A decision-making method and system for the resource utilization direction of multi-source coal-based solid waste

Abstract

The present invention discloses a decision-making method and system for the resource utilization direction of multi-source coal-based solid waste, which relates to the field of environmental engineering technology. It includes constructing a collaborative characteristic model of coal-based solid waste, constructing a solid waste characteristic spectrum through multi-dimensional physical and chemical characteristic detection, analyzing key collaborative factors, and generating an interaction relationship matrix; based on the interaction relationship matrix, establishing a matching relationship between the combination of coal-based solid waste and the resource utilization technology chain, and outputting a cluster of technical solutions; generating a scheme priority sequence through dynamic influencing factor analysis and policy response mechanism; generating a configuration strategy set based on the scheme priority sequence and regional spatio-temporal constraints; establishing a closed-loop feedback engine, and triggering the dynamic update of the configuration strategy set when external conditions change are monitored. The present invention solves the problems of lack of systematic evaluation and overall planning for the treatment of coal-based solid waste in traditional methods by accurately quantifying the complex synergistic effects between solid waste components, improves the resource conversion efficiency, and optimizes the economic benefits and environmental performance.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental engineering, and particularly to a decision-making method and system for the resource utilization direction of multi-source coal-based solid wastes. Background Art

[0002] Coal-based solid wastes (such as coal gangue, fly ash, desulfurized gypsum, etc.) are a large amount of solid wastes generated during coal mining, washing and combustion. Their cumulative stacking not only occupies a large amount of land resources, but also may cause environmental pollution problems. At present, the resource utilization of coal-based solid wastes mainly focuses on single fields such as building material production, road base, and land reclamation, lacking a systematic comprehensive utilization decision-making method. Traditional solid waste treatment usually adopts a simple linear decision-making model, relying on empirical judgment or single-index evaluation, and it is difficult to cope with the complex component changes and diverse utilization requirements of multi-source coal-based solid wastes.

[0003] In recent years, with the tightening of environmental protection policies and the promotion of the concept of circular economy, the resource utilization of coal-based solid wastes has received extensive attention. However, there are still obvious deficiencies in the existing technology in the decision-making process of coal-based solid waste resource utilization: on the one hand, the understanding of the complex synergistic effects among solid waste components is limited, and it is difficult to achieve efficient component matching; on the other hand, there is a lack of a dynamic response mechanism, and it is difficult to adapt to external factors such as policy changes and market fluctuations. These problems lead to low resource conversion efficiency and poor economic benefits, restricting the depth and breadth of coal-based solid waste resource utilization. Summary of the Invention

[0004] The present invention provides a decision-making method and system for the resource utilization direction of multi-source coal-based solid wastes, which are used to solve the technical problems that the decision-making of coal-based solid waste resource utilization in the existing technology lacks systematicness and is difficult to cope with complex component changes and dynamic adjustment of the external environment.

[0005] In view of this, the first aspect of the present invention provides a decision-making method for the resource utilization direction of multi-source coal-based solid wastes, including:

[0006] Construct a coal-based solid waste collaborative characteristic model, construct a solid waste characteristic spectrum through multi-dimensional physical and chemical property detection, analyze key collaborative factors, and generate an interaction relationship matrix;

[0007] Based on the interaction relationship matrix, establish a matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm, and output a cluster of technical solutions;

[0008] Based on the cluster of technical solutions, generate a scheme priority sequence through dynamic influencing factor analysis and policy response mechanism;

[0009] Based on the scheme priority sequence and regional spatio-temporal constraints, generate a configuration strategy set through a constraint optimization method;

[0010] Establish a closed-loop feedback engine to trigger the dynamic update of the configuration policy set when external condition changes are detected.

[0011] Optionally, constructing a collaborative characteristic model of coal-based solid waste includes:

[0012] Preprocess the original samples of multi-source coal-based solid waste to obtain a standardized sample set;

[0013] Perform multi-dimensional physical and chemical property detections on the standardized sample set to obtain physical and chemical property data;

[0014] Construct a feature space based on the physical and chemical property data and perform dimensionality reduction processing to generate a solid waste feature spectrum;

[0015] Calculate and generate a set of key collaborative factors based on the physical and chemical property data;

[0016] Based on the set of key collaborative factors and the solid waste feature spectrum, construct an interaction relationship matrix between the components of coal-based solid waste and generate a collaborative conversion characteristic data set.

[0017] Optionally, establishing the matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm includes:

[0018] Construct a knowledge base for the resource utilization technology chain, which includes a set of resource utilization technology nodes and technology chain connection rules;

[0019] Determine the coal-based solid waste combination based on the interaction relationship matrix and the collaborative conversion characteristic data set, calculate the collaborative matching degree of different coal-based solid waste combinations, and generate a set of solid waste combination schemes through an incremental clustering algorithm;

[0020] Utilize the set of key collaborative factors to establish an adaptability evaluation model for the coal-based solid waste combination and the resource utilization technology nodes;

[0021] Based on the output results of the adaptability evaluation model, construct a bipartite graph model between the set of solid waste combination schemes and the set of resource utilization technology nodes;

[0022] Apply a multi-objective evolutionary algorithm to optimize the path on the bipartite graph model to generate a Pareto optimal solution set;

[0023] Conduct hierarchical clustering analysis on the Pareto optimal solution set to obtain a family of technical solutions, and optimize and configure the process parameters of the family of technical solutions in combination with the collaborative conversion characteristic data set, and integrate them to form a cluster of technical solutions.

[0024] Optionally, generating a scheme priority sequence through dynamic influencing factor analysis and policy response mechanism includes:

[0025] Construct a dynamic influencing factor evaluation system to conduct a temporal evaluation of each scheme in the cluster of technical solutions;

[0026] Establish a policy impact matrix, quantify the impact coefficients of policy factors on each evaluation index, and calculate the policy sensitivity;

[0027] Analyze the synergy relationship among the technical solutions within the technical solution cluster, construct a synergy effect evaluation model, and generate a synergy effect score;

[0028] Design a dynamic coupling evaluation function, integrate the dynamic score, policy sensitivity, and synergy effect score to generate a comprehensive evaluation value of the solution;

[0029] Based on the comprehensive evaluation value of the solution, apply the time series TOPSIS method to generate a solution priority sequence.

[0030] Optionally, generating a configuration policy set through a constraint optimization method includes:

[0031] Construct a dynamic spatio-temporal constraint system, and quantify geographical distribution, resource flow, seasonal fluctuations, and infrastructure conditions into a set of constraint parameters;

[0032] Based on the solution priority sequence, establish a technical solution deployment time series diagram, and set the implementation period and dependency relationship of each solution;

[0033] Design a multi-stage constraint optimization model, with the comprehensive regional resource benefit as the objective function, and use the set of constraint parameters as boundary conditions;

[0034] Apply the piecewise linear programming method to solve the multi-stage constraint optimization model and generate a configuration plan;

[0035] Perform a robustness analysis on the configuration plan, simulate the scenario of constraint condition fluctuations, and generate a set of configuration policies.

[0036] Optionally, triggering the dynamic update of the configuration policy set includes:

[0037] Real-time monitor the content of key solid waste components, environmental protection policy information, and regional carbon emission data;

[0038] Calculate the fluctuation value of the content of key solid waste components, the update flag of the environmental protection policy standard version, and the deviation value of the regional carbon emission quota;

[0039] Establish an event handling mechanism, monitor external conditions and determine whether the trigger standard is reached;

[0040] Based on the bipartite graph model reconstruction algorithm, locate the affected resource utilization technology chain nodes and the solid waste combination relationship according to the type of trigger event, and perform local reconstruction and parameter re-optimization on the matching relationship.

[0041] Optionally, the physical and chemical property data includes element composition data, specific surface area data, pore structure parameters, and pyrolysis characteristic parameters;

[0042] The key cofactor set includes substance conversion cofactors, element migration cofactors, and reaction kinetics cofactors.

[0043] In a second aspect of the present invention, a decision-making system for the resource utilization direction of multi-source coal-based solid waste is provided, including:

[0044] A characteristic analysis module for constructing a coal-based solid waste collaborative characteristic model, constructing a solid waste characteristic spectrum through multi-dimensional physical and chemical property detection, analyzing key cofactors, and generating an interaction relationship matrix;

[0045] A solution generation module for establishing a matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm based on the interaction relationship matrix and the collaborative conversion characteristic data set, and outputting a cluster of technical solutions;

[0046] A priority evaluation module for generating a solution priority sequence based on the cluster of technical solutions through dynamic influencing factor analysis and a policy response mechanism;

[0047] A strategy optimization module for generating a configuration strategy set through a constraint optimization method based on the solution priority sequence and regional spatio-temporal constraints;

[0048] A dynamic adjustment module for establishing a closed-loop feedback engine to trigger the dynamic update of the configuration strategy set when external condition changes are detected.

[0049] The beneficial effects of the present invention are as follows: By accurately identifying the synergistic effect between the components of coal-based solid waste, the present invention improves the conversion efficiency of traditional single-component treatment, optimizes the substance conversion process, and improves the level of comprehensive resource utilization; The combination of accurately matched technical paths and multi-objective optimization algorithms reduces ineffective conversion and resource waste, improves the input-output ratio, and enhances the economic efficiency of solid waste treatment; Through process parameter optimization and technical chain collaborative configuration, harmful substance emissions and energy consumption are reduced, and the environmental friendliness of the solid waste treatment process is improved; The application of multi-dimensional physical and chemical property analysis and collaborative models enables the system to handle more complex and diverse coal-based solid waste combinations, overcoming the limitations of traditional methods in dealing with complex components. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0051] Figure 1 It is a flowchart of a decision-making method for the resource utilization direction of multi-source coal-based solid waste.

[0052] Figure 2 It is a flow chart for constructing a collaborative characteristic model of coal-based solid waste for a decision-making method in the direction of resource utilization of multi-source coal-based solid waste.

[0053] Figure 3 It is a flow chart for generating a sequence of scheme priorities for a decision-making method in the direction of resource utilization of multi-source coal-based solid waste. Specific implementation manners

[0054] In order to make the invention objectives, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0055] Example 1, referring to Figures 1 to 3 , which is the first embodiment of the present invention. This embodiment provides a decision-making method for the direction of resource utilization of multi-source coal-based solid waste. The flow chart of this method is as Figure 1 shown, and this method includes:

[0056] S1: Construct a collaborative characteristic model of coal-based solid waste. By detecting multi-dimensional physical and chemical characteristics, construct a solid waste characteristic spectrum, analyze key collaborative factors, and generate an interaction relationship matrix.

[0057] In the specific implementation manners of the present invention, the flow chart for constructing the collaborative characteristic model of coal-based solid waste is as Figure 2 shown, and specifically includes the following steps:

[0058] S1.1: Uniquely identify and preprocess the original samples of multi-source coal-based solid waste to obtain a standardized sample set.

[0059] Specifically, the identification uses a five-tuple code, including a geographical area code, a source power plant code, a solid waste type classification code, a production date (in the format of YYYYMMDD), and a batch number. The preprocessing includes crushing, drying, screening, and homogenization.

[0060] S1.2: Perform multi-dimensional physical and chemical characteristic detection on the standardized sample set to obtain physical and chemical characteristic data, and record the process environment parameters at the same time.

[0061] Among them, the physical and chemical property data include elemental composition data (major elements, trace elements), specific surface area data, pore structure parameters, and pyrolysis property parameters. Specifically, the content data of major elements, including Al, Si, Ca, Fe, S, etc., are determined by double verification using X-ray fluorescence spectrometry (XRF) and inductively coupled plasma optical emission spectrometry (ICP-OES); the content data of trace elements, including Na, K, Mg, etc., are determined by inductively coupled plasma mass spectrometry (ICP-MS); the specific surface area data are obtained by the BET nitrogen adsorption method; the pore structure parameters are determined by the nitrogen adsorption-desorption method; under a nitrogen atmosphere (100 mL / min), starting from room temperature, the temperature is increased to 900 °C at three heating rates of 5 °C / min, 10 °C / min, and 15 °C / min respectively, and the pyrolysis property parameters, including the maximum weight loss rate, the maximum weight loss temperature, and the residual mass percentage, are obtained through a synchronous thermal analyzer.

[0062] Furthermore, the process environment parameters include process operation parameters (temperature, pressure, gas flow rate, residence time) and external constraint parameters (environmental protection standard limits, resource utilization rate requirements). The process operation parameters are collected through the DCS system, and the external constraint parameters are obtained through the enterprise environmental protection management system. The system sets up a data validity verification mechanism: when the detected data deviates from the mean range, the data review process is triggered; if the review confirms that the data is abnormal, re-sampling and detection are carried out.

[0063] S1.3: Construct a feature space based on the physical and chemical property data and perform dimensionality reduction processing to generate a solid waste feature spectrum.

[0064] Specifically, organize the physical and chemical property data into a feature matrix; after performing Z-score standardization on the feature matrix, use the principal component analysis method (PCA) for dimensionality reduction: first calculate the covariance matrix, solve the eigenvalues and eigenvectors, arrange the eigenvalues in descending order, select the first k principal components whose cumulative contribution rate reaches the preset contribution rate threshold, generate the dimensionality-reduced solid waste feature spectrum, and evaluate the dimensionality reduction effect by calculating the reconstruction error, with the error threshold set to 0.15. Preferably, the preset contribution rate threshold is 85%.

[0065] S1.4: Calculate and generate a set of key synergistic factors based on the physical and chemical property data.

[0066] Among them, the key cofactor set includes substance conversion cofactors, element migration cofactors, and reaction kinetics cofactors. Specifically, based on the major element content data, the molar ratios of oxides such as aluminum-silicon ratio (Al2O3 / SiO2), calcium-sulfur ratio (CaO / SO3), and iron-calcium ratio (Fe2O3 / CaO) are calculated as substance conversion cofactors; based on the trace element content data, the standardized molar ratios of characteristic element pairs such as Na / K and Mg / Ca are calculated as element migration cofactors; the weighted geometric mean method is used to integrate specific surface area data, pore structure parameters, and pyrolysis characteristic parameters to construct reaction kinetics cofactors, and the weight coefficients are determined by analysis of variance.

[0067] S1.5: Based on the key cofactor set and the solid waste characteristic spectrum, construct an interaction relationship matrix among the components of coal-based solid waste.

[0068] Specifically, the key cofactor set is associated with the dimension-reduced solid waste characteristic spectrum to construct an enhanced feature matrix. This matrix constructs the interaction relationship through the following steps: First, partial correlation analysis is used to quantify the correlation strength among the substance conversion cofactor, element migration cofactor, and reaction kinetics cofactor; then, a symmetric interaction relationship matrix is constructed, and the matrix element values are represented by standardized correlation coefficients, with a value range of [-1, 1], where a positive value indicates a synergistic promotion effect, a negative value indicates an inhibitory competition effect, and an absolute value of the correlation coefficient greater than 0.5 indicates a significant correlation; finally, the interaction relationship matrix is subjected to Z-score standardization processing and a significance test is performed through Bootstrap resampling (the number of resampling times is 500 times, and the confidence level is 95%).

[0069] Preferably, the method for constructing the interaction relationship matrix innovatively integrates three types of cofactors: substance conversion, element migration, and reaction kinetics, breaking through the limitations of only relying on the characteristics of single components or simple linear superposition in traditional solid waste treatment. By establishing an enhanced feature matrix and a standardized correlation coefficient system, the precise quantification and characterization of the complex synergistic effect among the components of coal-based solid waste are realized. In particular, the method combining covariance analysis and partial least squares regression effectively reveals the synergistic promotion and inhibitory competition mechanisms among the components. At the same time, the significance test based on Bootstrap resampling ensures the statistical reliability of the interaction relationship matrix, providing reliable data support for the subsequent orthogonal test scheme design and technical path optimization.

[0070] S1.6: Establish a coal-based solid waste co-conversion experimental platform, conduct gradient combination experiments, and generate a co-conversion characteristic data set.

[0071] Specifically, select the pairs of coal-based solid waste components that are significantly correlated in the interaction relationship matrix (through Bootstrap significance test and the absolute value of the correlation coefficient is greater than 0.5), design an orthogonal test scheme, use temperature, gas flow rate, residence time, and mixing ratio as the investigation factors, and carry out a co-conversion experiment within the range of external constraint parameters to obtain the co-conversion characteristic data set of each component under different process conditions.

[0072] S2: Based on the interaction relationship matrix and the co-conversion characteristic data set, establish the matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm, and output a cluster of configurable technology solutions.

[0073] In the specific implementation manner of the present invention, S2 specifically includes the following steps:

[0074] S2.1: Construct a knowledge base for the resource utilization technology chain, and the knowledge base for the resource utilization technology chain includes a set of resource utilization technology nodes and technology chain connection rules.

[0075] Specifically, the set of resource utilization technology nodes includes thermochemical conversion nodes (pyrolysis, gasification, combustion), physicochemical conversion nodes (materialization, extraction), and environmental protection treatment nodes (flue gas purification, wastewater treatment). Each technology node includes raw material adaptability parameters (substance composition range, particle size requirements), process operation parameters (temperature, pressure, gas flow rate, residence time), product characteristic parameters (yield, quality index), and technical and economic parameters (investment cost, operation cost). The technology chain connection rules define the pre-order dependence relationship, material flow direction constraint, and energy balance requirement between technology nodes, are represented by a directed graph structure, and mark the compatibility conditions of each connection.

[0076] S2.2: Determine the coal-based solid waste combination based on the interaction relationship matrix and the co-conversion characteristic data set, calculate the co-matching degree of different coal-based solid waste combinations, and generate a set of solid waste combination solutions through an incremental clustering algorithm.

[0077] Specifically, extract the pairs of coal-based solid waste components that are significantly correlated (the absolute value of the correlation coefficient is greater than 0.5 and passes the Bootstrap resampling significance test) from the interaction relationship matrix to construct a candidate pool for coal-based solid waste combinations; combine the co-conversion characteristic data set to construct an evaluation index system (including conversion efficiency, co-effect intensity, economic feasibility), and establish a fuzzy evaluation matrix; calculate the co-matching degree for each coal-based solid waste combination, and the co-matching degree is defined as the fuzzy comprehensive evaluation result; apply the incremental clustering algorithm (the initial clustering center is the coal-based solid waste combination with the highest co-matching degree, the clustering radius is 0.3, and the incremental step size is 0.05) to classify the coal-based solid waste combinations to form a set of solid waste combination solutions.

[0078] S2.3: Using the key co-factor set, establish an adaptability evaluation model for the matching between coal-based solid waste combinations and resource utilization technology nodes using a BP neural network.

[0079] Specifically, construct a three-layer neural network model. The input layer includes the material conversion co-factor, element migration co-factor, and reaction kinetics co-factor of the solid waste combination. The hidden layer uses the ReLU activation function, and the output layer corresponds to the adaptability score of the resource utilization technology node. Use the co-transformation characteristic data set as the training set and optimize the model parameters using five-fold cross-validation. For each coal-based solid waste combination, predict its adaptability score with each resource utilization technology node and calculate the 95% confidence interval. When the adaptability score is higher than the preset adaptability threshold (0.7), establish a preliminary matching relationship between the coal-based solid waste combination and the corresponding resource utilization technology node.

[0080] S2.4: Based on the output results of the adaptability evaluation model, construct a bipartite graph model between the set of solid waste combination solutions and the set of resource utilization technology nodes.

[0081] Specifically, the left nodes of the bipartite graph model represent each coal-based solid waste combination in the set of solid waste combination solutions, and the right nodes represent each technology node in the set of resource utilization technology nodes. The weight of the edge is determined by the adaptability score and is assigned three attributes: resource conversion attribute (based on the co-transformation efficiency data in S1.6), economic attribute (based on the cost-benefit data of the technology node), and environmental impact attribute (based on the emission data of the technology node). When the adaptability score is lower than the critical value (0.4), no connection edge is established.

[0082] S2.5: Apply the improved NSGA-Ⅲ algorithm to the bipartite graph model, and perform path optimization with the goals of maximizing resource conversion efficiency, optimizing economic benefits, and minimizing environmental impact to generate a Pareto optimal solution set.

[0083] Specifically, set the population size to 100 individuals, the number of generations to 500, the crossover probability to 0.8, and the mutation probability to 0.1. Use the adaptive reference point generation strategy to enhance the coverage of the search space and introduce the crowding distance to maintain the diversity of the solution set. For each resource utilization technology chain path, comprehensively evaluate its performance in the three objective dimensions and use the dominance sorting to determine the non-dominated solution set. Introduce the technology chain connection rule as a constraint condition to eliminate the paths that do not meet the process logic, and finally form a Pareto optimal solution set.

[0084] S2.6: Perform hierarchical clustering analysis on the Pareto optimal solution set to obtain technology solution families, optimize and configure the process parameters of the technology solution families in combination with the co-transformation characteristic data set, and integrate them into configurable technology solution clusters.

[0085] Specifically, an affinity matrix is constructed based on the similarity of the technology chain path. The optimal number of clusters K is determined using the Ward minimum variance method, and the hierarchical clustering algorithm is applied to generate technology solution families. For each technology solution family, combined with the co-transformation characteristic dataset in S1.6, the response surface method is used to optimize the process parameter configuration (temperature, gas flow rate, residence time, mixing ratio). The Box-Behnken design is selected to construct the experimental scheme, and a second-order polynomial regression equation is established. The significance of the model is evaluated using analysis of variance, and cross-validation is adopted to prevent overfitting. The standardized sensitivity coefficients of each parameter and their interaction effects are calculated to determine the parameter regulation priority. Each technology solution is marked with the applicable type of coal-based solid waste, the range of operating parameters, and environmental constraint conditions. Finally, all technology solution families and their optimized parameter configurations are integrated to form a complete configurable technology solution cluster.

[0086] Preferably, in this step, a knowledge base of resource utilization technology chains and a multi-objective optimization framework are established to achieve a systematic mapping from solid waste characteristics to technology solutions. In particular, a method combining fuzzy comprehensive evaluation and BP neural network is used to establish an accurate matching mechanism between solid waste combinations and technology nodes, and the improved NSGA-Ⅲ algorithm is used to achieve multi-objective optimization of the technology chain. The finally generated configurable technology solution cluster not only ensures the resource conversion efficiency but also takes into account economic benefits and environmental impacts, reflecting the systematicness and scientificity of solution selection.

[0087] S3: Based on the configurable technology solution cluster, generate a sequence of solution priorities through dynamic influencing factor analysis and policy response mechanism.

[0088] In the specific implementation manner of the present invention, the flowchart for generating the sequence of solution priorities is as Figure 3 shown, and specifically includes the following steps:

[0089] S3.1: Construct a dynamic influencing factor evaluation system to conduct a temporal evaluation of each solution in the configurable technology solution cluster.

[0090] Specifically, the dynamic influencing factor evaluation system includes a market fluctuation index group (raw material supply stability, product demand change, price elasticity), a regional development index group (industrial planning fit, infrastructure completeness, technology support ability), and a social effect index group (employment driving ability, industrial chain contribution degree, public acceptance degree). The time series analysis method is used to dynamically track each index, a quarterly rolling evaluation mechanism is established, the tracking period is set to one year, and using the historical database and market research data, combined with Monte Carlo simulation, calculate the dynamic scores of each solution and their 95% confidence intervals to form a dynamic scoring matrix of technology solutions.

[0091] S3.2: Establish a policy impact matrix, quantify the influence coefficients of policy factors on each evaluation index, and calculate the policy sensitivity of each technology solution.

[0092] Specifically, first identify the key policy factors, including environmental protection emission standards, comprehensive utilization policies of resources, energy structure adjustment policies, and regional industrial plans, and mark the trends of policy intensity; secondly, construct an impact matrix of policy factors on evaluation indicators, and use the expert scoring method to determine the impact coefficient (range [-1, 1]), where a positive value indicates a promoting effect, a negative value indicates an inhibitory effect, and 0 indicates no impact, and set a dynamic time coefficient for trend policies; finally, calculate the comprehensive impact score based on the implementation progress and impact degree of each policy as the policy sensitivity of the plan.

[0093] S3.3: Analyze the synergy relationships among the solutions within the configurable technology solution cluster, construct a synergy effect evaluation model, and generate a synergy effect score.

[0094] Specifically, based on the classification results of the technology solution family in S2.6, analyze the possible synergy types among the solutions, mainly including material synergy type (closed-loop material utilization), energy synergy type (cascaded utilization of waste heat), and process synergy type (sharing of equipment and facilities); construct a benefit ratio function to quantify the synergy effect, and measure the synergy intensity by calculating the ratio of the comprehensive benefit of the solution combination to the sum of the benefits of individual solutions; calculate the marginal contribution rate of each solution in different combinations; construct a comprehensive synergy effect evaluation model, use the benefit ratio and marginal contribution rate as the core indicators, and use the entropy weight method to determine the weight coefficients of each indicator to obtain the synergy effect score.

[0095] S3.4: Design a dynamic coupling evaluation function, integrate the dynamic score, policy sensitivity, and synergy effect score to generate a comprehensive evaluation value of the solution.

[0096] Specifically, construct a time-series weighted coupling evaluation function, and the specific formula is as follows:

[0097]

[0098] where E t is the comprehensive evaluation value at time t, D t is the dynamic score at time t, is the risk adjustment term of the dynamic score, S is the policy sensitivity, C is the synergy effect score, is the time-varying weight coefficient at time t, is the risk aversion coefficient, is the policy sensitivity weight index, is the synergy effect weight index. Use the dynamic programming algorithm to optimize the time-varying weight coefficient, use the historical case data as the training set, and minimize the prediction error; when the policy sensitivity is higher than the sensitivity threshold (0.7) or the synergy effect score is lower than the synergy threshold (0.3), adjust the risk aversion coefficient to punish and adjust the evaluation value.

[0099] S3.5: Based on the comprehensive evaluation value of the solutions, apply the time series TOPSIS method to generate the solution priority sequence.

[0100] Specifically, after standardizing the time series comprehensive evaluation value, construct a time-varying weighted standardized decision matrix; calculate the positive and negative ideal solutions considering the time lag effect, and solve the time series distances from each solution to the positive and negative ideal solutions; sort the solutions based on the time integral value of the relative closeness to generate the solution priority sequence.

[0101] Preferably, this step breaks through the limitations of traditional static evaluation. By introducing dynamic influencing factors such as market fluctuations, regional development, and social effects, a time series evaluation system for technical solutions is established. In particular, through time-varying policy sensitivity analysis and the time series TOPSIS method, a dynamic response mechanism for solution evaluation is realized, improving the forward-looking and adaptability of decision-making. This evaluation system complements the technical and economic environment evaluation in S2 to jointly construct a comprehensive solution optimization framework.

[0102] S4: Based on the solution priority sequence and regional spatio-temporal constraints, generate a configuration strategy set through a constraint optimization method.

[0103] In the specific implementation manner of the present invention, S4 specifically includes the following steps:

[0104] S4.1: Construct a dynamic spatio-temporal constraint system, and quantify geographical distribution, resource flow, seasonal fluctuations, and infrastructure conditions into a constraint parameter set.

[0105] Specifically, use a geographic information system to construct a digital map of the distribution of regional coal-based solid waste, mark the solid waste generation points, potential treatment points, and transportation networks; construct a resource flow topology map, mark the upper limit of the processing capacity and the minimum economic scale of each node; analyze the seasonal fluctuation pattern based on the recorded process environment parameters and historical data of the past 5 years, and apply the Fourier transform method to predict seasonal supply and demand changes; quantify the infrastructure conditions through infrastructure evaluation indicators (including access level, stability coefficient, expansion potential value); all parameters form a constraint parameter set after standardization processing.

[0106] S4.2: Based on the solution priority sequence, establish a time series deployment diagram of technical solutions, and set the implementation cycle and dependency relationship of each solution.

[0107] Specifically, take the solution priority sequence as the input, assign a three-stage time frame (preparation period, construction period, operation period) to each technical solution; analyze the dependency chain among the technical solution families to determine the necessary precedence relationship; use the critical path method to calculate the shortest implementation cycle, and optimize the resource allocation time series through resource load balancing technology; integrate the implementation cycle and dependency relationship of each solution to form a complete time series deployment diagram of technical solutions.

[0108] S4.3: Design a multi-stage constraint optimization model, with the comprehensive regional resource benefit as the objective function and the constraint parameter set as the boundary conditions.

[0109] Specifically, construct a multi-stage objective function, where the objective of each stage is set as the sum of the present values of the comprehensive stage resource benefits; use the analytic hierarchy process to determine the weight coefficients of economic net income, environmental benefits, and social value for the quantitative evaluation of the comprehensive resource benefits; transform the constraint parameter set into linear constraints and identify non-linear constraints; design inter-stage connection constraints to ensure the continuity of resource flow and the smooth transition of facility utilization rates.

[0110] S4.4: Apply the piecewise linear programming method to solve the multi-stage constraint optimization model and generate a configuration plan.

[0111] Specifically, transform the non-linear constraints into a linear form through piecewise linear approximation, and the breakpoints are selected based on the results of sensitivity analysis; use the rolling horizon method to decompose the multi-stage problem into a series of short-term optimization problems, set the rolling window width to 3 years and the step size to 1 year; use the Lagrangian relaxation method to handle the constraints that are difficult to linearize directly, and iteratively update the Lagrange multipliers through the subgradient method; use a commercial solver to solve the multi-stage constraint optimization model, set the relative gap tolerance to 0.5%, and obtain the optimal configuration plan under the satisfaction of all constraint conditions.

[0112] S4.5: Perform a robustness analysis on the configuration plan, simulate the scenarios of constraint condition fluctuations, and generate a set of configuration strategies.

[0113] Specifically, use an improved scenario tree construction method to analyze the uncertainty of constraint conditions and design typical fluctuation scenarios; conduct a simulation evaluation on the configuration plan under each scenario and calculate the fluctuation range of key performance indicators; determine the robustness index of the plan based on reliability theory and screen out the configuration plans with good adaptability under different scenarios; formulate emergency adjustment strategies for high-frequency fluctuation factors, and finally form a set of configuration strategies including multiple alternative plans.

[0114] Preferably, in this step, by constructing a dynamic spatio-temporal constraint system and a multi-stage optimization model, the transformation from the priority sequence to the specific configuration strategy is realized. In particular, the piecewise linear programming and scenario analysis methods are used, which not only ensure the feasibility of the plan but also improve the robustness of the configuration strategy. This method fully considers the regional characteristics and resource endowments, and provides a systematic implementation path for the resource utilization of coal-based solid waste.

[0115] S5: Establish a closed-loop feedback engine to trigger the dynamic update of the set of configuration strategies when external condition changes are detected.

[0116] In the specific implementation manner of the present invention, S5 specifically includes the following steps:

[0117] S5.1: Build a multi-source data acquisition network to monitor the content of key components of solid waste, environmental protection policy information, and regional carbon emission data in real time.

[0118] Specifically, deploy online monitoring equipment to collect the content of key components in the physicochemical characteristics data. The sampling frequency is adaptively adjusted according to the process fluctuation characteristics, and the sampling interval ranges from 1 to 24 hours. Establish an environmental protection policy information scraping system that covers the policy release channels at the central, provincial, and local levels, and set keyword triggers and semantic analysis modules. Connect to the data interface of the regional carbon emission trading platform to update the carbon emission quota and price information daily. After all data is preprocessed (denoising, outlier detection, missing value filling), it is stored in a distributed time series database, and the data retention period is 3 years.

[0119] S5.2: Calculate the fluctuation value of the content of key components of solid waste, the update flag of the environmental protection policy standard version, and the deviation value of the regional carbon emission quota.

[0120] Specifically, calculate the fluctuation value of the content of key components of solid waste, which is defined as the relative deviation percentage of the current value from the moving average of the base period (the previous 30 days). Through policy text similarity comparison and key parameter extraction algorithms, generate the update flag of the environmental protection policy standard version. A flag value of 0 indicates no update, and a non-zero value code indicates the impact degree and type of the update. The regional carbon emission quota deviation value is calculated as the percentage of the difference between the actual cumulative emissions and the quota to the quota.

[0121] S5.3: Establish an event handling mechanism to monitor external conditions and determine whether the trigger criteria are met.

[0122] Among them, the trigger criteria include that the fluctuation of the content of key components of solid waste exceeds the threshold, the update of the environmental protection policy standard version, or the breakthrough of the warning value of the regional carbon emission quota.

[0123] Furthermore, set hierarchical trigger criteria: the first-level threshold for the fluctuation of the content of key components of solid waste is ±10%, the second-level threshold is ±20%, and the third-level threshold is ±30%; the update flag of the environmental protection policy standard version is triggered when it is non-zero, and it is divided into three levels: low, medium, and high according to the impact degree; the warning value of the regional carbon emission quota deviation value is set to 80%, the early warning value is 90%, and the critical value is 95%; establish a trigger event priority evaluation mechanism to allocate processing resources according to the impact scope and urgency; implement the event correlation analysis function to identify the cross-impacts of multiple triggers.

[0124] S5.4: Based on the bipartite graph model reconstruction algorithm, locate the relationship between the affected resource recovery technology chain nodes and solid waste combinations according to the trigger event type, and perform local reconstruction and parameter re-optimization on the matching relationship.

[0125] Furthermore, for the trigger event, identify the affected nodes and edges of the mapping recognition bipartite graph model; for the trigger caused by the fluctuation of the key component content of solid waste, recalculate the collaborative matching degree of the coal-based solid waste combination and update the connection weights with the resource utilization technology nodes; for the trigger caused by the update of the environmental protection policy standard version, adjust the correlation coefficients of the policy impact matrix and re-evaluate the compliance of the solution; for the trigger of regional carbon emission quotas, preferentially adjust the configuration ratio of the high-carbon emission technology chain; use a gradient-based incremental optimization method for local search, only reconstruct the affected area, keep other parts stable, and achieve high computational efficiency.

[0126] S5.5: Execute the policy verification process, perform consistency checks and feasibility evaluations on the reconstructed policy, and generate an updated configuration policy set.

[0127] Furthermore, perform consistency checks on the reconstructed solution, verify the resource balance constraint, process compatibility constraint, and timing logic constraint; evaluate the short-term response performance and long-term stability of the reconstructed solution through rapid simulation; calculate the ratio of the solution adjustment cost to the expected benefit, and set a minimum benefit threshold value (a configurable parameter, defaulting to 1.5 times the adjustment cost); prioritize the solutions that pass the verification to form a hierarchical response strategy; generate a policy implementation roadmap, including decision nodes, implementation steps, and effectiveness evaluation indicators; finally, update the configuration policy set and feedback the update result to the system knowledge base to complete the closed-loop optimization.

[0128] S5.6: Establish a continuous learning and optimization module, and feedback the execution results to the relevant models.

[0129] Furthermore, use the newly collected data to update the solid waste characteristic spectrum and the key collaborative factor set; evaluate the prediction accuracy of the adaptability evaluation model according to the actual operation effect, and use an online learning algorithm to update the model parameters; optimize the parameters of the dynamic coupling evaluation function using the newly added time-series samples; regularly evaluate the solution efficiency and solution quality of the multi-stage constraint optimization model, and improve the model performance through adaptive algorithm parameter adjustment; establish a quarterly report mechanism, summarize the system response effect, identify the optimization direction, and automatically generate the monitoring focus for the next cycle.

[0130] Preferably, the present invention breaks through the limitations of the static configuration of traditional solid waste treatment technologies, innovatively constructs a multi-source data acquisition network and a multi-level triggering mechanism, can monitor the fluctuations of solid waste components, changes in the policy environment, and carbon emission constraints in real time, and accurately judge the timing of system adjustment. In particular, the local optimization strategy implemented through the bipartite graph model reconstruction algorithm greatly improves the response efficiency and computational performance of the system to external changes, and achieves precise adjustment of key nodes on the premise of maintaining overall stability. At the same time, based on the closed-loop feedback mechanism of continuous learning and optimization, the entire resource utilization system has the ability of self-iteration and evolution, can continuously accumulate experience and knowledge, optimize model parameters and decision-making logic, and form an increasingly precise strategy configuration ability.

[0131] Further, this embodiment also provides a decision-making system for the resource utilization direction of multi-source coal-based solid waste, including: a characteristic analysis module, used to construct a collaborative characteristic model of coal-based solid waste, construct a solid waste characteristic spectrum through multi-dimensional physical and chemical characteristic detection, analyze key collaborative factors, and generate an interaction relationship matrix; a solution generation module, used to establish a matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm based on the interaction relationship matrix and the collaborative conversion characteristic data set, and output a cluster of configurable technology solutions; a priority evaluation module, used to generate a solution priority sequence based on the cluster of configurable technology solutions through dynamic influencing factor analysis and policy response mechanism; a strategy optimization module, used to generate a configuration strategy set through a constraint optimization method based on the solution priority sequence and regional spatio-temporal constraints; a dynamic adjustment module, used to establish a closed-loop feedback engine, and trigger the dynamic update of the configuration strategy set when external conditions are monitored to change.

[0132] In summary, the present invention improves the conversion efficiency of traditional single-component treatment by accurately identifying the synergistic effect between coal-based solid waste components, optimizes the material conversion process, and improves the level of comprehensive resource utilization; the combination of accurately matched technical paths and multi-objective optimization algorithms reduces ineffective conversion and resource waste, improves the input-output ratio, and improves the economy of solid waste treatment; through process parameter optimization and collaborative configuration of the technology chain, harmful substance emissions and energy consumption are reduced, and the environmental friendliness of the solid waste treatment process is improved; the application of multi-dimensional physical and chemical characteristic analysis and collaborative models enables the system to handle more complex and diverse coal-based solid waste combinations, overcoming the limitations of traditional methods in dealing with complex components.

[0133] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A decision-making method for the resource utilization direction of multi-source coal-based solid waste, characterized in that Including: Construct a collaborative characteristic model of coal-based solid waste. Through multi-dimensional physical and chemical characteristic detection, construct a characteristic spectrum of solid waste, analyze key collaborative factors, and generate an interaction relationship matrix; Based on the interaction relationship matrix and the collaborative transformation characteristic data set, establish a matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm, and output a cluster of technical solutions; Based on the cluster of technical solutions, generate a sequence of scheme priorities through dynamic influencing factor analysis and policy response mechanism; Based on the sequence of scheme priorities and regional spatio-temporal constraints, generate a configuration strategy set through a constraint optimization method; Establish a closed-loop feedback engine. When external condition changes are detected, trigger the dynamic update of the configuration strategy set; The construction of the collaborative characteristic model of coal-based solid waste includes: Preprocess the original samples of multi-source coal-based solid waste to obtain a standardized sample set; Conduct multi-dimensional physical and chemical characteristic detection on the standardized sample set to obtain physical and chemical characteristic data; Based on the physical and chemical characteristic data, construct a feature space and perform dimensionality reduction processing to generate a characteristic spectrum of solid waste; Based on the physical and chemical characteristic data, calculate and generate a set of key collaborative factors; Based on the set of key collaborative factors and the characteristic spectrum of solid waste, construct an interaction relationship matrix between the components of coal-based solid waste, and generate a collaborative transformation characteristic data set.

2. The decision-making method for the multi-source coal-based solid waste resource utilization direction according to claim 1, wherein, The establishment of the matching relationship between the coal-based solid waste combination and the resource utilization technology chain through a multi-objective optimization algorithm includes: Construct a knowledge base of the resource utilization technology chain, and the knowledge base of the resource utilization technology chain includes a set of resource utilization technology nodes and technology chain connection rules; Based on the interaction relationship matrix and the collaborative transformation characteristic data set, determine the coal-based solid waste combination, calculate the collaborative matching degree of different coal-based solid waste combinations, and generate a set of solid waste combination solutions through an incremental clustering algorithm; Using the set of key collaborative factors, establish an adaptability evaluation model between the coal-based solid waste combination and the resource utilization technology nodes; Based on the output results of the adaptability evaluation model, construct a bipartite graph model between the set of solid waste combination solutions and the set of resource utilization technology nodes; Apply a multi-objective evolutionary algorithm to the bipartite graph model for path optimization to generate a Pareto optimal solution set; Conduct hierarchical clustering analysis on the Pareto optimal solution set to obtain a family of technical solutions, and optimize and configure the process parameters of the family of technical solutions in combination with the collaborative transformation characteristic data set, and integrate them to form a cluster of technical solutions.

3. The decision-making method for the multi-source coal-based solid waste resource utilization direction according to claim 1, characterized in that, The generation of the sequence of scheme priorities through dynamic influencing factor analysis and policy response mechanism includes: Construct a dynamic influencing factor evaluation system to conduct a temporal evaluation on each scheme in the cluster of technical solutions; Establish a policy influence matrix, quantify the influence coefficients of policy factors on each evaluation index, and calculate the policy sensitivity; Analyze the collaborative relationship between each scheme in the cluster of technical solutions, construct a collaborative effect evaluation model, and generate a collaborative effect score; Design a dynamic coupling evaluation function to fuse the dynamic score, policy sensitivity, and collaborative effect score to generate a comprehensive evaluation value of the scheme; Based on the comprehensive evaluation value of the scheme, apply the temporal TOPSIS method to generate a sequence of scheme priorities.

4. The decision-making method for the resource utilization direction of multi-source coal-based solid waste according to claim 1, characterized in that The generation of the configuration strategy set through a constraint optimization method includes: Construct a dynamic spatio-temporal constraint system, and quantify geographical distribution, resource flow, seasonal fluctuations, and infrastructure conditions into a set of constraint parameters; Based on the above-mentioned solution priority sequence, establish a timing diagram for the deployment of technical solutions, and set the implementation cycle and dependency relationship of each solution; Design a multi-stage constraint optimization model, with the comprehensive regional resource benefit as the objective function, and use the above-mentioned constraint parameter set as the boundary condition; Apply the piecewise linear programming method to solve the multi-stage constraint optimization model and generate a configuration plan; Perform a robustness analysis on the configuration plan, simulate the scenario of constraint condition fluctuations, and generate a set of configuration strategies.

5. The decision-making method for the multi-source coal-based solid waste resource utilization direction according to claim 1, wherein Triggering the dynamic update of the above-mentioned configuration strategy set includes: Real-time monitoring of the content of key components of solid waste, environmental protection policy information, and regional carbon emission data; Calculating the fluctuation value of the content of key components of solid waste, the update flag of the environmental protection policy standard version, and the deviation value of the regional carbon emission quota; Establish an event handling mechanism to monitor external conditions and determine whether the trigger standard is met; Based on the bipartite graph model reconstruction algorithm, locate the affected resource utilization technology chain nodes and the solid waste combination relationship according to the trigger event type, and perform local reconstruction and parameter re-optimization on the matching relationship.

6. The decision-making method for the resource utilization direction of multi-source coal-based solid waste according to claim 1, characterized in that, The above-mentioned physical and chemical property data includes elemental composition data, specific surface area data, pore structure parameters, and pyrolysis characteristic parameters; The key synergy factor set includes substance conversion synergy factors, element migration synergy factors, and reaction kinetics synergy factors.

7. A decision-making system for the resource utilization direction of multi-source coal-based solid waste, characterized in that, Including: A characteristic analysis module for constructing a coal-based solid waste synergy characteristic model, constructing a solid waste characteristic spectrum through multi-dimensional physical and chemical property detection, analyzing key synergy factors, and generating an interaction relationship matrix; A solution generation module for establishing a matching relationship between coal-based solid waste combinations and resource utilization technology chains through a multi-objective optimization algorithm based on the above-mentioned interaction relationship matrix and synergy conversion characteristic data set, and outputting a cluster of technical solutions; A priority evaluation module for generating a solution priority sequence based on the cluster of technical solutions through dynamic influencing factor analysis and policy response mechanisms; A strategy optimization module for generating a set of configuration strategies through a constraint optimization method based on the above-mentioned solution priority sequence and regional space-time constraints; A dynamic adjustment module for establishing a closed-loop feedback engine to trigger the dynamic update of the above-mentioned configuration strategy set when external condition changes are detected; The above-mentioned construction of the coal-based solid waste synergy characteristic model includes: Preprocess the original samples of multi-source coal-based solid waste to obtain a standardized sample set; Perform multi-dimensional physical and chemical property detection on the above-mentioned standardized sample set to obtain physical and chemical property data; Construct a feature space based on the above-mentioned physical and chemical property data and perform dimensionality reduction processing to generate a solid waste characteristic spectrum; Based on the above-mentioned physical and chemical property data, calculate and generate a key synergy factor set; Based on the above-mentioned key synergy factor set and the solid waste characteristic spectrum, construct an interaction relationship matrix between the components of coal-based solid waste and generate a synergy conversion characteristic data set.

Images