that The construction of a distributed consensus network coordinates multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy, specifically including: Set pollutant concentration warning thresholds. When a single sensor detects that the pollutant concentration exceeds the warning threshold, a three-level warning mechanism is triggered; Build a distributed consensus network, synchronize node states, and coordinate multi-node zeolite units to establish a directional adsorption barrier, where peripheral nodes act as annular airflow barriers. A gradient regeneration strategy was triggered to restore the adsorption capacity of the zeolite purification system through thermal regeneration and solution chemical regeneration, respectively.

6. The method according to claim 4, characterized in that The zeolite performance attenuation curve fed back by the dielectric sensor is combined to dynamically adjust the regeneration cycle and energy consumption ratio to establish a closed-loop control mechanism, specifically including: A zeolite performance attenuation model was established to calculate the remaining adsorption capacity of zeolite: q(t)=q0·e -λt ; Where q(t) is the adsorption capacity at time t, q0 is the initial adsorption capacity, and λ is the attenuation coefficient; Dynamically adjust the regeneration cycle to obtain the optimal regeneration cycle at each moment: Among them, T reg is the regeneration period, T max is the maximum regeneration period, T0 is the minimum regeneration period, λ max is the maximum attenuation coefficient, k is the adjustment coefficient; Quantify the energy consumption during system operation and optimize the ratio of fan power to regeneration temperature: From total =E fan +E reg =ρ·Q·H+σ·m·C p ·ΔT; Among them, E total is the total energy consumption, E fan is the fan energy consumption, E reg is the regeneration energy consumption, Q is the air volume, H is the wind pressure, m is the regeneration mass, C p is the specific heat capacity, ΔT is the temperature rise, ρ and σ are the energy consumption coefficients A deep reinforcement learning algorithm is used to continuously optimize control parameters.

7. The intelligent linkage control system of the Internet of Things terminal and the zeolite purification system is characterized by: The system comprises: The parameter data acquisition module is used to synchronously collect environmental parameters and zeolite module status indicators through the distributed sensor array carried by the IoT terminal, and to perform spatiotemporal alignment and noise suppression on multi-source data to construct a three-dimensional pollution situation map; The treatment priority analysis module is used to analyze the dynamic coupling relationship between pollutant diffusion characteristics and zeolite adsorption capacity, calculate the pollutant treatment priority, and generate a purification control instruction queue containing the priority; A control instruction issuing module is used to issue instructions to the distributed zeolite purification unit based on the purification control instructions, dynamically adjusting the airflow distribution ratio of the fan matrix and the activation strategy of the molecular sieve module; The pollution event response module is used to build a distributed consensus network, coordinate multi-node zeolite units to establish a directional adsorption barrier and trigger a gradient regeneration strategy when a sudden pollution event is detected; The energy consumption optimization closed-loop control module is used to dynamically adjust the regeneration cycle and energy consumption ratio based on the zeolite performance attenuation curve fed back by the dielectric sensor, thus establishing a closed-loop control mechanism.

Images