Chemical exposure dynamic simulation evaluation method and system for wastewater treatment process

Abstract

The invention provides a chemical exposure dynamic simulation evaluation method and system for a wastewater treatment process, and the method comprises the steps: firstly obtaining a chemical adding data set and a real-time monitoring data set of a target wastewater treatment process, and then carrying out the dynamic exposure parameter extraction; generating a dynamic exposure parameter set containing chemical diffusion rate, exposure contact time and environment interaction influence parameters, and then performing multi-stage iterative simulation on the dynamic exposure parameter set based on a preset dynamic simulation strategy network; a chemical exposure dynamic simulation result containing exposure concentration distribution and exposure risk probability of each processing node is obtained, exposure risk assessment is carried out according to the chemical exposure dynamic simulation result, and a chemical exposure optimization strategy set containing dosing concentration adjustment, dosing time sequence optimization and emergency response triggering strategies is generated; and feeding back to a wastewater treatment control system so as to dynamically adjust the chemical adding process and realize more accurate and safer wastewater treatment chemical management.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and system for dynamic simulation assessment of chemical exposure in a wastewater treatment process. Background Art

[0002] In wastewater treatment, chemicals play a vital role in removing pollutants and purifying water. However, improper chemical addition and management can pose serious environmental and health risks, such as excessive chemical residues, damage to aquatic ecosystems, and potential harm to operators. Therefore, there is an urgent need to accurately assess and optimize chemical exposure management in wastewater treatment processes.

[0003] Existing technologies primarily rely on static models and empirical parameters to determine chemical dosage. These models are unable to adapt to real-time changes in water quality and equipment operating conditions during wastewater treatment. In actual treatment, factors such as wastewater flow, pollutant composition and concentration, and equipment performance fluctuate constantly. Static models are unable to adjust dosing strategies based on this dynamic information, making it difficult to accurately match chemical dosage to actual needs. This can easily lead to over- or underdosing and increase the risk of chemical exposure.

[0004] At the same time, existing chemical exposure assessments are mostly post-tests, assessing exposure through laboratory analysis of water samples. This approach fails to monitor chemical exposure in real time, making it difficult to identify potential risks and implement effective measures promptly. Once a problem is detected, environmental damage or health hazards may have already occurred. Furthermore, due to the lack of dynamic simulation and predictive capabilities, it is impossible to develop optimization strategies to adjust the dosing process in advance. Summary of the Invention

[0005] In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides a dynamic simulation assessment method for chemical exposure in a wastewater treatment process, the method comprising:

[0006] Obtain a chemical dosing data set and a real-time monitoring data set for the target wastewater treatment process, wherein the chemical dosing data set includes chemical type identification, dosing concentration, and dosing time interval at different treatment stages, and the real-time monitoring data set includes water quality parameters, equipment operating parameters, and chemical residual concentrations at each treatment stage;

[0007] Performing dynamic exposure parameter extraction processing on the chemical addition data set and the real-time monitoring data set to generate a dynamic exposure parameter set for each treatment stage, the dynamic exposure parameter set including a chemical diffusion rate parameter, an exposure contact time parameter, and an environmental interaction parameter;

[0008] Based on a preset dynamic simulation strategy network, a multi-stage iterative simulation process is performed on the dynamic exposure parameter set to generate a dynamic simulation result of chemical exposure for the target wastewater treatment process, wherein the dynamic simulation result of chemical exposure includes the exposure concentration distribution and exposure risk probability of each treatment node;

[0009] Performing exposure risk assessment processing based on the chemical exposure dynamic simulation results to generate a chemical exposure optimization strategy set for the target wastewater treatment process, wherein the chemical exposure optimization strategy set includes a dosing concentration adjustment strategy, a dosing timing optimization strategy, and an emergency response triggering strategy;

[0010] The chemical exposure optimization strategy set is fed back to the wastewater treatment control system to trigger dynamic adjustment operations of the chemical dosing process.

[0011] On the other hand, an embodiment of the present invention also provides a dynamic simulation and assessment system for chemical exposure in a wastewater treatment process, comprising a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0012] Based on the above aspects, the embodiment of the present invention can accurately depict the key dynamic parameters of chemicals such as diffusion rate, exposure contact time and environmental interaction at different treatment stages by integrating the chemical addition data and real-time monitoring data of the target wastewater treatment process, and then perform multi-stage iterative simulation based on the preset dynamic simulation strategy network, thereby achieving a comprehensive, dynamic and refined simulation of the chemical exposure process, effectively overcoming the limitation of traditional static assessment methods that cannot reflect the complex dynamic changes in the actual treatment process. By generating dynamic simulation results of chemical exposure that include the exposure concentration distribution and exposure risk probability of each treatment node, the actual situation and potential risks of chemical exposure in the wastewater treatment process can be truly reflected. On this basis, a set of chemical exposure optimization strategies is further generated, covering multi-dimensional strategies such as addition concentration adjustment, addition timing optimization and emergency response triggering, providing a targeted and operational dynamic adjustment scheme for the wastewater treatment control system, realizing closed-loop management from simulation evaluation to optimization control, significantly improving the safety, efficiency and intelligence level of chemical management in the wastewater treatment process, and effectively reducing the risk of chemical exposure. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a schematic diagram of the execution flow of the dynamic simulation assessment method for chemical exposure in a wastewater treatment process provided by an embodiment of the present invention.

[0014] Figure 2Schematic diagram of exemplary hardware and software components of a dynamic simulation and assessment system for chemical exposure in a wastewater treatment process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The present invention will be described in detail below with reference to the accompanying drawings. Figure 1 This is a flow chart of a method for dynamic simulation assessment of chemical exposure in a wastewater treatment process provided by an embodiment of the present invention. The method for dynamic simulation assessment of chemical exposure in a wastewater treatment process is introduced in detail below.

[0016] Step S110: Obtain a chemical addition data set and a real-time monitoring data set for the target wastewater treatment process, wherein the chemical addition data set includes the chemical type identification, addition concentration, and addition time interval of different treatment stages, and the real-time monitoring data set includes the water quality parameters, equipment operation parameters, and chemical residual concentration of each treatment stage.

[0017] In order to dynamically simulate and assess chemical exposure during wastewater treatment, it is necessary to obtain a data set related to the target wastewater treatment process. The target wastewater treatment process often includes multiple treatment stages, with different chemicals added at different stages based on the wastewater characteristics and treatment requirements.

[0018] The chemical dosing data set covers the chemical type identification, dosing concentration, and dosing interval for different treatment stages. Chemical type identification is used to distinguish different types of chemicals. For example, different types of chemicals such as flocculants, disinfectants, and pH regulators may be used in wastewater treatment. Each chemical has its own unique chemical properties and treatment effects. Dosing concentration refers to the amount of chemical added at each treatment stage. Different dosing concentrations will have a significant impact on the wastewater treatment effect and chemical exposure. Dosing interval reflects the frequency of chemical addition, which is closely related to the wastewater treatment rhythm and the action time of the chemical.

[0019] Real-time monitoring data sets include water quality parameters, equipment operating parameters, and residual chemical concentrations at each treatment stage. Water quality parameters reflect the quality of wastewater at each treatment stage. Common water quality parameters include pH, turbidity, dissolved oxygen, and chemical oxygen demand. Changes in these parameters provide a visual indicator of the effectiveness and progress of wastewater treatment. Equipment operating parameters, such as pump flow rate, agitator speed, and treatment tank temperature, reflect the operating status of treatment equipment. The stability of equipment operating parameters directly impacts the efficiency and quality of wastewater treatment. Residual chemical concentrations directly indicate the amount of chemicals remaining in the wastewater after treatment.

[0020] Step S111: extracting chemical addition records within a preset time window from a historical database of the wastewater treatment control system, wherein the chemical addition records include a chemical type identifier, an addition concentration, an addition timestamp, and an addition equipment identifier.

[0021] Wastewater treatment control systems keep detailed records of chemical dosing and store these records in a historical database. To obtain accurate and targeted chemical dosing data, records within a preset time window must be extracted from the historical database. Determining this preset time window requires considering multiple factors, such as the stability of the wastewater treatment process, the timeliness of the data, and its availability.

[0022] When extracting records, the set query statement or data extraction tool is used to filter out qualified records according to the preset time window. The chemical type identifier is a code or name that uniquely identifies each chemical, which facilitates the subsequent classification and analysis of the addition of different chemicals. The addition concentration records the specific content of each chemical added, which can be expressed in letters such as C1, C2, C3, etc. Different subscripts represent different addition times. The addition timestamp is accurate to a specific time point, represented by t1, t2, t3, etc., and is used to accurately determine the time sequence and interval of chemical addition. The addition equipment identifier indicates the equipment number that performs the addition operation, which helps to analyze the addition effect and reliability of different equipment.

[0023] Step S112: performing outlier cleaning processing on the dosing records to generate a cleaned chemical dosing data set, wherein the outlier cleaning processing includes removing records with dosing concentrations exceeding a preset safety range or with discontinuous dosing timestamps.

[0024] Since errors or anomalies may occur during data collection and recording, it is necessary to clean the extracted dosing records to ensure the accuracy and reliability of the data. The preset safety range is determined based on the properties of the chemical, wastewater treatment requirements, and relevant safety standards.

[0025] For dosing concentration, assuming the preset safety range is [C_min, C_max], if the dosing concentration C_i in a record is less than C_min or greater than C_max, the record is considered an outlier and needs to be removed. For example, for a chemical with a specific toxicity, too high a dosing concentration may cause harm to the environment and subsequent treatment, while too low may not achieve the desired treatment effect. Therefore, dosing concentration records that exceed the safety range do not conform to the actual situation and should be excluded.

[0026] Regarding dosing timestamps, under normal circumstances, dosing operations should follow a set time pattern, meaning that the dosing timestamps should be continuous. If dosing timestamps are discontinuous, for example, if dosing should occur at regular intervals but the time interval between two additions is significantly outside the normal range and there is no reasonable explanation, then this record is also considered an outlier and needs to be removed. This outlier cleaning process yields a cleaned chemical dosing data set.

[0027] Step S113: Acquire water quality parameters, equipment operating parameters, and chemical residual concentrations corresponding to the preset time window from the real-time monitoring equipment to generate an original monitoring data set.

[0028] Real-time monitoring equipment collects water quality parameters, equipment operating parameters, and chemical residual concentrations at each stage of the wastewater treatment process. To match and correlate data with chemical dosing data, it's necessary to obtain monitoring data from the real-time monitoring equipment corresponding to a pre-set time window.

[0029] Water quality parameters are typically monitored using various sensors, such as pH sensors for measuring the pH value of wastewater, turbidity sensors for detecting the turbidity of wastewater, and dissolved oxygen sensors for monitoring the dissolved oxygen content in water. These sensors measure water quality parameters at set intervals and record the measurement results.

[0030] The acquisition of equipment operating parameters depends on the monitoring devices installed on the processing equipment. For example, the flow sensor can measure the flow rate of the water pump, the temperature sensor can monitor the temperature of the treatment tank, and the speed sensor can obtain the speed of the agitator.

[0031] Chemical residual concentrations are typically measured using chemical analysis methods or sensor technology to accurately determine the remaining levels of chemicals in wastewater. By collecting these monitoring data corresponding to a pre-set time window, a raw monitoring data set is generated.

[0032] Step S114: performing time synchronization processing on the original monitoring data set to generate a time-aligned real-time monitoring data set, wherein the time synchronization processing includes: interpolating and aligning water quality parameters, equipment operating parameters, and residual concentrations according to injection timestamps.

[0033] Since the monitoring time of water quality parameters, equipment operating parameters and chemical residual concentration may not be completely consistent with the chemical addition time, in order to match the monitoring data with the addition data in the time dimension, the original monitoring data set needs to be time synchronized.

[0034] Specifically, water quality parameters, equipment operating parameters, and residual concentrations are interpolated and aligned based on the dosing timestamps. For each dosing timestamp, the parameter value at the closest monitoring time point is found in the original monitoring data set. If there is no corresponding monitoring data at the dosing timestamp, i.e., data is missing, interpolation is required.

[0035] Step S1141: for each dosing timestamp, extract the water quality parameters and equipment operation parameters at the corresponding time point from the original monitoring data set.

[0036] When performing time synchronization, for each dosing timestamp, the original monitoring data set is first searched for the water quality parameters and equipment operating parameters at the corresponding time point. Because the monitoring data collection time and dosing time may differ, it may not be possible to directly find an exact match. In this case, it is necessary to find the monitoring time point closest to the dosing timestamp and extract the water quality parameters and equipment operating parameters at that time point.

[0037] Step S1142: If the water quality parameters or equipment operating parameters are missing at the injection timestamp, the change rate of the equipment operating parameters at adjacent time points is first detected. When the change rate exceeds the preset threshold, the nearest neighbor interpolation method is used. Otherwise, the linear interpolation method is used to generate the supplementary data to generate the supplemented water quality parameters and equipment operating parameters.

[0038] When water quality parameters or equipment operating parameters are missing at the injection timestamp, an appropriate interpolation method is required to generate supplementary data. The specific steps are as follows: First, detect the rate of change of the equipment operating parameters at adjacent time points. The rate of change of the equipment operating parameters can be obtained by calculating the ratio of the difference between the equipment operating parameters at adjacent time points to the time interval. Assuming that two adjacent time points are t_j and t_{j+1}, and the corresponding equipment operating parameters are P_j and P_{j+1}, then the rate of change of the equipment operating parameters R = (P_{j+1}-P_j) / (t_{j+1}-t_j).

[0039] The calculated rate of change R is then compared with the preset threshold R_threshold. When the rate of change R exceeds the preset threshold R_threshold, it indicates that the equipment operating status has significantly changed between adjacent time points. In this case, the nearest neighbor interpolation method is used. The nearest neighbor interpolation method selects the parameter value at the monitoring time point closest to the injection timestamp as the supplementary data. For example, if the injection timestamp t_i is between t_j and t_{j+1}, and |t_i-t_j| < |t_i-t_{j+1}|, the parameter value at time t_j is selected as the supplementary data.

[0040] When the rate of change R does not exceed the preset threshold R_threshold, it indicates that the equipment's operating status is relatively stable, and linear interpolation is used at this time. The linear interpolation method calculates the parameter value at the injection timestamp using a linear proportional relationship based on the parameter values and time intervals of two adjacent monitoring time points. Assuming that the injection timestamp t_i is between t_j and t_{j+1}, and the corresponding parameter values are P_j and P_{j+1}, the linear interpolation formula can be used to calculate the completed parameter value at the injection timestamp t_i: P_i = P_j + (P_{j+1} - P_j) * (t_i - t_j) / (t_{j+1} - t_j). Through this process, the completed water quality parameters and equipment operating parameters are generated.

[0041] Step S1143: Smoothing and filtering are performed on the completed water quality parameters and equipment operation parameters, and the smoothed and filtered water quality parameters, equipment operation parameters and corresponding chemical residual concentrations are integrated according to the injection timestamp to generate a real-time monitoring data set with a consistent time dimension.

[0042] To further improve data quality and stability, smoothing filters are applied to the completed water quality parameters and equipment operating parameters. Smoothing filters can remove noise and fluctuations in the data, making it smoother and more stable. Common smoothing filter methods include moving average filtering and Gaussian filtering.

[0043] Taking moving average filtering as an example, assuming an N-point moving average filter is used, for each parameter value at a time point, the average of the parameter values at the N time points before and after it is taken as the smoothed parameter value at that time point. For example, for the pH value of the water quality parameter, at time point t_i, its smoothed pH value pH_i_smooth = (pH_{i-(N-1) / 2}+...+pH_i+...+pH_{i+(N-1) / 2}) / N (assuming N is an odd number).

[0044] After smoothing and filtering, the filtered water quality parameters, equipment operating parameters, and corresponding chemical residual concentrations are integrated according to the injection timestamp. This creates a real-time monitoring data set with consistent temporal dimensions, ensuring accurate temporal correspondence between monitoring data and chemical injection data, providing a reliable data foundation for subsequent dynamic exposure parameter extraction and simulation assessment.

[0045] Step S120: performing dynamic exposure parameter extraction processing on the chemical addition data set and the real-time monitoring data set to generate a dynamic exposure parameter set for each processing stage, wherein the dynamic exposure parameter set includes a chemical diffusion rate parameter, an exposure contact time parameter, and an environmental interaction parameter.

[0046] After obtaining the chemical dosing data set and the real-time monitoring data set, in order to gain a deeper understanding of chemical exposure during the wastewater treatment process, it is necessary to extract dynamic exposure parameters from these two data sets to generate a dynamic exposure parameter set for each treatment stage. These dynamic exposure parameters can reflect the diffusion of chemicals during the wastewater treatment process, their contact time with wastewater, and their interactions with environmental factors, and are crucial for accurately simulating the exposure dynamics of chemicals.

[0047] Step S121: performing time series alignment processing on the chemical addition concentrations and addition time intervals in the chemical addition data set to generate a chemical addition time series feature.

[0048] The concentration and time interval of chemical dosing data are important information reflecting the chemical dosing situation. To better analyze the dosing patterns and temporal characteristics of chemicals, it is necessary to perform time series alignment on the concentration and time interval.

[0049] First, the dose concentrations and dose intervals are sorted by the dose timestamp to ensure the data is in the correct chronological order. Then, the dose concentrations are interpolated based on the fixed time intervals so that each time point has a corresponding dose concentration value. This creates a continuous dose concentration time series.

[0050] For the dosing interval, we also use a fixed time interval as a benchmark to calculate the dosing interval corresponding to each time point. If no dosing occurs at a certain time point, the dosing interval is calculated as the time difference between the last dosing operation and that time point. This process generates a chemical dosing time series feature that clearly demonstrates the temporal patterns of chemical dosing and concentration changes.

[0051] Step S122: performing environmental state correlation analysis on the water quality parameters and equipment operation parameters in the real-time monitoring data set to generate environmental state correlation features.

[0052] The water quality parameters and equipment operating parameters in the real-time monitoring data set can reflect the environmental status of the wastewater treatment process. In order to explore the correlation between these parameters, it is necessary to perform environmental status correlation analysis on the water quality parameters and equipment operating parameters.

[0053] First, data preprocessing is performed on water quality parameters and equipment operating parameters, including normalization to eliminate the dimensional effects of different parameters. Normalization can be done using common methods, such as subtracting the minimum value from each parameter value and then dividing it by the difference between the maximum and minimum values.

[0054] Next, correlation analysis is used to calculate the correlation coefficient between water quality parameters and equipment operating parameters. The correlation coefficient measures the degree of linear relationship between two parameters. For example, by calculating the correlation coefficient between pH and agitator speed, it is possible to determine whether changes in pH are related to the agitator's operating status.

[0055] Based on the correlation coefficient, parameter pairs with strong correlations are selected. For these highly correlated parameter pairs, a correlation model can be constructed to describe their relationship. This process generates environmental state correlation features that reflect the mutual influence and correlation between water quality parameters and equipment operating parameters during the wastewater treatment process.

[0056] Step S123: performing diffusion path modeling based on the chemical addition timing characteristics and the environmental state association characteristics to obtain the chemical diffusion rate parameters, wherein the diffusion path modeling includes determining the diffusion path weight based on the correlation between the time gradient change of the addition concentration and the spatial distribution change of the water quality parameters.

[0057] After obtaining the chemical addition time series characteristics and environmental state correlation characteristics, diffusion path modeling is required to determine the diffusion path and rate of chemicals in wastewater.

[0058] First, analyze the time gradient of the dosed concentration. This can be calculated by calculating the ratio of the difference in dosed concentration between adjacent time points to the time interval. For example, for dosed concentrations C_i and C_{i+1}, with a time interval of Δt, the time gradient of the dosed concentration is ΔC / Δt = (C_{i+1} - C_i) / Δt.

[0059] At the same time, the spatial distribution of water quality parameters is analyzed. This can be determined by setting up monitoring points at different locations within the wastewater treatment tank, measuring the water quality parameters at these locations, and calculating the ratio of the water quality parameter difference to the distance between adjacent monitoring points. For example, for the values of water quality parameter P at locations x_j and x_{j+1}, P_j and P_{j+1}, with a distance Δx, the spatial distribution change of the water quality parameter is ΔP / Δx = (P_{j+1} - P_j) / Δx.

[0060] Next, calculate the correlation between the temporal gradient of the dose concentration and the spatial distribution of the water quality parameters. Correlation analysis methods, such as the Pearson correlation coefficient, can be used to measure the degree of correlation. Based on the magnitude of the correlation, the diffusion path weight is determined. The stronger the correlation, the greater the impact of the dose concentration change on the spatial distribution of the water quality parameters, and the greater the corresponding diffusion path weight.

[0061] Finally, a chemical diffusion path model was established based on the diffusion path weights, combined with fluid mechanics principles and the structural characteristics of the wastewater treatment tank. This model can be used to calculate the diffusion rate parameter of the chemical in the wastewater, which reflects the speed of chemical diffusion in the wastewater.

[0062] Step S124: determining the exposure contact time parameter according to the flow rate parameter in the equipment operation parameter and the volume parameter of the treatment tank, wherein the exposure contact time parameter is the ratio of the treatment tank volume to the flow rate parameter, representing the hydraulic retention time of the chemical in the corresponding treatment stage.

[0063] Equipment operating parameters, including flow rate and treatment tank volume, are crucial for determining the chemical's exposure time during the treatment phase. This parameter reflects the length of time the chemical remains in contact with wastewater in the treatment tank and is crucial for assessing treatment effectiveness and exposure risk.

[0064] The volume parameter of a treatment tank refers to its effective volume, represented by V. The flow rate parameter refers to the flow rate of the wastewater in the treatment tank, represented by v. According to the principles of fluid mechanics, the hydraulic retention time of chemicals in the treatment tank can be calculated by the ratio of the treatment tank volume to the flow rate parameter, that is, the exposure contact time parameter t = V / v.

[0065] For example, in a wastewater treatment tank with an effective volume of V cubic meters and a wastewater flow rate of v cubic meters per hour, the chemical's exposure time parameter, t, is V / v hours. This calculation accurately determines the chemical's exposure time at each treatment stage.

[0066] Step S125: performing interaction influence coefficient calculation processing on the temperature parameter, pH parameter and redox potential parameter in the environmental state association feature to generate the environmental interaction influence parameter, the interaction influence coefficient calculation processing includes: first normalizing the temperature parameter and the redox potential parameter based on a preset safety range to convert them into dimensionless parameters, and then determining the environmental interaction coefficient matrix based on the linear combination of the normalized temperature parameter and the pH parameter and the exponential transformation of the normalized redox potential parameter.

[0067] Temperature, pH, and redox potential, among the environmental state association characteristics, have a significant impact on the reaction and diffusion of chemicals in wastewater. To comprehensively consider the interactions between these parameters, it is necessary to calculate the interaction coefficients to generate environmental interaction parameters.

[0068] First, the temperature parameter and redox potential parameter are normalized based on the preset safety range. The preset safety range is determined according to the properties of the chemical and the requirements of wastewater treatment. For the temperature parameter T, assuming its preset safety range is [T_min, T_max], the normalized temperature parameter T_normalized = (T-T_min) / (T_max-T_min). For the redox potential parameter ORP, assuming its preset safety range is [ORP_min, ORP_max], the normalized redox potential parameter ORP_normalized = (ORP-ORP_min) / (ORP_max-ORP_min). Through this normalization process, the temperature parameter and redox potential parameter are converted into dimensionless parameters, eliminating the influence of dimension.

[0069] Then, the environmental interaction coefficient matrix is determined based on the linear combination of the normalized temperature parameter and the pH parameter and the exponential transformation of the normalized redox potential parameter. Specifically, for the linear combination of the normalized temperature parameter and the pH parameter, it is necessary to first clarify the relative importance of the two in the environmental interaction. This can be achieved through in-depth analysis of a large amount of historical data. During the long-term operation of wastewater treatment, information such as temperature, pH value and corresponding chemical reaction effect at different times will be recorded. By mining these data and applying data analysis techniques, such as correlation analysis, the degree of influence of temperature and pH on the reaction and diffusion of chemicals in wastewater is determined. Based on this analysis result, the normalized temperature parameter and pH parameter are assigned corresponding weights.

[0070] Assume that, after the above analysis, the weight of the normalized temperature parameter is determined to be A, and the weight of the pH parameter is determined to be B. Then, the linear combination of the normalized temperature parameter and the pH parameter is calculated by multiplying the normalized temperature parameter by the weight A, and then multiplying the pH parameter by the weight B. This linear combination value comprehensively reflects the initial impact of the combined effects of temperature and pH on environmental interactions.

[0071] The exponential transformation of the normalized redox potential parameter takes into account the fact that the effect of redox potential on chemicals often exhibits nonlinear characteristics. First, the base and exponential portion of the exponential transformation must be determined. The choice of the base is generally based on an understanding of the nature of redox reactions and the relevant chemical principles, with reference to existing chemical research results and empirical summaries from actual wastewater treatment. The exponential portion is related to the normalized redox potential parameter. Generally, an appropriate coefficient C is determined based on experimental or simulation results, and this coefficient C is multiplied by the normalized redox potential parameter to obtain the exponential portion. The exponential transformation of the normalized redox potential parameter is then obtained through exponential calculation. This exponential transformation amplifies the differences in the impact of redox potential on environmental interactions at different levels.

[0072] Next, we construct the environmental interaction coefficient matrix. The dimensions of this matrix are determined by the specific structure of the wastewater treatment system and the analysis requirements. For example, if the wastewater treatment process is divided into n distinct treatment units, and each unit considers the interactive effects of m environmental factors, then the environmental interaction coefficient matrix is a matrix with n rows and m columns. Each element in the environmental interaction coefficient matrix represents the interaction coefficient for a specific combination of environmental factors in a specific treatment unit.

[0073] To determine each element in the environmental interaction coefficient matrix, a mapping relationship is established. The linear combination of the normalized temperature and pH parameters and the exponential transformation of the normalized redox potential parameter obtained previously are used as inputs, and the value of each element is calculated using a predefined mapping function. The construction of this mapping function requires a combination of chemical kinetics, fluid mechanics, and practical wastewater treatment experience. In actual applications, the mapping function may be adjusted and optimized multiple times to ensure that the matrix elements accurately reflect the interactions between environmental factors.

[0074] When calculating matrix elements, special attention should be paid to dimensional consistency. Since the normalized temperature parameter, pH parameter, and normalized redox potential parameter are all dimensionless, the values obtained after linear combination and exponential transformation are also dimensionless. Therefore, the elements of the environmental interaction coefficient matrix finally calculated are also dimensionless, which ensures the rationality of the entire calculation process in terms of physical meaning and mathematical logic. In this way, the process of determining the environmental interaction coefficient matrix based on the linear combination of the normalized temperature parameter and the pH parameter and the exponential transformation of the normalized redox potential parameter is completed. The obtained environmental interaction coefficient matrix can provide a key basis for the subsequent analysis of the environmental interaction effects of chemicals in the wastewater treatment process.

[0075] Step S130: Based on a preset dynamic simulation strategy network, a multi-stage iterative simulation process is performed on the dynamic exposure parameter set to generate a dynamic simulation result of chemical exposure of the target wastewater treatment process, wherein the dynamic simulation result of chemical exposure includes the exposure concentration distribution and exposure risk probability of each processing node.

[0076] After obtaining the dynamic exposure parameter set for each treatment stage, in order to fully understand the exposure dynamics of chemicals in the target wastewater treatment process, it is necessary to use a preset dynamic simulation strategy network to perform multi-stage iterative simulation processing on the dynamic exposure parameter set. The target wastewater treatment process is usually composed of multiple continuous treatment nodes, each of which has its own unique treatment function and environmental conditions. The diffusion, reaction and residue of chemicals in different treatment nodes will vary. Through multi-stage iterative simulation processing, the dynamic changes of chemicals in the entire wastewater treatment process can be simulated, thereby generating chemical exposure dynamic simulation results that include the exposure concentration distribution and exposure risk probability of each treatment node.

[0077] Step S131: calling the dynamic simulation strategy network to perform diffusion path simulation processing on the chemical diffusion rate parameters to generate an initial diffusion concentration distribution, wherein the diffusion path simulation processing includes determining the initial diffusion concentration of each processing node using a random walk algorithm based on diffusion path weights.

[0078] The dynamic simulation strategy network is trained based on extensive historical data and actual experimental results. It can simulate the diffusion paths of chemicals in wastewater treatment systems based on input chemical diffusion rate parameters. A random walk algorithm based on diffusion path weights is used to simulate the diffusion paths.

[0079] Diffusion path weights were determined in the previous step based on the correlation between the temporal gradient of the dose concentration and the spatial distribution of water quality parameters. These diffusion path weights reflect the likelihood of the chemical diffusing in different directions and along different paths. The basic idea of the random walk algorithm is to treat the chemical as a randomly moving particle, with each particle choosing a possible direction of movement at each time step based on the diffusion path weight.

[0080] Specifically, for each processing node, the algorithm assigns different probabilities to particle movement based on the weight of the diffusion paths surrounding that node. For example, if the diffusion path weight in a certain direction is larger, the probability of the particle moving in that direction is higher. By simulating the random movement of particles multiple times, the distribution of particles at each processing node can be statistically analyzed, thereby determining the initial diffusion concentration at each processing node.

[0081] During the simulation, an initial chemical injection point is set, from which particles begin a random walk. Over time, the particles gradually diffuse to various processing nodes. Each simulation records the number of times the particles stop at each processing node. A higher number of stops indicates a higher chemical concentration at that node. After a large number of simulations, the concentration distribution at each processing node is statistically calculated and used as the initial diffusion concentration distribution.

[0082] Step S132: performing time decay correction processing on the initial diffusion concentration distribution according to the exposure contact time parameter to obtain a corrected diffusion concentration distribution, wherein the time decay correction processing includes: performing exponential decay simulation on the initial diffusion concentration based on the chemical decay rate coefficient with the inverse time dimension and the exposure contact time parameter, wherein the decay rate coefficient is obtained by measuring the hourly concentration decay rate in the laboratory.

[0083] Chemicals in wastewater naturally decay over time due to a variety of factors, including chemical reactions and physical adsorption. To more accurately simulate the actual concentration changes of chemicals during treatment, a time-decay correction is applied to the initial diffusion concentration distribution based on the exposure contact time parameter.

[0084] The chemical decay rate coefficient is a parameter with an inverse time dimension that reflects the degree of chemical decay per unit time. This chemical decay rate coefficient is obtained by measuring the hourly concentration decay rate in the laboratory. The specific experimental process involves adding a certain concentration of chemical to the simulated wastewater under laboratory conditions, simulating wastewater treatment conditions. The chemical concentration is then measured at different time points. By calculating the ratio of the concentration difference between adjacent time points to the time interval, the hourly concentration decay rate is obtained, and the chemical decay rate coefficient is then determined.

[0085] When performing time decay correction, an exponential decay simulation method is used. The initial diffusion concentration of each processing node is corrected based on the chemical decay rate coefficient and the exposure contact time parameter. Specifically, as the exposure contact time increases, the chemical concentration decays exponentially. Assuming the initial diffusion concentration of a processing node is C0, the chemical decay rate coefficient is k, and the exposure contact time is t, the concentration C after time decay correction can be calculated using the exponential decay formula. By performing this correction on the initial diffusion concentration of each processing node, a corrected diffusion concentration distribution is obtained.

[0086] Step S133: Dynamically coupling the modified diffusion concentration distribution with the environmental interaction parameters to generate a stage coupled concentration distribution. The dynamic coupling process includes: performing nonlinear dynamic coupling on the modified diffusion concentration and the environmental interaction coefficient matrix based on a multi-layer perceptron model to capture the combined effects of temperature, pH, and redox potential.

[0087] Environmental factors such as temperature, pH, and redox potential can significantly influence the concentration of chemicals in wastewater, and these effects are often interrelated and nonlinear. To fully account for the combined impact of these environmental factors, it is necessary to dynamically couple the modified diffusion concentration distribution with the environmental interaction parameters.

[0088] The multilayer perceptron model is a commonly used artificial neural network model with powerful nonlinear mapping capabilities and the ability to handle complex input-output relationships. In this step, the modified diffusion concentration and environmental interaction coefficient matrix are used as inputs to the multilayer perceptron model.

[0089] The corrected diffusion concentration reflects the current concentration distribution of the chemical after correction for time decay, while the environmental interaction coefficient matrix comprehensively reflects the interactive effects of environmental factors such as temperature, pH, and redox potential. The multilayer perceptron model processes input data through multiple hidden layers. Each hidden layer contains multiple neurons, which are connected by weights.

[0090] During the model training phase, a large amount of historical data is used as training samples. This data includes different modified diffusion concentrations, environmental interaction coefficient matrices, and corresponding actual concentration data. The model is trained by adjusting the weight parameters of the multi-layer perceptron model to make the model output as close as possible to the actual concentration data.

[0091] During dynamic coupling processing, the current modified diffusion concentration and environmental interaction coefficient matrix are input into a trained multi-layer perceptron model. The model then calculates, based on its internal weight parameters, the output of the concentration distribution after environmental influences, known as the stage-coupled concentration distribution. This stage-coupled concentration distribution takes into account the combined effects of environmental factors such as temperature, pH, and redox potential, more accurately reflecting the actual concentration of the chemical at the current processing stage.

[0092] Step S134: performing risk probability mapping processing on the coupling concentration distribution in the stage according to a preset exposure risk threshold, and generating an exposure risk probability of each processing node.

[0093] To assess the exposure risk of chemicals at each treatment node, a risk probability mapping process is performed on the stage-coupled concentration distribution based on a preset exposure risk threshold. This exposure risk threshold is determined based on the toxicity of the chemical, environmental standards, and relevant safety regulations, and represents the upper limit of the safe concentration of the chemical in wastewater.

[0094] The stage-coupled concentration at each processing node is compared with a preset exposure risk threshold. If the stage-coupled concentration is lower than the exposure risk threshold, the chemical exposure risk at that node is low; if the stage-coupled concentration is higher than the exposure risk threshold, the node has a certain exposure risk.

[0095] To more accurately assess the degree of risk, a risk probability mapping approach is employed. Specifically, a risk probability mapping function is established whose input is the ratio of the stage-coupled concentration to the exposure risk threshold, and whose output is the corresponding exposure risk probability. This risk probability mapping function can be derived through analysis and statistics of a large number of real-world cases, or it can be set based on expert experience.

[0096] For example, when the ratio of the stage-coupled concentration to the exposure risk threshold is small, the exposure risk probability is low; when the ratio approaches or exceeds 1, the exposure risk probability increases significantly. By performing this risk probability mapping process on the stage-coupled concentration of each processing node, the exposure risk probability of each processing node is obtained.

[0097] Step S135: Using the stage-coupled concentration distribution of the current processing stage as the initial diffusion concentration input of the next processing stage, repeatedly performing the diffusion path simulation processing, time attenuation correction processing, dynamic coupling processing, and risk probability mapping processing until all processing stages are traversed to generate the chemical exposure dynamic simulation results.

[0098] The target wastewater treatment process consists of multiple sequential stages, with chemicals undergoing diffusion, decay, and interaction with the environment at each stage. To simulate the dynamic changes of chemicals throughout the process, a multi-stage iterative simulation is required.

[0099] After completing the simulation of a processing stage, the stage-coupled concentration distribution of that stage is used as the initial diffusion concentration input for the next processing stage. Then, the diffusion path simulation process, time decay correction process, dynamic coupling process and risk probability mapping process are repeated.

[0100] In the diffusion path simulation process, a random walk algorithm is used to determine the new initial diffusion concentration for each processing node based on the new initial diffusion concentration and diffusion path weight. In the time decay correction process, an exponential decay simulation is performed on the new initial diffusion concentration based on the new exposure contact time parameter to obtain a corrected diffusion concentration distribution. In the dynamic coupling process, the corrected diffusion concentration distribution is nonlinearly dynamically coupled with the new environmental interaction parameters based on a multi-layer perceptron model to generate a new stage-coupled concentration distribution. In the risk probability mapping process, the new stage-coupled concentration distribution is subjected to risk probability mapping based on a preset exposure risk threshold to generate a new exposure risk probability for each processing node.

[0101] The above process is repeated until all treatment stages are traversed. Finally, the exposure concentration distribution and exposure risk probability of each treatment node in each treatment stage are integrated to generate the dynamic simulation results of chemical exposure in the target wastewater treatment process, which can fully reflect the dynamic exposure of chemicals in the entire wastewater treatment process.

[0102] Step S140: performing exposure risk assessment processing based on the chemical exposure dynamic simulation results to generate a chemical exposure optimization strategy set for the target wastewater treatment process, wherein the chemical exposure optimization strategy set includes a dosing concentration adjustment strategy, a dosing timing optimization strategy, and an emergency response triggering strategy.

[0103] After obtaining the dynamic chemical exposure simulation results for the target wastewater treatment process, an exposure risk assessment is performed based on these simulation results to reduce chemical exposure risks and improve the safety and efficiency of wastewater treatment. The corresponding chemical exposure optimization strategy set is generated. The dynamic chemical exposure simulation results include the exposure concentration distribution and exposure risk probability at each treatment node. This information can intuitively reflect the exposure status and potential risks of chemicals in the wastewater treatment process.

[0104] Step S141: performing risk level classification processing on the exposure risk probability in the chemical exposure dynamic simulation result to generate a risk node set and a secondary risk node set.

[0105] To conduct more targeted risk assessments and develop optimization strategies, the exposure risk probabilities from dynamic chemical exposure simulations need to be categorized into risk levels. First, different risk level thresholds are determined based on the chemical's properties, environmental standards, and relevant safety regulations. For example, low, medium, and high risk thresholds can be set.

[0106] The exposure risk probability of each processing node is compared to these risk level thresholds. If the exposure risk probability is above the high risk threshold, the node is classified as a risky node; if the exposure risk probability is between the medium and high risk thresholds, the node is classified as a secondary risk node; if the exposure risk probability is below the medium risk threshold, the node is considered to have low risk and is not currently considered a priority.

[0107] This risk classification process generates a set of risk nodes and a set of secondary risk nodes. Nodes in the risk node set present a higher risk of chemical exposure and require specific attention and optimization measures. Nodes in the secondary risk node set, while presenting a relatively lower risk, also require adjustments and optimization to prevent further escalation of risk.

[0108] Step S142: performing reverse tracing analysis on the exposure concentration distribution in the risk node set to determine the risk source node and the corresponding dosing stage identifier.

[0109] To fundamentally address exposure risks at risk nodes, reverse tracing analysis is required to analyze the exposure concentration distribution within the risk node set, identifying the source node and corresponding dosing stage that caused the risk. The basic idea behind reverse tracing analysis is to start from the risk node and, based on the chemical's diffusion path and reaction patterns, work backwards to deduce the source of the chemical.

[0110] First, we extract spatial gradient variation data on the exposure concentration distribution of the risk node set from the dynamic chemical exposure simulation results. Spatial gradient variation data reflects the variation in chemical concentration between different locations. By analyzing this data, we can determine the diffusion direction of the chemical.

[0111] Then, based on the spatial gradient change data, diffusion direction retracement is performed. Along the direction of maximum change in chemical concentration gradient, upstream processing nodes are traced back and identified as potential source nodes.

[0112] To further verify whether the potential source node is a true risk source node, the dynamic simulation strategy network is called to perform gradient verification processing on the potential source node. The gradient verification processing includes executing the dosing concentration zero simulation and the dosing concentration gradient decreasing simulation respectively. In the dosing concentration zero simulation, the dosing concentration of the potential source node is set to zero, and then the dynamic simulation strategy network is re-run to obtain new chemical exposure simulation results. In the dosing concentration gradient decreasing simulation, the dosing concentration of the potential source node is gradually reduced according to the set gradient, and the dynamic simulation strategy network is also re-run to obtain chemical exposure simulation results under different dosing concentrations.

[0113] Comprehensive verification is performed by comparing the difference in concentration distribution between the two simulation results. This difference can be calculated using methods such as structural similarity indices, combined with an assessment of the spatial correlation of concentration gradients. If the exposure concentration at a risk node significantly decreases after the dosed concentration is returned to zero or decreases along a gradient, this indicates that the potential source node is likely a risk source node.

[0114] Finally, the dosing phase identifier corresponding to the risk source node is matched from the chemical dosing data set and associated with the subsequent dosing concentration adjustment strategy. This completes the reverse traceability analysis and identifies the risk source node and the corresponding dosing phase identifier.

[0115] Step S143: Extract the corresponding dosage concentration and dosage time interval from the chemical dosage data set according to the dosage stage identifier of the risk source node, and generate a dosage concentration adjustment strategy. The dosage concentration adjustment strategy includes: under the premise of ensuring that the total dosage remains unchanged, adjusting the dosage concentration decrease amplitude according to the ratio of the processing time interval between the risk source node and the adjacent nodes; under the constraint of ensuring the conservation of the total dosage, dynamically allocating the reduced dosage to the adjacent processing stages according to the weight ratio of the processing time interval of each adjacent node.

[0116] After identifying the risk source node and the corresponding dosing stage identifier, a dosing concentration adjustment strategy is needed to reduce the exposure risk of the risk node. The core idea of the dosing concentration adjustment strategy is to reasonably adjust the dosing concentration of the risk source node and adjacent nodes while ensuring that the total dosing amount remains unchanged.

[0117] First, we extracted the corresponding chemical dosing concentrations and dosing intervals from the chemical dosing data set based on the dosing stage identifiers of the risk source nodes. We then analyzed the treatment interval ratios between the risk source nodes and their adjacent nodes. These ratios reflect the relative residence time of each node in the wastewater treatment process.

[0118] Adjust the dosage reduction rate based on the treatment time interval ratio. If the treatment time interval of the risk source node is relatively long, then the dosage reduction rate of this node can be appropriately increased; if the treatment time interval of adjacent nodes is relatively long, then consider distributing the reduced dosage to more of these nodes.

[0119] When adjusting the dosage concentration, it is important to ensure that the total dosage is conserved. Specifically, the dosage reduction at the risk source node is calculated and then dynamically allocated to adjacent treatment stages based on the weighted ratio of the treatment intervals of each adjacent node. The weighted ratio of the treatment intervals of adjacent nodes can be calculated by dividing the treatment interval of each adjacent node by the sum of the treatment intervals of all adjacent nodes.

[0120] For example, assume there is a risk source node A and adjacent nodes B and C. Node A's reduced dosage is ΔC, node B's treatment interval is tB, and node C's treatment interval is tC. Then, the increased dosage allocated to node B is ΔC × (tB / (tB + tC)), and the increased dosage allocated to node C is ΔC × (tC / (tB + tC)). This dosage concentration adjustment strategy can reduce the exposure risk of risk nodes while ensuring the total dosage remains unchanged.

[0121] Step S144: performing time series optimization processing on the exposure contact time parameters in the secondary risk node set to generate a dosing timing optimization strategy, wherein the time series optimization processing includes adjusting the distribution of dosing time intervals according to the correlation between the exposure contact time parameters and the equipment operating parameters.

[0122] When performing time series optimization on the exposure contact time parameters in the set of secondary risk nodes to generate an optimization strategy for dosing timing, the distribution of dosing time intervals is adjusted according to the correlation between the exposure contact time parameters and the equipment operating parameters, which requires detailed analysis and operation in many aspects.

[0123] First, it's important to clarify the correlation between exposure time parameters and equipment operating parameters. In wastewater treatment systems, equipment operating parameters such as pump flow rate and agitator speed directly affect the flow and mixing of wastewater within the treatment unit, which in turn influences the diffusion and reaction processes of chemicals, ultimately affecting exposure time. For example, a faster pump flow rate shortens the residence time of wastewater in the treatment tank, reducing chemical exposure time. A higher agitator speed can accelerate chemical diffusion and may also alter exposure time.

[0124] To accurately grasp this correlation, it is necessary to collect a large amount of historical data. This data should include the exposure time parameters and corresponding equipment operating parameters for each node in the secondary risk node set. In-depth data analysis of this data can be performed using methods such as correlation analysis to calculate the correlation coefficient between the exposure time parameters and each equipment operating parameter. The size of the correlation coefficient reflects the closeness of the relationship between the two. The closer the absolute value of the correlation coefficient is to 1, the stronger the association.

[0125] Based on the results of the correlation analysis, select equipment operating parameters that are strongly correlated with the exposure time parameter. Assume that the selected equipment operating parameters are parameters A and B. Next, analyze the variation patterns and value ranges of these equipment operating parameters. For example, parameter A may have an optimal value range within which the exposure time can reach a relatively ideal state, which is conducive to reducing the exposure risk of secondary risk nodes.

[0126] Then, based on the correlation between equipment operating parameters and exposure contact time parameters, formulate adjustment rules for the dosing interval. If changes in equipment operating parameters result in a shortened exposure contact time, the dosing interval can be appropriately shortened to ensure that the chemical has sufficient time to act in the wastewater. Conversely, if the exposure contact time is prolonged, the dosing interval can be appropriately extended to avoid excessive accumulation of chemicals.

[0127] When adjusting the distribution of dosing intervals, the continuity and stability of the entire wastewater treatment process must be considered. Adjustments should not be made solely based on the performance of a single secondary risk node, but rather should be comprehensively considered, considering the operational status of adjacent nodes and the entire treatment system. For example, when adjusting the dosing interval at a secondary risk node, the impact on subsequent nodes must be assessed to avoid a chain reaction that could increase risk at other nodes.

[0128] To more precisely adjust the distribution of dosing intervals, a dosing interval adjustment model can be established. This model uses the selected equipment operating parameters and exposure time parameters as input and, using pre-defined algorithms and rules, outputs an optimized dosing interval. This model can be based on machine learning algorithms, such as neural networks, trained using historical data to accurately predict the optimal dosing interval under different equipment operating parameters.

[0129] In practice, changes in equipment operating parameters and exposure time parameters are monitored in real time and fed into a dosing interval adjustment model. The model then calculates and outputs the optimized dosing interval based on the input data. This information is then fed back to the wastewater treatment control system, which automatically adjusts the chemical dosing schedule based on this feedback.

[0130] Through the above steps, the distribution of dosing time intervals was adjusted according to the correlation between the exposure contact time parameters and the equipment operating parameters, which achieved the optimization of the dosing timing, reduced the exposure risk of secondary risk nodes, and improved the safety and efficiency of the entire wastewater treatment process.

[0131] Step S145: generating an emergency response triggering strategy according to the spatial distribution characteristics of the risk node set and the secondary risk node set, wherein the emergency response triggering strategy includes an emergency processing node identifier and an emergency processing operation instruction.

[0132] The spatial distribution characteristics of risk nodes and secondary risk nodes contain important information that can be used to generate effective emergency response triggering strategies. First, a detailed analysis of the spatial location of risk nodes and secondary risk nodes within the wastewater treatment system was conducted. Different spatial locations determine the risk propagation path and impact range. For example, a risk node at the front end of the wastewater treatment process may affect multiple subsequent nodes, while a node at the end may primarily affect the discharge water quality.

[0133] When analyzing spatial distribution characteristics, it's important to consider the connectivity between nodes and the direction of water flow. Wastewater flows along specific pathways within the treatment system, with nodes connected by pipelines and other infrastructure. Based on the direction of water flow, we can determine the direction of risk propagation and identify which nodes are likely to be affected by the risk node. For example, if a risk node is located near the inlet of a treatment tank, that tank and subsequent connected treatment units may be affected.

[0134] Emergency response nodes are identified based on spatial distribution characteristics and risk transmission paths. Emergency response nodes are the nodes where emergency measures must be implemented first in the event of a risk event. Generally speaking, nodes that are close to risk nodes and located along the risk transmission path should be designated as emergency response nodes. The importance and processing capacity of the nodes should also be considered. For example, nodes with stronger processing capabilities can be prioritized as emergency response nodes to ensure a quick and effective response to risks.

[0135] After identifying the emergency response nodes, develop corresponding emergency response instructions. These instructions should be tailored to the type and severity of the risk. Different emergency response measures may be required for different chemical exposure risks. For example, if there is a risk of a chemical leak, the emergency response instructions may include closing relevant valves, activating a leak collection device, and administering a neutralizer. If there is a risk of excessive chemical concentration, the dosage of the corresponding treatment agent may need to be increased, the treatment time may need to be extended, and so on.

[0136] Emergency response instructions must be specific and clear, including the content, sequence, and timing of the operations. For example, instructions could specify how long to close a valve or activate a leak collection device after a risk is discovered. Furthermore, the impact of emergency response operations on the entire wastewater treatment system must be considered to avoid causing other problems due to emergency operations.

[0137] To ensure the effectiveness of the emergency response triggering strategy, regular drills and evaluations of emergency handling instructions are also required. Through drills, the feasibility of emergency handling operations and the proficiency of operators can be verified; through evaluations, emergency handling instructions can be adjusted and optimized according to actual conditions, thereby improving the efficiency and effectiveness of emergency response.

[0138] By comprehensively considering the spatial distribution characteristics of risk node sets and secondary risk node sets, determining emergency treatment node identification and formulating emergency treatment operation instructions, a complete emergency response trigger strategy was generated, which can respond promptly and effectively when chemical exposure risks occur, reducing the impact of risks on wastewater treatment systems and the environment.

[0139] Step S150: Feedback the chemical exposure optimization strategy set to the wastewater treatment control system to trigger a dynamic adjustment operation of the chemical dosing process.

[0140] After generating a set of optimized chemical exposure strategies, in order for these strategies to truly take effect, they need to be fed back to the wastewater treatment control system to trigger dynamic adjustments to the chemical dosing process. The wastewater treatment control system is responsible for monitoring and regulating the operation of wastewater treatment equipment and the dosing of chemicals.

[0141] First, the chemical exposure optimization strategy set, including the dosing concentration adjustment strategy, dosing timing optimization strategy, and emergency response trigger strategy, is organized and formatted so that it can be identified and processed by the wastewater treatment control system. The dosing concentration adjustment strategy contains the dosing concentration adjustment information for the risk source node and adjacent nodes, the dosing timing optimization strategy contains the optimized dosing interval information, and the emergency response trigger strategy contains the emergency treatment node identifier and emergency treatment operation instructions.

[0142] The optimized strategy is transmitted to the wastewater treatment control system via a data interface. After receiving the strategy, the wastewater treatment control system analyzes and verifies it. The analysis process converts the information in the strategy into instructions that the system can understand and execute. The verification process verifies the rationality and feasibility of the strategy, for example, by checking whether the adjusted dosing concentration is within a safe range and whether the adjusted dosing interval meets the equipment's operating requirements.

[0143] For the concentration adjustment strategy, the wastewater treatment control system automatically adjusts the chemical dosage concentration at the risk source node and adjacent nodes based on the information provided in the strategy. By controlling devices such as flow control valves in the dosing equipment, the system precisely regulates the amount of chemical added to ensure that it is added at the adjusted concentration.

[0144] For dosing timing optimization strategies, the wastewater treatment control system adjusts the chemical dosing schedule based on the optimized dosing intervals. The system updates the dosing equipment's operating schedule and triggers dosing operations according to the new intervals, ensuring that the chemicals are added to the wastewater at the appropriate time.

[0145] For emergency response triggering strategies, the wastewater treatment control system monitors the status of risk nodes and sub-risk nodes in real time. When the risk indicator at a node reaches the emergency trigger condition, the system immediately executes the corresponding emergency response measures according to the emergency response instructions. For example, if the chemical concentration at a risk node exceeds the safety threshold, the system automatically closes the relevant valves and activates the leak collection device.

[0146] During the dynamic adjustment of the chemical dosing process, the wastewater treatment control system monitors various parameters of the wastewater treatment process in real time, such as water quality parameters and equipment operating parameters. By monitoring these parameters, problems that arise during the adjustment process can be promptly identified, and the optimization strategy can be further adjusted and optimized based on actual conditions.

[0147] To ensure the stability and reliability of adjustment operations, the wastewater treatment control system records and logs the entire process. This log includes information such as the time of adjustment, the parameters used, and the operating status before and after the adjustment. These records provide data support for subsequent analysis and evaluation, and help analyze lessons learned and continuously improve the dynamic adjustment strategy for the chemical dosing process.

[0148] By feeding back the chemical exposure optimization strategy set to the wastewater treatment control system, dynamic adjustment operations of the chemical dosing process are triggered, achieving effective control and optimization of the chemical exposure risks in the wastewater treatment process, and improving the safety and efficiency of wastewater treatment.

[0149] Step S210: Construct a simulated training data set and corresponding chemical exposure verification results, wherein the simulated training data set includes historical wastewater treatment process data, synthetic data generated based on fluid mechanics simulation, and adversarial sample data generated by adding random perturbations to historical parameters, wherein the synthetic data includes extreme flow and abnormal water quality conditions.

[0150] In order to train an accurate and effective dynamic simulation strategy network, it is necessary to construct a comprehensive and representative simulation training dataset and corresponding chemical exposure verification results.

[0151] Historical wastewater treatment process data is a crucial component of simulation training datasets. It originates from actual wastewater treatment processes and records various operating parameters and chemical dosing patterns of wastewater treatment systems over different time periods. When collecting historical wastewater treatment process data, it is important to cover multiple different wastewater treatment projects to ensure data diversity and comprehensiveness. This data includes chemical type identification, dosing concentration, and dosing intervals in the chemical dosing dataset, as well as water quality parameters, equipment operating parameters, and chemical residual concentrations in the real-time monitoring dataset.

[0152] Synthetic data generated based on fluid dynamics simulation can make up for the shortcomings of historical data. Fluid dynamics simulation can simulate the flow and mixing process of wastewater in the treatment system, as well as the diffusion and reaction of chemicals therein. By setting different boundary conditions and parameters, synthetic data containing extreme flow and abnormal water quality conditions can be generated. For example, simulate the flow state of wastewater under high flow conditions, or simulate the reaction and diffusion of chemicals under abnormal water quality (such as high concentrations of pollutants, the presence of special chemicals). These synthetic data can help the dynamic simulation strategy network learn the exposure patterns of chemicals under extreme and abnormal conditions.

[0153] Adversarial data is generated by adding random perturbations to historical parameters. In real-world applications, wastewater treatment systems may be affected by various uncertainties, leading to parameter fluctuations. By adding random perturbations to historical parameters, we can simulate the impact of these uncertainties and generate adversarial data. The amplitude and method of random perturbations should be appropriately set based on the actual situation to ensure that the generated adversarial data is both random and conforms to actual physical laws.

[0154] While constructing the simulated training dataset, the corresponding chemical exposure verification results must also be determined. These can be determined using actual monitoring data or rigorously validated simulation results. For historical data, actual monitored chemical exposure can be used as verification results; for synthetic data and adversarial sample data, high-precision simulation models or experimental results can be used as verification results.

[0155] Historical wastewater treatment process data, synthetic data, and adversarial sample data are integrated to form a simulated training dataset. At the same time, the corresponding chemical exposure verification results are linked to the simulated training dataset to provide complete data support for subsequent model training.

[0156] Step S220: calling the initial dynamic simulation network to perform dynamic exposure parameter extraction processing on the chemical addition data and the real-time monitoring data in the simulation training data set to generate a training dynamic exposure parameter set.

[0157] After constructing the simulation training dataset, the initial dynamic simulation network is used to extract dynamic exposure parameters from the chemical dosing data and real-time monitoring data. The initial dynamic simulation network is a pre-built neural network model with a specific structure and parameters, but it still requires training to optimize.

[0158] First, the chemical dosing data and real-time monitoring data from the simulation training dataset were input into the initial dynamic simulation network. The chemical dosing data included the chemical type, dosing concentration, and dosing interval for each treatment stage. The real-time monitoring data included water quality parameters, equipment operating parameters, and chemical residual concentrations for each treatment stage.

[0159] The initial dynamic simulation network processes input data using steps similar to those used in real-world applications. For chemical dosing data, the network aligns the dosing concentrations and dosing intervals to generate a temporal signature of the chemical dosing. Specifically, the network sorts and interpolates the dosing concentrations and dosing intervals based on the dosing timestamps, ensuring data continuity across the time dimension.

[0160] For real-time monitoring data, the network performs environmental state correlation analysis on water quality parameters and equipment operating parameters to generate environmental state correlation features. The network uses methods such as calculating correlation coefficients to identify correlations between water quality parameters and equipment operating parameters and converts these relationships into feature vectors.

[0161] Based on the temporal characteristics of chemical dosing and the correlation between environmental conditions, the network models the diffusion pathways and generates chemical diffusion rate parameters. The network then determines diffusion path weights based on the correlation between the temporal gradient of dosing concentration and the spatial distribution of water quality parameters. The network then uses these weights to simulate the chemical's diffusion pathways and calculate the diffusion rate parameters.

[0162] The network determines the exposure contact time parameter based on the flow rate parameter in the equipment operating parameters and the volume parameter of the treatment tank. By calculating the ratio of the treatment tank volume to the flow rate parameter, the hydraulic retention time of the chemical at the corresponding treatment stage is obtained, which is the exposure contact time parameter.

[0163] The network calculates the interaction coefficients of the temperature, pH, and redox potential parameters in the environmental state association features to generate environmental interaction coefficients. The network first normalizes the temperature and redox potential parameters based on a preset safety range to convert them into dimensionless parameters. The network then determines the environmental interaction coefficient matrix based on the linear combination of the normalized temperature and pH parameters and the exponential transformation of the normalized redox potential parameter.

[0164] Through this series of processes, the initial dynamic simulation network generates a training dynamic exposure parameter set, which includes chemical diffusion rate parameters, exposure contact time parameters, and environmental interaction parameters. These parameters will be used in subsequent multi-stage iterative simulation processing and model training.

[0165] Step S230: performing multi-stage iterative simulation processing based on the training dynamic exposure parameter set to generate predicted chemical exposure results.

[0166] After obtaining the training dynamic exposure parameter set, a multi-stage iterative simulation process is performed based on this set to generate predicted chemical exposure results. This multi-stage iterative simulation process is similar to the simulation process based on dynamic simulation strategy networks in actual applications.

[0167] First, the initial dynamic simulation network is invoked to simulate the diffusion paths of the chemical diffusion rate parameters in the training dynamic exposure parameter set, generating an initial diffusion concentration distribution. The network then determines the initial diffusion concentrations for each treatment node using a random walk algorithm based on diffusion path weights. During this process, the network simulates the diffusion of chemicals in the wastewater treatment system and assigns initial chemical concentrations to each treatment node based on the diffusion path weights.

[0168] The initial diffusion concentration distribution is then corrected for time decay based on the exposure contact time parameter in the training dynamic exposure parameter set, resulting in a modified diffusion concentration distribution. The network simulates exponential decay of the initial diffusion concentration based on the chemical decay rate coefficient, which has a reciprocal time dimension, and the exposure contact time parameter. The chemical decay rate coefficient is obtained from hourly laboratory measurements of concentration decay rates. The network uses this chemical decay rate coefficient and the exposure contact time parameter to calculate the chemical concentration decay at each treatment node.

[0169] Next, the modified diffusion concentration distribution is dynamically coupled with the environmental interaction parameters from the training dynamic exposure parameter set to generate a stage-coupled concentration distribution. The network uses a multilayer perceptron model to dynamically and nonlinearly couple the modified diffusion concentration with the environmental interaction coefficient matrix, capturing the combined effects of temperature, pH, and redox potential. The multilayer perceptron model processes the input modified diffusion concentration and environmental interaction coefficient matrix, outputting a stage-coupled concentration distribution that accounts for environmental factors.

[0170] The network then maps the phase-coupled concentration distribution to a risk probability based on a preset exposure risk threshold, generating an exposure risk probability for each processing node. The network then establishes a risk probability mapping function that takes the ratio of the phase-coupled concentration to the exposure risk threshold as input and outputs the corresponding exposure risk probability.

[0171] The coupled concentration distribution of the current processing stage is used as the initial diffusion concentration input for the next processing stage. The diffusion path simulation, time decay correction, dynamic coupling, and risk probability mapping processes are repeated until all processing stages are traversed to generate predicted chemical exposure results. These predicted chemical exposure results include the exposure concentration distribution and exposure risk probability for each processing node, which will be used for subsequent model evaluation and parameter updates.

[0172] Step S240: constructing a simulation loss function according to the difference between the predicted chemical exposure result and the chemical exposure verification result.

[0173] To evaluate the performance of the initial dynamic simulation network and optimize its parameters, a simulation loss function is constructed based on the difference between the predicted chemical exposure results and the verified chemical exposure results. The simulation loss function measures the deviation between the network's predictions and the actual situation. Minimizing this loss function can make the network's predictions more accurate.

[0174] First, the mean squared error (MSE) between the predicted chemical exposure results and the verified chemical exposure results was calculated as the baseline loss. MSE is a commonly used metric to measure the difference between predicted and true values. It is calculated by averaging the squares of the differences between the predicted and true values. For each treatment node's exposure concentration and exposure risk probability, the difference between the predicted and verified values was calculated. These differences were then squared, summed, and divided by the number of treatment nodes to obtain the MSE.

[0175] The concentration points in the predicted chemical exposure results that exceed the preset safety threshold are weighted by a penalty coefficient to generate a penalty loss. The preset safety threshold is determined based on the properties of the chemical and environmental standards. Concentration points that exceed the safety threshold may cause harm to the environment and human health, so these points need to be penalized. The specific approach is to extract all concentration point data from the predicted chemical exposure results and compare each concentration point data with the preset safety threshold. If the concentration point data exceeds the safety threshold, the difference between the part exceeding the safety threshold and the safety threshold is calculated, and a local penalty term is generated based on the square of the difference and the product of the preset penalty coefficient. The local penalty terms of all concentration point data are summed to obtain the penalty loss. The preset penalty coefficient is a dimensionless weight parameter used to balance the risk levels of different over-limit concentrations.

[0176] A gradient consistency loss is constructed based on the correlation between the spatial gradients of the predicted chemical exposure results and the spatial gradients of the chemical exposure verification results. The spatial gradient reflects the spatial variation of chemical concentrations. The stronger the correlation between the spatial gradients of the predicted and verified results, the more accurately the network simulates the chemical diffusion and distribution. The gradient consistency loss is obtained by calculating the correlation coefficient between the two and converting it into a loss value.

[0177] The basic loss, penalty loss, and gradient consistency loss are weighted and summed to generate a simulation loss function. During the weighted summation process, it is necessary to assign corresponding weights to each loss term. The determination of these weights needs to be adjusted according to the actual situation to balance the importance of different loss terms. For example, if you pay more attention to the penalty for concentration points that exceed the safety threshold, you can appropriately increase the weight of the penalty loss. The generated simulation loss function plays a key role in the entire model training process. It is an important indicator to measure the prediction accuracy of the initial dynamic simulation network. By continuously adjusting the network parameters to minimize the loss function, the network's prediction results can gradually approach the actual chemical exposure situation.

[0178] In order to more accurately adjust the parameters of the initial dynamic simulation network, it is necessary to analyze in detail the contribution of each loss term in the simulation loss function. The base loss reflects the overall numerical difference between the predicted chemical exposure results and the chemical exposure verification results, and focuses on measuring the prediction error in an average sense. By analyzing the base loss, we can understand the network's ability to predict chemical exposure in most cases. For example, if the base loss is large, it means that the network's overall prediction of the chemical concentration distribution and exposure risk probability has a large deviation, and the network structure or training data may need to be adjusted.

[0179] Penalty loss focuses on concentration points in the prediction results that exceed preset safety thresholds. These concentration points may represent potentially high-risk situations, posing a significant threat to the environment and human health. When analyzing penalty loss, it is necessary to examine which processing nodes' concentrations exceed the safety threshold and to what extent. If the penalty term for a processing node contributes significantly, this indicates a serious problem with the network's prediction at that node, and targeted optimization of the relevant features and parameters for that node may be necessary. Furthermore, the choice of preset penalty coefficient also affects the magnitude of the penalty loss. The penalty coefficient needs to be adjusted appropriately based on actual conditions to balance the level of concern regarding different exceeding-limit concentrations.

[0180] Gradient consistency loss primarily examines the correlation between the spatial gradients of the predicted results and the spatial gradients of the verified results. Spatial gradients reflect the spatial variation in chemical concentrations and are crucial for understanding chemical diffusion and distribution patterns. A high gradient consistency loss indicates that the network is inaccurately modeling the spatial diffusion and distribution of chemicals, possibly due to issues in diffusion path modeling or environmental interaction processing. In this case, it is necessary to review relevant parameters and model structure to ensure that the network accurately captures the spatial variation in chemical concentrations.

[0181] When determining the weights for each loss term, it's important to consider their importance and the actual application scenario. If, in actual wastewater treatment, the focus is on early warning and control of situations exceeding safety thresholds, then the weight of the penalty loss can be appropriately increased. If the network is desired to more accurately simulate the diffusion and distribution of chemicals, then the weight of the gradient consistency loss can be increased. Adjusting these weights requires multiple experiments and verifications to find the optimal combination that allows the simulation loss function to effectively guide the update of network parameters.

[0182] During model training, the backpropagation algorithm is used to update the parameters of the initial dynamic simulation network. Backpropagation is an optimization algorithm based on gradient descent. It calculates the gradient of the simulated loss function with respect to each parameter in the network and then adjusts the parameter values based on the direction and magnitude of the gradient, gradually reducing the loss function. Specifically, the gradient of the simulated loss function with respect to the network's output layer parameters is first calculated. These gradients are then backpropagated to the network's hidden and input layers, and the gradients of the parameters in each layer are calculated in turn. Based on the calculated gradients, the parameters are updated according to the set learning rate. The learning rate controls the step size of the parameter update. An excessively large learning rate may lead to overly rapid parameter updates, preventing the model from convergence; an excessively small learning rate may result in excessively slow training. Therefore, it is important to choose a reasonable learning rate based on the actual situation.

[0183] After each parameter update, the simulation loss function is recalculated and a check is made to determine whether the loss function has converged. The convergence criteria can be customized. For example, if the loss function changes by less than a preset threshold over multiple consecutive iterations, the loss function is considered to have converged. If the loss function has not converged, parameter updates are continued until convergence criteria are met.

[0184] During the training process, some strategies can be employed to improve training efficiency and stability. For example, the simulated training dataset can be divided into a training set, a validation set, and a test set. The training set is used to train the network, while the validation set is used to evaluate network performance during training and adjust model hyperparameters such as the learning rate and the weights of various loss terms. The test set is used for the final evaluation of network performance after training is complete. This approach can prevent model overfitting and improve the model's generalization ability.

[0185] To prevent vanishing or exploding gradients, optimization algorithms such as Adagrad, Adadelta, and Adam can be used. These algorithms can adaptively adjust the learning rate, making parameter updates more stable and effective. They can also normalize the network's input data, making it more evenly distributed and improving network training results.

[0186] Through the above steps, the parameters of the initial dynamic simulation network are continuously adjusted to make the simulation loss function converge, and finally a trained dynamic simulation strategy network is obtained, which can more accurately dynamically simulate and evaluate the chemical exposure of the target wastewater treatment process.

[0187] Step S250: updating the parameters of the initial dynamic simulation network through a back-propagation algorithm until the simulation loss function converges, thereby obtaining a trained dynamic simulation strategy network.

[0188] After constructing the simulated loss function, the backpropagation algorithm is used to update the parameters of the initial dynamic simulation network. Its core goal is to converge the simulated loss function and obtain the trained dynamic simulation policy network. The backpropagation algorithm is an efficient and widely used method for neural network training. It can accurately adjust various parameters in the network based on the gradient information of the loss function.

[0189] First, we need to understand the basic principles of the backpropagation algorithm. Backpropagation is based on the chain rule. Starting from the network's output layer, it reverses and calculates the gradient of the loss function with respect to each parameter in the network. The gradient represents the rate of change of the loss function in parameter space. By updating the parameters in the opposite direction of the gradient, the loss function can be gradually reduced.

[0190] Before starting backpropagation, the parameters of the initial dynamic simulation network must be initialized. The parameter initialization method affects the network's training performance and convergence speed. Common initialization methods include random initialization, Xavier initialization, and He initialization. Random initialization assigns random values to parameters. This method is simple but may result in slower network convergence. Xavier initialization and He initialization provide more reasonable parameter initialization based on the network structure and activation function characteristics, helping to improve network training efficiency.

[0191] In each training iteration, a batch of data from the simulated training dataset is first fed into the initial dynamic simulation network. The network then undergoes a forward propagation process to obtain predicted chemical exposure results. Forward propagation is the process by which data moves from the network's input layer through each hidden layer to the output layer. During this process, the network processes and transforms the input data based on its current parameters to produce a predicted result.

[0192] Then, based on the predicted chemical exposure results and the corresponding chemical exposure verification results, the value of the simulation loss function is calculated. The simulation loss function includes the base loss, penalty loss, and gradient consistency loss, which are combined together through a weighted summation.

[0193] Next, we perform the backpropagation process. Starting from the output layer, we calculate the gradient of the loss function with respect to the output layer parameters. For each parameter in the output layer, we calculate the partial derivative of the loss function with respect to that parameter using the chain rule. This partial derivative represents the rate of change of the loss function with respect to that parameter, also known as the gradient. By calculating the gradient of the output layer parameters, we can determine the update direction of the output layer parameters.

[0194] Next, the gradients of the output layer are backpropagated to the hidden layer. In the hidden layer, the gradients of the loss function with respect to the hidden layer parameters are calculated using the chain rule and the gradients passed from the output layer. Again, these gradients represent the rate of change of the loss function with respect to the hidden layer parameters, providing a basis for updating the hidden layer parameters.

[0195] By analogy, the gradient is back-propagated to the input layer of the network to calculate the gradient of the loss function with respect to the input layer parameters. Through this back-propagation process, the gradient of the loss function with respect to all parameters in the network is obtained.

[0196] After obtaining the gradients of each parameter, the parameters are updated according to the principle of gradient descent. The basic idea of the gradient descent algorithm is to update the parameters in the opposite direction of the gradient, so that the loss function gradually decreases. Specifically, for each parameter, the updated parameter value is obtained by subtracting the gradient from its current value and multiplying it by a learning rate. The learning rate is an important hyperparameter that controls the step size of the parameter update. If the learning rate is too large, the parameter update may skip the optimal solution, resulting in the loss function failing to converge. If the learning rate is too small, the parameter update will be very slow, and the training time will be very long.

[0197] After each parameter update, the simulation loss function is recalculated and checked for convergence. The convergence criteria can be customized. A common method is to observe how the loss function changes over multiple iterations. If the change in the loss function over multiple iterations is less than a preset threshold, the loss function is considered to have converged.

[0198] If the loss function does not converge, the next round of training iterations is continued. The above process of forward propagation, loss function calculation, backpropagation and parameter update is repeated until the loss function converges.

[0199] During training, several techniques can be employed to improve training effectiveness and efficiency. For example, batch normalization can be used to normalize the inputs of each layer in the network, making the data distribution more stable and helping to accelerate network convergence. Additionally, regularization methods, such as L1 and L2 regularization, can be employed to prevent overfitting and improve generalization.

[0200] When the simulation loss function converges, the parameters of the initial dynamic simulation network have been adjusted and optimized multiple times, and the resulting network is the trained dynamic simulation strategy network, which can more accurately dynamically simulate and evaluate the chemical exposure of the target wastewater treatment process.

[0201] Figure 2 A schematic diagram illustrates exemplary hardware and software components of a system 100 for dynamically simulating and assessing chemical exposure in a wastewater treatment process, according to some embodiments of the present invention, that can implement the concepts of the present invention. For example, a processor 120 can be used in the system 100 for dynamically simulating and assessing chemical exposure in a wastewater treatment process to perform the functions of the present invention.

[0202] The system 100 for dynamically simulating and assessing chemical exposure in wastewater treatment processes can be a general-purpose server or a special-purpose server, both of which can be used to implement the dynamic simulation and assessment method for chemical exposure in wastewater treatment processes of the present invention. Although only one server is shown in this disclosure, for convenience, the functions described in this disclosure can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0203] For example, the chemical exposure dynamic simulation assessment system 100 for wastewater treatment processes may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and storage media 140 in different forms, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the chemical exposure dynamic simulation assessment system 100 for wastewater treatment processes may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present invention can be implemented according to these program instructions. The chemical exposure dynamic simulation assessment system 100 for wastewater treatment processes also includes an I / O interface 150 between the computer and other input and output devices.

[0204] For ease of explanation, only one processor is described in the dynamic simulation and assessment system 100 for chemical exposure to wastewater treatment processes. However, it should be noted that the dynamic simulation and assessment system 100 for chemical exposure to wastewater treatment processes in the present invention may also include multiple processors, so the steps performed by one processor described in the present invention may also be performed jointly or individually by multiple processors. For example, if the processor of the dynamic simulation and assessment system 100 for chemical exposure to wastewater treatment processes executes step A and step B, it should be understood that step A and step B may also be executed jointly by two different processors or individually in one processor. For example, the first processor executes step A, the second processor executes step B, or the first processor and the second processor execute steps A and B together.

[0205] In addition, an embodiment of the present invention also provides a readable storage medium, which has computer-executable instructions preset in the readable storage medium. When the processor executes the computer-executable instructions, the above-mentioned dynamic simulation assessment method for chemical exposure in the wastewater treatment process is implemented.

[0206] It should be noted that in order to simplify the description of the present invention and thus help understand one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, multiple features are sometimes combined into one embodiment, figure or description thereof.

Claims

1. A dynamic simulation assessment method for chemical exposure in wastewater treatment processes, characterized in that: The method comprises: Obtain a chemical dosing data set and a real-time monitoring data set for the target wastewater treatment process, wherein the chemical dosing data set includes chemical type identification, dosing concentration, and dosing time interval at different treatment stages, and the real-time monitoring data set includes water quality parameters, equipment operating parameters, and chemical residual concentrations at each treatment stage; Performing dynamic exposure parameter extraction processing on the chemical addition data set and the real-time monitoring data set to generate a dynamic exposure parameter set for each treatment stage, the dynamic exposure parameter set including a chemical diffusion rate parameter, an exposure contact time parameter, and an environmental interaction parameter; Based on a preset dynamic simulation strategy network, a multi-stage iterative simulation process is performed on the dynamic exposure parameter set to generate a dynamic simulation result of chemical exposure for the target wastewater treatment process, wherein the dynamic simulation result of chemical exposure includes the exposure concentration distribution and exposure risk probability of each treatment node; Performing exposure risk assessment processing based on the chemical exposure dynamic simulation results to generate a chemical exposure optimization strategy set for the target wastewater treatment process, wherein the chemical exposure optimization strategy set includes a dosing concentration adjustment strategy, a dosing timing optimization strategy, and an emergency response triggering strategy; The chemical exposure optimization strategy set is fed back to the wastewater treatment control system to trigger dynamic adjustment operations of the chemical dosing process.

2. The dynamic simulation assessment method for chemical exposure in wastewater treatment process according to claim 1, characterized in that: The step of extracting dynamic exposure parameters from the chemical dosing data set and the real-time monitoring data set to generate a dynamic exposure parameter set for each processing stage includes: Performing time series alignment processing on the chemical addition concentration and the addition time interval in the chemical addition data set to generate a chemical addition time series feature; Performing environmental state correlation analysis on the water quality parameters and equipment operating parameters in the real-time monitoring data set to generate environmental state correlation features; Performing diffusion path modeling based on the chemical addition time sequence characteristics and the environmental state correlation characteristics to obtain the chemical diffusion rate parameter, the diffusion path modeling including: determining a diffusion path weight based on the correlation between the time gradient change of the addition concentration and the spatial distribution change of the water quality parameter; Determining the exposure contact time parameter based on the flow rate parameter and the volume parameter of the treatment tank in the equipment operation parameters, wherein the exposure contact time parameter is the ratio of the treatment tank volume to the flow rate parameter, representing the hydraulic retention time of the chemical in the corresponding treatment stage; The temperature parameter, pH parameter and redox potential parameter in the environmental state association feature are subjected to interaction influence coefficient calculation processing to generate the environmental interaction influence parameter. The interaction influence coefficient calculation processing includes: firstly normalizing the temperature parameter and the redox potential parameter based on a preset safety range to convert them into dimensionless parameters, and then determining the environmental interaction coefficient matrix based on the linear combination of the normalized temperature parameter and the pH parameter and the exponential transformation of the normalized redox potential parameter.

3. The dynamic simulation assessment method for chemical exposure in wastewater treatment process according to claim 1, characterized in that: The preset dynamic simulation strategy network is based on which a multi-stage iterative simulation process is performed on the dynamic exposure parameter set to generate a dynamic simulation result of chemical exposure of the target wastewater treatment process, including: Invoking a dynamic simulation strategy network to perform diffusion path simulation processing on the chemical diffusion rate parameter to generate an initial diffusion concentration distribution, wherein the diffusion path simulation processing includes determining the initial diffusion concentration of each processing node using a random walk algorithm based on diffusion path weights; Performing a time decay correction process on the initial diffusion concentration distribution according to the exposure contact time parameter to obtain a corrected diffusion concentration distribution, the time decay correction process comprising: performing an exponential decay simulation on the initial diffusion concentration based on a chemical decay rate coefficient having a time inverse dimension and the exposure contact time parameter, the decay rate coefficient being obtained by measuring an hourly concentration decay rate in a laboratory; Dynamically coupling the modified diffusion concentration distribution with the environmental interaction parameters to generate a stage coupled concentration distribution, the dynamic coupling processing comprising: performing nonlinear dynamic coupling on the modified diffusion concentration and the environmental interaction coefficient matrix based on a multi-layer perceptron model to capture the combined effects of temperature, pH, and redox potential; Performing risk probability mapping processing on the coupled concentration distribution of the stage according to a preset exposure risk threshold to generate the exposure risk probability of each processing node; The stage-coupled concentration distribution of the current processing stage is used as the initial diffusion concentration input of the next processing stage, and the diffusion path simulation processing, time attenuation correction processing, dynamic coupling processing and risk probability mapping processing are repeatedly performed until all processing stages are traversed to generate the dynamic simulation results of chemical exposure.

4. The dynamic simulation assessment method for chemical exposure in wastewater treatment processes according to claim 1, characterized in that: The exposure risk assessment process is performed based on the chemical exposure dynamic simulation results to generate a set of chemical exposure optimization strategies for the target wastewater treatment process, including: Performing risk level classification processing on the exposure risk probability in the dynamic simulation result of chemical exposure to generate a risk node set and a secondary risk node set; Performing reverse tracing analysis on the exposure concentration distribution in the risk node set to determine the risk source node and the corresponding dosing stage identifier; Extracting the corresponding dosage concentration and dosage time interval from the chemical dosage data set according to the dosage stage identifier of the risk source node, and generating a dosage concentration adjustment strategy, wherein the dosage concentration adjustment strategy includes: adjusting the dosage concentration reduction amplitude according to the ratio of the treatment time interval between the risk source node and the adjacent nodes while ensuring that the total dosage remains unchanged; and dynamically allocating the reduced dosage to adjacent treatment stages according to the weight ratio of the treatment time interval of each adjacent node while ensuring the conservation constraint of the total dosage; Performing time series optimization processing on the exposure contact time parameters in the secondary risk node set to generate a dosing timing optimization strategy, wherein the time series optimization processing includes: adjusting the distribution of dosing time intervals according to the correlation between the exposure contact time parameters and the equipment operating parameters; An emergency response triggering strategy is generated according to the spatial distribution characteristics of the risk node set and the secondary risk node set. The emergency response triggering strategy includes an emergency processing node identifier and an emergency processing operation instruction.

5. The dynamic simulation assessment method for chemical exposure in wastewater treatment process according to claim 4, characterized in that: The reverse tracing analysis of the exposure concentration distribution in the risk node set to determine the risk source node and the corresponding dosing stage identifier includes: Extracting spatial gradient change data of exposure concentration distribution of a risk node set from the chemical exposure dynamic simulation results; Perform diffusion direction backtracking based on the spatial gradient change data, and determine the upstream processing node corresponding to the direction of maximum concentration gradient change as a potential source node; Calling the dynamic simulation strategy network to perform gradient verification processing on the potential source node to generate a verification concentration distribution, wherein the gradient verification processing includes: performing a simulation of the addition concentration returning to zero and a simulation of the addition concentration decreasing according to the gradient, and performing a comprehensive verification by comparing the concentration distribution difference between the two simulation results; Determine the risk source node based on the spatial gradient similarity between the verified concentration distribution and the original exposure concentration distribution, wherein the difference is an indicator of the structural similarity between the verified concentration distribution and the original exposure concentration distribution, and a comprehensive assessment is performed in combination with the spatial correlation of the concentration gradient; The dosing phase identifier corresponding to the risk source node is matched from the chemical dosing data set and associated with the dosing concentration adjustment strategy.

6. The method for dynamic simulation assessment of chemical exposure in wastewater treatment process according to claim 1, characterized in that: The step of obtaining the chemical dosing data set and the real-time monitoring data set of the target wastewater treatment process includes: Extracting chemical addition records within a preset time window from a historical database of the wastewater treatment control system, wherein the chemical addition records include a chemical type identifier, an addition concentration, an addition timestamp, and an addition device identifier; Performing an outlier cleaning process on the dosing records to generate a cleaned chemical dosing data set, wherein the outlier cleaning process includes: removing records with dosing concentrations exceeding a preset safety range or with discontinuous dosing timestamps; Acquire water quality parameters, equipment operating parameters, and chemical residual concentrations corresponding to the preset time window from real-time monitoring equipment to generate an original monitoring data set; The original monitoring data set is subjected to time synchronization processing to generate a time-aligned real-time monitoring data set, wherein the time synchronization processing includes: interpolating and aligning water quality parameters, equipment operating parameters and residual concentration according to injection timestamps.

7. The method for dynamic simulation assessment of chemical exposure in wastewater treatment process according to claim 6, characterized in that: The performing time synchronization processing on the original monitoring data set to generate a time-aligned real-time monitoring data set includes: For each injection timestamp, extracting the water quality parameters and equipment operating parameters at the corresponding time point from the original monitoring data set; If the water quality parameters or equipment operating parameters are missing at the injection timestamp, the rate of change of the equipment operating parameters at adjacent time points is first detected. When the rate of change exceeds a preset threshold, the nearest neighbor interpolation method is used. Otherwise, the linear interpolation method is used to generate supplementary data to generate the supplemented water quality parameters and equipment operating parameters; The completed water quality parameters and equipment operation parameters are smoothed and filtered, and the smoothed and filtered water quality parameters, equipment operation parameters and corresponding chemical residual concentrations are integrated according to the addition timestamp to generate a real-time monitoring data set with consistent time dimension.

8. The method for dynamic simulation assessment of chemical exposure in wastewater treatment process according to claim 1, characterized in that: The dynamic simulation strategy network is trained by the following steps: Constructing a simulated training dataset and corresponding chemical exposure verification results. The simulated training dataset includes historical wastewater treatment process data, synthetic data generated based on fluid dynamics simulation, and adversarial sample data generated by adding random perturbations to historical parameters. The synthetic data includes extreme flow rates and abnormal water quality conditions. Calling the initial dynamic simulation network to perform dynamic exposure parameter extraction processing on the chemical addition data and the real-time monitoring data in the simulated training data set to generate a training dynamic exposure parameter set; Performing multi-stage iterative simulation processing based on the training dynamic exposure parameter set to generate predicted chemical exposure results; constructing a simulation loss function based on the difference between the predicted chemical exposure result and the chemical exposure verification result; The parameters of the initial dynamic simulation network are updated by a back-propagation algorithm until the simulation loss function converges, thereby obtaining a trained dynamic simulation strategy network.

9. The method for dynamic simulation assessment of chemical exposure in wastewater treatment process according to claim 8, characterized in that: The constructing a simulation loss function according to the difference between the predicted chemical exposure result and the chemical exposure verification result includes: Calculating the mean square error between the predicted chemical exposure result and the chemical exposure verification result as a basic loss; Performing weighted processing on the concentration points exceeding the preset safety threshold in the predicted chemical exposure results by applying a penalty coefficient to generate a penalty loss, wherein the penalty coefficient is a dimensionless weight parameter used to balance the risk levels of different exceeding limit concentrations; constructing a gradient consistency loss based on the correlation between the spatial gradient of the predicted chemical exposure results and the spatial gradient of the chemical exposure verification results; Performing a weighted summation of the base loss, penalty loss, and gradient consistency loss to generate the simulated loss function; The step of weighting the concentration points exceeding the preset safety threshold in the predicted chemical exposure results with penalty coefficients to generate penalty losses includes: extracting all concentration point data from the predicted chemical exposure results; Compare each concentration point data with a preset safety threshold, and if the concentration point data exceeds the safety threshold, calculate the difference between the portion exceeding the safety threshold and the safety threshold; Generate a local penalty term according to the product of the square of the difference and a preset penalty coefficient; The local penalty terms of all concentration point data are summed to generate the penalty loss.

10. A dynamic simulation and assessment system for chemical exposure in wastewater treatment processes, characterized in that: It includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the dynamic simulation assessment method for chemical exposure in the wastewater treatment process as described in any one of claims 1 to 9.

Images