Method and system for evaluating full-cycle treatment process of pollutants with high carbon-nitrogen ratio

Abstract

The invention relates to the technical field of data processing, in particular to a high-carbon-nitrogen-ratio pollutant full-cycle treatment process evaluation method and system, and the method comprises the steps: building a multi-objective optimization model; monitoring an environmental load contribution rate through real-time life cycle list data, and dynamically determining an environmental index weight; coupling operation parameters influencing upstream and downstream unit load balance are determined as model input; solving the model by adopting a hotspot sensing particle swarm optimization algorithm, wherein the algorithm introduces a dynamic penalty term related to an environment index weight and a contribution rate into a fitness function; and generating an operation strategy decision report according to the optimal solution set. According to the method, the defects of static state and low efficiency of the existing evaluation model are overcome, the real-time balance of the environmental and economic targets and the rapid generation of the operation strategy are realized, and the perception capability and the real-time guidance value of the evaluation method to the environmental risk are obviously enhanced. The method is suitable for guiding operation scheduling and parameter optimization of a high-carbon-nitrogen-ratio pollutant treatment process.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for evaluating the entire lifecycle treatment process of pollutants with a high carbon-to-nitrogen ratio. Background Technology

[0002] With increasing global focus on resource recycling and environmental sustainability, anaerobic digestion of food waste has become a mainstream resource recovery technology. The biogas slurry treatment process involves the configuration, scheduling, and operational decisions of multiple units, making its evaluation objectives complex. It must consider not only environmental impact but also ensure the maximization of economic benefits and resource recovery value. Existing process evaluations mostly employ static life-cycle assessments and cost accounting methods, lacking dynamic adaptability to real-time operating conditions and market fluctuations, making it difficult to support efficient management planning and optimization decisions.

[0003] In the field of multi-objective optimization, while existing technologies have proposed evaluation optimization models, most are based on simplified linear programming or employ staged optimization strategies, failing to fully construct and solve a complete process model involving large-scale mixed-integer nonlinear programming. Such simplified models cannot effectively capture the complex nonlinear coupling relationships in the process, resulting in resource allocation and scheduling schemes lacking global optimality. Furthermore, the models face technical barriers in integrating heterogeneous data to form a unified objective function, making accurate cost accounting and benefit assessment difficult. Existing solution algorithms, such as standard genetic algorithms or particle swarm optimization, exhibit poor robustness to discrete decision variables and nonlinear constraints, slow convergence speed, and low computational efficiency. These algorithms lack adaptive parameter adjustment mechanisms for the dynamic characteristics of the biogas slurry treatment process, making it difficult to quickly find a high-quality set of Pareto optimal strategies in a complex high-dimensional variable space.

[0004] In existing assessment models, environmental indicator weights are typically fixed, making it impossible to dynamically perceive and adjust environmental load hotspots generated by processes based on real-time data. This static optimization focus results in a strategy that lacks responsiveness to real-time environmental risks and struggles to quickly locate the optimal operating point that balances environmental benefits and economic costs at the Pareto front of the project.

[0005] Therefore, how to construct an efficient and robust multi-objective optimization evaluation method to achieve a dynamic trade-off between environmental and economic goals, and to use adaptive optimization algorithms to quickly generate process operation strategies with real-time guidance significance has become an urgent problem to be solved. To address this, a method and system for evaluating the entire lifecycle treatment process of pollutants with a high carbon-to-nitrogen ratio are proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for evaluating the full-cycle treatment process of pollutants with high carbon-to-nitrogen ratios. By establishing a dynamic multi-objective optimization model that integrates environmental hotspot perception and bidirectional coupled variables, and using the hotspot perception particle swarm optimization algorithm to efficiently solve the model, the invention achieves real-time trade-offs between environmental and economic objectives and rapid generation of operational strategies.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for evaluating the entire lifecycle treatment process of high carbon-to-nitrogen ratio pollutants, comprising: Establish a multi-objective optimization model to describe the configuration, allocation, and process operation variables of technical units in the processing flow; By using real-time lifecycle inventory data, the contribution rate of each technical unit in the processing flow to the total environmental load is monitored; when the contribution rate exceeds the preset environmental hotspot threshold, environmental indicator weights are dynamically determined and added to the technical unit. The process operation variables are used as dynamic decision variables in the multi-objective optimization model; by collecting performance index data of a set area in real time, the coupling operation parameters that affect the load balance of upstream and downstream units are determined and used as inputs to the multi-objective optimization model. A hotspot-aware particle swarm optimization algorithm is used to solve the multi-objective optimization model. The hotspot-aware particle swarm optimization algorithm introduces a dynamic penalty term in the fitness function that is related to the environmental index weights and the contribution rate. The dynamic penalty term drives the optimization direction of the particle swarm and converges to the Pareto optimal engineering solution set that minimizes the environmental hotspot load. Based on the Pareto optimal solution set, generate an operational strategy decision report.

[0008] Preferably, the technical units in the multi-objective optimization model include a pretreatment unit, a resource recovery unit, and a terminal treatment unit. The multi-objective optimization model uses binary integer variables to describe the configuration and allocation of these technical units, including the start / stop status of each unit and the flow distribution ratio of biogas slurry among different technical units. Simultaneously, the multi-objective optimization model uses continuous variables to describe the operational variables of the process. These operational variables encompass the feed flow rate of each unit, the dosage of chemical agents, and coupling control variables affecting the upstream and downstream load balance, thereby comprehensively characterizing the process structure and dynamic operating parameters.

[0009] Preferably, the real-time lifecycle inventory data is a set of quantitative indicators used to describe various inputs, outputs, and emissions in the processing flow. The set of quantitative indicators includes real-time input data, real-time output data, and real-time environmental emission data. The real-time input data includes the energy consumption, material input, and water consumption of each technical unit. The real-time output data includes the amount of recovered energy output and resource-based product output. The real-time environmental emission data includes the amount of pollutants discharged into the environmental media by the system.

[0010] Preferably, the environmental load contribution rate is an indicator used to determine the degree of influence of each technical unit on the total environmental load. The calculation of the indicator includes: selecting at least one environmental impact indicator as the basis for calculation; for each technical unit, calculating the unit contribution value of each technical unit to the selected environmental impact indicator based on real-time life cycle inventory data and preset environmental impact factors; dividing the unit contribution value by the total contribution value of all technical units to obtain the environmental load contribution rate of the unit, and the contribution rate serves as the basis for dynamically determining the weight of environmental indicators.

[0011] Preferably, the determination of the weights of the environmental indicators is as follows: A nonlinear increasing function is used as the weight assignment function. The weight assignment function takes the contribution rate of each technical unit as input and outputs the environmental objective function coefficients of the technical unit in the multi-objective optimization model.

[0012] When the contribution rate of the technical unit is lower than the preset hotspot threshold, the basic environmental objective function coefficients output by the weight assignment function are used to maintain the basic weight of the technical unit in the optimization evaluation.

[0013] When the contribution rate of the technical unit exceeds the preset environmental hotspot threshold, the weight assignment function calculates the environmental objective function coefficients in an accelerated incrementing manner according to the degree to which the contribution rate exceeds the threshold, so that the technical unit obtains a higher penalty weight in the optimization solution.

[0014] Preferably, the process operation variables are a set of continuous variables that affect the performance and energy consumption of the technical units. The set of continuous variables includes the feed flow rate, reaction temperature, pressure, dosage of chemical reagents, and reaction time of each technical unit. The process operation variables are all limited to a pre-defined feasible region, which is determined by the process design parameters and safe operating limits. The coupling operation parameters are determined by real-time acquisition of energy conversion efficiency data of the upstream reaction unit and initial processing load data of the downstream processing unit; based on the initial processing load data, the control variables required to maximize the upstream energy conversion efficiency under the condition of satisfying the downstream load constraints are calculated in real time; the control variables are used as inputs to the multi-objective optimization model to guide the optimization of the hotspot-aware particle swarm optimization algorithm.

[0015] Preferably, the hotspot-aware particle swarm optimization algorithm uses a hybrid encoding strategy to initialize the particle swarm, where the particle position vector simultaneously contains binary integer variables and continuous floating-point variables; the position of each particle in the particle swarm is evaluated, and the values ​​of the environmental objective function, economic objective function, and resource utilization objective function corresponding to the particle are calculated. When evaluating the environmental objective function, the hotspot-aware particle swarm optimization algorithm integrates a dynamic penalty term, which is composed of the environmental indicator weights and contribution rates, and is used to immediately penalize process paths where the environmental hotspot load exceeds a preset threshold. The hotspot-aware particle swarm optimization algorithm performs non-dominated sorting of the particle swarm based on the fitness value after integrating a dynamic penalty term, identifies and stores the currently found Pareto optimal solution set; and uses crowding distance as an auxiliary indicator to ensure the diversity and uniform distribution of the Pareto optimal solution set; updates the particle velocity and position based on the individual best position and the global best position of the particles; during the update process, the hotspot-aware particle swarm optimization algorithm dynamically adjusts the inertia weight and learning factor, so that the particles prioritize exploring a wide space in the early stage of optimization and converge to the Pareto front in the later stage of optimization; after updating the particle position, the hotspot-aware particle swarm optimization algorithm discretizes the binary integer variables in the position vector using a probability correction mechanism to ensure that the configuration and allocation of solutions are engineering feasible.

[0016] A full-cycle treatment process assessment system for high carbon-to-nitrogen ratio pollutants includes: Data acquisition and preprocessing module: Establishes a multi-objective optimization model to describe the configuration, allocation, and process operation variables of technical units in the processing flow; monitors the contribution rate of each technical unit in the processing flow to the total environmental load through real-time lifecycle inventory data; when the contribution rate exceeds the preset environmental hotspot threshold, dynamically determines and adds environmental indicator weights to the technical unit. Weight determination module: It uses process operation variables as dynamic decision variables for the multi-objective optimization model; by collecting performance index data of a set area in real time, it determines the coupling operation parameters that affect the load balance of upstream and downstream units, and uses them as inputs to the multi-objective optimization model. Hotspot sensing module: The hotspot sensing particle swarm optimization algorithm is used to solve the multi-objective optimization model. The hotspot sensing particle swarm optimization algorithm introduces a dynamic penalty term in the fitness function that is related to the environmental index weight and the contribution rate. The dynamic penalty term drives the optimization direction of the particle swarm and converges to the Pareto optimal engineering solution set that minimizes the environmental hotspot load. Results output module: Generates an operational strategy decision report based on the Pareto optimal solution set.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention monitors the environmental load contribution rate of each technical unit in the process using real-time lifecycle inventory data and dynamically determines and adjusts the weights of environmental indicators using a nonlinear increasing function. This mechanism solves the problem of statically fixed weights in multi-objective optimization, enabling the optimization model to respond to and penalize environmental hotspot loads generated in the process in real time, significantly enhancing the assessment method's ability to perceive environmental risks and its real-time guidance value for operational strategies.

[0018] 2. This invention employs a hotspot-aware particle swarm optimization algorithm to solve large-scale multi-objective optimization models. The algorithm integrates dynamically determined environmental indicator weights and contribution rates into a dynamic penalty term in the fitness function. This penalty term efficiently drives the particle swarm to quickly avoid regions with high environmental hotspot loads, improving the algorithm's robustness in complex high-dimensional spaces. Compared to the standard particle swarm optimization algorithm, this algorithm effectively solves the problems of slow convergence speed and low solution set quality in mixed-integer nonlinear programming models, achieving rapid and accurate identification of Pareto optimal policy sets.

[0019] 3. This invention uses process operation variables as dynamic decision variables and determines the coupled operation parameters affecting load balance by collecting key performance indicator data from upstream (e.g., gas production efficiency) and downstream (e.g., ammonia nitrogen concentration) in real time. This feature enables bidirectional engineering coupling optimization between core units such as anaerobic digestion and downstream treatment units. The optimization ensures that the final operating strategy is not only economically optimal but also adaptive and highly stable to process fluctuations, comprehensively improving the energy conversion efficiency and environmental compliance of the entire system. Attached Figure Description

[0020] Figure 1 This is a flowchart of a method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants proposed in this invention; Figure 2 This is a schematic diagram of the steps in the evaluation method for the whole-cycle treatment process of high carbon-to-nitrogen ratio pollutants proposed in this invention; Figure 3 This is a system structure diagram of an evaluation system for the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants proposed in this invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1 Please see Figures 1 to 2 This invention provides a method and system for evaluating the entire lifecycle treatment process of pollutants with a high carbon-to-nitrogen ratio. The technical solution is as follows: A method for evaluating the entire lifecycle treatment process of pollutants with high carbon-to-nitrogen ratios, such as Figure 1 - Figure 2 As shown, it includes: Establish a multi-objective optimization model to describe the configuration, allocation, and process operation variables of technical units in the processing flow; By using real-time lifecycle inventory data, the contribution rate of each technical unit in the processing flow to the total environmental load is monitored; when the contribution rate exceeds the preset environmental hotspot threshold, environmental indicator weights are dynamically determined and added to the technical unit. The process operation variables are used as dynamic decision variables in the multi-objective optimization model; by collecting performance index data of a set area in real time, the coupling operation parameters that affect the load balance of upstream and downstream units are determined and used as inputs to the multi-objective optimization model. A hotspot-aware particle swarm optimization algorithm is used to solve the multi-objective optimization model. The hotspot-aware particle swarm optimization algorithm introduces a dynamic penalty term in the fitness function that is related to the environmental index weights and the contribution rate. The dynamic penalty term drives the optimization direction of the particle swarm and converges to the Pareto optimal engineering solution set that minimizes the environmental hotspot load. Based on the Pareto optimal solution set, generate an operational strategy decision report. Furthermore, the technical units in the multi-objective optimization model include a pretreatment unit, a resource recovery unit, and a terminal treatment unit. The multi-objective optimization model uses binary integer variables to describe the configuration and allocation of these technical units, including the start / stop status of each unit and the flow distribution ratio of biogas slurry among different technical units. Simultaneously, the multi-objective optimization model uses continuous variables to describe the operational variables of the process. These operational variables encompass the feed flow rate of each unit, the dosage of chemical agents, and coupling control variables affecting the upstream and downstream load balance, thereby comprehensively characterizing the process structure and dynamic operating parameters. The objective function set of the multi-objective optimization model includes minimizing the environmental objective function, minimizing the economic objective function, and maximizing the resource utilization objective function. It achieves an integrated dynamic trade-off between environmental, economic, and resource utilization objectives, breaking through the limitations of traditional single-objective optimization. The multi-objective optimization model also includes a series of constraints, which at least cover: nonlinear model constraints describing the performance of each technical unit, mass and energy conservation constraints, and regulatory constraints on environmental emissions and product quality. This model fully defines multi-objective functions and full-process constraints through mixed-integer programming, covering conservation laws, nonlinear performance, and regulations. It overcomes the simplification problem of existing models and ensures that the optimization results simultaneously possess structural feasibility, physical robustness, and environmental compliance. Furthermore, the real-time lifecycle inventory data is a set of quantitative indicators used to describe various inputs, outputs, and emissions in the processing flow. The set of quantitative indicators includes real-time input data, real-time output data, and real-time environmental emission data. The real-time input data includes the energy consumption, material input, and water consumption of each technical unit. The real-time output data includes the amount of recovered energy output and resource-based product output. The real-time environmental emission data includes the amount of pollutants discharged into the environmental media by the system. The real-time lifecycle inventory data is correlated with the environmental objective function in the multi-objective optimization model through environmental impact factors. These environmental impact factors are used to convert physical quantities in the real-time input and emission data into environmental load indicators. The assessment method is periodically updated or uses pre-set environmental impact factors from an empirical database to maintain the accuracy of the environmental assessment. The real-time lifecycle inventory data is linked to the economic objective function in the multi-objective optimization model through economic factors. These economic factors include market price factors and fixed asset depreciation factors. These economic factors are used to quantify real-time input and output data into cost accounting indicators, ensuring that the model can make comprehensive economic trade-offs. This invention establishes a dynamic mathematical mapping relationship between real-time physical quantities of a process and environmental and economic objective functions by defining a detailed set of real-time input, output, and emission data and introducing environmental and economic factors. This overcomes the deficiency of data disconnect between data and model objectives in existing assessments, ensuring that the multi-objective optimization model is based on comprehensive, real-time, and quantified cost and environmental indicators. This mechanism provides accurate and robust real-time input for subsequent hotspot-aware optimization, significantly improving the real-time accuracy and economic feasibility of decision-making. Furthermore, the environmental load contribution rate is an indicator used to determine the degree of influence of each technical unit on the total environmental load. The calculation of the indicator includes: selecting at least one environmental impact indicator as the basis for calculation; for each technical unit, calculating the unit contribution value of each technical unit to the selected environmental impact indicator based on real-time life cycle inventory data and preset environmental impact factors; dividing the unit contribution value by the total contribution value of all technical units to obtain the environmental load contribution rate of the unit, which serves as the basis for dynamically determining the weight of environmental indicators. The calculation of the unit contribution value is based on the selected environmental impact indicator (e.g., greenhouse gas emission potential). Specifically, for each technology unit, the consumption of all input resources recorded in the real-time lifecycle inventory data (e.g., electricity consumption, raw material usage) is first obtained. Then, each consumption is multiplied by the preset environmental impact factor corresponding to that resource (i.e., the environmental load generated per unit of consumption, e.g., the carbon dioxide equivalent per kilowatt-hour). The values ​​obtained by multiplying all resource consumption of the technology unit by the corresponding environmental impact factor are summed to obtain the unit contribution value of the technology unit to the selected environmental impact indicator. The calculation of the total contribution value is determined by calculating the average contribution value of all technology units within a rolling time window. This rolling time window is a preset time range, and the total contribution value is the average of the sum of the contribution values ​​of all technology units within this window. Finally, the current unit contribution value of a single technology unit is divided by the average of the total contribution values ​​calculated in the rolling time window to obtain the environmental load contribution rate of that technology unit. The assessment method employs a periodic triggering mechanism for calculating the environmental load contribution rate. The period can be set to be triggered based on a predetermined time interval or an event-driven approach. After the environmental load contribution rate is calculated, it is used as the direct input of a nonlinear increasing function to dynamically determine the weight of the environmental indicators. The dynamic contribution rate calculation mechanism constructed in this invention achieves real-time and accurate quantification of environmental hotspot loads in the process by selecting environmental impact indicators and LCI data. The core innovation lies in the use of periodic triggering and rolling window averaging, effectively eliminating the impact of real-time data fluctuations and ensuring the stability and representativeness of weight updates. This mechanism establishes a direct feedback between real-time environmental status and nonlinear weight assignment, which is a key technical foundation for improving the adaptability of the hotspot perception optimization algorithm and ensuring the robustness of the strategy. Furthermore, the determination of the weights of the environmental indicators: A nonlinear increasing function is used as the weight assignment function. The weight assignment function takes the contribution rate of each technical unit as input and outputs the environmental objective function coefficients of the technical unit in the multi-objective optimization model. When the contribution rate of the technical unit is lower than the preset environmental hotspot threshold, the basic environmental objective function coefficients output by the weight assignment function are used to maintain the basic weight of the technical unit in the optimization evaluation. When the contribution rate of the technical unit exceeds the preset environmental hotspot threshold, the weight assignment function calculates the environmental objective function coefficients in an accelerated incrementing manner according to the degree to which the contribution rate exceeds the threshold, so that the technical unit obtains a higher penalty weight in the optimization solution. The environmental hotspot threshold is set based on the selected environmental impact indicators (e.g., water pollution emissions, energy consumption potential, etc.) and is determined comprehensively by combining historical operating data and industry regulatory requirements. First, for each selected environmental impact indicator, based on stable operating data of the process over a period of time or best practice data in the industry, the average environmental load contribution rate or average unit contribution value of each technical unit is calculated, which serves as the baseline contribution rate. Considering the process optimization objective, this threshold is usually set to a preset percentage higher than the baseline contribution rate (e.g., 10% to 20% higher than the average baseline contribution rate). This percentage represents the critical exceedance point that needs to be focused on and penalized by the optimization algorithm. The environmental hotspot thresholds are not fixed, but are adjusted and calibrated periodically (e.g., quarterly) or through event-driven mechanisms based on the pace of technological advancement in the industry, in order to ensure the advancement and effectiveness of the optimization objectives. The weights of the environmental indicators are determined using an exponentially increasing function or a higher-order power function as the weight assignment function. This weight assignment function includes at least one penalty coefficient, the value of which increases exponentially or non-linearly with the degree to which the contribution rate exceeds the environmental hotspot threshold. This achieves quantitative and enhanced penalty for environmental hotspots, significantly accelerating the optimization convergence. The basic environmental objective function coefficients are determined based on the average historical contribution rate of the process or a preset minimum objective function value, ensuring that the model can maintain a continuous assessment of environmental objectives even when no environmental hotspots occur. This invention achieves precise and enhanced adaptive punishment for environmental hotspot risks by constructing a nonlinear weighting mechanism, significantly improving the efficiency and reliability of multi-objective optimization. The method employs an exponentially increasing function or a higher-order power function as the weighting function, transforming "accelerated incrementing" into a quantifiable mechanism with a clear numerical form, overcoming the problem of punishment intensity relying on fuzzy experience. This enhanced punishment ensures that process paths with high environmental hotspot loads receive extremely high mathematical weights, efficiently driving the particle swarm to converge rapidly to the Pareto front that minimizes the environmental hotspot load, thus improving the accuracy and efficiency of solving large-scale MINLP models. Furthermore, by defining the coefficients of the basic environmental objective function, the continuity and stability of the model's evaluation are ensured when there are no environmental hotspots. Furthermore, the process operation variables are a set of continuous variables that affect the performance and energy consumption of the technical units. The set of continuous variables includes the feed flow rate, reaction temperature, pressure, dosage of chemical reagents, and reaction time of each technical unit. All process operation variables are limited to a pre-defined feasible region, which is determined by process design parameters and safe operating limits. The coupling operation parameters are determined by real-time acquisition of energy conversion efficiency data of the upstream reaction unit and initial processing load data of the downstream processing unit; based on the initial processing load data, the control variables required to maximize the upstream energy conversion efficiency under the condition of satisfying the downstream load constraints are calculated in real time; the control variables are used as inputs to the multi-objective optimization model to guide the optimization of the hotspot-aware particle swarm optimization algorithm. The determination of the coupling operation parameters is based on an online sub-optimization model or a model predictive control (MPC) strategy. The sub-optimization model takes maximizing energy conversion efficiency as its objective function and transforms the initial load index of the downstream processing unit into a hard constraint. The values ​​of the control variables are obtained by solving the sub-optimization model in real time. The calculated control variables serve as the assignments for the corresponding dynamic decision variables or the boundaries of the feasible region in the multi-objective optimization model. These assignments or boundaries are used in each iteration or periodic update of the hotspot-aware particle swarm optimization algorithm to ensure that the optimization solution is obtained within a dynamically feasible region that satisfies upstream and downstream coupling constraints. This invention achieves deep dynamic coupling of upstream and downstream unit load balancing by establishing an online sub-optimization model or model predictive control strategy, overcoming the problem of disconnect between process control and optimization. The method uses maximizing energy conversion efficiency as the sub-optimization objective and downstream load constraints as a hard limit, calculating the required control variables in real time. This mechanism ensures that the final decision strategy simultaneously meets the dual requirements of upstream energy output and downstream environmental compliance. Furthermore, by using these control variables as dynamic assignments to the master optimization model or as boundaries of feasible regions, effective connection is achieved between the real-time control level and the global multi-objective decision-making level, significantly improving the real-time adaptability, control accuracy, and economy of the full-cycle evaluation. Furthermore, the hotspot-aware particle swarm optimization algorithm employs a hybrid encoding strategy to initialize the particle swarm, where the particle position vector simultaneously contains binary integer variables and continuous floating-point variables. The position of each particle in the swarm is evaluated, and the values ​​of the corresponding environmental objective function, economic objective function, and resource utilization objective function are calculated. When evaluating the environmental objective function, the hotspot-aware particle swarm optimization algorithm integrates a dynamic penalty term, which is composed of the environmental indicator weights and contribution rates, and is used to immediately penalize process paths where the environmental hotspot load exceeds a preset threshold. The core steps of the hotspot-aware particle swarm optimization algorithm include: First, the hotspot-aware particle swarm optimization algorithm randomly generates an initial particle swarm within a defined solution space. The position vector of each particle uses a hybrid encoding (containing both discrete variables reflecting the configuration of technical units and continuous variables reflecting parameter optimization). Next, for each particle in the swarm, based on its current position vector (i.e., a specific solution), the values ​​of the environmental objective function, the economic objective function, and the resource utilization objective function are calculated respectively. When evaluating the environmental objective function, the hotspot-aware particle swarm optimization algorithm integrates a dynamic penalty term. Specifically, if the environmental load contribution rate of a certain technical unit exceeds a preset threshold, the product of the environmental indicator weight and the contribution rate is added as a penalty value to the particle's environmental objective function value to quantify the environmental hotspots in the penalty process. Then, the three objective function values ​​with integrated dynamic penalty terms are combined through non-dominated sorting, and the crowding distance is calculated to obtain the particle's fitness value. The hotspot-aware particle swarm optimization (PSO) algorithm updates and stores the currently found set of Pareto optimal solutions (i.e., non-dominated solutions) based on the particle's fitness value. Crowding distance is used to select solutions that are evenly distributed on the Pareto front to maintain the diversity of the solution set. The algorithm calculates the particle's new velocity using dynamically adjusted inertia weights and a learning factor, based on each particle's individual historical best position (i.e., individual optimal solution) and the currently found global best position (i.e., global optimal solution). This new velocity determines the particle's direction and distance of movement in the solution space. Subsequently, the particle's new position is updated based on the new velocity. After updating the particle positions, a probabilistic correction mechanism is applied to the binary integer portion of the position vector, mapping its values ​​to discrete integers (e.g., 0 or 1), ensuring that the solution is configurable and assignable in practical engineering. The above steps are repeated until the preset number of iterations or convergence criteria are reached. The fitness evaluation of the hotspot-aware particle swarm optimization algorithm involves integrating the environmental objective function, economic objective function, and resource utilization objective function through non-dominated ranking and crowding distance to form a fitness value. The dynamic penalty term is added to the environmental objective function as a product of the environmental indicator weights and contribution rates, quantifying the negative impact caused by exceeding environmental hotspot limits. The hotspot-aware particle swarm optimization algorithm dynamically adjusts the inertia weights using a non-linear decreasing function based on the number of iterations when updating particle positions. This ensures the algorithm has strong global search capabilities in the early stages of optimization and strong local exploitation capabilities in the later stages. The learning factor employs a dynamic adjustment strategy based on particle fitness levels to balance the weights of individual particle learning and group collaboration. The hotspot-aware particle swarm optimization algorithm, after updating particle positions, employs a Sigmoid probability correction mechanism to discretize the binary integer variables in the position vector. This mechanism maps continuous position variables to... The probability value within the interval is used to determine the final value of the binary variable, thereby ensuring the engineering feasibility of the configuration and allocation. This invention significantly improves the efficiency and accuracy of solving large-scale mixed-integer nonlinear programming multi-objective optimization models by constructing a hotspot-aware particle swarm optimization algorithm that integrates a hybrid encoding strategy and a dynamic adaptation mechanism. It achieves immediate and quantified penalty for environmental hotspot load paths. This mechanism efficiently drives the particle swarm to quickly avoid high environmental load regions, ensuring the algorithm's directional convergence and high efficiency. Simultaneously, by dynamically adjusting inertia weights and learning factors and employing a sigmoid probability correction mechanism to handle binary variables, this algorithm successfully solves the problems of slow convergence speed, poor solution set diversity, and incompatibility with discrete variable handling in traditional PSO when dealing with MINLP models. This ensures that the final output Pareto optimal policy set has engineering feasibility, high diversity, and accurate convergence quality. This invention achieves a dynamic and efficient trade-off between environmental benefits and economic costs. First, a multi-objective optimization model of mixed-integer nonlinear programming is constructed, incorporating all configuration, allocation, and operational variables in the process into the decision space. Second, environmental hotspots are dynamically perceived through real-time lifecycle data, and environmental indicator weights are adaptively added to the model, ensuring that decisions can respond to environmental risks in real time. Finally, a hotspot-aware particle swarm optimization algorithm is used to efficiently solve the dynamic model. This algorithm, utilizing a dynamic penalty term mechanism, successfully solves the convergence problem of large-scale mixed-integer optimization models, ultimately outputting an operational strategy report with real-time guidance and engineering feasibility. Example 2 like Figure 3 As shown in the figure, this embodiment 2 describes in detail the application of the method of the present invention in the dynamic operation scheduling of the biogas slurry treatment process of a large-scale anaerobic digestion plant for kitchen waste. The scenario description and objective setting of this facility's biogas slurry treatment system includes: a chemical flocculation dewatering unit (pretreatment), a membrane filtration unit and a struvite sedimentation unit (resource recovery), and a biochemical treatment unit (terminal treatment). This embodiment aims to utilize the system of the present invention to dynamically optimize and schedule the process operation for the next 24 hours, in order to minimize the total lifecycle cost. and global warming potential At the same time, maximize the recovery rate of phosphorus resources . The system establishes a mixed-integer nonlinear programming (MINLP) model. The decision variables of the model include: Binary integer variable: Start-up and shutdown status of the guano sedimentation unit . Continuous variable: Flow distribution ratio of biogas slurry among different units Real-time dosage of chemical flocculants Operating temperature of the membrane filtration unit The model's constraints cover material conservation, energy balance, the maximum processing capacity of the MBR unit, and the final effluent discharge. Regulatory limits. During the t=4-hour operation, the data acquisition and preprocessing module monitored in real time that, due to changes in the composition of food waste, the dosage of chemical flocculant added by the flocculation and dewatering unit was adjusted to ensure the quality of the filtrate. It temporarily exceeded the historical average by 30%. Real-time lifecycle inventory data calculations show that the chemical and energy consumption of this unit causes its environmental load contribution to the total global warming potential to exceed the preset 65% environmental hotspot threshold. The dynamic weighting module immediately adopts an exponentially increasing function to adjust the environmental objective function. The penalty coefficient associated with the flocculation unit is dynamically increased by 2.5 times. This enhanced penalty will be applied immediately to the fitness assessment of the optimization model; the penalty coefficient is the weight of the environmental indicator. To ensure process stability, the system collects real-time data on the biogas conversion efficiency of the upstream anaerobic digester and the initial chemical oxygen demand (COD) load of the downstream MBR unit. The coupling parameter determination module runs an online sub-optimization model with the goal of maximizing biogas conversion efficiency while satisfying the downstream MBR initial COD load constraint. The sub-optimization model calculates the optimal cumulative stirring time for the anaerobic digester in real time. This value is then assigned as a dynamic control variable in the master multi-objective optimization model to ensure that the master optimization is carried out within a dynamically feasible process window. The system initiates a hotspot-aware particle swarm optimization algorithm to solve the MINLP model with integrated dynamic penalties: the algorithm uses hybrid encoding to initialize the particle swarm, and the position vector simultaneously contains... and Variables such as [variable name missing]. In fitness evaluation, a dynamic penalty term (consisting of a penalty weight multiplied by the contribution rate, equal to 2.5 times the penalty weight) is integrated to drive the particle swarm to avoid paths with high chemical flocculant dosages. During the algorithm's iteration process, the inertia weight ([variable name missing]) is used. A non-linear decreasing function based on the number of iterations is used for dynamic adjustment to balance global search and local exploitation capabilities. Binary variables are discretized using a Sigmoid probability correction mechanism. The hotspot-aware particle swarm optimization algorithm efficiently converges to a new Pareto optimal engineering solution set. A running strategy report is generated based on the new Pareto optimal solution set: Immediately adjust the strategy: Despite the low cost of flocculation, the system decision will allocate more flow to the flocculation unit due to the amplified penalty weight of global warming potential. Reduce by 15% and redirect the flow to the struvite settling unit ( Maintain at 1). Operating parameter adjustment: The operation report indicates that the cumulative stirring time of the anaerobic tank should be adjusted to [the specified value]. This dynamic strategy successfully achieves a dynamic balance between minimizing GWP (avoiding chemical hotspots) and maximizing resource recovery (increasing phosphorus recovery) to maintain high biogas production. Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the entire lifecycle treatment process of pollutants with high carbon-to-nitrogen ratios. Includes, characterized in that: Establish a multi-objective optimization model to describe the configuration, allocation, and process operation variables of technical units in the processing flow; By using real-time lifecycle inventory data, the contribution rate of each technical unit in the processing flow to the total environmental load is monitored; when the contribution rate exceeds the preset environmental hotspot threshold, environmental indicator weights are dynamically determined and added to the technical unit. The process operation variables are used as dynamic decision variables in the multi-objective optimization model; by collecting performance index data of a set area in real time, the coupling operation parameters that affect the load balance of upstream and downstream units are determined and used as inputs to the multi-objective optimization model. A hotspot-aware particle swarm optimization algorithm is used to solve the multi-objective optimization model. The hotspot-aware particle swarm optimization algorithm introduces a dynamic penalty term in the fitness function that is related to the environmental index weights and the contribution rate. The dynamic penalty term drives the optimization direction of the particle swarm and converges to the Pareto optimal engineering solution set that minimizes the environmental hotspot load. Based on the Pareto optimal solution set, generate an operational strategy decision report.

2. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, characterized in that: The technical units in the multi-objective optimization model include a pretreatment unit, a resource recovery unit, and a terminal treatment unit. The multi-objective optimization model uses binary integer variables to describe the configuration and allocation of the technical units, including the start-up and shutdown status of each unit and the flow distribution ratio of biogas slurry among different technical units. At the same time, the multi-objective optimization model uses continuous variables to describe the operational variables of the process. The operational variables cover the feed flow rate of each unit, the dosage of chemical agents, and the coupling control variables that affect the load balance between upstream and downstream, thereby comprehensively characterizing the structure and dynamic operating parameters of the process.

3. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, characterized in that: The real-time lifecycle inventory data is a set of quantitative indicators used to describe various inputs, outputs, and emissions in the processing flow. The set of quantitative indicators includes real-time input data, real-time output data, and real-time environmental emission data. The real-time input data includes the energy consumption, material input, and water consumption of each technical unit. The real-time output data includes the amount of recovered energy output and resource-based product output. The real-time environmental emission data includes the amount of pollutants discharged into the environmental media by the system.

4. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, characterized in that: The environmental load contribution rate is an indicator used to determine the degree of influence of each technical unit on the total environmental load. The calculation of the indicator includes: selecting at least one environmental impact indicator as the basis for calculation; for each technical unit, calculating the unit contribution value of each technical unit to the selected environmental impact indicator based on real-time life cycle inventory data and preset environmental impact factors; dividing the unit contribution value by the total contribution value of all technical units to obtain the environmental load contribution rate of the unit. The contribution rate serves as the basis for dynamically determining the weight of environmental indicators.

5. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, wherein the determination of the environmental indicator weights is characterized in that: A nonlinear increasing function is used as the weight assignment function. The weight assignment function takes the contribution rate of each technical unit as input and outputs the environmental objective function coefficients of the technical unit in the multi-objective optimization model. When the contribution rate of the technical unit is lower than the preset hotspot threshold, the basic environmental objective function coefficients output by the weight assignment function are used to maintain the basic weight of the technical unit in the optimization evaluation. When the contribution rate of the technical unit exceeds the preset environmental hotspot threshold, the weight assignment function calculates the environmental objective function coefficients in an accelerated incrementing manner according to the degree to which the contribution rate exceeds the threshold, so that the technical unit obtains a higher penalty weight in the optimization solution.

6. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, characterized in that: The process operation variables are a set of continuous variables that affect the performance and energy consumption of the technical units. The set of continuous variables includes the feed flow rate, reaction temperature, pressure, dosage of chemical reagents, and reaction time of each technical unit. The process operation variables are all limited to a pre-defined feasible region, which is determined by the process design parameters and safe operating limits. The coupling operation parameters are determined by real-time acquisition of energy conversion efficiency data of the upstream reaction unit and initial processing load data of the downstream processing unit; based on the initial processing load data, the control variables required to maximize the upstream energy conversion efficiency under the condition of satisfying the downstream load constraints are calculated in real time; the control variables are used as inputs to the multi-objective optimization model to guide the optimization of the hotspot-aware particle swarm optimization algorithm.

7. The method for evaluating the full-cycle treatment process of high carbon-to-nitrogen ratio pollutants according to claim 1, characterized in that, The hotspot-aware particle swarm optimization algorithm uses a hybrid encoding strategy to initialize the particle swarm, where the position vector of each particle contains both binary integer variables and continuous floating-point variables. The algorithm evaluates the position of each particle in the swarm and calculates the values ​​of the environmental objective function, economic objective function, and resource utilization objective function corresponding to that particle. When evaluating the environmental objective function, the hotspot-aware particle swarm optimization algorithm integrates a dynamic penalty term, which consists of the environmental indicator weights and contribution rates, and is used to immediately penalize process paths where the environmental hotspot load exceeds a preset threshold. The hotspot-aware particle swarm optimization algorithm performs non-dominated sorting of the particle swarm based on the fitness value after integrating a dynamic penalty term, identifies and stores the currently found Pareto optimal solution set; and uses crowding distance as an auxiliary indicator to ensure the diversity and uniform distribution of the Pareto optimal solution set; updates the particle velocity and position based on the individual best position and the global best position of the particles; during the update process, the hotspot-aware particle swarm optimization algorithm dynamically adjusts the inertia weight and learning factor, so that the particles prioritize exploring a wide space in the early stage of optimization and converge to the Pareto front in the later stage of optimization; after updating the particle position, the hotspot-aware particle swarm optimization algorithm discretizes the binary integer variables in the position vector using a probability correction mechanism to ensure that the configuration and allocation of solutions are engineering feasible.

8. A system for evaluating the entire lifecycle treatment process of pollutants with a high carbon-to-nitrogen ratio, characterized in that, include: Data acquisition and preprocessing module: Establish a multi-objective optimization model to describe the configuration, allocation, and process operation variables of technical units in the processing flow; By using real-time lifecycle inventory data, the contribution rate of each technical unit in the processing flow to the total environmental load is monitored; when the contribution rate exceeds the preset environmental hotspot threshold, environmental indicator weights are dynamically determined and added to the technical unit. Weight determination module: It uses process operation variables as dynamic decision variables for the multi-objective optimization model; by collecting performance index data of a set area in real time, it determines the coupling operation parameters that affect the load balance of upstream and downstream units, and uses them as inputs to the multi-objective optimization model. Hotspot sensing module: The hotspot sensing particle swarm optimization algorithm is used to solve the multi-objective optimization model. The hotspot sensing particle swarm optimization algorithm introduces a dynamic penalty term in the fitness function that is related to the environmental index weight and the contribution rate. The dynamic penalty term drives the optimization direction of the particle swarm and converges to the Pareto optimal engineering solution set that minimizes the environmental hotspot load. Results output module: Generates an operational strategy decision report based on the Pareto optimal solution set.

Images