Method and system for improving saline-alkali soil by using fly ash-based multi-element solid waste mineralized product
By using precise design and efficient improvement methods for fly ash-based multi-element solid waste mineralization products, the problems of unstable effects and low resource recycling efficiency in saline-alkali land improvement have been solved, achieving precise improvement and cost optimization of saline-alkali land.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-02-09
- Publication Date
- 2026-03-31
AI Technical Summary
In existing technologies, fly ash-based solid waste materials have unstable improvement effects and poor targeting in the improvement of saline-alkali land. They lack a precise quantitative mapping relationship between the key performance parameters of mineralized products and the soil improvement effect, have long research and development cycles and lack intelligent decision support, resulting in low efficiency of resource recycling.
A method for improving saline-alkali land using fly ash-based multi-element solid waste mineralization products is proposed. By measuring the core parameters of saline-alkali land, a product performance prediction model is established using machine learning algorithms. The preparation parameters are solved in reverse, and combined with the material preparation optimization model, precise design and efficient improvement are achieved.
It has enabled precise improvement of saline-alkali land, increased the success rate and effectiveness of improvement, reduced production costs, reduced investment risks, and promoted resource recycling technology.
Smart Images
Figure CN121753568A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of soil improvement and remediation technology, and in particular to a method and system for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement. Background Technology
[0002] Saline-alkali land is a significant obstacle restricting agricultural production and threatening ecological security. Its improvement hinges on reducing soil salinity, replacing excess sodium ions to lower alkalinity, and improving the poor soil physical structure. Traditional improvement methods mainly include salt leaching through water conservancy projects, applying chemical amendments (such as gypsum and humic acid), and planting salt-tolerant plants. While these methods have some effect, they generally suffer from high costs, unstable results, long cycles, and potential secondary pollution. On the other hand, a massive amount of industrial solid waste, such as fly ash, steel slag, and desulfurization gypsum, is generated annually. The accumulation of this solid waste not only occupies large amounts of land but also poses environmental pollution risks. Using it for saline-alkali land improvement can achieve "waste treatment of waste, turning waste into treasure," representing a promising circular economy path. Existing research shows that fly ash and other siliceous-aluminous solid waste, after treatment with technologies such as alkali activation, can form mineralized products with certain cementing and ion exchange capabilities, possessing the potential to improve saline-alkali land and effectively enhance soil properties.
[0003] However, current technologies for applying industrial solid waste to improve saline-alkali land mostly rely on fixed formulas or empirical formulations, simply mixing various solid wastes. Due to the wide range of solid waste sources, large fluctuations in composition, and the complex and diverse types of saline-alkali land, fixed material formulas are difficult to adapt to the diverse improvement needs, resulting in unstable and poorly targeted improvement effects. In fact, they may even produce negative effects due to inappropriate product pH or salinity. Furthermore, there is a lack of effective models that can accurately describe the quantitative mapping relationship between key performance parameters of mineralized products (such as cation exchange capacity, cementing activity, and microstructure) and the final improvement effects (such as reduced alkalinity and aggregate formation) in complex soil systems. Material development and field application are severely disconnected, relying mainly on time-consuming and labor-intensive trial-and-error methods for screening, resulting in long development cycles and low efficiency. Moreover, after determining the improvement targets, current technologies cannot reverse-engineer and efficiently derive the optimal product formula and production process "tailor-made" for specific saline-alkali land. Simultaneously, there is a lack of scientific and intelligent decision support methods for multi-objective optimization that comprehensively considers production costs, carbon sequestration benefits, and environmental safety while meeting soil improvement requirements.
[0004] In summary, the current technology's inability to systematically couple and intelligently decide on solid waste characteristics, material design, soil processes, and economic benefits limits the large-scale, precise, and efficient application of fly ash-based solid waste materials in saline-alkali land improvement. Therefore, there is an urgent need to develop a new method and system capable of achieving "precise diagnosis, intelligent design, and targeted improvement." To this end, a method and system for using fly ash-based multi-element solid waste mineralization products in saline-alkali land improvement is proposed, aiming to widely promote resource recycling technologies. Summary of the Invention
[0005] The main objective of this invention is to provide a method and system for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement, which can effectively solve the problems in the background art.
[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0007] A method for using fly ash-based multi-component solid waste mineralization products for saline-alkali land improvement includes the following steps:
[0008] S1: Determine the core parameters of the saline-alkali land to be improved, including at least soil electrical conductivity, pH value, alkalinity, soluble sodium ion content, and soil texture; determine the improvement target value based on the core parameters;
[0009] S2: Input the improvement target value and the core parameters into the preset product performance prediction model. The model is used to establish the mapping relationship between the preparation parameters and performance parameters of the mineralized product and the soil improvement effect. The model is then used to solve in reverse to obtain the target performance parameter set of the mineralized product that meets the improvement target value. The target performance parameter set includes at least cation exchange capacity, available calcium and magnesium content, product pH value, gelation activity index and specific surface area.
[0010] S3: Based on the target performance parameter set, the optimized preparation parameters of fly ash-based multi-component solid waste mineralization products are determined through a material preparation optimization model. The optimized preparation parameters include at least the ratio of multi-component solid waste raw materials, the type and dosage of activator, and the curing regime.
[0011] S4: Based on the optimized preparation parameters, the fly ash-based multi-element solid waste mineralization product is prepared and applied to the saline-alkali land to be improved.
[0012] Furthermore, in step S2, the product performance prediction model is a surrogate model trained based on a machine learning algorithm. Its input variables include the performance parameters of the mineralized product, the initial soil core parameters, and the product application rate, and the output variable is the predicted improved soil parameters.
[0013] Furthermore, in step S2, the product performance prediction model is a multi-level feedback network model, which defines different functional components of mineralized products as material agents and different components in the soil as soil agents. By simulating the dynamic interaction network between material agents and soil agents, a macroscopic soil improvement effect emerges. The reverse solution process adjusts the distribution of the group attributes of material agents through optimization algorithms, so that the emerging macroscopic soil improvement effect approaches the improvement target value.
[0014] Furthermore, in step S3, the material preparation optimization model is a machine learning model trained based on experimental data, with the preparation parameters as input variables and the predicted product performance parameters as output variables; the process of determining the optimized preparation parameters is to use the target performance parameter set as constraints, production cost or environmental benefits as optimization objectives, and solve the material preparation optimization model using an optimization algorithm.
[0015] Furthermore, in step S3, the multi-component solid waste raw materials include at least fly ash, steel slag, and desulfurized gypsum;
[0016] In the optimized preparation parameters, the mass ratio range of fly ash, steel slag and desulfurized gypsum is (40-60): (20-5): (15-30);
[0017] The activator is composed of water glass with a modulus of 1.5 to 1.8 and solid sodium hydroxide in a mass ratio of (4:1) to (5:1), and its total dosage is 9% to 12% of the total mass of the solid raw materials.
[0018] The maintenance system includes carbonization maintenance, with a carbon dioxide concentration of 10% to 50%, a maintenance pressure of 0.1 to 0.5 MPa, and a maintenance duration of 2 to 24 hours.
[0019] Furthermore, in step S4, the mineralized product obtained has the following performance parameters:
[0020] Cation exchange capacity ≥ 30 cmol / kg;
[0021] Effective calcium content (calculated as CaO) ≥ 15%;
[0022] The product's pH value is in the range of 8.5 to 10.5;
[0023] Specific surface area ≥ 20 m² / g;
[0024] The gelling activity index was determined by wet sieving after being mixed with standard loam at an application rate of 2% of the soil mass and cultured for 28 days. Compared with the untreated control soil, the percentage increase in the content of water-stable aggregates with a particle size greater than 0.25 mm was not less than 30%.
[0025] Furthermore, the material smart body includes at least an active calcium release source smart body, a porous unit smart body, and a cementation network unit smart body;
[0026] The soil agent includes at least a sodium ion unit agent and a clay unit agent;
[0027] The dynamic interaction network is a dynamically weighted interaction network formed by defining local interaction rules between the agents, and the network state evolves over time; wherein, the interaction rules include at least displacement reaction rules, physical adsorption rules, structural cementation rules and environmental feedback rules, and the environmental feedback rules enable the state of the soil agents to dynamically adjust the properties or interaction intensity of the material agents.
[0028] Furthermore, in the dynamically weighted interaction network, the weights of the connecting edges are adaptively adjusted according to the following formula: = + In the formula, For connecting edges With connecting edge exist Weight of time, For learning rate, It is a feedback signal based on changes in the local environmental state.
[0029] A system for using fly ash-based multi-component solid waste mineralization products for saline-alkali land improvement, the system comprising a method for implementing the use of fly ash-based multi-component solid waste mineralization products for saline-alkali land improvement, including:
[0030] The data acquisition module is used to collect initial parameters and improvement targets for saline-alkali land.
[0031] The model library module is used to train and store the product performance prediction model and the material preparation optimization model.
[0032] The optimization solution module is used to call the models in the model library module, run the optimization algorithm for reverse design, and output the optimized preparation parameters;
[0033] A production control interface is used to send the optimized preparation parameters to the production line to control product preparation.
[0034] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement.
[0035] The present invention has the following beneficial effects:
[0036] Compared with existing technologies, this solution establishes a mapping relationship between the preparation parameters and performance parameters of mineralized products and the soil improvement effect through the proposed product performance prediction model. By solving the model in reverse, the target performance parameter set of mineralized products that meet the improvement target value is obtained. Moreover, the performance parameters of the products are precisely matched and designed according to the specific salinization type and degree of the target plot, thereby realizing precise improvement of "one policy for one site" and greatly improving the success rate and effectiveness of improvement measures.
[0037] Compared with existing technologies, this solution proposes a material preparation optimization model that takes preparation parameters as input variables and predicted product performance parameters as output variables. It uses the target performance parameter set as constraints and production cost or environmental benefits as optimization objectives. The optimization algorithm is used to solve for the optimized preparation parameters, which can directly output the formula and process parameters with the best economic benefits. This transforms the production process from relying on experience to making scientific decisions based on data and algorithms, and significantly reduces production costs.
[0038] Compared to existing technologies, this solution transforms saline-alkali land improvement from a highly uncertain engineering problem into a quantifiable, predictable, and optimizable scientific decision-making process. This allows users to obtain clear predictions of improvement effects, cost budgets, and timelines before investment, thereby significantly reducing their investment risk. It also promotes the widespread adoption of environmental protection and resource recycling technologies. Attached Figure Description
[0039] Figure 1 This is a schematic diagram of the process for using fly ash-based multi-component solid waste mineralization products for saline-alkali land improvement according to the present invention.
[0040] Figure 2 This is a schematic diagram of the structure of the fly ash-based multi-component solid waste mineralization product system for saline-alkali land improvement according to the present invention. Detailed Implementation
[0041] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0042] See Figure 1 The flowchart shown is a method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to the present invention, which includes the following steps:
[0043] Step 1: Determine the core parameters of the saline-alkali land to be improved.
[0044] Among them, the core parameters include at least soil electrical conductivity, pH value, alkalinity, soluble sodium ion content, and soil texture.
[0045] In one feasible operational method, the following steps can be taken:
[0046] Step 11: On-site investigation and sampling
[0047] Determine the boundaries of the saline-alkali land to be improved and conduct systematic sampling. Typically, "S" shaped or grid methods are used to collect soil samples from the 0-20cm and 20-40cm layers.
[0048] Step 12: Laboratory determination of core parameters
[0049] Salt damage indicator: Soil leachate electrical conductivity (EC, used to determine total salt content).
[0050] Alkali damage indicators: pH value, sodium adsorption ratio (SAR), and alkalinity (ESP, the core indicator).
[0051] Salt composition: determination of soluble cations (Na+) + ,K + Ca 2+ Mg 2+ ) and anions (Cl) - SO4 2- CO3 2- HCO3 - This is to determine the type of salting-alkali treatment (such as chloride type, sulfate type, soda type).
[0052] Soil texture (sand / loam / clay).
[0053] Step 2: Determine the target improvement value based on the core parameters.
[0054] In an feasible operational approach, specific and measurable target values are set based on measurement results and crop growth requirements. For example: "Within 12 months, reduce the ESP of the 0-20cm soil layer from 35% to below 15%, reduce EC to below 4dS / m, and stabilize the pH below 8.5."
[0055] Step 3: Input the target improvement value and core parameters into the preset product performance prediction model.
[0056] Among them, the product performance prediction model establishes the mapping relationship between the preparation parameters and performance parameters of mineralized products and the soil improvement effect. It is a surrogate model trained based on machine learning algorithms. Its input variables include the performance parameters of mineralized products, the initial soil core parameters, and the product application rate. The output variable is the predicted soil parameters after improvement.
[0057] Specifically, the product performance prediction model is a multi-level feedback network model, which defines different functional components of mineralized products as material intelligent agents and different components in the soil as soil intelligent agents. By simulating the dynamic interaction network between material intelligent agents and soil intelligent agents, macroscopic soil improvement effects emerge.
[0058] Material smart agents include active calcium release source smart agents, porous unit smart agents, and cementation network unit smart agents;
[0059] Soil intelligent agents include at least sodium ion unit intelligent agents and clay unit intelligent agents;
[0060] Dynamic interaction networks are dynamically weighted interaction networks formed by defining local interaction rules between agents, and the network state evolves over time.
[0061] In one feasible operational approach, the product performance prediction model has the following structure, including:
[0062] 1) Environmental layer
[0063] The target soil profile (e.g., 0-20cm) is discretized into a two-dimensional or three-dimensional grid world. Each grid cell has a basic environmental state vector E(x,y,t), including: moisture content, local pH, local salt ion concentration, temperature, etc. The environmental state is the common background for all agent activities and is changed by the agent's activities.
[0064] 2) Intelligent Agent Layer
[0065] 2.1) Material intelligent agent: a functional unit representing mineralized products, with attributes including:
[0066] Types: Ca-Source, Porous-Block, Product Performance Prediction Model (Cementitious Network Unit).
[0067] Internal state:
[0068] For Ca-Source, there are "calcium reserves" and "release rate function";
[0069] For Porous-Block, there are "porosity", "surface charge", and "adsorption capacity";
[0070] For product performance prediction models, there are "bonding strength" and "degree of hydration".
[0071] Behavioral rules: Based on its own type and local environment E, it performs specific behaviors (such as Ca-Source releasing Ca according to a function). 2+ The product performance prediction model identifies nearby soil particles for bonding.
[0072] 2.2) Soil Intelligent Agent: Represents the constituent unit of soil, with attributes including:
[0073] Types: Clay-Particle, Na-Ion, Salt-Crystal, Organic-Matter, Microbe-Cluster.
[0074] Internal state:
[0075] For Clay-Particle, there are "exchangeability Na⁺ quantity" and "dispersion";
[0076] For Na-Ion, there are "location" and "binding state (dissolved / exchanged / fixed)".
[0077] Behavioral rules: For example, the dispersion of Clay particles is affected by the ionic strength and pH of the surrounding solution; Na-Ion migrates between the solution and the surface of the clay particles.
[0078] 3) Interactive network layer
[0079] Agents are connected through a dynamic neighborhood interaction network. Network relationships are established and dissolved in real time during the simulation based on the spatial proximity and functional complementarity of the agents. Each interaction edge has a weight and an interaction rule; the weight of the connecting edge is adaptively adjusted according to the following formula: = + In the formula, For connecting edges With connecting edge exist Weight of time, For learning rate, Feedback signals based on changes in local environmental state
[0080] Interaction rules include displacement reaction rules, physical adsorption rules, structural cementation rules, and environmental feedback rules. Environmental feedback rules enable the state of soil agents to dynamically adjust the properties or interaction intensity of material agents.
[0081] The following are specific examples of interaction rules:
[0082] Example 1:
[0083] Ca-Source↔Na-Ion: The weights represent the exchange reaction rates of Ca²⁺ and Na⁺, and the protocol is "ion exchange".
[0084] Example 2:
[0085] Product performance prediction model ↔ Clay-Particle: The weights represent the bonding probability, and the protocol is "glue".
[0086] Example 3:
[0087] Porous-Block↔Na-Ion: The weight represents the adsorption efficiency, and the protocol is "physical adsorption".
[0088] Example 4: Microbe-Cluster ↔ All agents: The weights represent the intensity of their life activities (such as acid production) on the local environment E.
[0089] 4) Emerging observation layer
[0090] During model operation, the collective state of all agents in the entire grid world is continuously monitored and statistically analyzed to calculate macroscopic, observable soil parameters, specifically:
[0091] Overall ESP = (Total exchange rate Na⁺ of all Clay-Particles) / (Total CEC of all Clay-Particles) * 100%.
[0092] Soil structure index = (number of Clay-Particles effectively cemented together by the product performance prediction model) / (total number of Clay-Particles).
[0093] The profile average EC / pH = spatial average of the corresponding component of E for all grid cells.
[0094] System orderliness: It is evaluated by calculating the spatial entropy of the distribution of agent types or the clustering coefficient of the interaction network.
[0095] The specific process for constructing the product performance prediction model is as follows:
[0096] Phase 1: Model Definition and Parameterization
[0097] Agent abstraction and parameter extraction:
[0098] Physicochemical analysis of mineralization products (XRD, BET, ion leaching experiments, etc.) decomposes the products into several functional units (MAs), and defines key state parameters and their value ranges for each functional unit (such as the calcium reserves and release kinetics parameters of the Ca-Source). Similarly, based on soil testing data, soil smart particles and their initial distribution are defined (such as initializing the exchangeable Na⁺ quantity of Clay-Particles based on ESP).
[0099] Formalizing interaction rules:
[0100] Known soil chemical and physical processes (such as ion exchange described by the Gapon equation, colloidal stability described by the DLVO theory, and the bonding mechanism of cementitious materials) are transformed into probabilistic or deterministic interaction rules between agents. For example: "A Ca-Source initiates an exchange interaction with Na-Ion in its own grid and adjacent grids within time step ∆t. The success probability is determined by the local Ca²⁺ / Na⁺ concentration ratio and the activation energy of the reaction."
[0101] Phase 2: Model Training and Calibration
[0102] Historical / Experimental Data Preparation:
[0103] Collect or design a set of control experimental data. For example, known product performance parameters (e.g., CEC=30, available Ca=18%) and measured data after improvement under different initial soils (ESP=30%, clay texture) and application rates (2%) (e.g., ESP after 90 days, aggregate content).
[0104] Parameter calibration:
[0105] Initialize the model by setting the initial states of MAs and soil agents according to the experimental conditions.
[0106] Run the model to obtain the predicted macroscopic improvement parameters.
[0107] Compare the predicted values with the experimentally measured values and calculate the loss function (such as mean squared error).
[0108] Genetic algorithms are used to automatically calibrate rule parameters in the model that are difficult to measure directly (such as interaction probability coefficients and environmental diffusion coefficients) and minimize the loss function.
[0109] Phase 3: Model Validation and Application
[0110] Independent verification:
[0111] The calibrated product performance prediction model was validated using another set of independent experimental data that were not involved in the calibration, and its accuracy and generalization ability were evaluated.
[0112] Deploy as a prediction engine:
[0113] The trained model is encapsulated as a "virtual laboratory" function. Its input is:
[0114] Product performance parameters Specs , used to generate the attributes of functional units;
[0115] Initial soil parameters Initial , used to generate the initial state and distribution of soil smart entities;
[0116] Application Rate This is used to determine the initial deployment density of functional units in the environmental grid.
[0117] Its output is: all predicted improved soil parameters read from the emergent observation layer after the model simulation has been performed for a specified time (e.g., 90 days).
[0118] Step 4: Obtain the target performance parameter set of the mineralized product that satisfies the improvement target value by solving the model in reverse.
[0119] Among them, the reverse solution process adjusts the distribution of the population attributes of the material intelligence through optimization algorithms, so that the emerging macroscopic soil improvement effect is close to the improvement target value. The target performance parameter set includes at least cation exchange capacity, available calcium and magnesium content, product pH value, gelling activity index and specific surface area.
[0120] Step 5: Based on the target performance parameter set, determine the optimized preparation parameters for fly ash-based multi-component solid waste mineralization products using a material preparation optimization model.
[0121] Among them, optimizing the preparation parameters includes at least the ratio of multiple solid waste raw materials, the type and dosage of activator, and the curing system;
[0122] Among them, the material preparation optimization model is a machine learning model trained based on experimental data. Its input variables are preparation parameters, and its output variables are predicted product performance parameters. The process of determining the optimized preparation parameters is to use the target performance parameter set as constraints, production cost or environmental benefits as optimization objectives, and solve the material preparation optimization model using optimization algorithms.
[0123] In one possible implementation, a material preparation optimization model with production cost as the optimization objective is given. The objective function of the model is defined as: total production cost C. total =Raw material cost C raw +Activator cost C act +Processing energy consumption cost C energy +Fixed processing costs C fixed Specifically, the objective function can be defined using the following function expression:
[0124] Minimize: C total = + + + +
[0125] In the formula,
[0126] Minimize is the objective function;
[0127] x i Let be the mass fraction of the i-th raw material in the formula, and have . =1;
[0128] P i Let be the unit cost (yuan / ton) of the i-th raw material.
[0129] y act The dosage of the activator;
[0130] P act Cost of the active ingredient in the activator (RMB / kg);
[0131] E grind The energy consumption per unit product (kWh / ton) required to grind the raw materials to the target fineness is the raw material ratio x i grinding efficiency index W i The function can be simplified to E grind = ;
[0132] P energy Industrial electricity price (RMB / kWh);
[0133] E cure The unit product energy consumption (kWh / ton) for carbonization curing is determined by the carbonization pressure and time (t). cure The function can be simplified to E cure = k × t cure k is a constant coefficient;
[0134] C fixed Fixed costs such as labor, depreciation, and maintenance allocated per ton of product (RMB / ton);
[0135] The constraints are defined by the proportions of the multi-component solid waste raw materials, the type and dosage of the activator, the curing regime, and the performance parameters of the prepared mineralized products, including:
[0136] Constraint 1: The mass ratio range of fly ash, steel slag and desulfurization gypsum is (40~60):(20~5):(15~30);
[0137] Constraint 2: The activator is composed of water glass with a modulus of 1.5 to 1.8 and solid sodium hydroxide in a mass ratio of (4:1) to (5:1), and its total dosage is 9% to 12% of the total mass of the solid raw materials;
[0138] Constraint 3: The curing regime includes carbonization curing, with a carbon dioxide concentration of 10% to 50%, a curing pressure of 0.1 to 0.5 MPa, and a curing duration of 2 to 24 hours;
[0139] The prepared mineralized product has the following performance parameters:
[0140] Constraint 4: Cation exchange capacity ≥ 30 cmol / kg;
[0141] Constraint 5: Effective calcium content (calculated as CaO) ≥ 15%;
[0142] Constraint 6: The pH value of the product is in the range of 8.5 to 10.5;
[0143] Constraint 7: Specific surface area ≥ 20 m² / g;
[0144] Constraint 8: The gelling activity index, applied at a rate of 2% of the soil mass, was mixed with standard loam and cultured for 28 days. The index was then determined using the wet sieving method. Compared with the untreated control soil, the percentage increase in the content of water-stable aggregates with a particle size greater than 0.25 mm was not less than 30%.
[0145] Step 6: Based on the optimized preparation parameters, fly ash-based multi-element solid waste mineralization products are prepared and applied to the saline-alkali land to be improved.
[0146] Among them, the performance parameters of the prepared mineralized products must meet the above-mentioned constraints 4-8.
[0147] Example 2:
[0148] This invention also provides a system for using fly ash-based multi-component solid waste mineralization products for saline-alkali land improvement, see [link to relevant documentation]. Figure 2 The structural diagram shown includes:
[0149] The data acquisition module is used to collect initial parameters and improvement targets for saline-alkali land.
[0150] The model library module is used to train and store product performance prediction models and material preparation optimization models;
[0151] The optimization and solution module is used to call the models in the model library module, run the optimization algorithm for reverse design, and output the optimized preparation parameters;
[0152] The production control interface is used to send optimized preparation parameters to the production line to control product preparation.
[0153] Example 3:
[0154] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement.
[0155] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.
Claims
1. A method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement, characterized in that, Includes the following steps: S1: Determine the core parameters of the saline-alkali land to be improved, including at least soil electrical conductivity, pH value, alkalinity, soluble sodium ion content, and soil texture; determine the improvement target value based on the core parameters; S2: Input the improvement target value and the core parameters into the preset product performance prediction model. The model is used to establish the mapping relationship between the preparation parameters and performance parameters of the mineralized product and the soil improvement effect. The model is then used to solve in reverse to obtain the target performance parameter set of the mineralized product that meets the improvement target value. The target performance parameter set includes at least cation exchange capacity, available calcium and magnesium content, product pH value, gelation activity index and specific surface area. S3: Based on the target performance parameter set, the optimized preparation parameters of fly ash-based multi-component solid waste mineralization products are determined through a material preparation optimization model. The optimized preparation parameters include at least the ratio of multi-component solid waste raw materials, the type and dosage of activator, and the curing regime. S4: Based on the optimized preparation parameters, the fly ash-based multi-element solid waste mineralization product is prepared and applied to the saline-alkali land to be improved.
2. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 1, characterized in that, In step S2, the product performance prediction model is a surrogate model trained based on a machine learning algorithm. Its input variables include the performance parameters of the mineralized product, the initial soil core parameters, and the product application rate. The output variable is the predicted improved soil parameters.
3. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 1 or 2, characterized in that, In step S2, the product performance prediction model is a multi-level feedback network model, which defines different functional components of mineralized products as material agents and different components in the soil as soil agents. By simulating the dynamic interaction network between material agents and soil agents, a macroscopic soil improvement effect emerges. The reverse solution process adjusts the distribution of the group attributes of material agents through optimization algorithms, so that the emerging macroscopic soil improvement effect approaches the improvement target value.
4. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 1, characterized in that, In step S3, the material preparation optimization model is a machine learning model trained based on experimental data. Its input variables are the preparation parameters, and its output variables are the predicted product performance parameters. The process of determining the optimized preparation parameters is to use the target performance parameter set as a constraint, production cost or environmental benefits as the optimization objective, and solve the material preparation optimization model using an optimization algorithm.
5. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 1, characterized in that, In step S3, the multi-component solid waste raw materials include at least fly ash, steel slag, and desulfurized gypsum; In the optimized preparation parameters, the mass ratio range of fly ash, steel slag and desulfurized gypsum is (40-60): (20-5): (15-30); The activator is composed of water glass with a modulus of 1.5 to 1.8 and solid sodium hydroxide in a mass ratio of (4:1) to (5:1), and its total dosage is 9% to 12% of the total mass of the solid raw materials. The maintenance system includes carbonization maintenance, with a carbon dioxide concentration of 10% to 50%, a maintenance pressure of 0.1 to 0.5 MPa, and a maintenance duration of 2 to 24 hours.
6. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 1, characterized in that, In step S4, the prepared mineralized product has the following performance parameters: Cation exchange capacity ≥ 30 cmol / kg; Effective calcium content (calculated as CaO) ≥ 15%; The product's pH value is in the range of 8.5 to 10.5; Specific surface area ≥ 20 m² / g; The gelling activity index was determined by wet sieving after being mixed with standard loam at an application rate of 2% of the soil mass and cultured for 28 days. Compared with the untreated control soil, the percentage increase in the content of water-stable aggregates with a particle size greater than 0.25 mm was not less than 30%.
7. The method for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement according to claim 3, characterized in that, The material smart body includes at least an active calcium release source smart body, a porous unit smart body, and a cementation network unit smart body; The soil agent includes at least a sodium ion unit agent and a clay unit agent; The dynamic interaction network is a dynamically weighted interaction network formed by defining local interaction rules between the agents, and the network state evolves over time; wherein, the interaction rules include at least displacement reaction rules, physical adsorption rules, structural cementation rules and environmental feedback rules, and the environmental feedback rules enable the state of the soil agents to dynamically adjust the properties or interaction intensity of the material agents.
8. The method according to claim 7, characterized in that, In the dynamically weighted interaction network, the weights of the connecting edges are adaptively adjusted according to the following formula: In the formula, For connecting edges With connecting edge exist Weight of time, For learning rate, This is a feedback signal based on changes in the local environmental state.
9. A system for using fly ash-based multi-element solid waste mineralization products for saline-alkali land improvement, characterized in that, The system is used to implement the method for improving saline-alkali land using fly ash-based multi-element solid waste mineralization products as described in any one of claims 1-8, comprising: The data acquisition module is used to collect initial parameters and improvement targets for saline-alkali land. The model library module is used to train and store the product performance prediction model and the material preparation optimization model. The optimization solution module is used to call the models in the model library module, run the optimization algorithm for reverse design, and output the optimized preparation parameters; A production control interface is used to send the optimized preparation parameters to the production line to control product preparation.