Intelligent matching and ventilation control method for hazardous waste temporary storage library in chemical industrial park

Abstract

The application discloses an intelligent compatibility and ventilation control method for a hazardous waste temporary storage library in a chemical industrial park, and belongs to the technical field of hazardous waste storage management. The method comprises the following steps: constructing a hazardous waste information database; constructing a three-dimensional digital twin space model of the temporary storage library and discretizing the model; generating a space compatibility scheme based on compatibility rules; predicting the spatial and temporal distribution of waste gas concentration in each region and generating a differentiated ventilation control strategy; comparing the real-time monitoring data with the predicted values in the spatial and temporal correlation; when the deviation exceeds the threshold, a hierarchical response is triggered, the first type of deviation triggers adaptive adjustment of ventilation, and the second type of deviation triggers re-optimization of the compatibility scheme and outputs a minimum disturbance adjustment scheme. The application breaks through the technical barriers of mutual fragmentation of compatibility decision and ventilation control, realizes two-way collaborative closed-loop control of storage layout optimization and precise matching of ventilation resources, and takes into account the safety and energy saving of the temporary storage library.

Description

Technical Field

[0001] This invention relates to the field of hazardous waste storage management technology, and more specifically, to an intelligent matching and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park. Background Technology

[0002] Hazardous waste temporary storage facilities in chemical industrial parks are key nodes connecting the generation, collection, and final disposal of hazardous waste, and their safety management level directly affects the park's environmental risk prevention and control capabilities. According to the requirements of standards such as the "Standard for Pollution Control of Hazardous Waste Storage" (GB 18597-2023), temporary storage facilities must simultaneously meet the dual control objectives of waste compatibility storage and effective collection and treatment of waste gas.

[0003] Currently, the management technologies surrounding hazardous waste temporary storage facilities are mainly divided into two independently developing branches:

[0004] Branch 1: Hazardous Waste Compatibility Storage Technology. This type of technology focuses on the chemical compatibility between wastes. By constructing a compatibility rule base or knowledge graph, it makes compatibility judgments and zoning decisions for hazardous waste entering the storage facility. For example, Chinese patent application CN119168631A discloses an intelligent compatibility method for hazardous waste disposal; Chinese patent CN118967117B discloses an intelligent treatment method for hazardous chemical waste based on a knowledge graph. However, the above technologies only address the problem of "how to store safely." The compatibility decision-making process does not consider the impact of different storage layouts on subsequent exhaust gas diffusion patterns and ventilation loads, and there is a lack of data correlation and coordination mechanisms between the compatibility scheme and ventilation control.

[0005] Branch Two: Temporary Storage Ventilation Control Technology. This type of technology focuses on the monitoring and discharge of waste gas within the temporary storage facility. It uses sensors to monitor environmental parameters in real time and activates ventilation equipment when standards are exceeded. For example, Chinese patent application CN121581633A discloses a real-time monitoring and risk warning system for hazardous waste storage environments based on digital twins. This system achieves environmental risk warnings by constructing a digital twin model and comparing measured data with virtual parameters. Furthermore, existing technologies disclose schemes for automatically triggering graded ventilation based on VOCs sensor arrays and intelligent ventilation methods based on CFD simulations. However, the ventilation control strategies of these technologies are either based on passive responses to real-time monitoring or on the fixed execution of preset schemes, lacking a correlation between the generation of ventilation strategies and the hazardous waste storage layout. When abnormal waste gas emissions occur, the system can only adjust ventilation parameters for end-point remediation, failing to address the root cause at the storage layout level.

[0006] In summary, a significant technological gap exists in existing technologies: compatibility decisions and ventilation control are disconnected. The decision-making results in the compatibility stage (the location and quantity of waste storage) directly affect the generation intensity, spatial distribution, and temporal evolution of waste gas within the temporary storage facility. However, existing compatibility methods do not transmit this information to the ventilation system. The ventilation system only passively responds to environmental monitoring data, unable to predict changes in waste gas load caused by changes in storage status, nor can it feed back anomalies detected during ventilation operation to the compatibility stage to optimize the storage layout. This disconnect leads to a dilemma in temporary storage facility management: insufficient safety redundancy and excessively high ventilation energy consumption. If high-intensity continuous ventilation is used to ensure safety, energy consumption is enormous; if intermittent ventilation is used to save energy, there is a risk of waste gas accumulation and amplified reaction risks.

[0007] Therefore, there is an urgent need for an intelligent management and control method that can break down the technical barriers between compatibility decision-making and ventilation control, and achieve two-way collaboration between storage layout optimization and precise matching of ventilation resources. Summary of the Invention

[0008] The purpose of this invention is to address the technical shortcomings of existing technologies where compatibility decision-making and ventilation control are disconnected and cannot be optimized synergistically, by providing an intelligent compatibility and ventilation control method for hazardous waste temporary storage facilities in chemical industrial parks. This invention establishes a quantitative predictive correlation between spatial compatibility schemes and waste gas generation characteristics, achieving a two-way collaborative closed-loop control where "storage layout decisions drive ventilation strategy generation, and ventilation operation data feedback optimizes storage layout," thus balancing the safety and energy efficiency of the temporary storage facility.

[0009] To achieve the above objectives, this invention provides an intelligent method for the compatibility and ventilation control of a hazardous waste temporary storage facility in a chemical industrial park, comprising the following steps:

[0010] Step S1: Construct a hazardous waste information database containing hazardous waste physicochemical property data and compatibility rules.

[0011] Hazardous waste physicochemical property data includes, but is not limited to: information on one or more hazardous characteristics of the waste, such as toxicity, corrosivity, flammability, reactivity, and infectivity, as well as physicochemical parameters such as pH value, flash point, calorific value, halogen content, heavy metal content, saturated vapor pressure, and boiling point. Compatibility rules include chemical compatibility rules, toxicity superposition rules, corrosivity conflict rules, and physical incompatibility rules.

[0012] Preferably, the hazardous waste information database also includes a hazardous waste compatibility knowledge graph. The hazardous waste compatibility knowledge graph uses different types of hazardous waste as nodes, the reaction risk types between each pair as edges, and includes weight values ​​reflecting the intensity of the reaction and the severity of the consequences.

[0013] Step S2: Obtain the attribute information of the hazardous waste to be stored, construct a three-dimensional digital twin spatial model of the temporary storage warehouse and discretize it into multiple grid cells. Based on the compatibility rule, with the goal of minimizing the proximity risk of incompatible hazardous waste in the spatial grid, perform spatial clustering of the hazardous waste to be stored to generate a spatial matching scheme that represents the correspondence between each batch of hazardous waste and each grid area.

[0014] Preferably, the objective function also includes a penalty term based on the risk level score of each grid region. The penalty term is used to increase the spatial clustering constraint strength of high-risk grid regions, preventing excessive accumulation of high-risk hazardous waste in space. The formula for calculating the risk level score is:

[0015]

[0016] in, To score the hazardous characteristics of waste, Rate the amount of waste. This represents the compatibility conflict index within the partition. The neighboring area influence coefficient, Score based on storage duration. To score the container's tightness, to These are the corresponding weighting coefficients.

[0017] More preferably, the three-dimensional digital twin spatial model of the temporary storage facility is discretized into multiple grid cells. Specifically, this includes: firstly, based on the attribute information and compatibility rules of the hazardous waste to be stored, a pre-assessment of the potential risk level of the initial coarse grid area within the temporary storage facility is performed; for areas with a potential risk level higher than a preset threshold, a first grid resolution is used for fine subdivision; for areas with a potential risk level lower than the preset threshold, a second grid resolution is used for coarse subdivision; wherein, the grid volume of the first grid resolution is smaller than the grid volume of the second grid resolution. The pre-assessment of the potential risk level can adopt a simplified form of the risk level scoring formula, for example, considering only the hazardous characteristics score of the waste. Waste quantity rating We will not consider factors that are yet to be determined, such as compatibility conflicts and the impact of neighboring areas.

[0018] Step S3: Based on the types and quantities of hazardous waste corresponding to each grid area in the spatial matching scheme, and combined with the volatile characteristics data of each hazardous waste in the hazardous waste information database, calculate the spatiotemporal distribution of exhaust gas concentration in each area within a preset time period in the future through a prediction model, and dynamically allocate differentiated target air exchange rates to different areas based on the prediction results, thereby generating and executing differentiated ventilation control strategies.

[0019] Preferably, the prediction model is a simplified computational fluid dynamics model or a neural network model trained based on historical monitoring data. The prediction model takes the volatile potential data of hazardous waste combinations in each region and the airflow organization parameters of the temporary storage tank as input, and outputs the predicted values ​​of the spatiotemporal distribution of waste gas concentration in each region.

[0020] Step S4: Collect environmental monitoring data in the temporary storage in real time, and match the environmental monitoring data with the predicted waste gas concentration time series in the time dimension and with the corresponding grid area location in the spatial dimension to perform spatiotemporal correlation comparison. The environmental monitoring data includes at least one of VOCs concentration, temperature, humidity, H2S concentration, combustible gas concentration, and zoned micro-negative pressure value.

[0021] Step S5: When the deviation of the comparison result exceeds the preset threshold, determine the type of deviation; if the deviation is the first type of deviation, trigger the adaptive adjustment of the ventilation strategy and optimize the execution parameters of the current ventilation control strategy; if the deviation is the second type of deviation, trigger the re-optimization of the compatibility scheme, feed the deviation data as a constraint to the hazardous waste information database to update the compatibility rules, and regenerate the space compatibility scheme based on the updated compatibility rules.

[0022] Preferably, the first type of deviation is a local, instantaneous deviation whose overall trend conforms to the prediction, the deviation amplitude does not exceed twice the upper limit of the preset threshold, and the duration is less than the preset duration threshold; the second type of deviation is a persistent, large-scale anomaly, or a deviation that has not converged after adaptive adjustment of the ventilation strategy, the deviation amplitude exceeds twice the upper limit of the preset threshold, or the duration is greater than or equal to the preset duration threshold.

[0023] More preferably, the matching scheme is re-optimized, and a new spatial matching scheme is generated based on the updated compatibility rules. Specifically, this includes: taking minimizing the change in storage location of the hazardous waste already in storage as the secondary optimization objective, and generating a physical location adjustment scheme under the premise of satisfying the updated compatibility rule constraints. This aims to minimize the total number of hazardous waste batches that need to be moved or to minimize the total movement path calculated based on the 3D digital twin spatial model, and outputting the physical location adjustment scheme to guide the physical location adjustment of the hazardous waste already in storage. The total movement path is based on the discretized grid map of the 3D digital twin spatial model, and a weighted A* algorithm is used to search for the collision-free shortest path from the current location to the target location for each hazardous waste batch. The total movement path is the sum of the shortest paths of all batches that need to be moved.

[0024] As a further optimization, when a request for the storage of a new batch of hazardous waste is received, a proactive ventilation pre-adjustment step is also included: obtaining the attribute information of the hazardous waste to be stored and simulating the state after the new batch of hazardous waste is added to the temporary storage in a virtual scenario of the current space matching scheme; running step S3 to generate a storage impact assessment report, which includes suggested storage partitions, predicted changes in exhaust gas concentration in each partition after storage, and expected changes in ventilation load; and adjusting the operating parameters of the temporary storage ventilation system to the pre-adjustment state in advance according to the storage impact assessment report before the actual storage operation begins.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. This invention is the first to deeply couple hazardous waste compatibility decision-making with ventilation control, establishing a positive quantitative correlation of "compatibility scheme → exhaust gas prediction → ventilation strategy" and a negative feedback loop of "monitoring data → deviation analysis → compatibility re-optimization", forming a complete two-way collaborative closed loop, effectively solving the core pain point of the separation of the two branches in the existing technology.

[0027] 2. This invention significantly reduces the spatial proximity risk of incompatible hazardous wastes through the dual constraints of spatial clustering optimization and risk scoring penalty terms. Simultaneously, prediction-based differentiated ventilation ensures that exhaust gases in each area receive collection and treatment capabilities commensurate with their generation intensity.

[0028] 3. Unlike the traditional continuous and uniform ventilation mode, this invention dynamically allocates different air exchange rates based on the predicted exhaust gas concentration in each area, achieving "ventilation on demand" and significantly reducing the total energy consumption of the ventilation system while ensuring safety.

[0029] 4. The dual-level closed-loop response mechanism enables the system to intelligently determine the type of deviation and take corresponding optimization measures: most short-term local anomalies can be quickly resolved through adaptive ventilation adjustment, and only a few root cause anomalies require triggering compatibility re-optimization, which greatly reduces the frequency of manual intervention and the amount of stacking operations in the warehouse.

[0030] 5. By updating the compatibility rules through feedback of deviation data, the system's compatibility decision knowledge base can continuously learn and evolve from actual operational data.

[0031] 6. The simulation and prediction mechanism and ventilation pre-adjustment mechanism before new waste is put into storage enable the ventilation system to adapt to changes in waste gas load in advance, effectively reducing the peak concentration of waste gas during the storage operation. Attached Figure Description

[0032] Figure 1 The overall flowchart of the intelligent matching and ventilation control method for the temporary storage of hazardous waste in chemical industrial parks provided in the embodiments of the present invention is shown.

[0033] Figure 2 This is a flowchart illustrating the deviation grading judgment and dual-path response in an embodiment of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of protection of this invention.

[0035] Example 1: Routine Warehousing and Collaborative Control Process

[0036] Please see Figure 1 This embodiment provides an intelligent configuration and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park. The complete execution process of this invention is described in detail below using a typical hazardous waste temporary storage facility in a chemical industrial park as an example. The storage facility measures 50m (length) × 30m (width) × 8m (height). It is equipped with shelving zones and independent air supply and exhaust systems, as well as a network of sensors for VOCs, temperature and humidity, H2S, combustible gases, and differential pressure.

[0037] Step S1: Construct a hazardous waste information database

[0038] First, a hazardous waste information database containing hazardous waste physicochemical property data and compatibility rules is constructed.

[0039] The sources of physicochemical property data for hazardous waste include hazardous waste information declared by enterprises, third-party testing reports, and chemical safety data sheets. Taking the three typical types of hazardous waste involved in this embodiment as examples, their key property data are shown in Table 1:

[0040] Table 1: Examples of Typical Physical and Chemical Properties of Hazardous Waste

[0041] HW06 Waste organic solvents Flammable and toxic 0.85 (High) Volatile VOCs HW34 waste acid corrosive 0.15 (Low) Acid mist corrosion HW35 Waste alkali corrosive 0.12 (low) It reacts with acid to produce heat.

[0042] Among them, the volatility potential coefficient is a normalized relative value, which can be determined comprehensively based on the physicochemical parameters of the waste, such as saturated vapor pressure and boiling point.

[0043] Compatibility rules are stored in the form of a compatibility knowledge graph. Different hazardous waste types are used as nodes, and the reaction risk types between each pair are used as edges. Each edge is assigned a weight value (ranging from 0 to 1, with higher values ​​indicating greater incompatibility) reflecting the intensity of the reaction and the severity of its consequences. For example, an edge for "acid-base neutralization reaction" is established between HW34 (waste acid) and HW35 (waste alkali), with a weight value of 0.9.

[0044] Step S2: 3D digital twin spatial modeling and spatial clustering matching

[0045] After obtaining the attribute information of the hazardous waste to be stored, a three-dimensional digital twin space model of the temporary storage warehouse is constructed. This model includes spatial elements such as the warehouse's geometric dimensions, shelf layout, column positions, ventilation duct routing, and sensor placement locations. The three-dimensional digital twin space model can be constructed using BIM modeling software, a 3D engine, or conventional 3D modeling tools.

[0046] The 3D spatial model is discretized into multiple mesh elements. This embodiment employs an adaptive resolution strategy, and the specific process is as follows:

[0047] First, an initial uniform coarse grid is created (e.g., a grid side length of 2m). Based on the attribute information of the hazardous waste to be stored, a pre-assessment of the potential risk level of the initial coarse grid area within the temporary storage facility is conducted. The pre-assessment uses a simplified risk scoring formula, considering only the hazardous characteristics score of the waste. And quantity score ,in Take 0.6, Take 0.4. For High-potential-risk areas are finely divided using a first-level grid resolution (0.5m × 0.5m × 0.5m); for For low-potential-risk areas, a coarse division with a second grid resolution (2m×2m×2m) is maintained. A preset threshold of 0.7 is used, based on analysis of historical incident data from this temporary storage facility. This involves statistically analyzing areas where minor leaks or reaction events occurred within the past three years, and then using a simplified risk score. All values ​​were higher than 0.65. To allow for a certain safety margin, the threshold was set to 0.7.

[0048] Based on the optimized grid, the final spatial clustering calculation is performed. At this point, the complete risk level score is used. As a penalty. Risk level rating. The calculation formula is:

[0049]

[0050] The meanings and specific calculation methods of each factor are as follows:

[0051] (Waste Hazard Characteristic Scoring): Referring to the "Hazardous Waste Identification Standard", a base score of 0.25 is set for each hazardous characteristic (toxicity T, flammability I, reactivity R, corrosivity C). When waste has multiple characteristics, the scores are accumulated, but the total score shall not exceed 1.0. In this embodiment, HW06 is scored as 0.7, HW34 as 0.6, and HW35 as 0.5.

[0052] (Waste quantity score): Determined based on the proportion of the total amount of hazardous waste to be stored in the grid area to the maximum allowable storage capacity of the grid, with a value range of 0-1.

[0053] (Intra-partition compatibility conflict index): For grids Intra-coexisting For each type of waste, query the compatibility knowledge graph to obtain the incompatibility weights between each pair of waste. Then the grid's That is, the average of the incompatible weights. If ,but .

[0054] (Neighboring cell influence coefficient): Defines the grid The neighborhood of a is the 6 adjacent grids (up, down, left, right, front, back) that share the same surface with it. ,in It is a grid Internal waste and neighborhood grid The maximum incompatibility weight between internal wastes.

[0055] (Storage Duration Rating): Normalized based on the expected storage days, scores above 0.8 are given for storage exceeding 30 days, and scores below 0.3 are given for storage within 7 days.

[0056] (Container sealing rating): 0.2 for sealed containers, 0.8 for open containers, and 0.5 for containers with a breather valve.

[0057] Weighting coefficient to The determination of the factors was performed using the Analytic Hierarchy Process (AHP): five hazardous waste management experts were invited to conduct pairwise comparisons and scores on the six factors, constructing a judgment matrix. After consistency testing, the weight vector was obtained. In this embodiment, the values ​​are as follows: , , , , , Etc. Generally speaking, (Hazardous characteristics) and (Compatibility conflict) has the greatest impact on the risk level, and a value of 0.2-0.3 is recommended; (Storage duration) and The impact on (sealing) is relatively small; a value of 0.1-0.15 is recommended. The above recommended range allows those skilled in the art to make appropriate configurations in practical applications.

[0058] Based on compatibility rules, spatial clustering calculations are performed on the hazardous waste to be stored, aiming to minimize the proximity risk of incompatible hazardous wastes in the spatial grid. The objective function is designed as follows:

[0059]

[0060] The first item is the adjacent risk item: Hazardous waste and Incompatible weights between them The first term represents the Euclidean distance between the two in space. The second term is the risk score penalty. For grid Risk level rating This is the penalty coefficient (0.3 in this example).

[0061] An improved K-means spatial clustering algorithm was used to solve the above objective function to generate spatial matching schemes. For example, HW34 (waste acid) and HW35 (waste alkali) were assigned to grid regions that were far apart from each other, while HW06 (waste organic solvent) was placed in a central region with good ventilation.

[0062] Step S3: Prediction of the spatiotemporal distribution of exhaust gas concentration and generation of differentiated ventilation strategies

[0063] Based on the types and quantities of hazardous waste corresponding to each grid area in the spatial allocation scheme, and combined with volatile characteristic data, the spatiotemporal distribution of exhaust gas concentration in each area within a preset time period is calculated using a prediction model.

[0064] This embodiment provides two methods for implementing the prediction model:

[0065] Method 1 (CFD Simplified Model): Using the hazardous waste volatilization potential coefficient and temporary storage airflow organization parameters of each grid region as input, a rapid solution is obtained based on mass conservation and component transport equations. The core simplification logic of the CFD simplified model is: ignoring the local turbulence effects of minor obstacles within the storage area, using ventilation opening boundaries and shelf boundaries as fixed boundary conditions, and using hazardous waste volatilization sources as area source terms, adopting standard... The turbulence model is solved in steady state with a solution step size of 10 minutes. The solution time for a single operating condition is controlled within 5 minutes to meet the requirements of real-time prediction. The model outputs the VOCs concentration values ​​for each grid node.

[0066] Method 2 (Neural Network Model): Using historical operational data as training samples. The training data comes from the operational records of this temporary repository over the past 12 months, collecting a total of 12,000 valid samples, which are divided into training, validation, and test sets in an 8:1:1 ratio. A three-layer backpropagation (BP) neural network is constructed. The input layer contains the hazardous waste type codes, quantities, volatile matter potential coefficients, and airflow parameters for each region, and the input features are standardized using Z-scores. The hidden layer has 12 nodes. The output layer contains the predicted concentration values ​​for each region. The mean squared error (MSE) loss function is used, the Adam algorithm is used as the optimizer, the learning rate is set to 0.001, and the number of training iterations is 1000. Training is stopped early when the validation set loss value does not decrease for 50 consecutive iterations. After training, the model's mean absolute percentage error on the test set is 8.3%, which meets the accuracy requirements for engineering applications.

[0067] After obtaining the spatiotemporal distribution of VOCs concentration in each region over the next 4 hours using a prediction model, the target air exchange rate is dynamically assigned to different regions based on the prediction results:

[0068] Predicted peak concentration > 10 mg / m³: Target air exchange rate 12 times / hour

[0069] Predicted peak concentration 5-10 mg / m³: Target air exchange rate 8 times / hour

[0070] Predicted peak concentration < 5 mg / m³: Target air exchange rate 4 times / hour

[0071] Based on the target air exchange rate, corresponding valve opening control commands and supply and exhaust fan operating frequency curves are generated to drive the ventilation system to execute differentiated ventilation strategies.

[0072] Step S4: Spatiotemporal correlation comparison between environmental monitoring data and predicted values

[0073] The system collects environmental monitoring data (including VOCs concentration, temperature, humidity, H2S concentration, combustible gas concentration, and zoned micro-negative pressure value) from various sensors in the temporary storage area in real time, with a sampling frequency of once per minute.

[0074] The specific method for spatiotemporal correlation comparison is as follows: align the timestamp of each monitoring data point with the corresponding time in the predicted time series, match the spatial coordinates of the sensor with the grid area location, and calculate the deviation between the measured value and the predicted value at the same spatiotemporal point. :

[0075]

[0076] when When the deviation exceeds the preset threshold (30% in this embodiment), it is determined that the deviation exceeds the threshold.

[0077] Step S5: Deviation classification and dual-path closed-loop response

[0078] When the deviation of the comparison results exceeds a preset threshold, the deviation grading judgment process is initiated. (Combined with...) Figure 2 As shown, the specific judgment logic is as follows:

[0079] The system determines whether the deviation exceeds twice the upper limit of the preset threshold (i.e., 60%) and whether the duration reaches the preset duration threshold (30 minutes). The preset duration threshold of 30 minutes is determined based on a combination of the ventilation system response time and the exhaust gas diffusion time constant: after the ventilation system in a temporary storage facility is adjusted, it generally takes about 15-20 minutes for the concentration field in the space to reach a new steady state; considering the sensor sampling period (1 minute) and the system judgment delay, setting the duration threshold to 30 minutes can effectively distinguish between temporary fluctuations and persistent anomalies.

[0080] Type 1 deviation: Deviation amplitude ≤ 60% and duration < 30 minutes. Triggers adaptive adjustment of ventilation strategy: Automatically increases the air exchange rate in the deviation area by one level temporarily, or adjusts the opening of the corresponding air valve by 10%-20%, continuously monitoring until the deviation falls back to within the threshold and then restores the original strategy.

[0081] Type II deviation: Deviation amplitude > 60%, or duration ≥ 30 minutes, or deviation still not converged after adaptive adjustment using the above ventilation strategy. Triggering compatibility scheme re-optimization:

[0082] First, the deviation data is fed back to the hazardous waste information database as a constraint to update the compatibility rules. The specific rule update algorithm is as follows: If the monitoring data continuously shows that the VOCs concentration in a certain area far exceeds the predicted value, and hazardous waste A and hazardous waste B are stored in that area, the system performs the following steps:

[0083] (1) Calculate the average extent to which the VOCs concentration exceeds the predicted value during the period of deviation. ;

[0084] (2) Query whether there is a reaction edge between A and B in the compatibility knowledge graph;

[0085] (3) If it does not exist, add a new edge, mark the edge type as "suspected volatile reaction", and set the initial weight. ;

[0086] (4) If an edge already exists, then follow the formula Update the weight values, where The smoothing coefficient is set to 0.7. Wherein, This is the preset control threshold for VOCs concentration in the corresponding area.

[0087] Then, based on the updated compatibility rules, the spatial clustering calculation in step S2 is re-executed to generate an optimized spatial matching scheme.

[0088] In this process, minimizing the change in storage location of already-stored hazardous waste is the secondary optimization objective. Specifically, among several candidate matching schemes generated while satisfying compatibility constraints, the total number of hazardous waste batches requiring movement and the total movement path length calculated based on a 3D digital twin spatial model are calculated for each scheme. The total movement path is based on a discretized grid map of the 3D digital twin spatial model, and a weighted A* algorithm is used to search for the shortest collision-free path from the current location to the target location for each hazardous waste batch (path weights comprehensively consider movement distance and number of turns). The total movement path is the sum of the shortest paths for all batches requiring movement. The scheme with the fewest moving batches or the shortest total movement path is selected as the final physical location adjustment scheme to be executed.

[0089] Finally, output the physical location adjustment plan, such as: "Batch 001 (HW34) is moved from area A3 to area B2; Batch 005 (HW06) is moved from area B1 to area C3", to guide on-site operators in adjusting the physical location of hazardous waste that has been put into storage.

[0090] Through the above-mentioned two-level closed-loop response mechanism, this invention achieves coordinated control of "tactical-level ventilation self-adjustment" and "strategic-level compatibility optimization".

[0091] Example 2: Proactive Ventilation Pre-conditioning Process

[0092] Upon receiving a new batch of hazardous waste storage request, proactive ventilation pre-adjustment is performed:

[0093] First, obtain the attribute information of this batch of hazardous waste. Then, in the virtual scenario of the currently executed spatial matching scheme, simulate the state after adding this new batch of hazardous waste to the temporary storage.

[0094] Run the prediction model in step S3 to generate an impact assessment report on the storage entry. This report includes: recommended storage zoning, predicted changes in exhaust gas concentration in each zone after storage (displayed in tabular or heat map form), and expected changes in ventilation load.

[0095] Before the actual warehousing operation begins (e.g., 15-30 minutes before the forklift transports waste into the temporary storage area), the ventilation system operating parameters should be pre-adjusted to the pre-adjusted state based on the warehousing impact assessment report. For example, if the report shows that the predicted VOCs concentration in area A will increase from 4 mg / m³ to 9 mg / m³ after warehousing, the target air exchange rate in area A should be increased from 4 times / hour to 8 times / hour in advance, and the corresponding air valves should be pre-opened to the appropriate opening degree.

[0096] Through this proactive pre-adjustment mechanism, the ventilation system can adapt to upcoming changes in exhaust gas load before new waste is put into storage, effectively suppressing peak exhaust gas concentrations during storage operations.

[0097] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Various changes made within the scope of knowledge possessed by those skilled in the art without departing from the concept of the present invention still fall within the scope of protection of the present invention.

Claims

1. An intelligent method for the matching and ventilation control of hazardous waste temporary storage facilities in chemical industrial parks, characterized in that, Includes the following steps: S1: Construct a hazardous waste information database that includes data on the physical and chemical properties of hazardous waste and compatibility rules; S2: Obtain the attribute information of the hazardous waste to be stored, construct a three-dimensional digital twin space model of the temporary storage warehouse and discretize it into multiple grid units. Based on the compatibility rules, with the goal of minimizing the proximity risk of incompatible hazardous waste in the spatial grid, perform spatial clustering on the hazardous waste to be stored to generate a spatial matching scheme that represents the correspondence between each batch of hazardous waste and each grid area. S3: Based on the types and quantities of hazardous waste corresponding to each grid area in the spatial matching scheme, and combined with the volatile characteristics data of each hazardous waste in the hazardous waste information database, the spatiotemporal distribution of exhaust gas concentration in each area within a preset time period is calculated by the prediction model, and differentiated target air exchange rates are dynamically allocated to different areas based on the prediction results, thereby generating and executing differentiated ventilation control strategies. S4: Collect environmental monitoring data in the temporary storage in real time, and match the environmental monitoring data with the predicted exhaust gas concentration time series in the time dimension and with the corresponding grid area location in the spatial dimension to perform spatiotemporal correlation comparison. S5: When the deviation of the comparison result exceeds the preset threshold, determine the type of deviation; If the deviation is a local, instantaneous deviation and the overall trend conforms to the predicted first type of deviation, then the ventilation strategy is triggered to adaptively adjust and optimize the execution parameters of the current ventilation control strategy. If the deviation is persistent, widespread, or a second type of deviation that has not converged after adaptive adjustment of the ventilation strategy, then the compatibility scheme re-optimization is triggered. The deviation data is fed back to the hazardous waste information database as a constraint to update the compatibility rules, and the space compatibility scheme is regenerated based on the updated compatibility rules.

2. The intelligent matching and ventilation control method for the hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, When performing spatial clustering with the goal of minimizing the proximity risk of incompatible hazardous waste in the spatial grid, the objective function also includes a penalty term based on the risk level score of each grid region. The penalty term is used to increase the spatial clustering constraint strength of high-risk grid regions and avoid the spatial aggregation of high-risk hazardous waste. The formula for calculating the risk level score is as follows: ， in, To score the hazardous characteristics of waste, Rate the amount of waste. This represents the compatibility conflict index within the partition. The neighboring area influence coefficient, Score based on storage duration. To score the container's tightness, to These are the corresponding weighting coefficients.

3. The intelligent matching and ventilation control method for the hazardous waste temporary storage facility in a chemical industrial park according to claim 2, characterized in that, The discretization of the three-dimensional digital twin spatial model of the temporary storage into multiple grid cells specifically includes: The current risk level of each grid area is assessed based on the aforementioned risk level score. For areas with a risk level higher than a preset threshold, a first grid resolution is used for fine-grained subdivision; for areas with a risk level lower than a preset threshold, a second grid resolution is used for coarse-grained subdivision; wherein, the grid volume of the first grid resolution is smaller than the grid volume of the second grid resolution.

4. The intelligent matching and ventilation control method for the hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, The prediction model is a simplified computational fluid dynamics model or a neural network model trained based on historical monitoring data. The prediction model takes the volatile potential data of hazardous waste combinations in each region and the airflow organization parameters of the temporary storage tank as inputs, and outputs the predicted values ​​of the spatiotemporal distribution of waste gas concentration in each region.

5. The intelligent matching and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, The deviation amplitude of the first type of deviation does not exceed twice the upper limit of the preset threshold, and the duration is less than the preset duration threshold; the deviation amplitude of the second type of deviation exceeds twice the upper limit of the preset threshold, or the duration is greater than or equal to the preset duration threshold.

6. The intelligent matching and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, The triggering of compatibility scheme re-optimization, and the regeneration of spatial compatibility schemes based on the updated compatibility rules, specifically includes: With minimizing the changes in the storage location of hazardous waste already in storage as a secondary optimization objective, and under the premise of satisfying the updated compatibility rule constraints, a physical location adjustment scheme is generated to minimize the total number of hazardous waste batches that need to be moved or to minimize the total movement path calculated based on the three-dimensional digital twin space model. The physical location adjustment scheme is then output to guide the physical location adjustment of the hazardous waste already in storage.

7. The intelligent matching and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, The hazardous waste information database includes a hazardous waste compatibility knowledge graph, which uses different types of hazardous waste as nodes, the reaction risk types between each pair as edges, and includes weight values ​​that reflect the intensity of the reaction and the severity of the consequences.

8. The intelligent matching and ventilation control method for a hazardous waste temporary storage facility in a chemical industrial park according to claim 1, characterized in that, When a new batch of hazardous waste is received for warehousing, a proactive ventilation pre-conditioning step is also included: Obtain the attribute information of the hazardous waste to be put into storage, and simulate the state of the new batch of hazardous waste after it is added to the temporary storage in the virtual scenario of the current spatial matching scheme; Running step S3 generates an inbound impact assessment report, which includes suggested storage partitions, predicted changes in exhaust gas concentration in each partition after inbound storage, and expected changes in ventilation load. Before the actual warehousing operation begins, the operating parameters of the temporary storage ventilation system are adjusted to a pre-adjusted state according to the warehousing impact assessment report.

Images