Intelligent analysis and comprehensive evaluation system for calcium, magnesium and sulfur elements in a phosphogypsum soil conditioner

Abstract

The present application relates to phosphorus chemical technology field, specifically to a kind of phosphogypsum soil conditioner in calcium, magnesium, sulfur element intelligent analysis and comprehensive evaluation system of detection, including hardware analysis module, data acquisition and processing module and intelligent evaluation core module.Hardware analysis module is used to determine the total amount of element, chemical form and sample physicochemical property;Intelligent evaluation core module is based on pre-training model, according to the form analysis result, the biological availability coefficient of element is calculated, and the comprehensive agronomic effect index and environmental adaptability index are calculated by combining environmental and crop parameters.Finally, the system outputs comprehensive evaluation report covering product performance classification, applicability diagnosis and optimization suggestion.The present application solves the problem that traditional detection method cannot associate element form and agronomic effectiveness, realizes scientific, accurate and comprehensive intelligent evaluation and application guidance of phosphogypsum soil conditioner.

Description

Technical Field

[0001] This invention relates to the field of phosphorus chemical technology, specifically to an intelligent analysis, detection and comprehensive evaluation system for calcium, magnesium and sulfur elements in phosphogypsum soil conditioner. Background Technology

[0002] Against the backdrop of the rapid development of the phosphorus chemical industry, the large amount of phosphogypsum generated during the wet-process phosphoric acid production has become an industrial solid waste that urgently needs to be addressed. Currently, my country's annual phosphogypsum production is enormous, and the accumulated stockpile volume remains high. Traditional stockpiling methods not only occupy land resources but also pose long-term environmental pollution risks due to the presence of harmful substances such as phosphorus and fluorine. Promoting the resource utilization of phosphogypsum is both an urgent need for environmental protection and an inevitable choice for the sustainable development of the industry. Converting phosphogypsum into a soil conditioner can achieve both solid waste reduction and harmless treatment, and also provide a feasible way for agriculture to improve soil structure, regulate pH, and supplement trace elements, resulting in significant environmental and economic benefits.

[0003] However, the widespread use of phosphogypsum soil conditioners still faces a series of technical bottlenecks. Currently, a systematic and scientific framework for quality evaluation, testing methods, and application guidance for related products has not yet been established. In particular, the detection of key elements such as calcium, magnesium, and sulfur in conditioners is mostly limited to total quantity analysis, lacking systematic identification and quantitative evaluation of their different chemical forms. Elemental forms directly affect their availability, mobility, and plant bioavailability in the soil; relying solely on total quantity indicators is insufficient to accurately reflect the agronomic efficacy and environmental safety of the product. Existing testing standards and evaluation methods fail to fully consider the unique characteristics of phosphogypsum matrix and fail to systematically integrate the comprehensive effects of soil, crops, and the environment, hindering the standardized production and precise application of such products. Simultaneously, traditional testing methods largely rely on single-instrument analysis and experience-based judgment, lacking a data-driven intelligent evaluation process. Although some studies have attempted to establish detection methods for some elements in phosphogypsum, a systematic platform integrating multi-source information—such as total elemental content, form distribution, physicochemical properties, harmful substance residues, and crop requirements—is still lacking for comprehensive efficacy assessment. How to use intelligent methods to achieve rapid, accurate, and comprehensive evaluation of phosphogypsum soil conditioners, and provide scientific basis and optimization suggestions for their agricultural application, has become an important issue in promoting technological progress and industrial implementation in this field.

[0004] Therefore, developing an integrated analysis system that combines automated detection, morphological recognition, data modeling, and intelligent evaluation can not only fill the current technological gap in the quality evaluation and application guidance of phosphogypsum soil conditioners, but also provide reliable technical support for their resource utilization, promote the transformation of phosphate chemical by-products into high-value-added agricultural products, and achieve coordinated development of environmental protection and agricultural production. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent analysis, detection and comprehensive evaluation system for calcium, magnesium and sulfur elements in phosphogypsum soil conditioners. By integrating morphological analysis, intelligent modeling and risk assessment, the system can achieve a comprehensive evaluation of the agronomic efficacy and environmental safety of the conditioner and generate customized application plans.

[0006] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution: An intelligent analysis, detection, and comprehensive evaluation system for calcium, magnesium, and sulfur elements in phosphogypsum soil conditioners includes: The hardware analysis module, used to perform physicochemical analysis on phosphogypsum soil conditioner samples, includes at least: The total elemental analysis unit is configured to determine the total content of calcium, magnesium, and sulfur in a sample. The elemental speciation analysis unit is configured to separate and determine the content of different chemical speciations of calcium and magnesium in the sample. The physicochemical property analysis unit is configured to measure the pH value, conductivity, and particle size distribution of samples. The data acquisition and processing module is communicatively connected to the hardware analysis module and is used to receive the raw detection signal and convert it into a structured elemental analysis data packet. The data packet includes at least an element total vector, an element morphology distribution matrix, and a physicochemical property vector. The intelligent evaluation core module is communicatively connected to the data acquisition and processing module. It internally stores a pre-trained evaluation model, which is configured to receive the element analysis data packet and execute the following calculation process: S1: Based on the element morphology distribution matrix, calculate the bioavailability morphology coefficient of each element; S2: Based on the total element vector and physicochemical property vector, environmental and crop parameters are introduced to calculate the comprehensive agronomic effect index and environmental adaptability index; S3: Based on the calculation results of steps S1 and S2, output a comprehensive evaluation report that includes product performance classification, applicability diagnosis and optimization suggestions.

[0007] Furthermore, the elemental speciation analysis unit specifically includes a sequential chemical extraction device for separating the target elements (Ca, Mg) in the sample into water-soluble, exchangeable, organically bound, and residual states, and determining the concentration of each state using atomic absorption spectrometry or inductively coupled plasma atomic emission spectrometry. The hardware analysis module also includes a sulfur speciation analysis unit, which uses ion chromatography-mass spectrometry to distinguish and determine the different occurrence forms of sulfur in the sample, including the concentrations of sulfate sulfur, sulfide sulfur, and organic sulfur.

[0008] Furthermore, in the intelligent assessment core module, the assessment model calculates the bioavailability morphology coefficient (BAF). i The formula for () is as follows: For element i (i is Ca or Mg) in, This indicates the concentration of element i in the kth chemical form as measured by the elemental speciation analysis unit, where k = 1 to n, corresponding to water-soluble state, exchange state, organically bound state, etc. This indicates the total concentration of element i as measured by the total element analysis unit; This represents the relative bioavailability weighting factor for the k-th form of element i, and its value is obtained by fitting a large amount of plant absorption experimental data, satisfying the following conditions: And for the same element, .

[0009] Furthermore, in the intelligent evaluation core module, the formula for calculating the comprehensive agronomic effect index (AEI) by the evaluation model is as follows: in, , These are the bioavailability speciation coefficients for calcium and magnesium, respectively. , , These are the total calcium, total magnesium, and effective sulfur content measured by the sulfur speciation unit, respectively. , , These are the reference content thresholds for calcium, magnesium, and sulfur, set according to the fertilizer requirements of the target crop. , , These are the regulatory coefficients for the nutritional importance of calcium, magnesium, and sulfur elements to the target crop, respectively, and their values ​​are provided by the crop nutrition database. This is an acid-base effect function based on the sample's pH value, used to characterize the contribution of the sample's own pH to the soil's acid-base regulation potential, and is a nonlinear function of pH. The soil background adaptability correction factor is obtained by looking up tables from a preset rule base based on the input target soil type, such as acidic soil or saline-alkali soil, and its initial physicochemical properties, such as initial pH and CEC.

[0010] Furthermore, in the intelligent evaluation core module, the formula for calculating the Environment Adaptability Index (EAI) by the evaluation model is as follows: in, This indicates the concentration of the j-th harmful substance in the sample detected by the hardware analysis module, where j = 1 to m, and may include soluble fluorine, heavy metal cadmium, lead, etc. This represents the safety limit standard value for the j-th hazardous substance in the soil conditioner; , where is the environmental risk weight coefficient for the j-th hazardous substance, determined based on its environmental toxicity and mobility; This indicates the organic matter content in a sample as measured by standard methods, or the estimated organic matter content predicted by a near-infrared spectroscopy model. This is a carbon fixation and emission reduction benefit adjustment coefficient, used to quantify the potential positive carbon fixation effect brought about by calcium / magnesium organic complexation.

[0011] Furthermore, the intelligent evaluation core module also includes: The data verification unit is used to verify the credibility of the detection results of the hardware analysis module. The verification methods include: checking the consistency of the measurement results of different methods, calculating the element recovery rate based on the material balance principle, and performing outlier analysis on the detection data with the historical database of similar products. An adaptive optimization unit, communicatively connected to the data verification unit and an external field effect feedback database, is used to dynamically fine-tune the parameters in the evaluation model based on the correlation analysis between actual field application effect data, such as crop yield increase rate and soil pH improvement, and the system's predicted AEI and EAI indices. This enables the model to continuously self-optimize.

[0012] Furthermore, the system also includes a decision visualization and report generation module, which is communicatively connected to the intelligent evaluation core module. The report generation module is configured as follows: Based on the calculated AEI and EAI values, the products are divided into four levels: high efficiency and low risk, high efficiency and risk control, low efficiency and safety, and low efficiency and high risk. For different levels, structured agronomic advice is automatically matched and output, including but not limited to: recommended application rate range, optimal application time, recommended fertilizer types, and precautions for application. The analysis data, index calculation results, product grades, and agronomic recommendations are integrated to generate a comprehensive evaluation report with illustrations.

[0013] On the other hand, the present invention proposes a method for quality assessment and application guidance of phosphogypsum soil conditioner using the above-mentioned system, comprising the following steps: S1: Prepare representative phosphogypsum soil conditioner samples and obtain their complete elemental analysis data packages using the hardware analysis module; S2: Input the data packet obtained in S1 into the intelligent evaluation core module, run the evaluation model, and calculate the bioavailability morphology coefficient (BAF). i The comprehensive agronomic effect index (AEI) and the environmental adaptability index (EAI); S3: Based on the AEI and EAI values ​​calculated in S2, and combined with the input target soil and crop information, the product performance level is determined and a customized agronomic application plan is generated through the decision visualization and report generation module. S4: Use the evaluation report and plan generated in S3 as the quality archive and field operation guide for product batches.

[0014] The beneficial effects of this invention are: This invention fundamentally overcomes the technical deficiency of traditional total quantity analysis indicators being disconnected from actual agronomic effects by establishing a mapping model from elemental chemical speciation to quantitative calculation of bioavailability. It stems from the physical separation and precise quantification of different binding forms of elements such as calcium and magnesium, including water-soluble, exchangeable, organically bound, and residual states, achieved by a sequential chemical extraction device and an online detection unit. Furthermore, the evaluation model assigns bioavailability weighting factors based on plant absorption test data to each form. The system outputs the bioavailability speciation coefficient (BAF). i This enables the essential quantification of the actual available potential of nutrients in products within the soil-plant system, allowing for the scientific differentiation of products with the same total amount of elements but different forms in terms of agronomic value, and providing a key basis for product quality control and grading.

[0015] This invention constructs a multi-factor coupled calculation model of the Comprehensive Agronomic Effect Index (AEI) and the Environmental Adaptability Index (EAI) to achieve collaborative simulation and forward-looking assessment of the application effects and environmental risks of products in complex field environments. It relies on the integrated calculation of element availability, sample pH properties, target soil background conditions, and crop requirement parameters in the assessment model, particularly by introducing a soil background adaptability correction factor (CFsoil) to dynamically correct environmental conditions. The environmental risk assessment section uses an exponential decay function based on safety limits to perform weighted risk superposition calculations for multiple hazardous substances and innovatively incorporates potential carbon fixation ecological benefit factors contributed by available calcium and magnesium. This enables the system to produce a balanced assessment that considers both agronomic efficacy and environmental safety, providing precise data-driven decision support for the targeted application and risk management of products.

[0016] This invention, by incorporating an adaptive optimization unit based on a machine learning framework and coupling it with a standardized report generation module, forms a continuous improvement closed loop of detection-evaluation-feedback-optimization, transforming from static experience-based judgment to dynamic intelligent decision-making. By correlating historical evaluation data with actual field feedback data, the system drives the self-iterative fine-tuning of key parameters of the evaluation model. Simultaneously, the system automatically transforms all analytical data and model calculation results into a structured report containing product efficacy levels and customized agronomic solutions, thus converting complex professional analysis into operational instructions that can directly guide production and application. This not only significantly enhances the practical value of a single detection and evaluation but also enables the entire system to continuously evolve its judgment accuracy and applicability with practical experience, providing a core tool for establishing and consolidating industry technical standards.

[0017] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1

[0021] This embodiment describes an intelligent analysis, detection, and comprehensive evaluation system for calcium, magnesium, and sulfur elements in phosphogypsum soil conditioners, comprising: The elemental speciation analysis unit within the hardware analysis module couples sample pretreatment speciation separation with high-sensitivity detection online. First, ultrapure water is used to extract phosphogypsum soil conditioner samples under specific solid-liquid ratios and oscillation intensities to non-destructively obtain water-soluble calcium and magnesium in their free ionic form, which is considered to have the highest immediate biological availability. After microporous membrane filtration, a portion of the extract is directly introduced into the graphite furnace or flame atomizer of an atomic absorption spectrometer for analysis, while the other portion enters an ion chromatography system for analysis of anions such as sulfate. The solid residue after water extraction is automatically transferred to the next extraction container, where a 1 mol / L ammonium acetate solution is added to exchange calcium and magnesium ions adsorbed on the surface of the sample colloidal particles, i.e., the exchanged form. This form is sensitive to the ion balance of the soil solution and is the main source of absorption by plant roots through ion exchange. Subsequent extractions involved sequentially using 0.1 mol / L sodium pyrophosphate solution to extract calcium and magnesium (organically bound) that were coordinated with organic matter. Finally, a strong acid was used to completely digest the final residue to obtain the residue present in the primary or secondary mineral lattice. Parameters for each extraction and detection step, such as extraction time, temperature, centrifugation speed, and instrument detection conditions, were precisely controlled by a central controller according to a preset standardized protocol to maximize the reproducibility and comparability of analytical results across different batches and laboratories.

[0022] It should be understood that the relative bioavailability weighting factor on which the bioavailability morphology coefficient is calculated in the core module of the intelligent assessment is... The determination of this value is not a fixed empirical constant, but rather stems from an effect-form relationship model developed through continuous learning and optimization. It is based on extensive data from controlled environmental cultivation experiments, such as hydroponics, sand culture, and field micro-plot experiments. For calcium, multiple rounds of hydroponic crop experiments were conducted using nutrient solutions with different calcium form ratios to accurately measure the calcium absorption of various crop parts, such as roots and leaves. Multivariate statistical analysis was used to fit dose-response curves between crop calcium absorption and the supply of calcium in different forms. By normalizing the slopes of these curves and considering the response differences among different crop species, such as tobacco, corn, and vegetables, water-soluble calcium was ultimately assigned the highest weight. Because it can be directly intercepted by the root system; exchangeable calcium, due to the need to undergo a desorption process, has the next highest weight. Organically bound calcium requires mineralization or ligand exchange for release, and has a lower weighting. Residual calcium is almost unusable during the crop growth cycle, and its weight is close to zero. The model database periodically updates the value range of these weighting factors based on newly entered experimental data, ensuring... The calculations reflect the latest understanding of agronomic research. Therefore, the formula is essentially an algorithm that translates chemical form concentration information into agronomic effectiveness potential: In some embodiments, the calculation model of the Integrated Agronomic Effect Index (AEI) reflects the complexity of multi-factor coupled decision-making. It is a composite function in product form: The first multiplier focuses on calcium and uses a logarithmic function. The logarithmic form is used to describe its effect because the promoting effect of calcium on plants exhibits diminishing returns after reaching a suitable range, and the logarithmic form can better fit this biological law. This threshold is set based on the target crop's calcium requirement at its potential optimal growth, combined with the soil's baseline calcium supply capacity. The second multiplier is for magnesium, using a linear proportional relationship. This reflects that, under common farmland conditions, the relationship between magnesium supply and effect is closer to a linear response in the range from scarcity to adequacy. The third multiplier integrates the acid-base regulation potential of available sulfur and products. It is an empirical function constructed based on the soil chemical buffering theory, and its typical form is: in It is the most suitable soil for the target crop. Value, parameter The width of the control function, which is applied to the sample. Close to the crop's optimal The maximum value was obtained at that time, quantifying the sample's direct potential as an acid-base regulator. Nutritional importance regulation coefficient. The assigned value is directly linked to the crop nutrition database. For tobacco, which has a high calcium requirement, Set to 1.2; for fruit trees that require more magnesium, Upward adjustment; while cruciferous crops grown in sulfur-deficient areas, This will result in a higher value. Soil background adaptability correction factor This is a lookup table function, whose rule base is built upon soil science principles and extensive field validation data. When the target soil is strongly acidic (… When the base saturation is low, apply Higher levels of conditioning agents are expected to significantly improve the condition. Greater than 1, such as 1.1-1.3; if the target soil is close to neutral, then Approximately equal to 1; if the target soil is alkaline and the sample is also alkaline, then A value less than 1 indicates poor compatibility. The entire AEI calculation process is a complex simulation that involves real-time, quantitative matching of the product's chemical properties, morphological effectiveness, and dynamically changing crop requirements and soil environment.

[0023] It should be understood that the calculation model of the Environmental Adaptability Index (EAI) reflects the technological innovation of this invention in environmental risk early warning and ecological benefit assessment. Its calculation formula is as follows: Formula Part 1 It is an exponential decay function based on the risk superposition model. By directly adopting the limits set in mandatory national or industry standards, the legal basis for the assessment is ensured. Environmental risk weighting coefficient. The introduction of this concept differentiates the potential environmental hazards arising from exceeding the limits for different hazardous substances. Soluble fluorides ( Due to its high mobility and biotoxicity, the weight of a certain heavy metal is set at 1.5; while a certain heavy metal has weaker mobility in a specific form, and its weight is... The summation term is 1.0. In essence, it calculates a weighted exceedance index. If the concentration of any hazardous substance approaches or exceeds the limit, the index will increase significantly, causing the EAI value to decrease exponentially. This aligns with the objective law of the environmental risk short-spot effect. (Formula Part Two) This is used to assess the potential positive environmental effects of a product. (Numerator) This represents the average level of available calcium and magnesium. This reflects the relative abundance ratio of inorganic cations to organic matter. (Coefficient) Based on research findings on the formation mechanism of soil organic-inorganic complexes, this adjustment parameter quantifies the potential efficiency of effective calcium and magnesium in promoting soil organic matter stability, reducing mineralization loss, and thus fixing carbon. A higher value in this component indicates a more significant potential carbon increase and emission reduction benefit of the product. As part of the denominator, it adjusts the overall EAI value towards an increasing direction, thereby incorporating positive ecological value considerations into risk assessment. The model structure, which includes both penalty (risk) and reward (benefit) terms, allows EAN to make a more comprehensive and balanced evaluation of the product's environmental attributes.

[0024] In some embodiments, the implementation of the adaptive optimization unit relies on a closed loop between a built-in machine learning framework and external feedback data. A dynamically growing training dataset is constructed using all historically evaluated product sample data, including input detection data, all intermediate variables calculated by the model such as BAF, AEI, and EAI, and corresponding real-world performance data obtained from the field feedback database, such as relative yield increase, soil pH changes, and crop tissue calcium and magnesium content. The unit uses algorithms such as gradient boosting trees or neural networks, featuring various indices and parameters calculated by the previous model, and labeling them with actual field performance, to train an effect prediction and correction model. When this correction model detects that, under specific conditions, such as when the initial soil conductivity > 0.5 dS / m and the product pH < 6, the AEI predicted by the original model systematically deviates from the actual observed value, the adaptive optimization unit analyzes which input parameters or formula coefficients caused this deviation. After obtaining expert system confirmation or meeting a preset confidence threshold, the unit can fine-tune specific parameters in the original evaluation model. For the aforementioned high-salinity soil conditions, f(pH) is automatically lowered. sample Peak or adjustment CF in the function soil A specific assignment in the rule base enables the model to make more accurate predictions in new, similar scenarios. This ensures that the system of this invention is not a static black box, but an intelligent agent that can continuously evolve and improve itself with the accumulation of application data, and its reliability and practicality increase over time.

[0025] Building upon the ion chromatography analysis of sulfate anions in samples, the system can be further expanded into a more comprehensive sulfur speciation analysis scheme. For example, by online coupling of ion chromatography and mass spectrometry (IC-MS) and integrating specific pretreatment steps, the system can distinguish and quantify different sulfur speciations in phosphogypsum. Specifically, in addition to directly measuring sulfate sulfur in the water-soluble extract, the solid residue or a separate sample can be acid-distilled to convert acid volatile sulfur compounds (AVS) into gaseous form, which is then captured by an absorption liquid and subsequently analyzed by IC-MS to quantify the content of sulfur in the sulfur-containing form. For organic sulfur bound to organic matter, it can be digested under controlled conditions using an oxidant (such as hydrogen peroxide) to convert it into sulfate before measurement. The estimated value of organic sulfur can be obtained through subtraction or direct quantification. This sulfur speciation analysis data, combined with the results of calcium and magnesium speciation analysis, constitutes a multi-dimensional information foundation for assessing the nutrient release kinetics, potential environmental transformation behavior, and overall agroecological effects of the product.

[0026] It should be understood that the technical solution of this invention ultimately transforms the complex algorithm and model calculation results into decision information with direct guiding significance for product development, quality control, and agronomic applications through a decision visualization and report generation module. The core of this module is a rule-based classification engine and a natural language generation template. The classification engine executes strict classification logic based on the values ​​of AEI and EAI, combined with preset thresholds. These thresholds are dynamically adjusted according to factors such as crop economic value and environmentally sensitive areas. If AEI ≥ 0.80 and EAI ≥ 0.90, it is classified as Category I: Priority Promotion Type; if AEI ≥ 0.80 but 0.70 ≤ EAI < 0.90, it is classified as Category II: High-Efficiency Monitoring Type, indicating good agronomic effects but requiring attention to environmental parameters. For each category, the natural language generation template embeds specific analytical data, index interpretations, and agronomic suggestion variables. For Category II: High-Efficiency Monitoring Type products, the generated report specifically states: The AEI value of this batch of products is 0.85, indicating significant potential for increasing tobacco yield on the target soil. Its EAI value is 0.75, and the main limiting factor is that the soluble fluoride content is close to 85% of the standard limit. Example 2

[0027] To objectively verify the effectiveness of the system described in this invention, this control experiment was designed. Four batches of typical phosphogypsum soil conditioner industrial products, designated AD, were selected and applied in a uniform application scenario for tobacco cultivation in acidic red soil in southern China. The results of the three evaluation methods were compared with the actual field effects.

[0028] Table 1 Comparison of results from different evaluation methods with actual field performance Experimental results show that traditional empirical evaluation methods based on total calcium and pH have significant misjudgments. Sample A was misjudged as high-quality due to its high total calcium and alkalinity, but the system calculated its calcium bioavailability coefficient through speciation analysis, indicating poor actual bioavailability. The optimized model further incorporated acidic soil characteristics and lowered the evaluation, which is consistent with the slight yield increase observed in the field. Sample B, on the other hand, was underestimated by traditional methods due to its unremarkable total calcium content. The system gave it a high-efficiency evaluation based on its excellent speciation distribution, which was ultimately verified by significant yield increases in the field.

[0029] The intelligent evaluation model of this invention demonstrates its ability to handle complex factors. The comprehensive agronomic effect index achieves multi-dimensional dynamic simulation by coupling bioavailability form coefficients, crop demand regulation coefficients, nonlinear acid-base effect functions, and soil adaptability factors. The evaluation optimization process of sample B is the precise adjustment made by the model to a specific scenario based on feedback data. The design of the environmental adaptability index is particularly crucial, as it quantifies the environmental bottleneck effect through a weighted risk index function. In the case of sample C, traditional methods could only provide a vague warning, while the EAI index of this system dropped sharply due to cadmium near the standard, issuing a clear high-risk warning, which was consistent with the subsequent detection of cadmium exceeding the standard in crops. The evaluation change of sample D is even more significant: traditional methods ignored its high organic matter characteristics, and the initial model did not pay attention to it either; while the optimized model, through learning from historical data, increased the weight of carbon sequestration and emission reduction benefits, thereby identifying the soil carbon fixation potential of this product in addition to providing nutrients, and upgrading its level from low-efficiency and safe to high-efficiency and low-risk. This judgment is consistent with the increase in soil organic matter observed in the field, reflecting the system's ability to balance agronomic utility and ecological benefits in its evaluation.

[0030] The adaptive optimization mechanism is the key difference between this system and the static model. A comparison of the outputs of the initial and optimized models shows that the system can continuously evolve based on field feedback. For acidic soils in southern China, the optimized model reduces the compatibility evaluation of strongly alkaline conditioners and shows greater acceptance of milder amendments; in particular, it discovers the synergistic ecological value of high organic matter and effective calcium and magnesium through data learning, and adjusts the evaluation weights accordingly. This demonstrates that the system of this invention can continuously improve itself with practical experience, and the dynamic optimization characteristics of the evaluation conclusions play an important supporting role in establishing up-to-date industry standards.

[0031] In summary, this invention proposes an intelligent analysis, detection, and comprehensive evaluation system for calcium, magnesium, and sulfur elements in phosphogypsum soil conditioners. The system includes a hardware analysis module, a data acquisition and processing module, and an intelligent evaluation core module. The hardware analysis module determines the total element content, chemical speciation, and physicochemical properties of the sample. The intelligent evaluation core module, based on a pre-trained model, calculates the bioavailability coefficient of the elements based on speciation analysis results and combines environmental and crop parameters to calculate the comprehensive agronomic effect index and environmental suitability index. Finally, the system outputs a comprehensive evaluation report covering product efficacy grading, applicability diagnosis, and optimization suggestions. This invention solves the problem that traditional detection methods cannot correlate element speciation with agronomic effectiveness, achieving a scientific, accurate, and comprehensive intelligent evaluation and application guidance for phosphogypsum soil conditioners.

[0032] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent analysis, detection, and comprehensive evaluation system for calcium, magnesium, and sulfur elements in phosphogypsum soil conditioner, characterized in that, include: The hardware analysis module, used to perform physicochemical analysis on phosphogypsum soil conditioner samples, includes at least: The total elemental analysis unit is configured to determine the total content of calcium, magnesium, and sulfur in a sample. The elemental speciation analysis unit is configured to separate and determine the content of different chemical speciations of calcium and magnesium in the sample. The physicochemical property analysis unit is configured to measure the pH value, conductivity, and particle size distribution of samples. The data acquisition and processing module is communicatively connected to the hardware analysis module and is used to receive the raw detection signal and convert it into a structured elemental analysis data packet. The data packet includes at least an element total vector, an element morphology distribution matrix, and a physicochemical property vector. The intelligent evaluation core module is communicatively connected to the data acquisition and processing module. It internally stores a pre-trained evaluation model, which is configured to receive the element analysis data packet and execute the following calculation process: S1: Based on the element morphology distribution matrix, calculate the bioavailability morphology coefficient of each element; S2: Based on the total element vector and physicochemical property vector, environmental and crop parameters are introduced to calculate the comprehensive agronomic effect index and environmental adaptability index; S3: Based on the calculation results of steps S1 and S2, output a comprehensive evaluation report that includes product performance classification, applicability diagnosis and optimization suggestions.

2. The system as described in claim 1, characterized in that, The element speciation analysis unit specifically includes a sequential chemical extraction device, which is used to separate the target element in the sample into water-soluble, exchangeable, organically bound, and residual states, and to determine the concentration of each speciation using atomic absorption spectrometry or inductively coupled plasma atomic emission spectrometry. The hardware analysis module also includes a sulfur speciation analysis unit, which uses ion chromatography-mass spectrometry to distinguish and determine the different forms of sulfur in the sample, including the concentrations of sulfate sulfur, sulfide sulfur, and organic sulfur.

3. The system as described in claim 1, characterized in that, In the intelligent assessment core module, the assessment model calculates the bioavailability morphology coefficient (BAF). i The formula is as follows: For element i (i is Ca or Mg) in, This indicates the concentration of element i in the kth chemical form as measured by the elemental speciation analysis unit. This indicates the total concentration of element i as measured by the total element analysis unit; This represents the relative bioavailability weighting factor for the k-th form of element i, and its value is obtained by fitting a large amount of plant absorption experimental data, satisfying the following conditions: And for the same element, .

4. The system as described in claim 1 or 3, characterized in that, In the core module of the intelligent assessment, the formula for calculating the comprehensive agronomic effect index (AEI) by the assessment model is as follows: in, , These are the bioavailability speciation coefficients for calcium and magnesium, respectively. , , These are the total calcium, total magnesium, and effective sulfur content measured by the sulfur speciation unit, respectively. , , These are the reference content thresholds for calcium, magnesium, and sulfur, set according to the fertilizer requirements of the target crop. , , These are the regulatory coefficients for the nutritional importance of calcium, magnesium, and sulfur elements to the target crop, respectively, and their values ​​are provided by the crop nutrition database. This is an acid-base effect function based on the sample's pH value, used to characterize the contribution of the sample's own pH to the soil's acid-base regulation potential, and is a nonlinear function of pH. The soil background adaptability correction factor is obtained by looking up a table from a preset rule base based on the input target soil type and its initial physicochemical properties.

5. The system as described in claim 4, characterized in that, In the core module of the intelligent assessment, the formula for calculating the Environment Adaptability Index (EAI) by the assessment model is as follows: in, This indicates the concentration of the j-th harmful substance in the sample as detected by the hardware analysis module. This represents the safety limit standard value for the j-th hazardous substance in the soil conditioner; , where is the environmental risk weight coefficient for the j-th hazardous substance, determined based on its environmental toxicity and mobility; This indicates the organic matter content in a sample as measured by standard methods, or the estimated organic matter content predicted by a near-infrared spectroscopy model. This is a carbon fixation and emission reduction benefit adjustment coefficient, used to quantify the potential positive carbon fixation effect brought about by calcium / magnesium organic complexation.

6. The system as described in claim 1, characterized in that, The intelligent assessment core module also includes: The data verification unit is used to verify the credibility of the detection results of the hardware analysis module. The verification methods include: checking the consistency of the measurement results of different methods, calculating the element recovery rate based on the material balance principle, and performing outlier analysis on the detection data with the historical database of similar products. An adaptive optimization unit is communicatively connected to the data verification unit and an external field effect feedback database. It is used to dynamically fine-tune the parameters in the evaluation model based on the correlation analysis between the actual field application effect data and the AEI and EAI indices predicted by the system, so as to achieve continuous self-optimization of the model.

7. The system as described in claim 1, characterized in that, The system also includes a decision visualization and report generation module, which is communicatively connected to the intelligent evaluation core module. The report generation module is configured as follows: Based on the calculated AEI and EAI values, the products are divided into four levels: high efficiency and low risk, high efficiency and risk control, low efficiency and safety, and low efficiency and high risk. For different levels, structured agronomic advice is automatically matched and output, including but not limited to: recommended application rate range, optimal application time, recommended fertilizer types, and precautions for application. The analysis data, index calculation results, product grades, and agronomic recommendations are integrated to generate a comprehensive evaluation report.

8. A method for quality assessment and application guidance of phosphogypsum soil conditioner using the system described in any one of claims 1-7, characterized in that, Includes the following steps: S1: Prepare representative phosphogypsum soil conditioner samples and obtain their complete elemental analysis data packages using the hardware analysis module; S2: Input the data packet obtained in S1 into the intelligent evaluation core module, run the evaluation model, and calculate the bioavailability morphology coefficient (BAF) of the sample. i The comprehensive agronomic effect index (AEI) and the environmental adaptability index (EAI); S3: Based on the AEI and EAI values ​​calculated in S2, and combined with the input target soil and crop information, the product performance level is determined and a customized agronomic application plan is generated through the decision visualization and report generation module. S4: Use the evaluation report and plan generated in S3 as the quality file and field operation guide for this product batch.

Images