Waste gas treatment method and system based on data acquisition control

Abstract

The invention provides a data acquisition control waste gas treatment method and system, and relates to the technical field of waste gas treatment.The method comprises the steps that multi-tower collaborative search is conducted on waste gas treatment process towers, and an initial waste gas treatment strategy is determined; performing multi-mode hidden danger tracing on the initial waste gas treatment strategy to obtain a waste gas treatment hidden danger graph network; performing multi-dimensional feed-forward adjustment combination on the initial waste gas treatment strategy according to the waste gas treatment hidden danger graph network to obtain a waste gas treatment adjustment space; and calling the waste gas treatment dynamic risk model, and performing multi-target linkage optimization under risk constraint on the waste gas treatment adjustment space in combination with a waste gas treatment twin tower to obtain a waste gas treatment parameter optimization result. The technical problems that in an existing waste gas treatment process, due to the fact that multiple tower treatment units lack a cooperative adjustment mechanism, the emission standard exceeding risk, the working condition fluctuation risk and the equipment operation hidden danger are likely to be caused are solved. And the technical effects of improving the emission stability, the operation safety and the waste gas treatment quality in the waste gas treatment process are achieved.

Description

A data acquisition and control method and system for treating exhaust gas Technical Field

[0001] This application relates to the field of waste gas treatment technology, and in particular to a waste gas treatment method and system with data acquisition and control. Background Technology

[0002] Industrial production processes generate waste gases containing pollutants such as volatile organic compounds and particulate matter. If these gases are emitted directly without effective treatment, they will cause air pollution. Therefore, waste gas treatment is a core aspect of industrial environmental protection. Existing waste gas treatment systems mostly operate on a fixed-parameter adaptation model, setting process parameters based on experience. This makes it difficult to dynamically adapt to real-time fluctuations in waste gas flow and pollutant concentrations during the treatment process, easily leading to poor treatment effects and inconsistent emission compliance. Furthermore, existing waste gas treatment systems lack the ability to anticipate potential equipment failures. Equipment performance degradation and localized malfunctions can easily trigger a chain reaction, causing a sharp drop in treatment efficiency. Delayed troubleshooting also increases the risk of exceeding emission standards.

[0003] In summary, current technologies suffer from a lack of coordinated adjustment mechanisms among multiple treatment units in waste gas treatment processes. This makes it impossible to make feedforward corrections to treatment strategies based on real-time waste gas data, which can easily lead to risks of exceeding emission standards, fluctuations in operating conditions, and potential equipment malfunctions. Summary of the Invention

[0004] This application provides a data acquisition and control-based waste gas treatment method and system, which solves the technical problem in existing waste gas treatment processes where the lack of a coordinated adjustment mechanism among multi-tower treatment units makes it impossible to make feedforward corrections to the treatment strategy based on real-time waste gas data. This easily leads to risks of exceeding emission standards, fluctuations in operating conditions, and potential equipment malfunctions. It achieves the technical effect of improving the emission stability, operational safety, and waste gas treatment quality of the waste gas treatment process by using data acquisition and twin modeling for feedforward risk perception and coordinated adjustment.

[0005] This application provides a data acquisition and control method for waste gas treatment. The method is applied to a data acquisition and control waste gas treatment system. The method includes: acquiring data on the waste gas to be treated in the waste gas treatment process tower to obtain initial waste gas characteristic data; performing multi-tower collaborative search on the waste gas treatment process tower based on the initial waste gas characteristic data to determine an initial waste gas treatment strategy; constructing a waste gas treatment twin tower and performing multimodal hazard tracing on the initial waste gas treatment strategy based on the waste gas treatment twin tower to obtain a waste gas treatment hazard network; performing multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy based on the waste gas treatment hazard network to obtain a waste gas treatment adjustment space; calling a waste gas treatment dynamic risk model and combining it with the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space to obtain waste gas treatment parameter optimization results; and controlling the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment based on the waste gas treatment parameter optimization results.

[0006] This application also provides a data acquisition and control system for waste gas treatment, comprising: a waste gas data acquisition module for acquiring data on the waste gas to be treated in the waste gas treatment process tower to obtain initial waste gas characteristic data; a multi-tower collaborative search module for performing multi-tower collaborative search on the waste gas treatment process tower based on the initial waste gas characteristic data to determine an initial waste gas treatment strategy; a waste gas treatment hazard tracing module for constructing a waste gas treatment twin tower and performing multi-modal hazard tracing on the initial waste gas treatment strategy based on the waste gas treatment twin tower to obtain a waste gas treatment hazard network; a waste gas treatment feedforward adjustment module for performing multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy based on the waste gas treatment hazard network to obtain a waste gas treatment adjustment space; and a multi-objective linkage optimization control module for calling a waste gas treatment dynamic risk model, combining the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space, obtaining waste gas treatment parameter optimization results, and controlling the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment based on the waste gas treatment parameter optimization results.

[0007] This application proposes a data acquisition and control method and system for waste gas treatment. It involves acquiring initial waste gas characteristic data from the waste gas treatment process towers; using this initial characteristic data, performing a multi-tower collaborative search of the waste gas treatment process towers to determine the initial waste gas treatment strategy; using a waste gas treatment twin tower, tracing multimodal potential hazards of the initial waste gas treatment strategy to obtain a potential hazard network; using this hazard network, performing multi-dimensional feedforward adjustment combinations of the initial waste gas treatment strategy to obtain the waste gas treatment adjustment space; and using a dynamic risk model and the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space to obtain the waste gas treatment parameter optimization results. Based on these optimization results, the system controls the waste gas treatment process towers to perform multi-tower collaborative waste gas treatment. This solves the technical problem in existing waste gas treatment processes where the lack of a collaborative adjustment mechanism between multiple treatment units and the inability to perform feedforward correction of the treatment strategy based on real-time waste gas data easily leads to risks of exceeding emission standards, fluctuations in operating conditions, and potential equipment malfunctions. The technology achieves the effect of feedforward risk perception and coordinated regulation of the waste gas treatment process based on data acquisition and twin modeling, thereby improving the emission stability, operational safety and waste gas treatment quality of the waste gas treatment process. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0009] Figure 1 is a schematic flowchart of a data acquisition and control method for treating exhaust gas provided in this application.

[0010] Figure 2 is a schematic diagram of a data acquisition and control exhaust gas treatment system provided in this application.

[0011] Figure labeling: 1. Exhaust gas data acquisition module; 2. Multi-tower collaborative search module; 3. Exhaust gas treatment hidden danger tracing module; 4. Exhaust gas treatment feedforward adjustment module; 5. Multi-objective linkage optimization control module. Detailed Implementation

[0012] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used herein is for the purpose of describing this application only.

[0015] Example 1: This application provides a data acquisition and control method for waste gas treatment. The method is applied to a data acquisition and control waste gas treatment system, as shown in Figure 1. The method includes: Step A100: Data acquisition of the waste gas to be treated in the waste gas treatment process tower to obtain initial waste gas characteristic data. The waste gas treatment process tower includes a pretreatment tower, a main treatment tower, and a post-treatment tower. Before the waste gas enters the waste gas treatment process tower, the waste gas is monitored in real time through an Internet of Things (IoT) array to obtain initial waste gas characteristic data. The waste gas to be treated refers to the raw waste gas that has not yet entered the waste gas treatment process tower for purification. The IoT array includes multiple gas flow sensors, multiple gas concentration sensors, and multiple temperature sensors, etc. The initial waste gas characteristic data includes parameters such as the waste gas flow rate, pollutant concentration, and waste gas temperature corresponding to the waste gas to be treated. In a specific embodiment, the acquired initial waste gas characteristic data includes a waste gas flow rate of 12000 m³ / h. 3 / h, VOCs concentration is 380mg / m³ 3 Data such as exhaust gas temperature of 45℃ are collected. Initial exhaust gas characteristic data allows for accurate understanding of exhaust gas load levels and trends, providing a reliable data foundation for subsequent multi-tower collaborative search, exhaust gas treatment strategy determination, and twin-tower modeling. This avoids the lag in treatment strategy adjustments caused by monitoring only at the emission end, improving the timeliness of the exhaust gas treatment process's response to operating condition fluctuations and its overall operational stability.

[0016] A waste gas treatment process tower refers to multiple tower-type treatment units that sequentially perform different treatment functions in the same waste gas treatment process. These multiple tower-type treatment units include a pretreatment tower, a main treatment tower, and a post-treatment tower. The pretreatment tower has a vertical tower structure with waste gas inlets at its bottom or side. An inlet distribution device is installed inside the inlet to diffuse the waste gas entering the tower. Inside the tower, flow equalization plates or guide components are sequentially installed from bottom to top to create a uniform airflow within the tower's cross-section. A spray assembly or coarse filter assembly can be installed above the flow equalization structure to preliminarily regulate large particles, soluble components, or temperature and humidity in the waste gas. A waste gas outlet is located at the top of the tower, connecting to the main treatment tower. The main treatment tower has an exhaust gas inlet at its bottom, sourced from the pretreatment tower, with a gas distribution device at the inlet. Inside the tower, at least one core treatment assembly is installed along its height. This core assembly is designed to fully contact the exhaust gas to remove major pollutants. A support structure is located below the core assembly, and a limiting or clamping structure is installed above it to ensure assembly stability. Inspection ports are provided on the tower's side walls for maintenance and replacement of the treatment medium. An exhaust gas outlet is located at the top of the tower, connecting to the post-treatment tower. The post-treatment tower has an exhaust gas inlet at its bottom, sourced from the main treatment tower. A buffer space is provided inside the tower to ensure a stable flow pattern for the exhaust gas before it enters the fine treatment area. A fine treatment assembly is installed above the buffer space to compensate for residual pollutants. A demisting or flow stabilizing structure is installed above the fine treatment assembly to reduce liquid carryover and airflow fluctuations. An exhaust port is located at the top of the tower for connection to an emission pipeline.

[0017] Step A200: Perform a multi-tower collaborative search on the waste gas treatment process towers based on the initial waste gas characteristic data to determine the initial waste gas treatment strategy. Step A200 further includes steps A210 to A240. Step A210: Load the pretreatment record block, main treatment record block, and post-treatment record block corresponding to the waste gas treatment process towers; Step A220: Perform a similarity evaluation search on the pretreatment record block based on the initial waste gas characteristic data to obtain a waste gas pretreatment matching scheme; Step A230: Perform a similarity evaluation search on the main treatment record block based on the initial waste gas characteristic data to obtain a waste gas main treatment matching scheme; Step A240: Perform a similarity evaluation search on the post-treatment record block based on the initial waste gas characteristic data to obtain a waste gas post-treatment matching scheme, and combine the waste gas pretreatment matching scheme and the waste gas main treatment matching scheme to generate the initial waste gas treatment strategy.

[0018] Specifically, the historical operation record blocks corresponding to the pretreatment tower, main treatment tower, and post-treatment tower are synchronously loaded from the edge computing storage module of the exhaust gas treatment system to obtain the pretreatment record block, main treatment record block, and post-treatment record block. The pretreatment record block includes multiple historical pretreatment schemes corresponding to multiple historical initial exhaust gas samples. Each historical initial exhaust gas sample includes historical parameters such as historical exhaust gas flow rate, historical pollutant concentration, and historical exhaust gas temperature. The main treatment record block includes the historical main treatment scheme corresponding to each historical initial exhaust gas sample. The post-treatment record block includes the historical post-treatment scheme corresponding to each historical initial exhaust gas sample. Each record block is stored using a partitioned index, with index keywords including exhaust gas flow rate, pollutant concentration, and exhaust gas temperature. By loading the historical record blocks corresponding to the pretreatment tower, main treatment tower, and post-treatment tower, complete historical data support is provided for subsequent tower similarity searches, avoiding matching scheme deviations caused by missing historical data. At the same time, the partitioned index design shortens data retrieval time, ensures the real-time performance of multi-tower collaborative searches, and meets the timely response requirements of the exhaust gas treatment process to operating conditions.

[0019] Using the initial exhaust gas feature data collected in step A100 as the retrieval benchmark, a cosine similarity algorithm is employed to calculate the similarity of multiple historical initial exhaust gas samples in the preprocessing record block. Specifically, the three core parameters of the initial exhaust gas feature data—exhaust gas flow rate, VOCs concentration, and exhaust gas temperature—are used to create vector models with the historical exhaust gas flow rate, historical VOCs concentration, and historical exhaust gas temperature parameters within the historical initial exhaust gas samples. The cosine of the angle between the feature vectors of the two models is calculated using the cosine similarity formula to obtain multiple exhaust gas feature similarities. Then, the historical pretreatment scheme corresponding to the highest exhaust gas feature similarity is selected as the exhaust gas pretreatment matching scheme. The exhaust gas pretreatment matching scheme includes control parameters for the spray assembly, coarse filter assembly, and flow equalization plate. For example, the exhaust gas pretreatment matching scheme includes starting two sets of high-pressure spray heads in the spray assembly, with a spray liquid (alkaline absorption liquid) flow rate of 8 m³ / s. 3 The coarse filtration component uses a non-woven fabric filter layer (5μm pore size), and the flow equalization plate is opened to the maximum flow angle of 30°. Through precise similarity search matching, the exhaust gas pretreatment matching scheme is highly adapted to the exhaust gas to be treated, which can remove more than 80% of large particulate matter (particle size ≥5μm) and more than 30% of soluble VOCs in the exhaust gas in advance. At the same time, the exhaust gas temperature is adjusted to 38-42℃ to avoid large particulate matter clogging the core components of the main treatment tower and high temperature damage to the treatment medium.

[0020] Similarly, using the same cosine similarity algorithm as in step A220, multiple historical main treatment schemes in the main treatment record block are matched and searched based on the initial exhaust gas characteristic data. The historical main treatment scheme corresponding to the largest exhaust gas characteristic similarity is selected as the exhaust gas main treatment matching scheme. The exhaust gas main treatment matching scheme includes data such as activated carbon adsorption control parameters and tower internal pressure control parameters. For example, the exhaust gas main treatment matching scheme includes the core treatment component opening three layers of activated carbon adsorption layers (each layer is 80cm thick, and the activated carbon particle size is 2-3mm), the gas distribution device adopts a ring-shaped gas distribution structure, the gas distribution uniformity is ≥90%, the activated carbon adsorption space velocity is controlled at 0.8m / s, and the tower internal pressure is maintained at 0.12MPa. Then, using the same cosine similarity algorithm as in step A220, a similarity search is performed on the post-treatment record block, and the historical post-treatment scheme corresponding to the largest exhaust gas characteristic similarity is selected as the exhaust gas post-treatment matching scheme. The exhaust gas post-treatment matching scheme includes data such as buffer space residence time, photocatalytic oxidation parameters, and exhaust port pressure parameters. For example, the exhaust gas after-treatment matching scheme includes controlling the residence time in the buffer space of the after-treatment tower to 15s to ensure the stability of the exhaust gas flow, using a photocatalytic oxidation module (power 3000W, reaction temperature 35-40℃) for the fine treatment component, using a baffle demister (demister efficiency ≥98%) for the demisting structure, adjusting the exhaust port pressure to 0.1MPa, and simultaneously activating the real-time exhaust gas monitoring module, with monitoring data uploaded to the system control center in real time.

[0021] The waste gas pretreatment matching scheme, waste gas main treatment matching scheme, and waste gas post-treatment matching scheme are output as the initial waste gas treatment strategy. Multi-tower collaborative search is performed on the waste gas treatment process towers using the initial waste gas characteristic data, effectively improving the treatment efficiency of each step in the waste gas treatment process (pretreatment, main treatment, and post-treatment). This achieves seamless connection and parameter linkage between the three tower waste gas treatment processes, solving the disconnect problem of independent operation of multiple towers.

[0022] Step A300: Construct a waste gas treatment twin tower, and perform multimodal hazard tracing of the initial waste gas treatment strategy based on the waste gas treatment twin tower to obtain a waste gas treatment hazard map network. It should be noted that this application uses 3D laser scanning technology to perform a comprehensive scan of the waste gas treatment process tower, obtaining the actual dimensions, structural layout, component assembly relationships, and surface geometric features of the three towers. The initial waste gas characteristic data and the point cloud data obtained from the scan are imported into the Unity3D industrial version digital twin modeling platform. Combined with the design drawings of the physical tower and component specifications, the waste gas treatment twin tower is dynamically modeled to obtain the waste gas treatment twin tower. The waste gas treatment twin tower is a 1:1 digital mirror model constructed based on the waste gas treatment process tower (pretreatment tower, main treatment tower, and post-treatment tower) as a physical prototype, relying on digital twin technology and combining the initial waste gas characteristic data. The exhaust gas treatment twin tower realizes a full-dimensional digital mapping of the physical tower body from its geometric features and operating status to the entire exhaust gas treatment process. This solves the problem that the operating status of the physical tower body is difficult to monitor intuitively in traditional exhaust gas treatment systems, greatly improving the visualization and reliability of exhaust gas treatment. At the same time, the exhaust gas treatment twin tower eliminates the need for repeated trial and error experiments on the physical tower body, effectively reducing the trial and error costs and operational risks of exhaust gas treatment.

[0023] In one possible implementation, step A300 further includes steps A310, A320, A330, and A340. Step A310: Based on the exhaust gas treatment twin tower, trace the emission compliance risks of the initial exhaust gas treatment strategy to obtain an emission compliance risk evolution chain. Specifically, step A310 includes: step A311, performing virtual control of the exhaust gas treatment twin tower according to the initial exhaust gas treatment strategy to obtain virtual exhaust gas treatment data; step A312, identifying emission compliance risks based on the virtual exhaust gas treatment data to obtain emission compliance risk characteristics; step A313, performing multi-parameter correlation analysis on the virtual exhaust gas treatment data based on the emission compliance risk characteristics to obtain an emission risk parameter relationship network; and step A314, performing causal tracing of the emission compliance risk characteristics based on the emission risk parameter relationship network to generate the emission compliance risk evolution chain.

[0024] The initial waste gas treatment strategy is transmitted to the simulation control module of the waste gas treatment twin tower via industrial Ethernet, enabling virtual control of the entire process of the waste gas treatment twin tower without physical losses. The control logic is completely synchronized with the waste gas treatment process tower, ensuring that the virtual control process closely matches the actual operating scenario. During the implementation of virtual control, the initial waste gas characteristic data collected in step A100 is used as the basic input condition. At the same time, various operating condition fluctuation scenarios that may occur during the waste gas treatment process under the initial waste gas treatment strategy are simulated, covering waste gas flow fluctuations, VOCs concentration fluctuations, waste gas temperature fluctuations, etc., comprehensively replicating the operating condition changes in actual waste gas treatment. The entire process of the waste gas to be treated in the pretreatment tower, main treatment tower, and post-treatment tower is simulated in real time, and complete virtual waste gas treatment data is output synchronously. The virtual waste gas treatment data covers three main categories of core information: First, waste gas parameters after each treatment stage, including waste gas temperature, VOCs concentration, and large particulate matter removal rate after pretreatment; VOCs concentration and adsorption efficiency after main treatment; and VOCs concentration and exhaust pressure after posttreatment. Second, virtual operating parameters of each tower component, including spray liquid flow rate in the pretreatment tower, activated carbon adsorption space velocity in the main treatment tower, and photocatalytic power in the posttreatment tower. Third, auxiliary reference data, including pollutant degradation rate, component wear status, and treatment time for each stage.

[0025] Furthermore, based on the exhaust gas emission standards, a comprehensive analysis of the virtual exhaust gas treatment data is conducted using a "threshold comparison + anomaly detection" approach. This focuses on screening for anomalies in post-treatment exhaust parameters and key intermediate parameters, while simultaneously linking the operational status data of the exhaust gas treatment twin towers to identify potential emission compliance issues. For example, the exhaust gas emission standards include a VOCs concentration of ≤2mg / m³ in the final exhaust gas. 3 The exhaust gas flow rate is stable at 10,000-14,000 m³ / h 3 The simulated exhaust gas flow rate is 35-45℃, with no significant pollutant residue. When identifying potential emissions compliance risks using virtual exhaust gas treatment data, priority is given to screening core emission parameters such as post-treatment VOCs concentration, exhaust pressure, and demisting efficiency. Simultaneously, intermediate parameters from the main treatment and pretreatment stages are correlated to avoid overlooking risks due to relying on a single parameter. Characteristics of potential emissions compliance risks include the type of risk, severity, scope of impact, and manifestations. For example, it was identified that when the simulated exhaust gas flow rate fluctuates to 14000 m³ / h... 3 / h, VOCs concentration fluctuated to 420mg / m³ 3 Under these operating conditions, the virtual exhaust gas treatment data showed that the post-treatment VOCs concentration temporarily increased to 2.5 mg / m³. 3 (Exceeding the standard by 0.5 mg / m³) 3An abnormality in exhaust pressure dropping to 0.095 MPa (below the benchmark threshold) was identified as a potential emission compliance hazard. Further analysis was conducted to determine the characteristics of this hazard. Specifically, the characteristics included: hazard type: VOCs emission exceeding the standard (minor exceedance); hazard severity: Level 1 warning (no actual emission exceedance, only potential risk); hazard impact range: photocatalytic module of the aftertreatment tower, adsorption component of the main treatment tower, and the three-tower linkage control system; hazard manifestations: continuous increase in VOCs concentration after aftertreatment, and failure to adjust photocatalytic power in a timely manner.

[0026] Furthermore, focusing on the emission compliance hazard characteristics identified in step A312, this study concentrates on virtual waste gas treatment data related to these hazards, filtering out parameters with strong correlations. These parameters include initial waste gas characteristic parameters (flow rate, VOCs concentration, etc.), tower operating parameters (spray liquid flow rate, adsorption space velocity, photocatalytic power, tower pressure, etc.), intermediate treatment parameters (VOCs concentration after pretreatment, VOCs concentration after main treatment, etc.), and component operating status parameters (activated carbon adsorption saturation, photocatalytic module operating power, spray component atomization effect, etc.). The Pearson correlation analysis algorithm is used to calculate the correlation strength between each parameter (correlation coefficients range from -1 to 1; the closer the absolute value is to 1, the stronger the correlation), obtaining the emission hazard parameter relationship network. This network includes information on the positive and negative correlations and correlation strengths between the emission compliance hazard characteristics and the parameters within the virtual waste gas treatment data, providing a clear parameter correlation basis for subsequent causal tracing and avoiding logical gaps during the tracing process. For example, the parameter relationship network of emission hazards includes: an increase in exhaust gas flow rate (correlation coefficient 0.89), an increase in initial VOCs concentration (correlation coefficient 0.85), and an increase in VOCs concentration after pretreatment are strongly positively correlated; an increase in VOCs concentration after pretreatment (correlation coefficient 0.91) and an increase in adsorption load in the main treatment tower are strongly positively correlated; an increase in adsorption load (correlation coefficient 0.87) and a failure to adjust the activated carbon adsorption space velocity in time (correlation coefficient -0.78) jointly lead to an increase in VOCs concentration after main treatment; an increase in VOCs concentration after main treatment (correlation coefficient 0.93) and a failure to increase the photocatalytic power of the post-treatment tower in time (correlation coefficient -0.82) are strongly negatively correlated, ultimately leading to an excess of VOCs concentration after post-treatment; at the same time, a decrease in tower pressure (correlation coefficient -0.71) will exacerbate the insufficient residence time of exhaust gas in the tower, further amplifying the risk of VOCs exceeding the standard.

[0027] Based on the emission hazard parameter relationship network constructed in step A313, and using a Bayesian network causal inference algorithm combined with the operation and adjustment logic of the exhaust gas treatment twin tower, this method traces the root cause and intermediate transmission process of emission hazard from the final manifestation of emission compliance hazard, clarifying the causal relationship of each link, and ultimately generating an emission compliance hazard evolution chain. The emission compliance hazard evolution chain includes the entire process from the generation of the triggering factor to the final manifestation of the emission compliance hazard, with each link corresponding to specific parameter changes, causal relationships, and the degree of impact. For example, the causal tracing process of emission compliance hazard based on the Bayesian network causal inference algorithm is as follows: First, the final manifestation of the emission compliance hazard is "VOCs concentration exceeding the standard after post-treatment (2.5 mg / m³)". 3 The direct cause was that "the photocatalytic power of the post-treatment tower was not adjusted in time (maintained at 2900W, not reaching the suitable power of 3200W)," resulting in residual VOCs (25.2 mg / m³) after the main treatment. 3 The VOCs concentration was not completely degraded; the reason why the photocatalytic power was not increased was that "the data transmission delay of VOCs concentration after the main treatment (delay of 150ms)" resulted in the adjustment command not being issued in time; the reason why the VOCs concentration increased after the main treatment was that "the adsorption space velocity of activated carbon in the main treatment tower was not increased in time (maintained at 0.8m / s, not adapted to 0.85m / s)", and the adsorption load was overloaded; the reason why the adsorption space velocity was not increased in time was that "the VOCs concentration rose sharply after the pretreatment (rising to 280mg / m³)". 3 The VOCs concentration surged after pretreatment, exceeding the preset fine-tuning threshold and exhibiting a delayed response to adjustment commands. The root cause was the fluctuation in initial exhaust gas characteristic parameters (exhaust gas flow rate increased to 14000 m³ / h). 3 / h, VOCs concentration rose to 420mg / m³ 3 Furthermore, the pretreatment tower spray liquid flow rate was not adjusted in a timely manner (it was not simultaneously increased to 8.8m). 3 / h). Based on the above tracing process, an evolution chain of emission compliance hazards is generated, specifically: fluctuations in initial exhaust gas characteristic parameters (flow rate 12000→14000m³ / h). 3 / h, VOCs 380→420mg / m³ 3 → The pretreatment tower spray liquid flow rate adjustment is lagging (8.2 → 8.2m) 3 / h, not synchronously adjusted) → VOCs concentration surged after pretreatment (266→280mg / m³) 3 → The adsorption space velocity in the main treatment tower was not adjusted in time (0.8 → 0.8 m / s, not increased) → The VOCs concentration increased after the main treatment (21.3 → 25.2 mg / m³). 3 → Data transmission delay after main treatment → Photocatalytic power of post-treatment tower not increased in time (2900 → 2900W, not increased) → VOCs concentration exceeds standard after post-treatment (1.8 → 2.5mg / m³) 3→ Emergence of potential emission compliance hazards. The evolution chain of potential emission compliance hazards clarifies the causal relationship, parameter change range, time nodes, and degree of impact at each stage of the hazard, providing a precise logical basis for subsequent hazard handling and strategy optimization.

[0028] Step A320: Based on the exhaust gas treatment twin towers, trace the potential risks of operational fluctuations in the initial exhaust gas treatment strategy to obtain the evolution chain of these risks. Potential risks of operational fluctuations refer to fluctuations in the initial exhaust gas characteristic data, or insufficient adjustment logic and parameter adaptability of the initial exhaust gas treatment strategy after fluctuations, leading to decreased exhaust gas treatment efficiency, potential risks to emission compliance, or abnormal operation of tower components. Tracing potential risks of operational fluctuations includes: filtering abnormal parameters related to operational fluctuations from the virtual exhaust gas treatment data, clarifying the fluctuation type (e.g., sudden increase in flow rate, sudden increase in concentration, temperature fluctuation, etc.), fluctuation amplitude, fluctuation duration, and the adjustment response of the initial exhaust gas treatment strategy when the fluctuation occurs (e.g., whether the adjustment is timely, whether the adjustment amplitude is appropriate, and whether the three-tower linkage is smooth). Subsequently, a Bayesian network causal reasoning algorithm was used, combined with the results of parameter correlation analysis, to trace the root causes, transmission paths and evolution process of the hidden dangers of operating condition fluctuations. The focus was on clarifying the causal relationship between operating condition fluctuations and the lag in the initial exhaust gas treatment strategy adjustment and insufficient parameter adaptation, as well as the transmission law of hidden dangers between the pretreatment tower, main treatment tower and post-treatment tower. At the same time, the manifestations, scope of influence and severity of hidden dangers in each link were clarified, and the evolution chain of hidden dangers of operating condition fluctuations was obtained.

[0029] With exhaust gas flow rate from 12000m 3 / h surged to 14000m 3 / h (exceeding the preset fluctuation range ±1500m) 3 / h), VOCs concentration from 380mg / m³ 3 The blood pressure suddenly rose to 430 mg / m³ 3 Taking the fluctuation of operating conditions as an example, the cause of the fluctuation was first identified through virtual waste gas treatment data (simulating a sudden increase in waste gas emission load in actual production). Then, the parameter transmission process after the fluctuation was traced: the sudden increase in waste gas flow rate and VOCs concentration caused the pretreatment tower inlet load to rise instantaneously, while the response of the pretreatment tower spray liquid flow rate adjustment command in the initial waste gas treatment strategy was lagging (delayed by 150ms), and did not start from 8.2m in time. 3 / h increased to 8.8m 3 / h (flow deviation exceeding the preset threshold of 5%) resulted in the VOCs concentration after pretreatment not decreasing to the preset 266mg / m³. 3 Instead, it rose to 290 mg / m². 3The removal rate of large particulate matter dropped to 75%. After this abnormal parameter was transmitted to the main treatment tower, the activated carbon adsorption space velocity in the main treatment tower did not adjust in time to keep up with the change in VOCs concentration after pretreatment (remaining constant at 0.8 m / s), resulting in overload of the main treatment tower and a drop in adsorption efficiency below 90%. The VOCs concentration after main treatment rose to 29 mg / m³. 3 Exceeding the preset 25mg / m 3 The control requirements were not met; however, after being transferred to the post-treatment tower, the photocatalytic power of the post-treatment tower did not adaptively increase in time (remaining constant at 2900W), thus failing to effectively degrade residual VOCs, resulting in a phased increase in VOCs concentration to 2.3 mg / m³ after post-treatment. 3 The pressure was approaching the emission compliance threshold, and the pressure inside the tower dropped to 0.095 MPa, showing slight fluctuations, creating a potential hazard to compliance. Based on the above full-process tracing, a complete evolution chain of the hazard caused by operating condition fluctuations was identified. This chain includes the causal relationship, parameter changes, time points, and degree of impact of the hazard from the occurrence of the fluctuation to its final manifestation. For example, the evolution chain of the hazard caused by operating condition fluctuations is: sudden increase in exhaust gas load → operating condition fluctuation (flow rate 12000 → 14000 m³ / h). 3 / h, VOCs 380→430mg / m³ 3 → Initial exhaust gas treatment strategy adjustment response lag (untimely adjustment of spray liquid flow rate, adsorption space velocity, and photocatalytic power) → Pretreatment parameters abnormal (excessive deviation of spray liquid flow rate, VOCs removal not meeting standards) → Main treatment parameters abnormal (adsorption load overload, decreased adsorption efficiency) → Post-treatment parameters abnormal (insufficient photocatalytic degradation, pressure fluctuations) → Potential hazards to emission compliance become apparent. The system also labels the parameter deviation values, adjustment lag times, and impact range of each stage. This provides precise logical basis and data support for optimizing the initial exhaust gas treatment strategy and improving the mechanism for handling operating condition fluctuations. It ensures that various hazards caused by operating condition fluctuations can be addressed specifically, improving the adaptability and operational stability of the exhaust gas treatment system to operating condition fluctuations.

[0030] Step A330: Based on the exhaust gas treatment twin tower, trace the potential equipment failure of the initial exhaust gas treatment strategy to obtain the equipment failure evolution chain. Equipment failure tracing refers to conducting a full-dimensional tracing of potential equipment failures caused by abnormal operation of core tower equipment and supporting components during the execution of the initial exhaust gas treatment strategy, resulting in a decrease in exhaust gas treatment effect and a risk of emission compliance. First, from the virtual exhaust gas treatment data, filter out abnormal feature data related to equipment failure, covering abnormal information such as operating parameter deviations, performance degradation, and functional failures of the core components (spray components, activated carbon adsorption layer, photocatalytic oxidation module, demister, gas distribution device) and auxiliary equipment (sensors, regulating valves, spray pumps) of the pretreatment tower, main treatment tower, and post-treatment tower. Identify the type of equipment abnormality (such as component performance degradation, local equipment failure, sensor data distortion, actuator response lag, etc.), the location of the abnormality, the initial manifestation of the abnormality, and the corresponding parameter offset amplitude. At the same time, correlate the parameter adjustment instructions and adaptation logic execution status of the initial exhaust gas treatment strategy under the abnormal equipment state to confirm whether the strategy has problems such as adjustment failure or parameter mismatch due to equipment abnormality. Subsequently, a Bayesian network causal reasoning algorithm was used to locate the root cause of the equipment anomaly, tracing the process from its initial emergence to its gradual development. At the same time, it clarified how the deviation of operating parameters caused by the equipment anomaly led to the inability of the preset adjustment logic of the initial exhaust gas treatment strategy to adapt, the failure of process parameter control, and how the hidden danger was transmitted within a single tower and spread across towers, ultimately causing the exhaust gas treatment process to fail and the emission compliance to be potentially threatened. The algorithm also clarified the manifestation, scope of impact, transmission time, and correlation with the initial exhaust gas treatment strategy of the hidden dangers in each stage.

[0031] For example, a specific tracing is carried out using a nozzle blockage failure in the pretreatment tower spray assembly: Regarding the nozzle blockage failure in the pretreatment tower spray assembly, the virtual waste gas treatment data shows that the actual flow rate of the spray liquid is lower than the preset 8m³ / h of the initial waste gas treatment strategy. 3 / h decreased to 5.5m 3The atomization effect decreased, and the temperature and VOC concentration of the pretreated exhaust gas exceeded the preset range. The root cause of the fault was found to be blockage of some nozzles in the spray assembly. Although the output pressure of the spray pump was within the preset threshold of the strategy, the failure of the actuator resulted in insufficient spray liquid flow. The logic of "linked adjustment of spray liquid flow and exhaust gas flow" in the initial exhaust gas treatment strategy failed due to nozzle blockage, causing a significant decrease in the removal rate of large particulate matter and the absorption efficiency of soluble VOCs in the pretreatment stage. Subsequently, the main treatment tower experienced accelerated blockage of the adsorption layer due to increased impurities in the inlet gas, further aggravating the decline in adsorption efficiency, forming a chain of hidden dangers: "pretreatment equipment failure → accelerated performance decline of main treatment equipment → decline in overall treatment effect". Finally, based on the above complete traceability process, an evolution chain of equipment failure hazards was formed. The equipment failure hazard evolution chain encompasses the entire causal relationship from the root cause of equipment anomalies → initial parameter deviation → strategy adaptation failure → hazard development within a single tower → hazard propagation across towers → final manifestation of emission hazards. It marks the equipment anomaly characteristics, parameter deviation values, strategy failure manifestations, hazard propagation time, and impact degree at each stage, while also identifying the adaptation shortcomings of the initial exhaust gas treatment strategy under equipment failure conditions. Through this traceable equipment failure hazard evolution chain, the adaptation defects of the initial exhaust gas treatment strategy under equipment failure conditions are accurately located, providing precise logical basis and data support for subsequent strategy optimization, equipment failure prediction, and the establishment of coordinated response mechanisms, thereby improving the adaptability of the initial exhaust gas treatment strategy to equipment failure conditions.

[0032] Step A340: Perform cross-modal correlation analysis and aggregation on the emission compliance hazard evolution chain, the operating condition fluctuation hazard evolution chain, and the equipment failure hazard evolution chain to generate the waste gas treatment hazard network. First, the three types of hazard evolution chains are uniformly standardized and deconstructed to extract core elements such as the root cause of the hazard, key abnormal parameters, link transmission nodes, hazard manifestations, impact range, severity, and causal relationship weights from each hazard evolution chain. A unified element labeling rule is established according to the entire waste gas treatment process (pretreatment → main treatment → posttreatment), transforming the core elements of each hazard evolution chain into standardized data units that can be correlated and compared, laying a unified analytical foundation for cross-modal correlation. Based on the standardized and deconstructed element data, key abnormal parameters overlapping in the evolution chains of the three types of hidden dangers are mined from the parameter dimension, and the correlation strength and causal weight are calculated to clarify the chain effect of core common parameters. From the transmission dimension, the overlap and mutual triggering relationship of the transmission paths of the three types of hidden dangers in each treatment stage are analyzed, clarifying the chain coupling relationship such as operating condition fluctuations inducing equipment failure, equipment failure aggravating emission compliance hazards, and emission compliance hazards exposing strategy adaptation defects. At the same time, from the strategy dimension, the execution failure points of the initial waste gas treatment strategy in the evolution of the three types of hidden dangers are correlated, and the common shortcomings and individual defects of the strategy in terms of parameter adjustment, equipment adaptation, and operating condition response are clarified. Subsequently, the various correlation relationships obtained from cross-chain parsing are quantified, and each group of correlation elements is assigned a correlation coefficient in the range of 0-1, a causal direction of positive triggering / reverse influence, and an influence weight reflecting the contribution of the hidden danger. Combined with the logic of the entire waste gas treatment process, a hidden danger correlation topology model is constructed. The hidden danger correlation topology model includes: standardized element data as topology nodes, and quantified correlation relationships as connecting edges between nodes, ensuring that each node and each connecting edge has corresponding quantitative data and logical support. Finally, based on the hazard association topology model, a complete waste gas treatment hazard map network is generated by structured aggregation according to the hierarchical structure of "core layer - intermediate layer - outer layer".

[0033] In the waste gas treatment hazard network, the core layer consists of common key hazard nodes overlapping in the three types of hazard evolution chains. It encompasses the core hazard root causes and core abnormal parameters, serving as the network's central hub. The middle layer represents the key points of hazard transmission and cross-stage connections in pretreatment, main treatment, and post-treatment stages, reflecting the entire process transmission pattern of hazards. The outer layer comprises unique, personalized hazard nodes within the three evolution chains, covering specific hazard manifestations, secondary influencing parameters, and hazard elements related to specific equipment / operating conditions. Each node is also labeled with its corresponding hazard type, severity, and triggering conditions. Furthermore, the waste gas treatment hazard network synchronously links the corresponding execution nodes of the initial waste gas treatment strategy, the corresponding operating nodes of equipment components, and the corresponding process nodes of waste gas treatment, achieving a full-dimensional association between hazards and processes, equipment, and strategies. This ensures that the waste gas treatment hazard network achieves the effects of locatable nodes, traceable associations, quantifiable weights, and predictable transmission.

[0034] The waste gas treatment hazard network generated through cross-modal correlation analysis and aggregation not only fully integrates all core information of the evolution chains of the three types of hazards, breaking through the limitations of analyzing single-type hazards, but also uncovers the hidden coupling relationships and common problems between chains. It clearly presents the independent evolution patterns of various hazards and more intuitively reflects the inherent connections of mutual triggering and transmission between different types of hazards. This provides visualized, quantifiable, and predictable core support for the comprehensive optimization of subsequent initial waste gas treatment strategies and the whole-process risk prevention and control of waste gas treatment systems. It significantly improves the pertinence and systematic nature of waste gas treatment hazard handling, fundamentally reduces the operational risks of waste gas treatment systems, and ensures long-term stable compliance with waste gas emission standards.

[0035] Step A400: Perform multi-dimensional feedforward adjustment and combination of the initial waste gas treatment strategy according to the waste gas treatment hazard map network to obtain the waste gas treatment adjustment space. In one possible implementation, step A400 further includes steps A410 to A440. Step A410: Perform feedforward adjustment of the waste gas pretreatment matching scheme according to the waste gas treatment hazard map network to obtain the pretreatment adjustment space; Step A420: Perform feedforward adjustment of the waste gas main treatment matching scheme according to the waste gas treatment hazard map network to obtain the main treatment adjustment space; Step A430: Perform feedforward adjustment of the waste gas post-treatment matching scheme according to the waste gas treatment hazard map network to obtain the post-treatment adjustment space; Step A440: Randomly combine the pretreatment adjustment space, the main treatment adjustment space, and the post-treatment adjustment space to generate the waste gas treatment adjustment space.

[0036] Based on the comprehensive hazard correlation information related to the pretreatment stage in the waste gas treatment hazard network, this study focuses on the root causes of hazards pointing to the pretreatment stage, key abnormal parameters, hazard transmission triggering conditions, and coupling relationships with subsequent processes. Targeted feedforward adjustments are made to the waste gas pretreatment matching scheme. The core focus is on optimizing the parameter ranges and adjustment rules of core equipment such as the pretreatment tower spray assembly, air inlet distribution device, coarse filter assembly, and flow equalization plate, as well as the linkage adjustment logic of the pretreatment process. This proactively avoids various hazards identified in the hazard network in the pretreatment stage (such as delayed spray liquid flow adjustment, poor flow equalization, coarse filter assembly blockage, and pretreatment parameter deviation leading to main treatment load overload). The pretreatment adjustment space includes multiple pretreatment adjustment decisions. Specifically, based on the waste gas pretreatment matching scheme and combined with the quantified correlation weights and hazard triggering thresholds in the waste gas treatment hazard network, reasonable feedforward adjustment ranges are set for each parameter, while clarifying the linkage adjustment rules between parameters. For example, to address the potential hazard of "delayed spray flow rate adjustment leading to excessive VOC concentrations after pretreatment," the spray flow rate adjustment range was increased from the original 8m... 3 / h expanded to 6-10m 3 / h, and the threshold for the linkage deviation with the exhaust gas flow rate has been optimized from ≤5% to ≤3%.

[0037] Similarly, based on the hazard association system of the main treatment stage in the hazard network of waste gas treatment, the core hazard nodes of the main treatment stage in the hazard network of waste gas treatment (such as activated carbon adsorption layer saturation, adsorption space velocity adjustment failure, uneven gas distribution, and post-treatment emission hazards caused by main treatment parameter deviation, etc.), their coupling and transmission relationships with pretreatment / post-treatment stages, and the parameter thresholds for hazard triggering are extracted. The waste gas main treatment matching scheme is then adjusted to obtain the main treatment adjustment space. The main treatment adjustment space includes multiple main treatment adjustment decisions. For example, for the hazard of "fixed adsorption space velocity causing overload / insufficient adsorption," the adsorption space velocity adjustment range is expanded from the original 0.8 m / s to 0.6-1.0 m / s, and the linkage fine-tuning step size with the VOCs concentration after pretreatment is optimized to 0.05 m / s. To address the potential issue of uneven gas release due to localized saturation of the activated carbon adsorption layer, the adjustment range for the number of activated carbon adsorption layers in operation has been increased. This allows for adaptive adjustment within the 1-3 layer range based on the inlet gas load. Simultaneously, the adjustment range for single-layer thickness has been expanded to 60-100 cm. To address the potential issue of decreased adsorption efficiency caused by pressure fluctuations within the tower, the tower pressure control range has been optimized from the original 0.12 MPa to 0.10-0.14 MPa.

[0038] Similarly, based on the comprehensive hazard correlation information of the post-treatment stage in the hazard network of waste gas treatment, the core hazard nodes of the post-treatment stage (such as photocatalytic power adjustment lag, insufficient buffer space residence time, decreased demisting efficiency, and excessive emission parameters) are extracted, along with their coupling and transmission relationships with the main treatment stage and the critical parameter values ​​for hazard triggering. Adjustments are then made to the waste gas post-treatment matching scheme, focusing on optimizing core parameters and process termination logic such as post-treatment tower buffer space control, photocatalytic oxidation module, demisting structure, and exhaust pressure control. This proactively eliminates various hazard stages in the post-treatment stage identified in the hazard network, ensuring final emissions meet standards. The post-treatment adjustment space includes multiple post-treatment adjustment decisions. For example, regarding the hazard of "incomplete VOCs degradation caused by photocatalytic power adjustment lag," the photocatalytic power adjustment range is expanded from the original 3000W to 2000-4000W, and the linkage deviation threshold with the VOCs concentration after main treatment is optimized from ≤10% to ≤5%. To address the potential issue of "insufficient residence time in the buffer space leading to unstable exhaust gas flow," the residence time adjustment range has been extended from the original 15s to 10-20s, achieving linkage and matching with the exhaust gas flow rate after main treatment.

[0039] Subsequently, the obtained pretreatment adjustment space, main treatment adjustment space, and post-treatment adjustment space are randomly selected and combined to generate the waste gas treatment adjustment space. This space includes multiple waste gas treatment adjustment schemes. Each scheme comprises one randomly selected pretreatment adjustment decision, one main treatment adjustment decision, and one post-treatment adjustment decision. Through a waste gas treatment hazard network, the initial waste gas treatment strategy is optimized through comprehensive, multi-stage feedforward adjustments and parameter combinations, forming a waste gas treatment adjustment space covering the entire process from pretreatment to main treatment and post-treatment. This upgrades the initial waste gas treatment strategy from "fixed parameter adaptation" to "dynamic range adjustment," mitigating the problem of untimely hazard response caused by fixed parameters in a single stage. This provides comprehensive, feasible, and scientific parameter support for the real-time optimization and dynamic control of subsequent waste gas treatment strategies, transforming the waste gas treatment system from "passively handling hazards" to "actively avoiding hazards," significantly improving the stability of waste gas treatment and the flexibility of process adjustments.

[0040] Step A500: Invoke the dynamic risk model for waste gas treatment, and combine it with the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space, obtain the optimization results of waste gas treatment parameters, and control the waste gas treatment process tower to perform multi-tower coordinated waste gas treatment based on the optimization results of waste gas treatment parameters. In one possible implementation, step A500 further includes steps A510 to A540. Step A510: Perform risk constraint optimization on the waste gas treatment adjustment space based on the dynamic risk model for waste gas treatment and the waste gas treatment twin tower to obtain the optimized waste gas treatment space. Step A510 includes: Step A511, extracting the r-th waste gas treatment adjustment scheme according to the waste gas treatment adjustment space, where r is a positive integer; Step A512, performing virtual control on the waste gas treatment twin tower according to the r-th waste gas treatment adjustment scheme to obtain the r-th waste gas treatment virtual data; Step A513, inputting the r-th waste gas treatment virtual data into the waste gas treatment dynamic risk model to obtain the r-th waste gas treatment dynamic risk set; Step A514, determining whether the r-th waste gas treatment dynamic risk set satisfies the dynamic risk constraint; Step A515, if the r-th waste gas treatment dynamic risk set satisfies the dynamic risk constraint, adding the r-th waste gas treatment adjustment scheme to the optimization waste gas treatment space.

[0041] The r-th waste gas treatment regulation scheme is randomly extracted from the waste gas treatment regulation space. This r-th scheme serves as the core control command and is transmitted via industrial Ethernet to the simulation control module of the waste gas treatment twin tower. Following the same operating logic as the physical tower, the twin tower undergoes full-process virtual closed-loop control. The control process completely replicates the parameter execution, equipment linkage, and process connection flow of the actual waste gas treatment process, obtaining the r-th virtual data for waste gas treatment. The r-th waste gas treatment regulation scheme is any one of the waste gas treatment regulation schemes in the waste gas treatment regulation space. The r-th virtual data includes the process simulation operating parameters corresponding to the r-th scheme (real-time simulation operating values ​​of each component of the three towers, parameter adjustment simulation response time, etc.), waste gas treatment simulation effect data (simulated VOCs concentration after each stage, simulated particulate matter removal rate, simulated exhaust temperature / pressure, etc.), and equipment operating simulation status data. Subsequently, the r-th virtual data is used as input information to the waste gas treatment dynamic risk model to obtain the r-th dynamic risk set for waste gas treatment. The r-th dynamic risk set for waste gas treatment includes the predicted emission compliance risk coefficient, the predicted operating condition fluctuation risk coefficient, and the predicted equipment failure risk coefficient corresponding to the r-th waste gas treatment adjustment scheme.

[0042] The dynamic risk model for waste gas treatment is a multi-dimensional quantitative risk assessment model for the entire waste gas treatment process. Its core analytical indicators are emission compliance risk, operating condition fluctuation risk, and equipment failure risk. Through comprehensive analysis of virtual waste gas treatment data, it outputs quantitative risk coefficients for each indicator, enabling accurate and real-time assessment of dynamic risks during the execution of waste gas treatment adjustment schemes. The dynamic risk model adopts a four-layer modular structure: input layer, feature extraction layer, indicator analysis layer, and output layer. The input layer interfaces with the virtual waste gas treatment data, performing standardized preprocessing through wavelet transform noise reduction and linear interpolation completion. The feature extraction layer pre-sets feature pools associated with the three risk indicators based on waste gas treatment process rules, extracting and filtering core risk-related features using feature importance ranking. The indicator analysis layer sets up three independent XGBoost regression algorithm quantization sub-modules, realizing one-to-one transformation of the three risk indicators from feature data to quantitative risk coefficients. The output layer integrates the quantification results of each sub-module into a dynamic risk set for waste gas treatment, including predicted emission compliance risk coefficients, predicted operating condition fluctuation risk coefficients, and predicted equipment failure risk coefficients.

[0043] The construction process of the dynamic risk model for waste gas treatment includes: First, clarifying the model construction objectives and quantification rules, defining the range of quantified risk coefficients as 0-1, and clarifying the correspondence between coefficients and risk levels to provide a basis for subsequent training and validation. Second, constructing standardized training, validation, and test datasets, based on virtual simulation data of the waste gas treatment twin tower and historical system operation data, integrating multi-dimensional data under multiple operating conditions and equipment states, and completing precise manual annotation, dividing the dataset according to a reasonable proportion to ensure data representativeness; then, using Python combined with a deep learning framework to build a four-layer modular basic framework for the model, configuring adaptation algorithms for each layer, configuring wavelet transform denoising and linear interpolation completion algorithms for the input layer, configuring feature importance ranking algorithms for the feature extraction layer, and configuring XGBoost for all three sub-modules of the index analysis layer. The regression algorithm is followed by a data integration algorithm configured in the output layer. Next, model training and hyperparameter tuning are performed. Training and validation datasets are input into the model, which is trained using mini-batch gradient descent. Model performance is then validated, testing the trained model in terms of quantization accuracy, generalization ability, and computational efficiency. If the model fails to meet preset standards, it is re-tuned or retrained with supplementary data. Finally, the dynamic risk model for waste gas treatment is deployed in an industrial-grade operating environment, and standardized application interfaces are developed. This enables seamless integration between the dynamic risk model, the waste gas treatment twin tower, and the waste gas treatment system. An iterative optimization mechanism is established to continuously import actual operating data and newly added virtual simulation data to retrain the dynamic risk model, updating the feature pool and algorithm parameters. This ensures that the dynamic risk model adapts to process adjustments, equipment updates, and changes in operating conditions of the waste gas treatment system, guaranteeing continuous improvement in assessment accuracy.

[0044] Furthermore, it is determined whether the dynamic risk set of the r-th exhaust gas treatment meets the dynamic risk constraints. The dynamic risk constraints include pre-defined emission compliance risk ranges, operating condition fluctuation risk ranges, and equipment failure risk ranges. If each risk coefficient within the r-th exhaust gas treatment dynamic risk set meets its corresponding risk range, then the r-th exhaust gas treatment dynamic risk set meets the dynamic risk constraints. The r-th exhaust gas treatment adjustment scheme is then recorded as the r-th optimal exhaust gas treatment scheme and added to the optimal exhaust gas treatment space. The optimal exhaust gas treatment space includes multiple optimal exhaust gas treatment schemes. An optimal exhaust gas treatment scheme is an exhaust gas treatment adjustment scheme that meets the dynamic risk constraints. Through a closed-loop verification process of "virtual simulation verification - quantitative risk assessment - risk constraint screening" for all schemes within the exhaust gas treatment adjustment space, invalid schemes with excessive emissions, insufficient operating condition adaptation, or potential equipment failures are effectively eliminated. This ensures that all schemes within the optimal exhaust gas treatment space meet the dynamic risk constraints, achieving a deep integration of risk assessment and virtual simulation. This makes the risk-constrained optimization process more scientific, precise, and efficient. The obtained optimization space for waste gas treatment provides high-quality and feasible solutions for the dynamic optimization of subsequent waste gas treatment strategies and the whole-process risk prevention and control. It significantly improves the adaptability of the waste gas treatment system to complex operating conditions and the reliability of emission compliance, promotes the upgrade of waste gas treatment strategies from "range adjustment" to "precise matching", and further strengthens the intelligent empowerment of the waste gas treatment process by digital twin technology.

[0045] Step A520: Weights are assigned to the waste gas treatment risk analysis indicators of the dynamic risk model for waste gas treatment to establish a waste gas treatment adaptation risk function; wherein, the waste gas treatment risk analysis indicators include emission compliance risk, operating condition fluctuation risk, and equipment failure risk. Step A530: Based on the waste gas treatment adaptation risk function, the optimal waste gas treatment space is optimized through waste gas treatment adaptation risk analysis to obtain candidate waste gas treatment spaces that meet the adaptation risk target; Step A540: Based on the candidate waste gas treatment spaces, energy consumption targets are optimized to generate the waste gas treatment parameter optimization results.

[0046] By assigning weights to the risk analysis indicators for waste gas treatment, an adaptive risk function for waste gas treatment is obtained. These indicators include emission compliance risk, operating condition fluctuation risk, and equipment failure risk. The expression for the adaptive risk function is: F(λ) = ω1×C1 + ω2×C2 + ω3×C3; where F(λ) represents the adaptive risk value for waste gas treatment, ω1 represents the emission compliance risk weight, ω2 represents the operating condition fluctuation risk weight, ω3 represents the equipment failure risk weight, the sum of ω1, ω2, and ω3 is 1, C1 represents the predicted emission compliance risk coefficient, C2 represents the predicted operating condition fluctuation risk coefficient, and C3 represents the predicted equipment failure risk coefficient.

[0047] Furthermore, the predicted emission compliance risk coefficient, predicted operating condition fluctuation risk coefficient, and predicted equipment failure risk coefficient corresponding to each optimized waste gas treatment scheme within the optimized waste gas treatment space are input into the waste gas treatment adaptation risk function to obtain multiple waste gas treatment adaptation risk values. Each optimized waste gas treatment scheme corresponds to one waste gas treatment adaptation risk value. Subsequently, the adaptation risk target includes a pre-set and determined waste gas treatment adaptation risk threshold. It is determined whether each waste gas treatment adaptation risk value is less than or equal to the waste gas treatment adaptation risk threshold. If the waste gas treatment adaptation risk value is less than or equal to the waste gas treatment adaptation risk threshold, then the optimized waste gas treatment scheme corresponding to that waste gas treatment adaptation risk value meets the adaptation risk target. The optimized waste gas treatment scheme corresponding to that waste gas treatment adaptation risk value is recorded as a candidate waste gas treatment scheme and added to the candidate waste gas treatment space. The candidate waste gas treatment space includes multiple candidate waste gas treatment schemes. The candidate waste gas treatment scheme is the optimized waste gas treatment scheme that meets the adaptation risk target. "Meeting the adaptation risk target" means that the waste gas treatment adaptation risk value is less than or equal to the waste gas treatment adaptation risk threshold.

[0048] Furthermore, energy consumption targets are optimized for the candidate waste gas treatment spaces. Specifically, energy consumption is predicted for each candidate waste gas treatment scheme within the space, and the scheme with the lowest energy consumption is output as the optimization result for waste gas treatment parameters. Finally, the optimization result is transmitted to the control terminal of the waste gas treatment process tower via industrial Ethernet. The control terminal then controls the pretreatment tower, main treatment tower, and post-treatment tower according to the optimization result, achieving multi-tower coordinated operation.

[0049] By linking the dynamic risk model of waste gas treatment with the waste gas treatment twin tower, multi-objective optimization under risk constraints is carried out on the waste gas treatment adjustment space. It takes into account emission compliance, operating condition adaptability, equipment stability and minimum energy consumption, and realizes the closed-loop transformation from virtual optimization to physical control. This significantly improves the intelligence level and operational stability of the waste gas treatment process, reduces the risk of exceeding emission standards, operating condition fluctuation risk and equipment operation hidden danger risk, and improves the waste gas treatment effect.

[0050] In summary, the data acquisition and control-based waste gas treatment method provided in this application has the following technical effects: It acquires initial waste gas characteristic data by collecting data from the waste gas to be treated in the waste gas treatment process tower; it uses this initial waste gas characteristic data to perform multi-tower collaborative search of the waste gas treatment process tower to determine the initial waste gas treatment strategy; it uses a waste gas treatment twin tower to trace multimodal potential hazards of the initial waste gas treatment strategy and obtain a waste gas treatment hazard network; it uses the waste gas treatment hazard network to perform multi-dimensional feedforward adjustment combination of the initial waste gas treatment strategy and obtain the waste gas treatment adjustment space; it uses a waste gas treatment dynamic risk model and a waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space, obtains the waste gas treatment parameter optimization results, and controls the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment based on the waste gas treatment parameter optimization results. This achieves the technical effect of improving the emission stability, operational safety, and waste gas treatment quality of the waste gas treatment process by performing feedforward risk perception and collaborative adjustment based on data acquisition and twin modeling.

[0051] Example 2: Based on the same inventive concept as the data acquisition and control waste gas treatment method in the previous examples, this invention also provides a data acquisition and control waste gas treatment system. Referring to Figure 2, the system includes: a waste gas data acquisition module 1, used to acquire data on the waste gas to be treated in the waste gas treatment process tower to obtain initial waste gas characteristic data; a multi-tower collaborative search module 2, used to perform a multi-tower collaborative search of the waste gas treatment process tower based on the initial waste gas characteristic data to determine the initial waste gas treatment strategy; and a waste gas treatment hidden danger tracing module 3, used to construct a waste gas treatment twin tower and, based on the waste gas treatment twin tower... The tower performs multimodal hazard tracing on the initial waste gas treatment strategy to obtain a waste gas treatment hazard map network; the waste gas treatment feedforward adjustment module 4 is used to perform multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy according to the waste gas treatment hazard map network to obtain the waste gas treatment adjustment space; the multi-objective linkage optimization control module 5 is used to call the waste gas treatment dynamic risk model, combine the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space, obtain the waste gas treatment parameter optimization result, and control the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment according to the waste gas treatment parameter optimization result.

[0052] Furthermore, the multi-tower collaborative search module 2 is also used to: load the pretreatment record block, main treatment record block, and post-treatment record block corresponding to the waste gas treatment process tower; perform a similarity evaluation search on the pretreatment record block based on the initial waste gas feature data to obtain a waste gas pretreatment matching scheme; perform a similarity evaluation search on the main treatment record block based on the initial waste gas feature data to obtain a waste gas main treatment matching scheme; perform a similarity evaluation search on the post-treatment record block based on the initial waste gas feature data to obtain a waste gas post-treatment matching scheme; and combine the waste gas pretreatment matching scheme and the waste gas main treatment matching scheme to generate the initial waste gas treatment strategy.

[0053] Furthermore, the exhaust gas treatment hazard tracing module 3 is also used for: tracing emission compliance hazards of the initial exhaust gas treatment strategy based on the exhaust gas treatment twin tower, and obtaining an evolution chain of emission compliance hazards; tracing operating condition fluctuation hazards of the initial exhaust gas treatment strategy based on the exhaust gas treatment twin tower, and obtaining an evolution chain of operating condition fluctuation hazards; tracing equipment failure hazards of the initial exhaust gas treatment strategy based on the exhaust gas treatment twin tower, and obtaining an evolution chain of equipment failure hazards; and performing cross-modal association analysis and aggregation on the emission compliance hazard evolution chain, the operating condition fluctuation hazard evolution chain, and the equipment failure hazard evolution chain to generate the exhaust gas treatment hazard graph network.

[0054] Furthermore, the exhaust gas treatment hazard tracing module 3 is also used for: virtually controlling the exhaust gas treatment twin tower according to the initial exhaust gas treatment strategy to obtain virtual exhaust gas treatment data; identifying emission compliance hazards based on the virtual exhaust gas treatment data to obtain emission compliance hazard characteristics; performing multi-parameter correlation analysis on the virtual exhaust gas treatment data based on the emission compliance hazard characteristics to obtain an emission hazard parameter relationship network; and performing causal tracing on the emission compliance hazard characteristics based on the emission hazard parameter relationship network to generate the emission compliance hazard evolution chain.

[0055] Furthermore, the waste gas treatment feedforward adjustment module 4 is also used to: perform feedforward adjustment on the waste gas pretreatment matching scheme according to the waste gas treatment hazard map network to obtain the pretreatment adjustment space; perform feedforward adjustment on the waste gas main treatment matching scheme according to the waste gas treatment hazard map network to obtain the main treatment adjustment space; perform feedforward adjustment on the waste gas post-treatment matching scheme according to the waste gas treatment hazard map network to obtain the post-treatment adjustment space; and generate the waste gas treatment adjustment space by randomly combining the pretreatment adjustment space, the main treatment adjustment space, and the post-treatment adjustment space.

[0056] Furthermore, the multi-objective linkage optimization control module 5 is also used to: perform risk constraint optimization on the waste gas treatment adjustment space based on the waste gas treatment dynamic risk model and the waste gas treatment twin tower, and obtain the optimal waste gas treatment space; assign weights to the waste gas treatment risk analysis indicators of the waste gas treatment dynamic risk model, and establish a waste gas treatment adaptation risk function; perform waste gas treatment adaptation risk analysis optimization on the optimal waste gas treatment space based on the waste gas treatment adaptation risk function, and obtain candidate waste gas treatment spaces that meet the adaptation risk objectives; and perform energy consumption target optimization based on the candidate waste gas treatment spaces to generate the waste gas treatment parameter optimization results.

[0057] Furthermore, the multi-objective linkage optimization control module 5 is also used for: extracting the r-th waste gas treatment adjustment scheme according to the waste gas treatment adjustment space, where r is a positive integer; performing virtual control on the waste gas treatment twin tower according to the r-th waste gas treatment adjustment scheme to obtain the r-th waste gas treatment virtual data; inputting the r-th waste gas treatment virtual data into the waste gas treatment dynamic risk model to obtain the r-th waste gas treatment dynamic risk set; determining whether the r-th waste gas treatment dynamic risk set satisfies the dynamic risk constraint; if the r-th waste gas treatment dynamic risk set satisfies the dynamic risk constraint, adding the r-th waste gas treatment adjustment scheme to the optimization waste gas treatment space.

[0058] The waste gas treatment process tower includes a pretreatment tower, a main treatment tower, and a post-treatment tower.

[0059] The risk analysis indicators for waste gas treatment include emission compliance risk, operating condition fluctuation risk, and equipment failure risk.

[0060] The waste gas treatment system with data acquisition and control provided in this embodiment of the invention can execute the waste gas treatment method with data acquisition and control provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0061] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0062] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A waste gas treatment method with data acquisition and control, characterized in that, The method includes: collecting data on the waste gas to be treated from the waste gas treatment process tower to obtain initial waste gas characteristic data; performing multi-tower collaborative search on the waste gas treatment process tower based on the initial waste gas characteristic data to determine the initial waste gas treatment strategy; constructing a waste gas treatment twin tower and performing multimodal hazard tracing on the initial waste gas treatment strategy based on the waste gas treatment twin tower to obtain a waste gas treatment hazard network; performing multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy based on the waste gas treatment hazard network to obtain the waste gas treatment adjustment space; calling the waste gas treatment dynamic risk model and combining it with the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space to obtain the waste gas treatment parameter optimization result; and controlling the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment based on the waste gas treatment parameter optimization result.

2. The method as described in claim 1, characterized in that, Based on the initial exhaust gas characteristic data, a multi-tower collaborative search is performed on the exhaust gas treatment process towers to determine the initial exhaust gas treatment strategy. This includes: loading the pretreatment record block, main treatment record block, and post-treatment record block corresponding to the exhaust gas treatment process towers; performing a similarity evaluation search on the pretreatment record block based on the initial exhaust gas characteristic data to obtain an exhaust gas pretreatment matching scheme; performing a similarity evaluation search on the main treatment record block based on the initial exhaust gas characteristic data to obtain an exhaust gas main treatment matching scheme; performing a similarity evaluation search on the post-treatment record block based on the initial exhaust gas characteristic data to obtain an exhaust gas post-treatment matching scheme; and combining the exhaust gas pretreatment matching scheme and the exhaust gas main treatment matching scheme to generate the initial exhaust gas treatment strategy.

3. The method as described in claim 1, characterized in that, Based on the exhaust gas treatment twin tower, multimodal hazard tracing is performed on the initial exhaust gas treatment strategy to obtain an exhaust gas treatment hazard network, including: tracing emission compliance hazards based on the exhaust gas treatment twin tower to obtain an emission compliance hazard evolution chain; tracing operating condition fluctuation hazards based on the exhaust gas treatment twin tower to obtain an operating condition fluctuation hazard evolution chain; tracing equipment failure hazards based on the exhaust gas treatment twin tower to obtain an equipment failure hazard evolution chain; and performing cross-modal correlation analysis and aggregation on the emission compliance hazard evolution chain, the operating condition fluctuation hazard evolution chain, and the equipment failure hazard evolution chain to generate the exhaust gas treatment hazard network.

4. The method as described in claim 3, characterized in that, The process of tracing emission compliance risks based on the initial exhaust gas treatment strategy using the exhaust gas treatment twin tower and obtaining an evolution chain of emission compliance risks includes: performing virtual control of the exhaust gas treatment twin tower based on the initial exhaust gas treatment strategy to obtain virtual exhaust gas treatment data; identifying emission compliance risks based on the virtual exhaust gas treatment data to obtain emission compliance risk characteristics; performing multi-parameter correlation analysis on the virtual exhaust gas treatment data based on the emission compliance risk characteristics to obtain an emission risk parameter relationship network; and performing causal tracing of the emission compliance risk characteristics based on the emission risk parameter relationship network to generate the emission compliance risk evolution chain.

5. The method as described in claim 1, characterized in that, Perform multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy according to the waste gas treatment hidden danger map network to obtain a waste gas treatment adjustment space, including: performing feedforward adjustment on the waste gas pretreatment matching plan according to the waste gas treatment hidden danger map network to obtain a pretreatment adjustment space; performing feedforward adjustment on the waste gas main treatment matching plan according to the waste gas treatment hidden danger map network to obtain a main treatment adjustment space; performing feedforward adjustment on the waste gas post-treatment matching plan according to the waste gas treatment hidden danger map network to obtain a post-treatment adjustment space; randomly combining the pretreatment adjustment space, the main treatment adjustment space and the post-treatment adjustment space to generate the waste gas treatment adjustment space.

6. The method as described in claim 1, characterized in that, Invoke the waste gas treatment dynamic risk model, and perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space in combination with the waste gas treatment twin towers to obtain the waste gas treatment parameter optimization result, including: performing risk constraint optimization on the waste gas treatment adjustment space according to the waste gas treatment dynamic risk model and the waste gas treatment twin towers to obtain an optimized waste gas treatment space; assigning weights to the waste gas treatment risk analysis indicators of the waste gas treatment dynamic risk model to establish a waste gas treatment adaptation risk function; performing waste gas treatment adaptation risk analysis optimization on the optimized waste gas treatment space according to the waste gas treatment adaptation risk function to obtain a candidate waste gas treatment space that meets the adaptation risk target; performing energy consumption target optimization according to the candidate waste gas treatment space to generate the waste gas treatment parameter optimization result.

7. The method as described in claim 6, characterized in that, Performing risk constraint optimization on the waste gas treatment adjustment space according to the waste gas treatment dynamic risk model and the waste gas treatment twin towers to obtain an optimized waste gas treatment space, including: extracting the r-th waste gas treatment adjustment plan from the waste gas treatment adjustment space, where r is a positive integer; performing virtual control on the waste gas treatment twin towers according to the r-th waste gas treatment adjustment plan to obtain the r-th waste gas treatment virtual data; inputting the r-th waste gas treatment virtual data into the waste gas treatment dynamic risk model to obtain the r-th waste gas treatment dynamic risk set; judging whether the r-th waste gas treatment dynamic risk set meets the dynamic risk constraint; if the r-th waste gas treatment dynamic risk set meets the dynamic risk constraint, adding the r-th waste gas treatment adjustment plan to the optimized waste gas treatment space.

8. The method as described in claim 1, characterized in that, The waste gas treatment process tower includes a pretreatment tower, a main treatment tower and a post-treatment tower.

9. The method as described in claim 6, characterized in that, The waste gas treatment risk analysis indicators include emission compliance risk, operating condition fluctuation risk and equipment failure risk.

10. A waste gas treatment system with data acquisition and control, characterized in that, The system is used to implement the method described in any one of claims 1 to 9. The system comprises: a waste gas data acquisition module for acquiring data on the waste gas to be treated in the waste gas treatment process tower to obtain initial waste gas characteristic data; a multi-tower collaborative search module for performing a multi-tower collaborative search on the waste gas treatment process tower based on the initial waste gas characteristic data to determine an initial waste gas treatment strategy; a waste gas treatment hazard tracing module for constructing a waste gas treatment twin tower and performing multi-modal hazard tracing on the initial waste gas treatment strategy based on the waste gas treatment twin tower to obtain a waste gas treatment hazard network; a waste gas treatment feedforward adjustment module for performing multi-dimensional feedforward adjustment combination on the initial waste gas treatment strategy based on the waste gas treatment hazard network to obtain a waste gas treatment adjustment space; and a multi-objective linkage optimization control module for calling the waste gas treatment dynamic risk model, combining the waste gas treatment twin tower to perform multi-objective linkage optimization under risk constraints on the waste gas treatment adjustment space, obtaining waste gas treatment parameter optimization results, and controlling the waste gas treatment process tower to perform multi-tower collaborative waste gas treatment based on the waste gas treatment parameter optimization results.

Images